Full text of "Proceedings of the Royal Irish Academy"

See other formats

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project
to make the world's books discoverable online.

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover.

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the
publisher to a library and finally to you.

Usage guidelines

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to
prevent abuse by commercial parties, including placing technical restrictions on automated querying.

We also ask that you:

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for
personal, non-commercial purposes.

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the
use of public domain materials for these purposes and may be able to help.

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find
additional materials through Google Book Search. Please do not remove it.

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner
anywhere in the world. Copyright infringement liability can be quite severe.

About Google Book Search

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web

at|http : //books . google . com/

■ ,.\ 'I

,\^

LiW^

**» ^

SEP 2 0 1904

l^arbarli College l^ibrar^

^"ikoj CLc^

CLouu^vur

PROCEEDINGS

OP THK

ROYAL IRISH ACADEMY

VOLUME XXIV

^^^S^

DUBLIN

PUBLISHED AT THE ACADEMY HOUSE, 19, DAWSON STREET

MOLD AL80^ BY

HODOES, FIGGIS, & CO.. Limitrd, 1C4, GRAFTON STREET
Aiip BT WILLIAMS k NORGATE, LONDON, EDINBURGH, AND OXFORD

1902-1904

The Aoaprmy denire. it to he understood that theij nro not
answerable for any opinion^ representation of facts^ or train of
reasoning that may appear in any of the following Papers, TJie
Authors of the several Essays are alone responsible for their
cmitents.

PROCEEDINGS

OF THK

ROYAL IRISH ACADEMY

VOLUME XXIV

SECTION A-MATHEMATICAL, ASTRONOMICAL^
AND PHYSICAL SCIENCE

DUBLIN

FOBLISHBD AT THS ACADEMY HOUSE, 19, DAWSOM 8TKEET

■OLD ALSO BT

HODOR8, FI00I8, tc CO., Lnuno, 104, OBAFTON STBEET
Am BT WILLIAMS * NOfiOATE, LONDON, EDINBUBOH, AND OXFOBD

1902-1904

The Academy desire it to be understood that they are not
answerable for any opinion^ representation of factSy or train of
reasoning that may appear in any of the foilotcing Papers. The
Authors of the several Essays are alone responsible /or their
contents.

CONTENTS

SECTION A -MATHEMATICAL, ASTKONOMICAL, AND
PHYSICAL SCIENCE

FiAsiE (John), M.A., F.T.C.D. :— page

A Method of Bedaotion of a Qnaxtio Sur&oe possessing
a Nodal Conic to a Canonical Form. With an
Appbcation of the same Method to the Reduction
of a Binodal Quartic Curve to a Canonical Form, 71

HiKTON (C. H.; :—

The Geometrical Meaning of Cayley's Formula of

Orthogonal Transformation, .... 69

J0H1I8TON (J. P.), M.A., D.Sc. : —

Method of obtaining the Cubic Curve having three

given Conies as Polar Conies, • ... 66

Jolt (Cbables Jasfeb), M.A., D.So., F.T.C.D. : —

Integrals depending on a Single Quaternion Variable, 6

* The Multi-linear Quaternion Function, ... 47

' Some New Relations in the Theory of Screws, • • 69

Jolt (John), D.Sc, F.B.S., F.G.S. :—

Some Experiments on Denudation by Solution in Fresh

and Salt Water, ...... 21

BoBSRTs (Rev. W. R. Westbopp), B.D., F.T.O.D. :—

Some Properties of a certain Quintic Curve, . - 84
On Bicursal Curves, ; . . • • . 58

Catitents

Tbouton (Pbrdkmok T.), D.So., P.R.S. :— pagr

On the Greeping of Liquids and on the Surfiace Tension

of Mixtures, 1

DATES OF PUBLICATION

Pabt 1. Pages 1 to 46. August, 1902.

„ 2. „ 47 „ 68. April, 1908.

„ 8. „ 59 ,, 68. September, 1908.

„ 4. „ 69 „ 84. January, 1904.

PROCEEDINGS

THE ROYAL IRISH ACADEMY.

PAPERS READ BEFORE THE ACADEMY.

OIT THE CBEEPING OF LIQUIDS AND ON THE 8XTEPACE
TENSION OF MIXTUKES.

By FREDEEICK T. TROTJTON, D.So., F.E.8.

[Read Juki 10, 1901.]

Cboaiv liquids, as is well known, when left in the open air, will
creep up over the sides of the containing vessel and escape. Ordi-
naiy commercial paraffin is a liquid which creeps in this waj to a
ronai^ble extent. The phenomenon is well known in connexion
with dom^tic lamps, often producing inconvenient results.

The effect is readily observed by standing a beaker full of ordi-
nary paraffin on paper in the open air, when in a few days a
GOTsiderable quantity of the liquid will be seen to have crept out on
to the paper.

Experiments were undertaken with the view of investigating
the conditions necessary for this creeping to occur, and the con-
clusions ultimately arrived at from these may be summed up in the
statement that in order that a liquid should creep it must he a mixture^
ond the iufface tension of this mixture must be less than that of its least
volatile constituent,

A simple form of experiment, to compare the creeping tendencies
of different liquids, may be arranged as follows. A long metal strip

ILXO. PBOC, VOL. XXIY., SEC. A.] A

Proceedings of the Royal Irish Academy.

Fio. 1.

is made to stand up in a beakex, bj suitably forming its end into
a base (fig. 1). The upper end of the strip is bent oyer and
touches the interior of another yessel.
The liquid is placed in the beaker and
creeps over, and is collected in the
second yessel. The phenomenon di-
yides itself naturally into two parts :
(1) The initial stage, while the liquid
is forming a layer oyer the surface of
the strips; and (2) when the con-
tinuous trauEfport of liquid across to
the second yessel takes place. Passing
by for the present the initial stage, it
will abundantly appear from what
follows that a mixture is necessary for
continuous creeping.

Though many attempts were made, no pure liquid could be found
which would creep. Various mixtures, howeyer, were found which
did 80 actiyely. For instance, a pure paraffin (or rather what was
sold as such) did not creep, but the addition of a small quantity of a
lighter paraffin enabled it to do so actively. Again, ordinary paraffin,
which is, as is well known, a mixture of a number of different mem-
bers of the paraffin series, when left in the open air, loses much of its
lighter constituents and at the same time it is found to lose its power
of creeping in a like proportion. The power of creeping may be
restored by the addition now of a small quantity of a lighter oil.
Thus a liquid which has once crept over will not creep again nearly
so actiyely, if at all. If the surface-tension of the portion of liquid
which has crept oyer and been collected in the expeiiment described
aboye be determined, it will be found to be always greater than that
of the original liquid, and it also naturally consists of the less
yolatile constituents of the mixture. This suggests that eyaporation,
in conjunction with change in surface tension, plays an important
role in the phenomenon of creeping. This is easily yerified by
coyering the whole arrangement with a bell-jar. Evaporation being
thus preyented, creeping ceases likewise.

We now can see where the energy comes from to enable liquid, as
shown in fig. 1, to be continuously elevated and carried into the upper
yessel. Eyaporation of the liquid lowers the temperature, and in
consequence energy can be obtained from the environments. The effect
is brought about, it must be observed, through the loss into the

Trouton — On the Creeping of Liquids^ etc, 3

smroimdiiig atmoephere of the more yolatile portions of the liquid,
and the process ia consequently an irreversible one. The condition
necessary to enable this influx of energy to produce a directed effect
of the kmd required is that the surface-tension of the liquid remaining
after the loss of the more volatile constituents shall be greater than
before. In this -way, as the liquid passes along the strip, its surface
tension increases, and more liquid is enabled to be drawn up and
ultimately to pass over.^

In accordance with this view of the phenomenon, the creeping
activity of paraffin should be increased by the addition of any liquid
which is more volatile and which has a lower surface tension. The
addition of benzoline, ether, and of a number of other liquids, was
found to increase the rate of creeping immensely.

On the other hand, the addition of a more volatile liquid, with a
higher surface-tension, which, on mixing with the paraffin, increases
its surface-tension, should tend to prevent creeping. Various liquids,
such as benzene and chloroform, which are more volatile, but have
a higher surface-tension, were tried with this view. These were
found, however, not to prevent, but to actually increase, the creeping
activity.

On account of these unexpected results, experiments were made
with the object of ascertaining the effect produced on the surface-
tension of paraffin by the addition of these liquids, and afterward
experiments were made with mixtures of various liquids to investi-
gate the law of the surface-tension of mixtures in general.'

The curve in fig. 2 exhibits the determinations made of the
surface-tension of mixtures of paraffin and chloroform; and on
examining the curve, the reason becomes obvious why the addition of
chloroform does not prevent but rather facilitates the creeping of
paraffin, for the value of the surface-tension is there seen to be
diminished by the addition of small quantities of chloroform. The
like was found to hold good for benzene.

Mixtures, then, of various liquids were examined, and it was
invariably found that the surface-tension of a given mixture was

^A angle liquid might conceiyably creep through a similar gradient in
•uxface-tension, hrought about by the cooling due to evaporation alone increasing
the soifaoe tension.

s Since then the author has had the opportunity, through the courtesy of
PraL W. Ramsay, of consulting a hitherto unpublished paper of his on the
furface-tension of mixtures of liquids.

Proceedings of the Royal Irish Academy,

less than would be calculated from the percentage present in the
mixture, on the supposition that the surface-tension was proportional
to the composition. The following liquids were used: — ^benzene,
chloroform, turpentine, paraffin, alcohol, henzolene, and ether. These
were examined, two and two, in nearly all those cases where mutual

Pan^iA*m%

Chlerofinn 0 %

4*1

Fio. 2.

00%

Fig.ra ^^

^
\

solution occurs. In addition, water and alcohol, water and glycerine,,
aniline and chloroform, olive oil and benzene, were examined, all
with like results.

The curve shown, as in fig. 3, typically exhibits the results
obtained in general for mixtures of liquids. The dotted line givea
the value the surface-tension would
have for all percentage-mixtures if the
surface-tension were proportional to
the composition. The ordinates to the
full line represent the observed values.

When the surface-tensions of the
pure liquids are the same or not very
different from each other, the surface-
tension of mixtures in all or in some
proportions may be less than either of
the given substances.

The depression of the surface-ten-
sion of mixtures of liquids below the calculated value seems to be
but a particular case of a general principle which underlies the
character of the effect produced on physical properties by admixture
or solution, and which ranges from thmgs so far asunder as melting
points and electrical conductivities.

IOO%A

0%B

50%A

50%B

Fig. 3.

0%A
iOO%B

Trouton — On the Creeping ofLiquids^ etc, 5

No simple relation, however, could be foand connecting the
depreflSLon with the properties of the pure eubstances. In some
few cases, it shonld be remarked, it was found that the efPect of
one substance on another was roughly in inverse proportion to their
molecular weights. Thus, the depression produced on the surface-
tension of chloroform below the calculated value by the addition of
onall quantities of alcohol is at the rate of about 1*7 for each
percentage added, while the corresponding depression produced by
chloroform on alcohol is only about *7. These are in the ratio of
about 2-4, which is also nearly the ratio of their molecular weights,
59*5/23 = 2-5. The effects produced are, in this case at all events,
amply proportional to the number of molecules added, and the
failure in general to find similar relationships holding with other
liquids may perhaps be due to the masking of the effect through
molecular association.

A number of experiments have been made to ascertain if there is
any specific effect produced by the kind of material over which the
creeping takes place : this both in the initial stages and for con-
tinuous creeping. Ko consistent quantitative results have been so
far reached. This is probably to be attributed to the difficulty of
presenting clean and unaltered surfaces for the liquids to creep over.
The experiments, however, undoubtedly point to a decided difference
between different metals, both in the rate the creeping goes on at
in the continuous stage, as well as in the initial stage.

That the latter should be the case is not surprising, but it is not
«asy to see how the kind of material the surface is made of can have
effect once the layer of liquid has become established, for its thickness
is found to be great compared to molecular distances.

It is possible, however, that the effect may be wholly due to
specific roughness or corrugosity incident or natural to surfaces pre-
pared from different materials, for the state of roughness of a given
metal is found to have a great influence on the effect.

Attraction between the solid and liquid is a necessary condition
for creeping to occur at all; thus, that the rate the liquid in the
first instance establishes the layer should be dependant upon the
material over which the creeping occurs, is not surprising.

[ 6 ]

raTEGRALS DEPENDING ON A SINGLE aUATERNION
VAKIAJBLE. Bt CHARLES JASPER JOLY, M.A., D. Sc.,
F.T.G.D., Royal Astronomer of Ireland, and Andrews' Professor
of Astronomy in the University of Dublin.

[Bead April 28, 1902.]

TABLE OF CONTENTS.

An, Page.

1. — ^EzplAsation of Hamilton'^
method for single Qnatenuon
Integrals. Mode of passage, 7

2. — Difference of integrals be-
tween fixed limits according
to different modes of pas-
sage expreBaed as a double
integral, .... 8

3. — Conditions for independence

of mode of passage, . 9

4. — Case in which the variable is

ayector. Stokes's Theorem, 10

6. — ^Double integrals with a single

quaternion Tariable, . .10

6. — ^Variation of double integral
corresponding to yariation
in mode of passage, . 11

7. — Difference of double integrals
for different modes expressed
as a triple integral. Con-
ditions for independence of
mode of passage, . 12

8. — Triple integrals in a single

quaternion Tariable, . .13

An. Pag*.

9. — ^Litegrals of higher order, 14

10. — Specification of the modes of
passage. Time and space
method, . .14

11. — Hydrodjnamical iUustration
for single quaternion in-
tegral, . .16

12. — ^Physical illustration of mean-
ing of quaternion double
integral, . . .16

13. — ^Electro-magnetic equations as
conditions of independence
of mode of passage for a
double integral, . .17

14. — Double integral expressed as
difference of two single in-
tegrals. Vector potential of
magnetic current. Quater-
nion potential, . .17

15.~Case in which the integral
depends on the mode of pas-
sage. Conducting dielectric, 18

16. — Physical illustration for triple

integral, . .19

iNTBODucnoir,

Lr the '' Lectures on QuatemionSy" Hamilton devotes a brief series
of Articles (625-630) to the investigation of quaternion integrals.
It does not seem to have been observed that his results lead directly to
the fundamental theorems of Oreen and Stokes and to the extensions

Jolt — Integrals depending on a Single Quaternion Variabh. 7

of tbeee theoTems dne to Tait and proved by him and other writers in
^iiioTiB ways. Indeed, Hamilton regards the subject as one of great
^f&colty and diflmisses it rather abruptly ; but his method is of wide
Boope and merits further developments,

1 propoae therefore to sketch some of the consequences of

Hmiiton'B method in relation to quaternion integrals depending on

a sn^ quaternion variable, and from certain general results I shall

deduce as particular cases the extensions of the theorems of Stokes

and Qreen. It is not my object to furnish short proofs of these

tiieoremB ; they can be readily supplied from the results of this Paper

bj substituting from the commencement vectors instead of quaternions

and by integrating round closed curves or over closed surfaces.

In the concluding articles it is shown that the quaternion integrals
are capable of physical applications, and the more concrete character
of these articles may assist in forming a clearer conception of the
nature of the general integrals considered in the earlier portion.

Ab the integrals discussed in this Paper depend essentially on the
combinatorial functions which I have called quaternion arrays (Trans.
R.I.A., vol. xxxii., p. 17), it maybe useful to recapitulate the formulas
which we shall require. (Compare ''Elements of Quaternions,"
Art 365 (6)). Ji a^h, o and d are any quaternions, the arrays are

{ah) » YJS« - YaSh ; [a5] - Y.YaYh ; (aJc) = Sa[*(^] ;

[a«c] = (abe) - [he] Sa - [^] 8* - [ah] Se

and

{abed) » Balhcd].

Transposition of contiguous symbols changes the sign of an array,
and an array vanishes if its constituents are linearly connected.
Also for any fifth quaternion e,

a{hede) + h{edea) + c{deah) + d{edbe) -f ${abod^ = 0
aad

${ahed) - [hed] Sae - [acd] She + [aM] See - [aJtf] 8*.

Art. 1. — If F{qy r) is any function of two quaternions, distributive
with respect to the second, so that

F{q,r^s)^F{q,r)-^F{q,8), (1)

8 Proceedings of the Royal Irish Academy.

fhe integral considered by Hamilton is

e = fV(^,d^), (2)

in which the Tariable qnatemion q changee by a determinate wnode
of passage from one fixed limit q^ to the other q^ . He BnpposeB the
mode of passage to nndergo a slight variation while the limits remain
fixed, and he denotes the corresponding yariation in the integral by

8« = 8 f ' F{q, d^) = r 8^(^, d^) . (3)

Now

hF{q, d^) = h,F{q, dj) + F{q, Sdq) (4)

in which 8, is a symbol of partial differentiation and relates to
q alone and not to 8^. Siimlarly

d F{q, Sq) - d,F{q, Sq) -f F{q, dSq) , (6)

and because the differentials d^ and 8^ are independent

Bdq = d85'. (6)

Therefore, subtracting (5) from (4) we find

BF{q, dq) - dF{q, Sq) = 8,^(^, d^) - d,F(q, Sq) , (7)

and when we integrate this between the fixed limits we obtain
Hamilton's result

SQ^[''{S,F(q,dq)-d,F{q,Sq)}, (8)

because 8^ vanishes at the limits.

Art. 2. — Hamilton contents himself with observing that the
elements of the integral (8) do not generally vanish, and therefore
the value of the integral (2) depends in general on the mode of
passage. We shall suppose that it is possible to pass by continuous
variation of the mode of passage from one given mode to another
without the introduction of infinite terms. In this case we shall

Jolt — Integrals (Upending an a Single Quaternion Variable. 9

lisre the integral Q, for the fleoond mode connected with the integral
Q for the first hy the relation

e, - e+jse = e+j|^8,/'(^,d^)-d,j»(^,8^)). (9)

The limits of the douhle integral are fixed and prescribed by the
modes of passage for the single integrals ; and if the single integrals
are single-yalned (their modes of passage being given) the value
of the donble integral is independent of the manner in which the
variation has been performed : in other words, the double integral is
independent of its mode of passage, provided always that no infinite
tenns arise. More generally even if the single integrals are multiple-
valned, the double integral is independent of its mode of passage
provided that mode is included in a determinate domain.

Art 3. — ^Introducing a quaternion operator D, analogous to V,
which operates on q alone, we may write symbolically

8, = S«^D, d, = Sd^D; (10)

and therefore we may may replace (9) by

e, - e + JJ-P(?,d^88^D-8^Sd^D), (11)

in which we repeat B operates on q alone. It will be noticed that

dqSiqJ) " dq&dqJ)

vanishes for iq s d^ and is consequently expressible in terms of
the anays^

(dqiq) ^ SqSdq-dqSiq and [dqSq"] ^ YYdqYiq . (12)
In fact
dj8«yD-«ySd^D - -(d^8jr)8D + 8VD(d^&?)-VYD[d5'8j'], (13)

and in order that the integral (2) should be independent of the mode
of passage we must have

-^(£,i8-o)BD + -P(^,SDGS-a))-^(^, WDVa/J) « 0 (14)

for all constant vectors a and )S as we see by replacing the vectors

^ Trans. B.I.A., vol. zzzii., p. 17.

10 Proceedings of the Royal Iriah Academy.

(dq^ Bq) and [dq^ 8^] by /J - a and Va/J respectively. (Compare
I^ans. RJ.A.^ zxxii., p. 5.) It is easy to see that the terms in
/} - a and Yaj3 in (14) must vanish separately, so we may replace this
condition by the pair

SDa^(?, y) - SyD. J'(^, 1) = 0, I (^, VyVD) - 0, (16)

in which y is any constant vector.

In particular if we write q ^ t ^r p so that D becomes

D = A_v (16)

the conditions for an exact differential are

-gj ^(?, y) + Sy V - F{q, 1) = 0 , F{q, Vy V) = 0 ; (17)

and for a scalar integral, or if F{q^ dq) » Sp dji^, the conditions
reduce at once to

^Vi? + VSp = 0, VvVi? = 0 (18)

Art. 4. — When the variable is a vector p, equations (2) and (11)
become

. G = f ' Z(p, dp) ; G, = e + llF{p, VVV dp ip) (19)

because on replacing ^ by p and D by - V we have

dfl'SS^D-Sj^Sdj^D = -dpSSpV + SpSdpV = VvVdp^.

The double integral is consequentiy taken over the surface generated
by the motion of the path of integration from the first to the second
mode of passage or path of integration.
For a closed circuit

Q = lF{p, dp) = - llF{p, V VVdpSp) ; (20)

and for a scalar integral we have Stokes's theorem

Q = JSo-dp = -/JSo-VWdpSp = J/SVV<rVdp8p. (21)

Art. 5. — The results of Art. 2 may be extended to a class of
integrals not considered by Hamilton,

(2 = J/J'(^d},d'j) (22)

in which the limits are fixed while F[j, dq, d'q) is distribatiTO with

Jolt — Integrals depending on a Single Quaternion Variable. 11

rwpeet to the two independent differentials d^ and d!q and satisfies
the general condition^

F(,q,r,i)^F{q,s,r)^Q. (23)

The limits being fixed, a yaiiation of the two-spread mode of passage
gives

8G-JJ8^(^,d^,d'^); (24)

ad writing as before (4)

lF{q, iq, d!q) - Z,F{q, Aq, d!q) + F{q, 8dy, d'j) + F{q. dy, Sd'y)
d/Xf, d'y, iq) = d,^(j, d!q, hq) + F{q, Mq, 8j) + F{q, d!q, diq) (25)
A'Fis, «,, Hq) = d!,F{q, hq, iq) + F{q, d^, df ) + /-(f, 8y, d'd,)

we find on addition by (6) and (23),

W(f . <Jf . dV) + ^^-(1 , d!q, iq) + d'-P(?, 8j, Aq) (26)

- 8^(y, d^, d'f ) + d^(?, d'y, 8f ) + d'^(j, 8?, dy).

The limits being fixed, integration gives in place of (24) the relation
»« - /J{S.^(f. dj, A'q) + d.J'(y, dV, 8y) + d'.^(y, 8?, d£)). (27)

Art. 6. — ^By Article 3, as a conseqnence of the relation (23), the
fanction jP(f , r, «) mnst involve r and t combinatorially, that is in
terms of the azrajs (r«) and [r»]. We may therefore write

^h r, 0 = F, {q, (r,)) + -P,(?, [r*]), (28)

the functions being distributive with respect to (r«) and [r«] re«
spectively. Or for the sake of brevity if we nse the notation

n<l.r,^)-F{q,{rB])

instead of the expanded relation (28), we may by (10) replace (27) by

«« » \\F{q, {d^d'^}88^D + {d'^8y}8d^D + {Sj^d^^jBd'^D). (29)

As in Article 3 the element under the signs of integration must be

^ It u apparently imponble to aaoign any meaning to an ezpreflsion of the
type of (22) in which thia condition Ib not satisfied.

12 Proceediiigg of the Royal Irish Academy,

a fimotion of the three-symbol arrays* [8<?, dj', 6!q] and (8j, dj', d!q).
In fact the relations

{he) Bap-^ {ea) ^hp^ {ah) ^cp ^^ [/;, [^ahe'\'\

[ J(j] S «i? + M S 5p + [aJ] 8 c/> = -(/>, [fl3(?])

may be proved without any difficulty, so that we have

«« = //^. {?. \P, [Sjdyd'j]]) - l]F, {q, (D, [8?d?dV])). (31)

Art. 7. — Thus, given any two modes of passage for the variable
of the double integral (22) between the fixed limits, if it is possible
to pass from the first to the second by continuous variation without
introducing infinite terms, the difference of the values of the double
integral is expressible as a triple integral whose limits are prescribed
by the two modes of passage, and, except in cases of multiple values
of the double integrals, the value of the triple integral is independent
of its three-spread mode of passage.

' If the double integral (22) is independent of the mode of passage,
the element of the integral (31) must vanish, or replacing [S^dg'd'fj
by an arbitrary quaternion a we must have

^1 k. [!>«]}- F, {q, (D a)) = 0 , (32)

or separately for the scalar and vector part of a,

Ft {q> VD) = 0 , Fi (q, WD a) - SD • i?i (i?, a) = 0 (33)

J)
a being an arbitrary vector. Or in terms of V and -^ by (16) this is

-Pi(?, V) = 0, F,(y, Wo) + ^F,{s, «) = 0, (34)

The general scalar doable integral is of the form

Sj 8 0-, (dy d'q) + /J 8<r. [dj d'g] , (36)

and for this the conditions reduce to

SVa, = 0, ^' = VD<ri. (86)

* These arrays are defined by the relations

{abe) ^QYaYbYe; lobe] = (abc) + [tf*] &a + [a?] Sb 4 [ba] Be,

in which a, b and 0 are any quaternions, and as {abe) e S [abc] any three-symbol
array can be expressed in terms of [abe].

Jolt — Integrals depending on a Singh Quatemian Variable. 1^

For a Tector Taiiable p, (22) reduces to

C = /J^a(p,Vdpd'p) (37)

and (31) to

«e = -JJJ^2(p,V)B8pdpd'p, (38)

80 that

e, = (2-JJJ-Pa(p, V)88pdpd'p; (89)

but a direct proof of the relation (38) by Hamilton's method is
probably quite as short for anyone not thoroughly familiar with the
notation of this paper as the process of deduction from the general
molt. This last result includes Green's theorem.

Alt. 8. — Finally, so far as quateinions are concerned, we have
tnple integrals of the type

« = /JJ^(^,[d^,d'^,d"^]) (40)

in which (compare Arts. 5 and 6) the three independent differentials
df , d'qy A"q enter combinatorially or in tenns of the three-symbol
array [Aqd'qd^q^* ^^ limits of this integral being fixed, exactly
as in Art. 5, we may reduce 8 Q to the form

»« = m{8.-f (?. [dy, i'i, d"9l) - i.I'iq, [8?, A'q, d"j])

+ d'.-P(f, liq, 8q, d"j]) - d",^(j, [dq, i'q, 8^])} ; (41)

and because for any quaternions ^ we have identically

p {abed) = Ihed'l Sap- [<w?J] Shp-^ [aW] Sep- [ahe'] 8 dp (42)

we obtain in tenns of D by relations such as (10) the simple
eqniTalent of (41),

Se - in^i2. J))^{Sqdqi'qi"q). (43)

The conclusions of Art. 2 and of the last Article apply to this
case, the difference of two triple integrals corresponding to two
different modes of passage between fixed limits being expressible as a
quadruple integral. The condition that the triple integral (40)

* Here (mM) s Sa \ied'] fto. is the sisgle four-syinbol array for four given
qoatemioiift.

14 Proceedings of the Royal Irish Academy,

should be independent of the mode of passage is simply (compare (15)
and (33))

F{q, D) » 0 (44)

or in terms of V , (compare (17) and (34))

^F{q,l)-F{i,V).Q. (45)

For the general scalar triple integral having for its element
Sp[d5'd'5'd"<?], the condition (45) is

As|,-SVVi? = 0. (46)

Art. 9. — At the commencement of the last Article we stated that
the triple integral completed the list for quaternions. A quadruple
quaternion integral has a four-spread mode of passage; in other
words the quaternion variable receives every possible value included
within the given limits, and the mode of passage is incapable of
variation. The methods of the present Paper apply however without
formal modification to integrals of a variable

^ = a? + a^ifi + «ai, + . . . + ar^i„ (47)

where the units ii , h • • • ^i obey the laws

♦."--1, •.•, + t|i. = 0 (48)

and where multiplication is associative.

Art. 10. — Analytically the conception of the modes of passage for
single, double and triple quaternion integrals presents no difficulty.
We have only to conceive the variable quaternion to be a function of
one, two or three variable parameters. The limits are defined when
a single restriction is imposed on the group of parameters for each
limit. Thus the limits for a double integral are defined by two
quaternions each of which is a function of a single parameter, and
for a triple integral^ the limits are two quaternions, functions of two
variable parameters.

It is worth while inquiring whether we cannot assign useful
interpretations for the modes of passage and for the limits when we
replace ^ by ^ + p and regard t as the time measured from a fixed

JohY—Integrab depending on a Single Qaatemion Variable. 15

epoch and p as the vector to a variable representative point at the
time t. For a single integral taken between fixed limits, the repre-
sentative point is obliged to leave a fixed position at a given time and
to reach another fixed position at another given time. The path it
describes and the rate at which it traverses that path are fuUy
specified by the mode of passage.

Art 11. — To give an illustration, take the case of the scalar
integral

Q = JS(J?4<r)(d^ + dp) = IJEdit + JSo-dp. (49)

We have seen (18) that the conditions that this integral should be
independent of the mode of passage are

^+V-E'=0, VVcr = 0. (50)

The form of these equations suggests an example from fluid motion,
60 we suppose o- to be the velocity of the fluid which the second
condition requires to be irrotational. To see what interpretation
▼e may assign to the scalar JE^ we write down the equation of
motion, the suffix denoting that ob is not operated on by V, (compare
the Appendix to voL n. of the ''Elements of Quaternions," p. 547),

^ = -5T- 8<roV-<r = ^vP^-Vp (51)

in which P is the potential of the impressed force, c the density
and p the pressure. But we have identically

V<r,VV<r = S<r,V*<r- VS<ro<r = SoroV»<r + V iTcr« = 0, (52)

ud therefore

^= .VP-ivi?-*VTcF*. (53)

ot e

Thus we may take

JS'-P+fidi? + *Ta« (54)

80 that J? is the energy of the fluid per unit mass.

In general in the case of fluid motion when ^ is the energy per
uxiit mass and o- the velocity of the fluid at the representative

16 JProeeedings of the Boyal Irish Academy.

Boi&t duiBg its motion from one limit to the other, the time integral
(rf the energy diminished by the time integral of the component of the
Telocity along the path of the repreBentatiye point into the element of
that path (- So-dp), depends on the manner in which the point mores
in the given interval of time from one limit to the other ; bnt when
the fluid motion is iirotational the. difference of these integrals (49)
is independent of the mode of passage of the representative point.

It should be observed that J So-dp is not now the fifw^ for the
velocities cr are taken snccessively in time and not at the same instant
along the path.

When we do not suppose the motion inotational, the difference of
the integrals (Art. 3.) for two different modes of passage is given by

G/= C+[[s(8/dp-d<8p/|^+VjE'V/JSVVcrVdpSp. (55)

In this case by (51) and (52)

^ + VJg'= VcToVVcr, (56)

and this relation may be employed to simplify (55). But the double
integral admits of further simpMcation, for if in the variation of the
mode of passage we suppose tiie curves 8p to be instantaneous or to
pass through the loci of representative points at every instant during
the passage, we shall have 8^ = 0, bo that

Qi = Q-/Jd<S8pVcroVV<r~/|SVV<rVdp8p. (57)

Art. 12. — To illustrate the meaning of the double integral we
take the simple case (35)* which becomes in terms of t and p

Q = //S<rx(d^d>-d'^dp)+/|S<r,Vdpd'p (58)

because

{dq d'q) = dt d'p - d'^ dp , [dq d'^] « Vdp d'p . (59)

The limits being fixed must consist of a closed curve composed of
pairs of corresponding points of departure and arrival, the times being
prescribed for every point. There is now a singly infinite system of
representative points, each of which leaves its point of departure at a
definite instant and reaches its point of arrival at another definite
instant, and when the mode of passage is given the path and the
rate of description of that path is given for every representative

Jolt — Integrab depending on a Single Quaternion Variable. 17

point. We may therefore conoeive a curve connecting the series
of repiesentatiye points to sweep across the closed curve in a manner
prescribed by the mode of passage. Let ns suppose the differentials
chosen so that dp is an element of this instantaneous curve while
d'p is an element of tiie path of a representative point. We shall
then have d^ = 0, and the integral becomes

Q = -Jd'//8iridp + JJS<r,dpd'p (60)

and -JScidp is now (compare Art. 11) the flow of the vectors o-i
along the instantaneous curve from one extremity to the other. In
like manner JSo-jdpd'p is the flux of the vectors cr, through the
elementazy strip between two consecutive instantaneous curves, the
integration being performed along an instantaneous curve ; but for
the reason stated in Art. 11, J/Scradpd'p is not the flux of the
vectors os through the surface generated by the instantaneous curve,
being rather the integral of the fluxes at successive intervals of
time through the strips determined by successive positions of the
instantaneous curve.

Art. 13. — ^We have seen (36) that the conditions that this integral
should be independent of the mode of passage are

8V«r.-0, ^»=VV«r.. («1)

l^ow these are precisely the equations which the electric displacement

[ — (Taj in a dielectric and the corresponding magnetic force (o-i)

satiflfy. It is therefore possible to give a physical illustration of the
integral (60). Any closed curve being taken in the dielectric, if a
variable carve is subject to the conditions that its extremities shall
move in a determinate manner along the fixed curve ; then the time
integral of the flows of the magnetic force from one extremity of the
variable curve to the other in every position of the curve added to
4t timet the integral of the displacement through the strips between
soooesaive positions of the variable curve, is independent of the nature
of the variable curve.

Art. 14. — ^When the double integral is independent of the mode of
passage, it may be expressed as the difference of two single integrals.

B.I.A. PBOC., VOL. ZZIV., BBC. A.] B

18 Proceedings of the Royal Irish Academy,

Beplacing Q by jP in (49) and (55) for the sake of greater clearness
and choosing the differentials so that 8^ is zero, we have

i>=|S(^+<r)(d^+dp); P,-i>--[[d<S8p^^ + V^)

-/JSVV<rVdp8p. (62)

Comparing the second of these with (60) and observing that 8 and d
correspond respectively to d and d', we may write

cri=^4V^; cr, = VV<r, (63)

and the conditions (61) are identically satisfied. Now - cr is the
vector potential of the magnetic current, and - ^ is the scalar
magnetic potential,^ so that if

i^ = -^-cr, (64)

we shall have the integral (60) equal to the difference of the two
values of the integral

P = -JSpdj' (65)

corresponding to the two modes of passage which together form the
limit for the integral (60). The quaternion p may be called the
quaternion magnetic potential; and the magnetic force cti and the
electric displacement -r— ir% are derived from p by the combinatorial
operations with 2),

(rx = -(2>,l'), <r, -[Z),i?]. (66)

Art. 15. — When the integral is not independent of the mode of
passage (31) gives

8e = ffs^^'- VV<riV8^Vdpd> + d^Vd'p8p + d7V8pdp)

-JJSV<r,S8pdpd>;
or supposing the differentials chosen so that dp and ^p are along

^ Oliver Heavside: Electrical Papers, vol. i., p. 467.

Jolt — Integrals depending an a Single Quaternion Variable. 19

instantaneouB curves (bo that d^ and 8^ are zero)^ wbile d!p is an
element of a path of a representatiye point, this reduces to

8C = Jd'^jsf^'-VVo-iWspdp-JJSVcr.SSpdpd'p. (67)

The difference between two integrals for different modes of passage is
therefore

e, = e-Jd^/JSo-sSpdp-JJJSVo-aSSpdpd'p (68)

^•_VV<r. .-a,. (69)

In the variation from one mode of passage to the other, the in-
stantaneous curve corresponding to a given value of t traces out a
surface — the instantaneous surface. The integral JJ Sa-^Spdp is
the flux through this surface, supposed momentarily fixed, and the
time integral of this is J d!t jj S 0-3 8p dp. In the electro-magnetic
illustration 73 is the conduction current when the medium is not a
perfect non-conductor. 8 V 0-9 is the electric volume-density. (Clerk
Maxwell, Electricity and Magnetism, Art. 619).

Art. 16. — Finally for the triple integral (40), we take as an
example

e=/JJ8i»[dyd'fd"y], (70)

or in terms of p and t since

Idqd'qdf'q} = 8 dp d'p d"p - d^ V d'p d"p - d'^ V d"p dp

-d'7Vdpd'p, (71)

and since we may choose the differentials so that d^ and d'^ are
zero, we have

Q - -Jd"<JJ8Vi?Vdpd'p+nj8p8dpd'pd"p. (72)

The limits now consist of a closed surface composed of pairs of
points of departure and arrival corresponding to prescribed times.
We may imagine a su^ace drawn through the representative points
to sweep through the closed surface. This variable instantaneous
surface must at every instant cut the limiting surface in a definite
curve corresponding to that instant, but the shape of the instantaneous
Borface is otherwise arbitrary until the mode of passage is prescribed.

20 Proceedings of the Royal Irish Academy.

If the integral is independent of the mode of passage, the con-
dition (46) is

|^8i^-SvVf? = 0, (73)

and the simplest physical illustration seems to be to take S^ » « to
be the density of a fluid and V^? = c<r to be the 'product of the
density and the velocity. The equation of continuity being

|^-8v(.a)-0, (74)

the condition is satisfied, and the integral

Q = -Jd'7JJtfS<rdpd'p + JJJ<?8dpd'pd"p (75)

is independent of the mode of passage.

The integral JJtfS(rdpd'p is the flux of the fluid through the
sui&ce with whidi the variable instantaneous surface momentarily
coincides, and the integral - Jd'7 JJcSo-dpd'p is the negative of
the time integral of this flux corresponding to the motion of the
instantaneous surface. The integral J JJ « S dp d'p d'^p is simply the
negative of the quantity of fluid which has passed through the
instantaneous surface in its motion.

[ 21 ]

III.

SOME EXPERIMENTS ON DENUDATION BY SOLUTION IN
FRESH AND SALT WATER. By J. JOLY, D.Sc, F.R.8.,
F.G.S., Professor of Geology and Mineralogy in the University
of Dublin.

[Read Fbbbvast 24, 1901.]

Thb following experiments are directed to throw light on the much-
neglected question of the solvent ejSects of sea- water on rocks and
rock-forming silicates. ^

Ifateriali dealt with. — ^Four substances are dealt with in these
preliminary experiments — ^basalt, hornblende, obsidian, and orthoclase.
The basalt is a typical specimen, black, fine-grained, compact, with
specks of olivine, from the Qiant's Causeway, Ireland. The hornblende
is the dark-green aluminous variety, well crystallized, cleavable, from
Friedrickshaabe. The obsidian is a typical rhyolite glass from Monte
Pelato, lipari. The orthoclase is highly cleavable, fresh, pale pink in
colour.

Mode of Experiment — ^The experiments are all comparative, equal
amounts being exposed to solution in distilled water and in sea-water
under Hke conditions. The sea-water used was taken from the rocky
coast of Killiney, County Dublin, a part of the coast sufficiently far
removed from any stream or river discharge.

The experiments are of two distinct types. In the one it was
sou^t to secure to the full the effects of aeration upon the rate of
solution.

To this end ten grammes of the mineral, finely powdered, are
placed along with 1000 c.cs. of the solvent in a Jena-glass flask of
the conical Erlenmeyer shape, the flask having a capacity of
1100 CCS. A continuous stream of air is directed by a Jena-glass
tube to the bottom of the flask, the air escaping in bubbles which rise
through the liquid, and with the help of occasional shaking preserve
the sediment in suspension. The entering air is filtered from dust by
passage through cotton wool, and damped by passage through towers

a.I.A. FHOC., VOL. Tin., 8BC. ▲.] C

22 Proceedings of the Royal Iruh Academy.

. of pebbleB, wetted with salt water in the case of the salt-water experi-
ments, and fresh water in the case of the fresh- water experiments.
Eight flasks were exposed in this manner, each to an equal stream of
air, four containing fresh water, and four sea- water. The duration
of the experiments was three months, during which time the current
of air continued, with only a very few days intermission, both during
night and day. On each occasion upon which the flasks were shaken
it was found that the salt-water solutions had almost cleared before
the next day, whereas the fresh-water solutions remained turbid, a
natural effect, which in nature is of much importance.

The second mode of experiment was applied to a specimen of the
basalt only. In this case tiie material in coarse grains and fragments,
to the weight of about 180 grammes, is placed in a U-tube, and by a
cup-of-Tantalus arrangement, which will be described in the appencQx,
the solvent (which in this case also was 1000 c.cs. in volume) was
compelled to travel in opposite directions through the tube, passing from
an Erlenmeyer Jena-glass flask placed beneath to one placed above,
and gravitating back again, continuously during the daytime. The
air from the room enters through damping tubes into the upper and
lower flasks alternately with the withdrawal of the solvent.

The action upon the material in the U-tube may be considered as
much like what goes on upon the sea beach or in the wash of a river,
for at the completion of each upward passage of the water through
the U-tube one limb of the tube is to a considerable extent drained
of water, air entering freely between the coarser particles. On Hie
oompletion of the downward movement the other limb of the tube is
drained out to a large extent. The particles are thus exposed to the
wash of the solvent in both directions, and to its periodic partial
withdrawal from around them. There is, however, no attrition.

The time occupied in the upward passage of the water is from
seven to eight minutes ; in the downward from about eight to nine
minutes. The flasks and U -tubes are in duplicate, the U -tubes being
attached side by side, the one holding basalt traversed by salt water,
and the other basalt traversed by fresh water, and the current is
maintained through each by the one hydraulic arrangement.

The duration of this experiment was four months. At night the
active motion of the water was stopped, but during this period the
particles remained submerged.

Surface Area exposed to Solution. — It is very certain that the rate
of motion of the solvent in such experiments has within limits only a
minor influence on the ultimate results. It is even doubtful if the

JoLY — Some Experiments on Denudation. 23

qnantitj of the Bolvent within wide limits, so long as the solid ia
maintained inunersedi seriouslj effects the results. The primary
factor appears to be the stability of the solid material, and hence
the extent of surface which this exposes to the solvent is the most
important quantitative measurement involved. It is too often the
practice to state in such experiments the amounts gone into solution
as a percentage of the mass of i;he entire solid. The latter quantity
is in itself of little importance.

In the case of the last described experiment the materials
introduced into the U -tubes were sifted through sieves of measured
meaL Thus in each U-tube the following quantities of basalt were
inserted: —

25 grammes, passed 0*55 m.m., stopped by 0*45 mm. mesh.
40 „ ., 0-45 „ „ „ 0-35 „ „

20 „ „ 0-35 „ „ „ 0-20 „ „

To each of these, 103 grammes of coarse fragments having a mean
diameter of about 5 mma. were added. A minimum value for the total
saiface area is arrived at by assuming the particles spherical in form,
and having diameters of the mean Wlues of the mesh which admits
and the mesh which stops. The assumption is also applied to the 103
gnunmes of larger fragments. Making the requisite calculations, we
anive at the result that the area exposed is not less than 0*509 square
metres. The actual area lies above this minimum value. The
particles are rarely rounded, more often rectangular or wedge-shaped
and rough. The assumption that the particles were cubical would
leave the area still below one square metre. We are probably not far
from the actual value in assuming, therefore, one square metre as
approximately the total surface area exposed within each tube.

In the case of the ten-gramme charges, on the conclusion of the
experiments, each was separated by suspension in water into five
degrees of coarseness. These parts were carefully weighed, and the
mean diameters estimated by micrometric measurements (from ten to
twenty measurements being applied to each assortment), and the total
areas calculated, on the assumption that the particles are cubic in
fonn.

The following are the results in square metres: — ^basalt, 1*209;
orthoclase, 1*799; obsidian, 1*163; hornblende, 0*791.

On the foregoing data the table which follows further on is
calcniated representing the amounts of material removed in each case

24 Proceedings of the Royal Irish Academy •

per annum from an area of one square metre, according to the experi-
ments.

Such calculations can of course give only approximate evaluation
of surface. We possess no definite knowledge as to the depths to
which solutions might in such cases penetrate beneath the surface of
the various minerals and exert a solvent action. If this depth is
considerable, which does not appear probable, the gain in surface
obtained by reducing the material to very fine particles is more
apparent than real. Doubtless, the rate of abstraction of dissolved
material is in any case much greater at the actual surface. If we
assume such abstraction of material to only go on for molecular
distances beneath the surface, or at least distances small compared
with the diameters of the particles, the calculations afford at least a
definite basis of comparison with purely solvent processes in Nature,
for here also a similar penetration of solvent influences occurs. In
each case, too, we may assume a somewhat similar protective effect due
to residual materials. The conclusion to be drawn is that under-
estimate of the surfaces exposed— even on the assumption of cubic
particles — is more probable than ovdr-estimate.

As regards chemical attack over the surface of glass exposed in
each vessel, it may be stated here that this can only cause small error.
The glass used throughout (save the U -tubes in the basalt experiments)
was Jena glass. This glass has been the subject of tests made at the
Technical Institute of Charlottenburgh, which the makers have
published, and which apply to the amount of NaaO liberated under
various conditions. These show that the area of 500 sq. cms. exposed
in the flask, even if the full rate of solution for water at 20° C.
continued for three months, would only liberate 0*00017 grammes of
Na^O. As it is not to be supposed that the primary rate of extraction
would continue, and as the mean temperatures were not higher than
12° C, in any case the error may be considered negligible. The fact
is the area of glass exposed .is small compared with the areas of the
mineral particles.

The Chemical Results. — The chemical analyses were carried out in
the Chemical Laboratory of the Eoyal Dublin Society, under the
supervision of, and in part by, Mr. E. J. Moss, ¥.C.S., Chemical
Analyst to the Society, to whom my best thanks are due. Mr. Stone,
the assistant, bestowed the utmost pains on the very difficult task of
evaluating the small quantities available for estimation.

Unfortunately the estimate of the alkalies in the case of the sea-
water solutions, owing to the indirect methods available and the

JoLY — Some ExperimenU on Denudation. 26

oTerwhelming amounts of j sodium and potassium already present,
could not be effected with sufficient accuracy. The lime was in those
cases estimated where the nature of the mineral rendered its solution
probable. A partial analysis of the sea-water used was also carried
out under like conditions to those obtaining in the case of the salt-
water analyses.

The procedure in analyses was the usual one. The lime was
weighed as oxide by ignition of the oxalate, no attempt being made to
Beparate further possible impurities, which may in the case of sea-
water, as pointed out by Dittmar, amount to as much as 9 per cent, of
MgO, NatO, &c. The presence of TiOa in the silica precipitate was
not sought for, the weighing after the usual precipitation with KCl
and ignition being entered as silica.

The colunuis headed i. refer to the basalt dealt with according to
the second method of experiment as described, the substance being
oomparatiYely coarse-grained, and subjected to an alternate flow of
water in opposite directions. In all cases the quantities in the columns
headed '' Salt " have already been reduced by the amounts of dissolved
silica, alumina, and lime detected in the unused sea-water, as given in
the last column.

The mean temperature prevailing during the progress of Experi-
ments ii., m., IV., and v., was 7° C. ; and during Experiment i., 12° C.
At the conclusion of the experiments the sea- water, in each case,
showed a distinctly increased alkaline reaction towards litmus; the
fresh water also showed a very faint alkaline reaction towards litmus.

dmsideraiian of Eentlta, — The principal issue which led to the
foregoing experiments is the broad and somewhat complex one, as to
whether the water of the sea is a more active solvent denuding agent
than fresh water. It seems to have been left an open question up to
tbe present. Daubr^*s well-known experiments^ with chloride of
sodium and water acting on orthodase exposed to violent attrition in
a rotating vessel admittedly applies to the activity of the one dissolved
substance only. Moreover, the negative result which Daubr6e appar-
ently obtained is not in agreement with the results obtained by Beyer,*
that the felspars decompose rapidly in water containing sodium
diloride. But Daubr6e does not appear to have gone beyond investi-
gating the final reaction, whether alkaline or not. Further direct

'"Oeologie Ezp^rimentale,'* vol. i., p. 237.

'Quoted by O. P. Merrill in " Bocks, Bock- Weathering, and SoOb," p. 178.

^ 1
1

•00100

•00130

aba.

•69701

. i

if 1 1 1 I I 1

Frcah.
•00406

trace.

aba.

trace.

aba.

•01692

trace.t

o
o

^ 1

Salt.
•00161

•00109

aba.

•01962

Fre«h.
•00366

aba.

trace.

•00169

aba.

li 1 1 1 I ' I

ll -i 1 ' ' if

o
o

" 1

Salt.
•00364

aba.

•22062

Freah.
•01181*

trace.

Basalt.

Is 1 •§ 1 1 ' 1

Fresh.
•01581

aba.

abs.

•03210

•00909

•01127

trace.

o
b

's < i s ^ ^ ^

I
I

I-

= -s

8 J

. « 1-1

•M C« CO

1 CO

1 1

OT9 9

1 a

1 1

)

' •<••

> s

ja«

11 '

1 1

1 ^

' 1

. » »>•

1 s

. »

(So S

' ?

1 1

^ I

11 ,

1 1

o»

. O fr«

Salt
•347

•001

1 1

III

HOC]

If '

1 1

d»

. ^^

•^

II 1

1 ^^

I *o

I 1

•

h4 i

s •

11 '

1 1

. «*• 00

1 00

H i

1 s

1 1

1? '

R g

<N CO

<? 9

2 »^

i "^

<s •<

& o

^ i^

28 Proceedings of the Royal Irish Academy.

evidence on the broad question at issue I have not found. Dana con-
sidered that basalt rocks were protected by sea- water, either where
quite covered or merely washed with spray, relatively to the same
rock exposed to the alternate wetting and drying of sub-aereal actions.
Merrill,^ commenting on this, remarks justly that erosive actions, in
such cases, preserve a deceptive appearance of freshness to the rock.
He, however, thinks that no exception can be taken to Dana's remarks
regarding rocks wholly immersed.

. Oustav Bischof in his well-known ** Chemical and Physical Geology"
has advanced reasons from the chemical point of view for believing
that the alkaline silicates of felspars, &c., will experience more active
dissolution in water containing dissolved salts of calcium and mag-
nesium. This view is based on the fact that alkaline silicates are
decomposed in presence of the sulphates and chlorides of calcium and
magnesium, the sparingly soluble earthy silicates being precipitated.*

If this applies to the naturally occurring crystallized silicates, in
which alumina forms part of the molecule, and which are, compara-
tively speaking, insoluble bodies, sea-water, containing MgSOi, GaSOi,
and MgCla, in abundance, should accelerate the decomposition of
felspars.

The results of the reaction with the alkaline silicates appear,
according to Bischof, to be the formation of the silicates of lime and
magnesium and the sulphates and chlorides of the alkalies. The
latter will, of course, be dissolved. The silicate of lime will again be
decomposed if carbonic acid is present, silica separating out and preci-
pitating, and carbonate of calcium being formed. The silicate of
magnesia will, however, not be decomposed by carbonic acid.

According to these reactions, wherever sea- water acts upon silicates
containing alkaline silicates, and is« as in the experiments, freely
exposed to the CO3 of the atmosphere, decomposition will be accele-
rated, silicate of magnesia being precipitated, bicarbonate of calcium
formed and retained in solution (or precipitated if the amount of GOi
is deficient), silica precipitated, and soluble chlorides and sulphates of
the alkalies formed. These reactions would alone not serve to explain
the presence of the comparatively large amount of silica in solution
revealed by the reaction with orthodase (Ex. m.), unless soluble
alkaline silicates remain in solution, or a hydrosol of silica is formed.

iXo0.0i<.,p. 258.

' '* Geologie Ezp^rimentale," vol. i., p. 12.

JoLY — Some Experiments on Denudation. 29

But they suggest forcibly that the final results in nature (or in the
experiments), as regards bringing the rock materials into solution,
represent but a part of the total reaction upon the rock. In other
w<«ds, the amount of decomposition actually effected is indicated only
by the liberation in solution of certain of the constituents. This fact —
which could be instanced by many well-known phenomena of rock-
weathering — involves a conservative effect of great importance in
nature, and which must also be borne in mind in considering any such
experiments as the present ones, effected on fresh material. The pzp-
0688 of leaching out soluble constituents, and leaving insoluble ones, or
those of secondary formation, behind, must lead to a rapid diminution
of the surface actirity of the solid.

With the conspicuous exception of the orthoclase the silica obtained
in the salt water solutions is either about equalled or actually largely
excelled in ao&ount in the case of the fresh- water solutions — as in the
experiments on basalt and obsidian.

The obsidian, it will be observed, proved to both solvents the most
resistant of the materials dealt with. Daubr6e records among his
resmlts that this same substance offered remarkable resistance to
attack.^ The final solutions in this case, he records, showed scarcely
any alkaline reaction.

A conspicuous feature of the results is the much greater quantities
of lime dissolved in the salt-water solutions. According to Bischof
this might be explained, as we have seen, as the result of the
seoondary reaction attending the liberation of alkaline silicates. The
result, which is more especially conspicuous in the case of the basalt
(both in fresh and salt water), is in keeping with the well-known
deposition of carbonate of calcium in some basic igneous rocks
undergoing decay. Alumina in solution was only detected in the
salt solutions, but in every case in minute quantity. The almost
complete absence of iron in the solutions is remarkable, the delicate
test by sulphocyanide of ammonium revealing no more in any instance
than a tnoe.

Looking at the figures at the foot of the columns of Tables i. or n.,
we observe that in every case the total amount removed in solution by
the salt water exceeds considerably what is removed by the fresh water.
If the alkalies (and magnesia in some cases) taken up by the sea water
were added, the preponderance would be still greater. As it stands the

> loc. eit,, p. 276.

30 Proceedings of the Royal Irish Academy,

sea-water disBolves from twice (obsidian) to fourteen times (orthodase)
the mass dissolved by fresh water.

The main question at issue is undoubtedly answered by these
esq^eiimentSy incomplete though they be. They show, indeed, that
under the conditions of experiment: — moderate temperature; fresh
material ; abundant aeration ; active circulation ; absence of attrition :
marine solvent denudation exceeds in activity fresh-water denudation
in the case of hasalt not less than three times ; in the case of hornblende
not less than eight times ; in the case of obsidian over four times ; and
in the case oiorthoolase not less than fourteen times. In short, taking
alkalies into account and some MgO (as we have seen there is some
reason to believe MgO will not enter largely into solution in the case
of sea water), the preponderance ranges from about four times (basalt)
to seventeen or eighteen times (orthoclase). With the lapse of time,
as the surface of the solids become exhausted of the 'more soluble
constituents, a convergence and approximation of the two rates will
probably occur.

It is interesting to place the figures applying to fresh- water solvent
denudation side by side with estimates whidi have been based on river-
water analyses.

Mr. T. Mellard Eeade has estimated that solvent denudation in
England and Wales is lowering the surface of the land at the rate of
one centimetre in 430 years. This represents the removal of about
60 grammes per square metre per annum. The Mississippi, drawing
its supplies from areas exposed to wide climatic extremes and from
every variety of rock and soil, is lowering its basin at the rate of one
centimetre in 833 years, which represents the removal of about
30 grammes per square metre per annum. Comparing these figures
with the experimental figures, we see that even a brisk continuous
washing of fresh rock-surface having the superficial area of the denuded
region would not be competent to supply more than a small percentage
of these amounts. The mean of the figures at foot of the columns apply-
ing to fresh-water denudation in Table ii. is just 0*08 grammes
removed per square metre per annum. This is 0*15 per cent, of the
amount estimated by Mr. Mellard Beade, and 0*3 per cent, of the
amount removed per square metre per annum by the Mississippi.
Herein we see the infiuence primarily of the great surface areas
exposed in the soils (as much as 500 square metres in the litre),
as well as the solvent influence of the acids originating in vege-
tation, the more rapid solution of the calcareous rocks, effects of
alternate wetting and drying, frost and sxmshine.

Jolt — Some jEktperimefits m Dentsdation. 31

A brief account of the apparatas used in the experiments on the
Bolyent denndation of basalt (coarse grain) in fresh and salt water may
be of Talue to anyone entering on such experiments. The arrange-
ment is snch as to utilize the motion of a continuous water supply
bcm any source to produce a reciprocating passage of a given quantity
of a liquid through a U-tube containing the substance being dealt
with.

In the diagram for deamess one U-tube only is shown, X, contain-
ing the basalt. In the actual experiments there were two U -tubes
attached side by side so as to be under like conditions of temperature,
both containing basalt of same assortment of grain ; but through the
one salt water, through the other fresh water circulated. It will thus
be understood that the flasks containing these solvents, Fi and F^, were
four in number, the diagram showing those required for the one solvent
only. Beyond them those for the other solvent may be supposed
eonoealed. Similarly, behind Xthe second U-tube is concealed. Th^
tube B bifurcates at d, one branch ascending to the top of Fi as
shown, the other ascending to the top of the flask concealed behind Fi.
At 8 IB A stop-cock controlling the city water supply. A stout
rubber connection, closed all but for a nearly capillary glass tube,
admits from this a eotUtnuaui small stream of water at high pressure
into the tube A. It is thus conveyed in a slow continuous stream
into the closed Wolfs bottle placed above the tap and above the flasks.
If we imagine the Wolfs bottle just full of air, and water flowing
into it from the tap, this air will escape by the second tube B into
Fi and at first pass through the mineral particles in Z, and escape
through the measured quantity of solvent in jP„ emerging by the
damping tubes T. When the Wolfs bottle gets quite full the siphon
C comes into operation and rapidly empties the bottle, like a cup-of-
TsntaluB, the siphon being in fact of sufficient bore to empty the
bottle in about eight minutes, although the stream is entering by tube
A aU the time. During this emptying process evidently the solvent
in Ft is sucked up and passes through X, rising flnaUy into ^i. When
aU is nearly drawn up the Wolfs bottle is quite empty. The siphon
breaks, and the current from A gradually refills the Wolfs bottle,
during which time the solvent flows back through X This takes
about nine minutes.

There is thus a tide upwards and downwards maintained through

32 Proceedings of the Royal Irish Academy.

the U-tube X; and it is easy to arrange that the sand in each limb
shall he alternately uncovered and exposed to the air when the supply
of solvent from the attached flask fails. It is necessary for safety to
take the tube B about 60 centimetres or thereabouts above the Wolfs
bottle.

Where the siphon opens within the bottle a peculiar arrangement
is adopted, absolutely essential to the success of the apparatus. This
consists in forming the lower part, t, of the siphon of soft-rubber
tubing, carrying at its lower end an open thistle funnel, D. The
object of this is to obviate a well-known difficulty in tiie cup-of-
Tantalus arrangements : viz. the failure of the siphon to " break "
at end of its discharge and the consequent formation of a chain of
gas-bubbles and water-bubbles, carrying off the water at the .same
rate as that at which it enters the bottle. Now the action of the
thistle funnel and rubber tube is as follows : — as the water sinks in
the bottle and at last begins to uncover the thistle funnel, the weight
of the water in this fimnel elongates the rubber tube a little, so that
finally, when the lower lip of the funnel uncovers and the water
spills out of it, the contraction of the rubber tube jerks the funnel
completely out of the water beneath and lets the whole siphon fill
with air.

This arrangement gave no trouble and worked with no more
attention than that required to re-moisten the gravel contained in the
damping tubes T. I may point out that ike adjustment of the
effective capacity of the Wolf's bottle is simply carried out by an
adjustment of the length of the arm of the siphon within the
bottle, that is, by adjusting the length of the rubber tube t.

[ 34 ]

IV.

SOME PROPERTIES OF A CERTAIN QUINTIC CURVE. Br
The Rev. W. R. VESTROPP ROBERTS, B.D., F.T.C.D.

[Read Janua&t 27, 1902.]

1. — The Curve, the properties of which I treat of in this Pat>er,
is a special case of the class of quintic carves having a triple point.
Such curves possess considerahle interest, as many of their properties
can he ascertained hy the known properties of Ahelian integrals
and functions, and they thus afford us geometrical interpretations
of complex mathematical formulsB. ;

The equation of a quintic curve haviug a triple point is readily
seen to he of the form \

-4«»-2^«+ C= 0; (1)

the triple point, which we shall denote hy 0, heing at the point of
intersection of the axes x and y, A^ heing a hinary cuhic in x and y,
By a quartic and C a quintic in the same variables, and s a line
which passes through the points in which the five lines through
0 whose equation is C7 = 0 meet the curve.

We shall now express the coordinates of a point on the curve
in terms of a parameter 6. In order to effect this we shall seek
expressions for the coordinates of the two points in which the line
X = Oy meets the curve.

Introducing this value of x into the equation of the curve, we
find, after dividing by y*,

Ji»-2^«y+ Cy = 0, (2)

where A Js what A becomes when we put x ^ 0 and y - 1 , and,
similarly, B and C are what B and C become for the same values
of X and y.

BoBBRTS— 0« wme Properties of a Certain Quintic Curve. 36

If we now solve the above quadratic for the ratio s : y , it is plain

we may write

px == Ae

py^ W

where jB = Jb^-AC.

It follows thus that the line x = Oy meets the curve in two
points P and P', which we call corresponding points^ and further, if
we denote their coordinates by x, y, %, and x', y', z', we shall have

px = A$ pa/ =A^$

py = A' Pt/ = T ... (4)

p« ^T^Jl pn! = 5-J^ ,

that is to say, B being given, to one point P there corresponds a
positive value of the radical ^Ry and to the other point P a
negative value. It is evident that if i2 = 0, the points P and P
coincide.

2. — ^The class of this curve is easily ascertained, since the triple
point O is to be counted as equivalent to three double points in
estimating the number of tangents which can be drawn from an
arbitrary point to the curve ; this number is then 5x4-3x2 = 14;
and if tiiia point be on the curve the number of tangents which can
be drawn from it will be 14 diminished by 2, or 12 ; but if the point
be Oy the triple point, the number of tangents which can be drawn
from it to the curve will be 14-8x2== 8: hence eight tangents can
be drawn from O to the curve. Now, a line drawn through 0 meets
the curve in corresponding points, and these points will coincide
when the line touches the curve, the eight points of contact of
tangents which can be drawn from 0 to the curve, are then points
which coincides with their corresponding points. Their equation is
consequently _

We shall call the roots of jS = 0, a,, 02, . . . oe, and we shall
sometimes write it in the form

2>i, i>}, 2)„ and L^ being four quadratic factors whose roots are

Proceedings of the Royal Irish Academy,

«i> <»i ; o«> <»4 ? «•> ««; ^> <»«•
as the R points.

We shall refer to these eight points

3. — ^Let s s= 2r + my he the equation of any line, and let us seek
the equation which determines the parameters of the five points of
section. Suhstituting for s, its value Ix + my in the equation of
the curve, we ohtain

-4(& + i»y)*-2J9(i;r + ff»y) + C = 0, .

and consequently the five 0s of the points of section of the curve
with the line, which we shall denote hy 0i, $2^ Ozy 04, 05, are
found from the equation

^(W + »»)2- 2^ (» + !«)+ C s <^(0) = 0.

(2)

If we now investigate the change in 0 due to changes dl and dm
in / and m, we ohtain, hy differentiating the ahove equation.

de^2{A{ie-\-m)-B\{fidl^-dm) = 0.

Now,

A{l6 + m)-B =^R,

and }'. = Mit>'{6), where if is a function of / and m; hence
id

dOi (di)dl + dm)

JSx

$1 heing one of the five roots of the equation

^ (tf) o 0 , and ^' {&,) m (tf , - tf,) [$^ - $,) (ft - ft) (ft - ft)

"We now write

<fft (e^dl + dm)

+ 2S „ =0,

(3)

ft'rfft

if^'(ft)

{ei*dl + $idm)

= 0,

(O^dl + e^dm)

where 2 denotes summation from 0i to 05.

(4)

BoBBRTS — On 9om$ Prcperties of a Certain Quintie Curve. 37
Bat, by the theory of partial fractionB; we have

1 ft A' ^1*

Consequently we obtain the following three relationa connecting the
fiye YBlnes of 0 and their differentials which correspond to the points
of section of a line with the curve,

S-^=0, S-^=0, S-^-0. . (6)

ji^ js^ js;

These differential equations we can integrate since they are true for
every line which can be drawn to meet the curve.

If we now write f 6^d0 ^

r being an integer, we obtain, by integration of the differential
system (5),

27.(6) = «»., («)

S/,(tf) = «b, ,

i denoting summation from ft to ^si and ^oi ^i and e^ being constants.

4. — ^We now proceed to determine the values of the constants
€t, Ci^ and et.

By reference to equations (4), Art, 1, we see that to a given value
of the parameter 6, corresponds but one point P, if we agree to affect
the radical JW with a positive sign, its corresponding point P' being
determined by giving the radical a negative sign. To the triple
point 0 will then correspond three values of the parameter 0, or, in
other words, there are three different values of 6 which will give us

4? = 0, y = 0,

these values of $ corresponding to the different branches of the curve
1-i.A. noa, VOL. vm., sbc. a.] D

Proceedings of the Royal Irish Academy.

which pass through 0. These values of 0 are plainly the roots of
the equation ^ = 0, which we shall denote hy ni, n%y »(.

The constants ^, ^i, ^ being the same for all lines, their yalue
wHl remain unaltered if we consider a line, x ^ Oy^ drawn through
O. Now, such a line meets the curve in 0 counted three times, and
whose parameters on each branch on which it lies are respectively
Ml, fti, and Miy and in two corresponding points for which 0 is the
same but the radical is equal in value and opposite in sign. The sum,
then, of the five integrals I^ reduce to ii(ni) + -^W + '^oW ■ -^ ^J*
We find then

€b » iVi , and by parity of reasoning '

(1)

Hence any line meets the curve in five points such that

27,(0) -iV,,

(2)

In precisely the same way we may prove that any conic through
the triple point 0 meets the curve in seven points, and so that

(3)

As these three equations will, in all cases we shall discuss, always
obtain together, we shall write them in the briefer form

it being always understood that this equation 2 !{&) = IT implies
tkrte equations, connecting S /,(*), 27,(0), and 2^(0), with
JV,, Ni, and N».

BoBB&TS — On 8om$ ProperHea of a Certain Quintte Curve. 39

5. — ^To any point lor which 0 is given, and also the sign of the
niicalJSf cozrespond the three integrals i^(0), li{0)t and lt{0\
which we shall refer to as the integrals of the point, to any three
poiats whose parameters are 0i, 0s, and B^, will correspond three
int^irsls io(0i), -^(^i)* and 1^(0^), three integrals of the class li{0)
and three of the class It{B). If we write

i;(«i)+/o(A)+^o(tf.)- ^0

we mighty not improperly, call 2^, Vu and C^, the fffgoments ot
the three points whose parameters are 0i, 6%, and 0^.

Tor sach qnintic curves we have a theoiy of residuation analogous
to Sylvester^s Theory of Eesiduatumfor the Cubic ; hut as I have already
disqutsed such a theory in a Paper puhlished in the ^* Proceedings ot
London Mathematical Society" some years ago, and as the treatment
for the qnintic is almost exactly similar to that I adopted in the case
of the uni-nodal quartic, I shall not do more than allude to it.

We oonmder the equation of a curve ot the mth degree of the
form

as - ft a Of

which we call an 0 curve, where a and h are hinary forms of the
n - 1th and mth degrees respectively. Such a curve has 0 for a
point ot the M - 1th order; and we find, hy reasoning of a precisely
similar nature to that we employed in discussing the relations ot the
parameters of the points of section of a line with the quinitic, tluA
the 2» + 3 values of d corresponding to the 2m -k- 3 points of section
are connected by three relations, viz..

If two systems of points on the quintic a and fi together make up
the eomplete intersection of an 0 curve and the quintic, these
syttems are said to be co-residual.

6. — We now tun to the equation of the curve
At^- 2^W4 0^0.

40 Proceedings of the Royal Irish Academy.

If we write for s, Si+/, / being the equation of a line through 0^
we find

^«,>-2(^-4/)«i + C-2Bf^Ap = 0,
or, -4«i» - 2Bx%i + (7i = 0 ,

where B^m B- Af, C^ = C - 2i?/+ Ap .

I9^ow it is clear since / contained but two constants, it will not in
general be possible to make C^ contain the binary cubic ^ as a factor.
The curve we discuss is that special case of the quintic with a triple
point in which it is possible to transform the equation of the curve
so that Cim AQy Q being a binary quadratic. This involves one
condition. The equation of this curve can then be written in
this form

A%*-2B% + -4Q = 0 (1)

This being the case, suppose we transform the above equation to a
new axis of s , which will be effected by writing « +/ for s . For
this transformation we have

^««-2^i«+ (7, = 0,
where Bi ^ B - Af,

Now, we say, since we have two constants at our disposal, that it ia
possible to determine / so that B - Af may contain Q -f* as a factor.
Let us write then

B-A/^{Q^p)F, (2)

F being a binary quadratic.

Let us now see what Ci becomes in this case. We have

C,^A{Q^r)-2Bf

^A{Q^r)^2f.[Af^{Q-P)F]
- {Q^r){A-2fF) s. (Q-/^)£rsay,
where Em Am 2fF.

BoBBRTS — On same Properties of a Certain Quintic Curve. 41

The equation of our cnrve can consequently be written

^i»-2(Q-/»)^a + (G-/»)jff= 0. • . (8)

Now the equation of the tangents from 0 to this curve will be the
discriminant of the aboye equation considered as a quadratic in s.

Hence R - (Q-p){{Q-f^)F^ AB] ^ D,jD,D,D,;

we have, consequently,

D] being the equation of a pair of tangents from 0 to the curve
multiplied by such a factor so that

There are eight jS points as we have stated, and there are con-
sequently twenty-eight ways of arranging the eight points in pairs;
and consequently twenty-eight ways of reducing the equation of the
curve to the form

Ai^' 2lXF%-\^ DB = 0 (4)

If we write A m OiOtOtj Oi, (h and 0, being the tangents at
triple point, we can show that our quintic can be regarded as the
envelope of a certain cubic curve. Let it be so chosen that

k Oso, + D H ^' say,

then we have, multipl3^g the equation (4) by h and substituting
for kotOi its value if^^- D,

(Hii^^'B)^? - 2DFk% + kBB - 0;

or, «i (^s - By - D{(h%* + 2s (* jP- a,<^) + 0,2) - ifejff }

showing that the curve is the envelope of the cubic

•iD0>-l)»-2(p-l)(4^-^0

+ {ai««+ 2«(iJ-P-«i<^) + ai2)-4jff) = 0.

or, a,i^ + 2«(4^-a,p<^) + p»<h2)-*jff- 0; . . . (6)

and this cubic touches the curve in five points where it meets

the conic

s^ B Dp .

\!i ^if$cHdmga of the BoytU Irish Academy,

7. — XtM •qvation of our quiatio being

At^-^Bt-^AQ » 0,

4k» ^utttioa whioh is obviouBlj satisfied by s = 0 , ^ » 0, it follows
IbukI the tangents at the triple point meet the curve again in three
yHUttta which lie on a right line, this then is the characteristic of
\>ur qointic, th(U the tangenU at 0 meet the curve in three pointe tohieh
^W en « right line ; these points, we call the A points, and the line
joining them meets the curve in twa points which we call the
Q points.

It is easy to see that each of the A points has 0 as a cor-
responding point, and consequently the arguments of the A points
are seen to be -iV^, -i^i, and -iVi. K now we call Q and C,
the two points in which the line joining the A points meets the
eurve, and J and J' their integrals, as above defined, we must have

or, /+/'-2i«r. j ^^^

The Q points play aa important part in the geometry of our quintic.

8. — ^Any conic drawn through the Q points and 0 meets the
curve in five points whose corresponding points lie on a line.
This theorem is readily proved as follows : we have

where 2 refers to the five values of the parameters of the points in
in which the conic meets the quintic. Kow we proved, in the last
Article, that

GonsequenUy,

or,

-SJ(tf)-Jf,

which proYes the theorem, as - 1{$) is the yalue of I{0) for the
points corresponding to those in wlddi the conic meets the curve.

BoBBRTS — On some Prt^perHea of a Certain QuinUc Curve. 43

9.-43T7en the curve, to determine l^e Q pointB. By aid of the
theorein of the last Article, we can find the Q points by drawing any
line meeting the curre in fiye points. By means of the ruler alone
we can determine their five corresponding points, and through these
latter points and 0 it will be found possible to draw a conic which
will meet the curve in the required points.

Since we know the Q points, we can draw the tangents at the
triple point.

All we have to do is to draw a line through the Q points meeting
the curve in three other points, the lines joining these to 0 will
touch the curve at 0 .

10. — ^H any line be drawn through one of the Q points, Q,
meeting the curve in four other points, their corresponding points
lie on a line which passes through Qf .

We have

3 referring to the four Ob of the points in which the line through Q
meet tiie curve. Now

therefore,

consequently,

which proves the theorem.

On account of the importance of this theorem, we give another
proof.

Let the factors of Q be ^ and /, so that

and let us seek the equation which determines the points in which
a fine s « Xf , drawn through one of the Q points meets the curve.
Substituting the value of s in the equation

Aifi'2B%^AQ » 0,
we find

AX'f - 2Skq + Aq^ - 0,

or, dividing by X^,

A^kq-^hr^f) - 2^ = 0 (1)

44 Proceedings of the Bayal Irish Academy.

Now this is exactly the equation we should find to determine the
points in which a line %-\~^^ meets the curve; hence it foUows
that the lines

s-Xj' s 0, s-X'Y = 0

meet the curre in points which correspond.

NoWj these two lines are ohyiously connected hy a 1 , 1 relation
and the locus of their intersections is the conic Q = s' which we
call the Q conic.

11. — ^The eight R points lie on the Q conic. For, if a line be
drawn through 0 meeting the curve in two corresponding points
P and P^ the lines QP, QP intersect on the Q conic ; consequently
the Q conic must meet the curve in points which coincide with their
corresponding points, or in other words the Q conic passes through
the R points.

We shall show how to construct, geometrically, the Q conicy and
consequently to determine, geometricidly, the R points.

We have already shown how to determine the Q points ; conse-
quently, if we draw any three lines through one of the Q points, and
through the other Q point the lines which correspond to them, we
determine hy their intersection three points, which, with the Q
points, enahles us to determine completely the Q conic.

The R points are then found hy descrihing the Q conic as above
indicated.

12. — ^To draw a tangent to the curve at a given point P. Join
P to Q and Q hy lines PQ and PQ! , meeting the curve in two sets
of three points, one on each line whose arguments are, say, u and r.

Then we have, if $ refe^r to the point P,

+ « + / = ilT, \

from which we obtain by addition,

2/(tf) + « + !; + /+/'« 2iV, ... (2)
or, sum J jf T ^ %N^

2I($) + « + © « 0,

BoBBBTs — On some Properties qfa Certain Quintic Curve. 45

we mmy writB this in the fonn

-«-a+{J!r-2/(tf)} = -y, .... (3)

the signiflcation of which is that the 0 cuhic through the triplets
conesponding to u and v passes through the residual of the pair of
consecittiTe points at P. We can, howerer, draw the tangent at P
by means of two conies as follows ~

Let

a, + a, + «,, V = 6i + &, + ^,

then through the points corresponding to the points Oi , ^ , ^i » ^
describe an O conic meeting the quintio again in three points ^, e^, e^.

2/(tf) + 11 + 17 = 0,

we haTe 2/(tf) + <?i + tf, + ^ + flj + ^ = JV,

which proTes that the five points Ci, Ct, e^^ a^, and 5^ lie on 0 conic
which touches the quintic at P. Hence the tangent required is
deteimined by drawing the tangent to this conic at P.

13. — ^If a line he drawn through one of the Q points to touch the
coTYe in P, then the tangent at P\ its conesponding point, will pass
through Q' , and the correspondence between the lines QP and Q[P
will be of the kind noted in Article 10.

Au Atf AzwnttieA poantB lying on a line, and Q and Q* the Q pointa.

Now twelve tangents can be drawn from each Q point to the
curre, and to each tangent from Q, such as QP, corresponds a tangent
(DTP, so that the anharmonic ratio of any four tangents from Q is
equal to that of the four corresponding tangents from Qf.

M.tJL nu>c., TOL. Till., aao. a.] K

46 Proceedings of the Royal Irish Academy.

The twelve points of intersection, then, of the tangents from Q
with the corresponding tangents from Q' lie on a conic through Q
and Q*.

As the treatment of this qnintic is so similar to that of the
uninodal qnartic, previously dealt with, it is only necessary to
indicate how this curve may he suhmitted to the same manner of
investigation with many similar results. We give a figure of the
curve on page 45.

S.,Sa^ lZ6<!i.lo

THE MULTI-LINfiAB QUATEBNION FUNCTION.

Bt GHARLBS jasper jolt, M.A., D.So., F.T.C.D., Royal
Aattonoiner of Ireland, and Andrews' Professor of Astronomy in
the Univenity of Dublin.

[Bead Notbmbbh 10, 1902.]

1. A bilinear quaternion function is symbolically defined by the
equation

/(• + », c-^d) -/(a, c) +/(a, d) +/(*, e) +/(ft, d), (1)

i, i, e, and d being any four quaternions. In otber words, the function
/{ff f ) IS distributiye with respect to its first quaternion p, and also
with respect to its second quaternion q.

The quaternions e being arbitrarily assumed constants, the function
may be expressed in the form

/(W) « ^fipfiq + ^^M + ^J^M + ^Sfi/lj' ; (2)

and is thus seen to involre sixty-four constants, sixteen in each linear
function /i, A /„/4.

2. Traaapoaing the quaternions alters the function into its p&r'
mtMif which may be distinguished by a sub-accent ; thus

/(«)-/(tf), /(»»)-/ (w). («)

If the linear functions in (2) are self-conjugate, the function is
fmmuUlhf and oonversely,

3. Introducing two new functions P and C defined by the equa-

/(ff)-I?(«) + C(i»f), /,iPt)-tipq)-C(pq). (4)

it is evident that P is a permutable fanotion, and that C changes sign
witik pennotation ; in bot (by (3) and (4)),

a.I«A. PKOC, TOL. TIU«, 0IC. A.] F

48 Proceedings of the Royal Irish Acadetny.

The function C may be called a eomhitMtorial function, for

C{p'^xq,q)^C{pq), C{pp)^0. (6)

Tbu8 an arbitrary bilinear function is reducible in one way to the
sum of a permutable and a combinatorial function.

4. For all quaternions jp, q^ r, we agree to write

8p/(jr) = SiApr) = Sr/"(2p) ; (7)

and we call the new functions f{pq), f'{pq)i the first and SBcmd
conjugates, respectively, of the given ftmction. The phraseology is
justified by the consideration that the first conjugate is the conjugate
if the bilinear function is regarded as a linsar function of its first
quateniion.

5. Permuting the quaternions in (7) according to the rule (3), and
taking the conjugates, we obtain the series of equal scalars

Bpfiqr) = Sif'ipr) = Br{f)"{pq) = 8?(/),(,y)
= »rf"{qp) = BrirUM) = 6i(rnrp)
= Spf,{rq) - 8rUy{pi) = 6i{/,nrp); (8)
and from these we obtain the relations

(A (Pi) = (/')' (M) - UTiPi) = f'ifp) ; (9)

where the brackets are employed to obviate any confusion as to the
order in which the operations indicated by the accents have been
performed.

These relations, taken in conjunction with the obvious relations

Aps) = ifMP9) = (/)'(w) - if'TiPi), (10)

enable us to reduce any multiply-accented function to one or other of
the six fundamental functions, — ^the function and its two conjugates,
the permutate and its two conjugates.

6. Having now explained the fundamental principles underlying
the manipulation of bilinear functions, we shall indicate some of the
uses to which they may be applied.

Jolt — On the MuUi-Linear QuaternUm Function. 49

A quaternion being interpreted as representing a point, the equa-
tion

r = /(p^) (11)

establishes a relation between three points p^ q^ and r.

Let ^ be supposed given and constant. In this case we have the
general homographic transformation in space for a set of points {q) to
another set (r). The nature of this transformation is changed for
everj change in the constant p ; and the equation may be taken to
represent what Sir Bobert Ball might have called a four-syitem of
homographic transformations. The four-system of transformations is
more clearly exhibited by writing

P = Ml + ^ + ^tf» + ^4, (12)

where the symbols a denote given and constant quaternions, while the
Bymbols t denote scalar parameters. Thus the linear transformation
is

r = t^/{a^q) + f,/(a^) + t^{a^) + tj{a,q) ; (13)

and it is compounded from the four given linear transformations

/{^9)f A^)f f{M\ and f[a^q).

7. In the second place, consider r to be a constant quaternion,
whiles and q are variable, subject to the condition (11).

The equation then represents the general space hamographjf con-
neeting two points p and q^ so that if one is given, the other is
generdly uniquely determinate.

Again, as in (12), we may replace r by

r » #1*, + 9^ + #A + «A ; (14)

and, aoeofding to the various values assignable to the scalars a, we
obtain a four-system of space homographies.

8. In the third place, let ^ = ^ ; then we have (Art. 3),

'■-/(w)-P(w); (i«)

sad this represents the general quadratic transformation in space, so
that to one point q corresponds one point r, and to one point r cor-
respond eight points q determined by the intersections of three
qnadrie surfaces

8r,r - 8riP(^^) = 0, Sr,r= 8r,P(yy) = 0, Srgf = 8r,P(jy) = 0, (16)
where Srjr » 0, &c. are any three planes through the point r.

50 ProeeeditigH of the Royal Irinh A cat ferny.

9. In the fotirth place, write

r^i/(jq)-m9P)'0{pq); (17)

and we find an equation which represents a one-to-one correspondence
between lines pq knd points r.

10. Again, consider the mutual relations of three points p^ q,
and r, which satisfy the equation

Sp/iqr) = 0. (18)

If pt^qt^r^ the equation

Sr/(fr) = 0. or SrP(rr) « 0, (19)

represents the general cubic surface, and with this surface are asso-
ciated systems of linear complexes,

Sp(/(^)-/(r)) = 0, or SpC(fr) = 0, (20)

so that to each value of p corresponds a linear complex represented
by (19).

This is quite analogous to the quadric surface and the correlated
linear complex

82^^ = 0, Sfi/i-SyjS^-O, (21)

obtainable from a linear function/

Further, by suitable permutation of the quaternions in (18), we
may obtain an equation of the form

{Ipqr) = 0, (22)

which is combinatorial with respect to p^ q^ and r, where / is a
constant quaternion determined by the nature of the function/. This
equation (22) represents a determinate fixed plane which contains the
points j9, 9, andr.

11. Similarly, for the trilinear function, Tarious analogous reaolta
may be] deduced; but there is one which deserres special mention.
The equation

P^f{^.hq\ (23)

in which a and h are quaternion constants arbitrarily assignable,
represents the camphte group of homographie trans/ormMtions in epaee, or

JoLY — On the MuliuTAnear QttateniioH F^tuclion. 51

tiie whole HgUm^tUm of snob transformatioiiB. This appears on
expTMring « and h in terms of sets of scalar parameters ; and then from
(28) we obtain sixteen distinct transformations corresponding to the
axteeo products of the scalars of one set by those of the other.

12. From the equation for a trilinear fanction

80f{hcd) = 0, (24)

it is easy to derive, by permutation and conjugation, scalar equations
of the form

F{{abl W})=0, (26)

whidi is combinatorial with respect to m and i, and also with respect to
emi d. Thus given a line ab, (25) represents a linear complex ;
ind in this equation there is a relation between line and complex
and complex and line, somewhat analogous to the relations connecting
geaerstors of opposite systems of a quadric.

18. It is possible by suitable permutation to derive from f{abe) a
eombbatorial function of a, b,e; or, in other words, a linear function
(compare (27))

Flabc] (26)

of the symbol of the plane [jabc] containing these points. And in like
manner fran &/(M), we may deduce a scalar combinatorial function
of the four points; but this is simply {ahcd) multiplied by a scalar
deteniined by the nature of the function.

Following out this line, it appears at once that the various per-
Butatea of a function of the fifth order are not independent. In fact,
for the trilinear function, we find the combinatorial function (26), or
more folly

f{M) ^fihca) ^ficab) -f{cba) -f{hac) -^nacb) ; (27)

and similarly from the permutates of a function of the fifth order, we
can obtain a combinatorial function of the five quaternions a, i, e, d, e,
and the function F{ahcde), But a combinatorial function of five
quaternions is aero; and consequently the 115 permutates of the
function are coonected by one identical relation. In like manner, for
fnaetiona of hi^^ order, the permutates obtained from any group of

52 Proceedings of the Royal Irinh Acadetnp.

five quaternions are linearly connected ; and the number of distinct
permutates is reduced in this way to

„ w.ii-1. 11-2. 11-3. 11-4 ,^^.

n„ ^-^ . (28)

Similarly the conjugates and permutates of a function of the
fourth order are connected because Sa/(bcde) is a particular case of
the function of the fifth order.

The number of conjugates and permutates formed on the plan of
Art. 5 is, in the first instance, n(ii + 1). This reduces, for the
reasons given, to

^^""^^^ ir5IHn-5) n4n(n-4)' ^ ^

[ 63 ]

VI.
01^ BICTTRSAL CURVES.

By rev. WILLIAM RALPH WESTROPP ROBERTS, M.A.,
Fellow Trinity College, Dublin.

[Read Janvart 26, 1003.]

It u well known that the coordinates of any point on a curve which
pofiBesBes its maximum number of double points can be expressed
as rational algebraic functions of a variable parameter. The converse
theorem is also true, namely, that if the coordinates of a curve are
expressed as rational functions of a parameter, such a curve possesses
the maximum number of double points. Curves of this nature are
termed unicursal curves, and to each value of the parameter
corresponds one and one point only lying on the curve.

We propose to consider in this paper curves which we shall call
Heunalj since to each value of the parameter, in terms of which the
coordinates of the curve are expressed, correspond two points lying
on the curve.

We suppose then, that the coordinates of such a curve are ex-
pressed in terms of a parameter in the following manner : —

(1) a? = ^1 t -», -/5.

s = -4, + Bi yB.

Where ^i, At^ Ai, are binary quantics of the m^ degree in
two variables X and fi, the ratio of X to /& being regarded as the
parameter determining the points on the curve, i2 is a binary quantic
of the 2ii* degree, and Bi, Bt, B^ binary quantics of (m - n)^
degree in X and ft.

Such equations obviously remain unchanged in form for any linear
transformation of the variables X and ft.

54 Proceedings of the Hoyal Irish Academy.

In order to determine the degree of a cunre given by the abore
equations we have only to ascertain the number of points in which
an arbitrary line meets it.

Let the equation of such a line be ilr + my + iis ; and we have
eyidently the following equation to determine the ratio of X to f&, or
the parameter of each point in which the line meets the curve : —

(2) lAx + m^a + nA^ + {IB^ + mBt + »A) \/5 = ^i
or {lAi + mAt + nAt)^ « {IBi + mBt + nBt)*R.

Now this being an equation of the 2m^ degree, the degree of the
curve is, in general, 2m.

We now proceed to investigate the number, and determine the
position of, the double points on the curve.

In order to make clear the spirit of our method we shall first
show how the double points on the unicursal curve given by the
equations,

(3) X = ^i,

y^Aty

« = A^.

Aiy Atf Ai, being binary quantics of the m*^ degree in k and fi,
may be determined.

Let U'^O he the equation of the curve in x, y, s, which results
from the elimination of the parameter from the above equations ; and
let us call Z, M, and i^, the differential coefficients of U with regard
to X, y, and s, respectively.

We have then, for any point on the curve Zx + My + iVk s 0,
and for the consecutive point £dx -f Miy + Ifds « 0.

Hence we easily see that we must have
(4)

since Xy y, and s are homogeneous functions of the m*^ degree
in X and /i.

BoBBRTS-rOii Bie^^l Cw-vea.

ft5

Bat these eqaations show ns that Z, Jf, and iV'are proporticnal to
the detenaiiuuitB

dX dX

ds dx
dX dX

dx dy
dX dX

df &
d,. df.

'dz dx

d/i dii

■

dx dy
1 d/i dft.

we may write

dt
d^-

t S--^^^-^')'

where / stands for the Jacphian of the quantics A^ and Ai^ and
where it is to be remembered that

J{A„A,) = -J{A,,A,).

Gonseqaently we may write

(5) L^KJ{A^,A,\

M*XJ{A,,A,\
Nr.KJ{A,,A,\

where A is some quantity yet to be determined.

dU
Now the equation of the curve being of the m^ degree, ^ is of

the (m - 1)*^ in x^ y, and s, and, as these are each of the degree m in
the parameter, Z, regarded as a function of the parameter, is of the
ii(m- I )^ degree.

The system of equations (5), however, shows us that Z, if, and N
are proportional to functions of the parameter o! the degree

2«i - 2, for J {An A^\ J {A,, A,), J{A,, A,)

SIB of the degree 2m - 2,

Hence, it follows that Z, if, JT, when expressed in terms of the
parameter, contain a common factor A whose degree in X and /i must
be equal to the difference between »i(m- 1) and 2(m- 1), or
M* - 3m ^ 2.

Z, M. and N can thus simultaneously vanish for the

(m - 1) C«» - 2) roots of A =^ 0.

K.I.A. PBOC., VOL. YItl., SBC. A.] O

56 Proceedings of the Royal Lnsh Academy.

liTow, we know that to each double point correspoDd two valaes of the
parameter, that is to say, one corresponding to each branch of the
cuTTe on which the double point lies. The number of double points
is consequently equal to half the number

(m - 1) (i» -.2) or i (m - 1) (w - 2).

Having found the equation of the curve, we are then to find the
greatest common measure of Z, Jf, and N which will give us the
function A, and consequently the double points on the curve. This
is most simply done by dividing L by Ji^At, A^).

We now turn to the discussion of the curve given by the
equations

a? = .4i + A v^5.

y = A^ + B^y/M.

Proceeding in the same manner as that adopted in the case of the
unicursal curve, we find

L is proportional to the determinant,

2y/R 2-v/^

'^uyj'^'.B,^ '4^^yjp^B,'4

d/jt. djjL dfi d/jt.

2y/R

dfi d/jL

2y/B

or.

(6) L = A{2RlJ{Au JS2) + A^„ A)] +fiA^u R) + RA^i ^a).
+ y/Rl2J{Ai, A2) + 2RJ{R, ^0 + R2J(:Bu R) + B,J{R, ^3)]}.

Now the degree of R is 2n, that of J{Ai B^) in
!» - 1 + «i - n - 1, or 2i» - n - 2,

and consequently the function multiplied by A is of the degree
2m + ft - 2 in X and /i.

BofiBKTS — On Bieureal Curves. 67

But the degree of the carve is 2m and consequently the degree of
Lj M, iV, considered as functions of the parameter, will be of the
(2m - l)m^ degree involving y/lt which is of the n*^ degree.

And we infer, aa before, that the degree of A is

(2i»- l)»t-(2»»+ »-2),
or

i{(2»i- l)2»i-4»t-2» + 4}.

Hence the degree of A is

i(l»-l)(i'-2)-(»-l),

wliere p = 2m.

It mnst be remembered, however, that as A involves \/^, the
equation A » 0 must be rationalised to solve it ; and consequently
the number of roots of this equation, when freed from radicals, is of
the degree (/» - 1) (^ - 2) - 2 (n - 1), and admits of as many
roots. But, since each double point has two values of the parameter <
corresponding to it according to the branph on which it Hes, the
number of the double points of the curve in question is

l(l,-l)(j»-2)-(»-l),

that is to say that the deficiency of the curve is (n - 1).

The degree of the curve may, however, be reduced if Ai, At, At
contain a common factor which is also common to i?. If the degree
of the factor be r, the degree of the curve will be reduced by r.

This is easily seen by determining the values of the parameters
corresponding to the points of sections of the curve with an arbitrary
rig^t line, when it will appear that the conmion factor will divide the
equation when made rational. The degree then of the curve is

p = 2m - r.

Such a curve is represented by the system of equations

X = AiU + JSi ^uSn

y = A^u + B^ ^u8.

% = A^u + B% ^/u8.

where ^i, A^^ and A^ are of the (m - r)^ degree in X and ^l,

« is of the r^ degree and ^i, B^^ J?,, of the (m - iCf" degree, the
function under the square root being of the degree 2n.

Bl^ PfX>eeedinffs of the Royai Irish 4ca^ny>

II we refer to the equation for L in equation 6, we see that,
in the present case, it contains the facUtf u^ so that Z, M^ and If
are proportional to functions of the degree

2m - 2 + n - r,

while Z, If, and N^ considered as functions of X and /i, ese of
the degree

when we reject the common factors, and consequently the degree
of A will he

(/^-l)(i^-2)

'-(«-l).

Hence, in general, a hicursal curve of the degree p has its
deficiency equal to (n ~ 1) wliere 2n is the degree of the quantic
under the radical.

In our next paper we shall treat of a special class of hicursal
curves.

[ 59 ]

YH.

THE GEOMETKICAL MEANING OF CAYLEY'S FORMULA
OF ORTHOGONAL TRANSFORMATION.

By C. H. HINTON, Patent Office, U.S.A.

[oOMXTnilCATlSI) BT PB0FES80B A. C. HADDOK, F.K.8.]

Read Noybmber 29, 1902.

Catlkt has given the geometrical significance of his formul® of
transfonnation in the case of three axes. In Vol. zn., American
Journal of Mathematics^ Cole has shown the geometrical significance
of the formalffi in a special case for four dimensions. In order to
discoas the question of the significance of the general case in four
dimensions, I will write down all the forms which can come from
any oomhination of axes and angles. It will he seen that Cayley's
forms in the general case are incapable of geometrical significancie in
tenns olaxes and angles.

A very convenient form of quaternion symbolism can he used of
space of 2* dimensions, which I will adopt in the present discussion.

Let *yj\ kf h})e unit vectors mutually perpendicular. Assume

t^=ya = Jt» = A» = +l.

Let a transposition be accompanied by a change of sign. Taking the
two possible Hamiltonian circuitings

t;H = + 1, ijkh^" 1,

let as distinguish between the two systems of equations derivable by
calling those derived from ijkh >= + 1 the ^ kind, and those from
ijkk » - 1 the ^ kind.

To distinguish when the latter equation is used, introduce a symbol
«, which has simply the significance that the vector symbols after it
are combined— according to the ifkh » - I laws.

From the A form ifkh « + 1,

▼e ^t ijk = A, i^' o hky j'i = M ;

from the B form ffkh » - 1,

▼e get ifk ^ -h, ij ^ -hky j% = hk.

Hence, introducing the symbol cd, we get cd^V » coM, &c,

aa.A. rsoc., tol. xxnr., sic. a.] H

60 Proceedings of the Boyal Irish Academy.

By using the equations f* = 1, &c., we get the following multipli-
cation tables : —

Multiplier »

k h

M =y» j

-♦

-h k'

whk = aft

-i

-k

ih = hi -h

-J »■

iohi = iokf

-J

-i

jh'^ik -k

-h

» y

mhj = wik

-k

-J-

Here the ci> simply denotes that expressions like co^V operate according
to the B system.

Now, consider the effect of multiplying/t into

xi-^yj ^-zk-k-tohi

we get + ^' - yt - lA + wk^

or the projections of the vector are turned by right angles in each of
the coordinate planes of ij and kh ; wji gives

+ ic; - y» + ssA - wAi,

which differs from the last result in that, in the plane of kh^ the rota-
tion is in the opposite direction. Hence

(cos ^ + sin ffmji) (cos fl + sin ^V)

will turn a vector by the angle 0 -v ff m the plane of y, and by the
angle fl - tf' in the perpendicular plane of kh. Hence jV can be repre-
sented as a plane pair — that of y» and of kh.

Now t, y, ky h are any unit lines mutually perpendicular in space :
hence this symbolism is perfectly general ; and, introducing the six
coordinates which define any plane, we have enough constants to
determine the next more general rotation — a rotation, namely, in the
plane of XjXa, where Ai and \i are any perpendicular unit vectors, by a
given angle, and in the plane perpendicular to XiA, by another given
angle.

The plane Xi\i has direction cosines yaj3, '/a'P! subject to the two
equations

a» + j8» + f + a'» + j3'* + y»= 1,
and

aa' + pp! + yy' = 0,
which give also

(a + a')»+(i8 + i8')» + (y + y)«=l.
Writing \i\% as

yji + a^' + P%k + y'kh + o!ih + /J^A,

we see that, by the equations derived from ijkh = + 1, it becomes

. (r + y)y» + (a + a')^>(^ + /3')»^,

HiNTON — Cayley^s FormultB of Orthogonal Transformation. 61

and represents a plane pair whose constituents are to a certain extent
indeterminate, the plane pair is derived from all those perpendicular
planes, the corresponding sums of whose direction cosines are the
same. By " perpendicular,'* I mean perpendicular like/t and kk, not
nonnal, like the planes of y» and ^'.

Any of these plane pairs will turn a vector round in the plane of
A|A«, and in the perpendicular plane.

Now, consid^ the co form of the plane XiXa,

»(XfV + a*;' + pik + ykh + a'ih + fffh) ;

we obtain the plane pair

(y - Z)*?^* + (a - a')«*y+ (P - P')io%k.

This plane pair rotates a vector in the plane of XiXa, and in the per-
pendicular plane by a right angle in the direction k to h.
Hence the product

[oos^ + 8infl{(y + y)yf + (a + a')i^'+(i3 + i8')tAn

X [costf + 8in^ {(y-y)oj;V+ (a - a')<ii;^ + (/J-i3')«»*}]

will rotate a vector by 20 in the plane of Xi\2 ; and its projection on
the perpendicular plane will be unaltered.

If, instead of taking both angles equal, we take different angles
^ and ^, we get the rotation of amplitude 0 + ^ in the plane of
Y, a, py y, afy p'f and 6 - ^ in the perpendicular plane. It is most
convenient to take n, /, m, n', T, m! as the direction cosines of a plane,
instead of y, a, /?, &c., and denote the sums n + n', &c., by y, a, /?,
the differences n - n', &c., by y', a', p'. The most general rotation is
given by letting both angles be different, and taking two plane pairs

yfi + a^' + piky and y'wfi + a'tokj + j3'(oi%,

which I will call vi and um^u where y and /, &c., are unrelated, and y
and y stand respectively for n + n' and iV~ N't and so on, where Imn,
fiiV, ZMNf LM'N* are the direction cosines of two independent
planes.

In order to present the multiplication in a conspicuous form, I
write out the multiplication table of the quaternions, with the coeffi-
cient adjacent. By assigning values to 0, ^, 0', ^', y, y, c, ^,&c.,
we can find the combined rotation equivalent to any pair of rotations.
For cos 0 1 write 0], and sin 0, 0. It will be found that taji axidji are
commutative as well as any other A and B pairs. The multiplier
occupies the two columns to the left.

Proceedings of the Boyal Irish Acadefny,

Table 1. — Showing Pboduci

MultipUer.

ye<t>i

y'Bifp
ao'i

i»kj

^i4>'i

oV'i

c^kj

«a

C^i^'l

;»

-1

iaji.ji

wkj.ji

-^•

cdk.J

<f0\<p'

wji

o»ji

uji.Ji

-1

ufji.kj

o»ik

t^'i . ik

-^k

ae^4>i

-U

«;» . kj

-1

wkj.kj

i^.il

o^^'kp'

ukj

wkj

ookjji

-uik

«^-.^-

-1

wkJ.iJk

bff<t>\

ofji.ik

-y»

u»kj.ik

-I

mik.i

b'e^i<t>'

talk .Ji

»ki;

i^k.kj

-wji

wik.ik

-1

c'ce'ip'

taji.ji

oiji.ji

-wji

-y»

ftj/i . ik

uik .Ji

-^Ji.kJ

-«*;••;

a{^0'(p'

vikj.hj

oikj, hj

-takj.ik

--mk.kj

-.kj

-hf

nhJ.Ji

^Ji.h)

bb'0<p'

wik.ik

uikAk

uik.kj

-o^kj.ik

- wik .Ji

-otji.ik

-«•*

-ik

<fa0<p'

c^i . Hj

u)ji,hii

-cojiAk

-kj

-»Ji

aaik . ^'

o^.Ji

-^A

cd^^

^kj .ji

u>kjji

-ckj

- talk .Ji

f^ij.ik

-J*

-mhj.kj

^ji.Ji

db^^

wkj. ik

wkj.ik

<.kj,kj

-cik.ik

-w^'.yi

-ik

-•^*

^ji.ik

ah'tf^

uik.kj

mk,1y

^mk.ik

^kj.kj

-«•*

-f^Ji.kj

uik Ji

-kJ

O'c&ip'

6oik ,ji

mk.ji

-mk

ukj.ji

aik.ik

-t^Ji

-mik.lij

-Ji

y^^'

tnji . ik

o^i . ik

<»Ji.kJ

-ik

-^ji.ji

taik.ik

-^Ji

-•xy.ii

Oi is written for cob 9, and e for Bin 9.

HiNTON— Cayley*9 Formulm of Orthogonal Tramfomiation. 63

BOTATQB BT A BoTATOB.

n •kj.kf

mik.ik

tckj.ji

<ttkj . ik

wik . kj

0y9ip
toik ,ji

y'$e<p

o^i . ik

> •kj . kj mik . ik

ix m^f.ik

H wik,^;

i.ii -mkf

I
i.;i ^kf

.h -mkf'.ji

j^ -•fi.kj

I mk.ik mkj.kj

mik.kj

-•mkf.ik

mik,ji

9ffi . ik

-mii

-ik

^ - mk

^■h ji
-;..• mji

•') -mji.ik

' '. -mik.ji

- m^\Ji

-mji^kj

-t^i

->»

•iy

t^i.kj

mji . ik

-*/

-mik.kj
mkj.kj

-i*
-mkj.ji

f»kj Ji
- nkj
mik.ji

- ctkj, ik

-J*
mkj,hy

- mji Ji

-mik

ik
-t^'i.kj

- mik . ik

-V

0^'t . ik

mji
mik . kj

mkj.ik

-mkj.kj

mik . ik

mkjji

-ik

-mkj

- ttfi . ik

— mik,Jy
->»

mik Ji

- •!/■» J^

c^'i . *y

-mik

mik.fy

mik . iA;

-mkj.kj
— mik
ttf'i.kj

- mik .ji

-mkj.ik

-ctf'i

m^j

mji . ik

-mji.ji

-ik

mltjji

mik .ji
- mik

-mkJJi

-mik.ik
t^iji

mxk.m

mkj

-o^i.ik

-*/

ml^' . ik
— ftj/i
mji.kj

tUJi . ik

-t^'i.kj

-ik

mjiji

-mik.ik

-^ji

mkj . ik

kj
-mik.ji

- mkj

mik . kj

mik
mhj.ji
-mkj.kj

- mkj . kj 1

#1 IS written for cos B^ and B for «in B.

Proceedings of the Royal Irish Academy,

From the foregoing Table 1, we can write down the coefficient of the
compounded rotator by simply collecting terms, and can, by a simple
calculation^ find the angle and axis planes of the compound rotation.

Table 2. — SnowiNa Multiplioation of Botatob into a Veotob.

z
k

w
h

«1*1

(><nr

-t

-A

•iW

9ffi

-t

-A

«^ia

-A

»,*a'

^kj

-y

— i

0^10

-k

-A

Oi^fl-

c^ik

-J

ntf'i.ji

— i

-J

e^aa'

f»^'.kj

-J

-it

0ip0P^

wik.ik

- %

e^y'

m-Hf

Bipa'y

f»kj\ji

-h

-J

0ipa'fi

f»1y,ik

6^0^

uAk.kJ

-h

-k

e^ffy

uik.ji

e^^y

c^'i . ik

-A

-i

01 is written for cos 0, and 9 for sin 0.

Table 2 diows the effect of multiplying the quaternion
{01 + ftr,) (^, + ^jr'i) into a vector.

The direction cosines of a transformation can be found from Table 2
by collecting the terms : zi + yj+tk + toh becomes

*[«{ft^-«^(7Y'-«»'+i8i8')}+y|-e^.y-tf,^ + tf<^(a'/8+aj8')}

+ *[«(- ft^ijS - 0i<l>P'+ 6<t> (ya'+ a'v) I + y {tf^io + e.<^'+ $<!, {pY+ p'y) }
+ «{tf,^. - «^(aa' + i8^ - n')} + wf^ii^iy - tf^ + ft^(a'/8 - aj8'))]

+ «{- e^,y + S,^/ + 0<l>{a'p - a/8') I + w {*,^ + fi^Cy/ + aa' + fifi')}].

HiHTON — Cayki^B Formula of Orthogonal Transformation. 65

These tonnnls agree with Cayley's if 0 » ^, and y « n + n^ &c.,
y^n-n'f &c., in which case the rotation is about a plane, by the

80^20.

In no case, howeyer, can they be made to assame his general form.
The reason appears to be that he starts from the determinants

/ 9

% h

^d -g -A %

and the form with the rows and columns interchanged in defining his
quantities a, ft, «,/, g^ h. Now an inspection of the form above shows
that the determinant of a transformation effected by any kind of
rotation cannot assume this form. Hence there is no geometrical
inteipretation of Cayley's constants with angles and axial planes.

[ 66 ]

VIII.

METHOD OF OBTAIOTNG THE CUBIC CURVE HAVING
THREE GIVEN CONICS AS POLAR CONICS.

By J. P. JOHNSTON, ScD.

Read Junb 22, 1903.

The problem to obtain the cubic which will have three given conies
as polar conicB was solved by Dr. Salmon (Conm, § 389 «) by finding
the equation of the cubic when the equations of each of the conies
was written in the form

OP* + 5y» + <»' -I- (^» = 0.

The form in which he obtained the cubic in this case enabled him to
state at once that in the general case the cubic was the Hessian of the
Jacobian of the three conies minus twice the Jacobian multiplied by
the invariant T, The following is a different method of investigating
the same problem, by which it is seen at once that the solution, where
possible, is unique.

Let the equations of the conies be

u a {a,h,o,f,g,hJxy%f = 0.

If these transform to
by means of substitution

then since

dV dW dW dU dU dV

dz^JY' dx^dz' ^y"5z'

(1)

Johnston — Method of Obtaining the Cubic Curve. 67

we must have

f d . d .,d\ ( d , d „d\

Squating the coefficients of «, y, s in these identities we get nine
tquations of the form

The eliminant of these equations with respect to X, /i, v, X', &c.,
Lb a skew symmetrical detenuinant of the ninth order, which con-
sequently Taniahes indentically. Therefore the nine equations are
not independent, but are connected by a linear relation. It is therefore
poadble, in general, to transform the conies to the required form, the
transioiination being given uniquely by any eight of the equations.
If the values obtained for X, /a, &c., are such that

= 0,

•

•'

the transfonnation fails, and it is not possible to obtain a cubic having
the three conies as polar conies, for the vanishing of this determinant
would imply that a? = 0, y = 0, t = 0, passed through a common point.
Let ^ be what * becomes when we replace XyYyZm it by the
corresponding values of a?, y, «. Then prince

we have
but

4» = XU\ Yr+ ZJF,

<li ^ Xu ■\- Tv -\- Zu?,
X = XZ + ziiF+vZ,
y = VJ + ^F+i/Z,
» = X"J + /i"F+v"-^.

%,VK. FBOC, VOL. XXIV., SEC. A.]

68 Proceedings of the JRoyal Irish

Therefore,

^ u V w
X \ iL y
y V / v'

= 0.

Therefore the equation of the cuhic having ti, p, tr, as polar conies is

0 u
X \

y A'

X" /t"

« 0.

and Uy t7, tr, are the polar oonics of the points whose coordinates are
(X, X', X"), (/i, /, /')» (•'j •''i »'")> respectively, since ZT, F, »^ are
the polar conies of * with respect to the points (1, 0, 0), (0, 1, 0),
(0, 0, 1).

The equations (1) show that there are three points, say P, Q, R^
associated with any three conies ti, v, io^ such that the polar line of R
with respect to f>, and Q with respect to fc' ; of P with respect to «?,
R with respect to u ; and Q with respect to ft, and P with respect to
V, are coincident. If P, Q, and i2 are not coUinear, it is possible to find
a cubic having u, t^, fc' as polar conies, and they are the polar conies of
the points P, Q, and R,

The algebraical statement of the problem under consideration is to
transform three ternary quadrics, so that they may be the first deriveds
of a ternary cubic. The corresponding problem for binary quantics is
to transform two binary cubics so that they may be the first deriveds
of a binary quantic ; and I would remark, in conclusion, that the above
method of investigation readily gives a solution of this latter problem
also.

[ 69 ]

IX.

SOME NEW RELATIONS IN THE THEORY OF SCBEWS.

•Bt PROFESSOR C. J. JOLY, M.A., D.Sc, F.T.CD.

ScAd DiCBMBU U, 1903. Published Jakuabt 26, 1904.

br a paper, communicated to the Academy two years ago, and since
published in the Transactions,^ I wrote : —

^« Ab another example, if fi, X represent a wrench, X being the
force, and /i the couple at the origin as base-point ; the ratios of the
independent terms of the array

(Ml, /*ti . . . M
(Xi, Xj, , . . X^)

incliide all the invariants of an n-system of screws."

In other words, the ratios of independent terms are the same for
idl screws of the system.

Naturally one seeks to reduce invariants to their simplest form,
but it is possible to overlook some important relations involved in the
erode and unreduced expressions. Consequently I did not notice the
following theorem until recently, in framing an example for a text-
book. It is one of many, but I have not leisure at present to examine
the subject in detail.

By the laws of quaternion arrays, two independent terms in the
•nay for a three-system of screws afford the invariant

S/AiX,Xs + S/i^XaXi + S/igXiXa ^ ^ , ^ ^

SX^X.^^ ""^^^'^

where «, &, and e are the pitches of the principal screws of the system,
la this replace /ii by piXi + YaXi, &c., where pi is the pitch and a the

1 Yd. zzzii, Section A, p. 30. See also p. 28.

a.l.A. PMOC., VOL. XXIV., SEC. A.] K

70 Proceedings of the Royal Irish Academy.

vector to any point a on the axis of the screw (fii, Xi). The invariant
reduces at once to

SVaXiVAAs+ 8V/9X,VXsX, + SVyXsVAiA,

^^^^^ ^ '^+j>i+j>,+j>, = g + & + <?;

and on expansion of the terms in the numerator it becomes

S(i3-y)X|8X,X,+ S(y~a)X,8X3Xi + S(a-i8)XsSX,X,

-^ — '- SaXx +i'i + f'»+i'j=»a+* + tf.

Let ▲, B, c be any points on the axis of three screws (the extre-
mities of the vectors a, ^, and y) ; let B|Cx denote the projection of the
line BG on the axis of the screw (/ii, \\) ; let (23) be the angle between
the axes of the screws (/as, A^) and (fi«, Xs) ; and let sin (123) denote
the sine of the solid angle determined by lines parallel to the three
axes (rotation from Xi to Xt to A« being supposed positive) ; then the
invariant is

B,C|C08(23) + C2A»C0s(81)-f Atf,C0s(12) . ^ . ^ . ^ .^IL^.

3i^ ^123) +i^i+i^.+^, = a + * + ^.

In particular if three axes intersect, and if the three points are taken
to be at the point of intersection,

jPi+i?«+jPj = « + * + <^;

while for any points on the axes,

BiCi cos (23) + CsAs cos (31) + AsB, cos (12) = 0 ;

and from this single theorem many other consequences may be de-
duced.

[ 71 ]

A METHOD OF REDUCTION OF A QUARTIC 8UEFACE FOB-
8ESSING A NODAL CONIC TO A CANONICAL FORM.
WITH AN AFFLICATION OF THE SAME METHOD TO
THE REDUCTION OF A BINODAL QUARTIC CURVE
TO A CANONICAL FORM.

Bt JOHN ERASER, M.A., F.T.C.D.

Bead Dscbmbbr 14, 1S103. Published Janita&t 26, 1904.

Thb equation of a quartic surface possessing a nodal conic may be
written in the form —

[ia» -h )8y» + y8»]» + w^ {[«, h, e, d, /, ^, A, /, m, n] [^wj] - 0.
We may write x for «v/a, &c. ; and then the equation becomes
[«» + y* + «*]• + tr* { labedfyh hnn] [«y»«^]*} = 0.
If the quadric

«' + y* + 2' + [«ur + /3y + y8 + Sw^w = 0

has double contact with the quartic, then it must have double contact
with the quadric

[ax -k- fy •¥-/* + Sf^]' + labedfyh Imn] [xyzw^ = 0 ;
and hence
[or + J5y + 71 + iwy + labedfyh Imn] [icysw?]*

+ 2X[4r» + y» + 8* + w{ax + )8y + y8 + Sic)] ■ ZJT,

where X « 0 and Jf «= 0 are two planes.

Hence, since every plane meets this quadric in a pair of lines,
every fiivt minor of the discriminating determinant of this quadric must
Tiniah.

Let A denote this determinant : then

«+2X4-a* h-k-afi y + ay /+o(X + 8)

h^fia a + 2X + i5» f+fiy f» + )3(X + 8)

y + ya f+yP tf + 2X + y» « + y(X + 8)

U a(X + «) w + )8(X + 8) « + y(X + 8) c + (X+ 8)»-X«

Proceedings of the Royal Irish Academy.

1 0

a » + 2X + a'

P h + pa

X + 8 /+o(X + S)

1
a

x+8

0
J + 2X + i8«

f^yp

0
y + ay

f^Py

tf + 2X + y*

/+a(X + 8)
ffi + )8(X + 8)
n + y(X + 8)

i» + i8(X + 8) « + y(X+8) c? + (X+8)»-X*

!» + 2X

-i8

J + 2X

/
m

-y

^ + 2X
n

-(X+8)
/

d-\*

= 0,

Now smce any first minor of the original form of A yanishes, then
any first minor of the latter form must vanish. Hence,

a + 2X A g

A * + 2X /

g / tf + 2X

I m n

which is a qnintic for X.

ii. oZ + /3if+yJ\r+(X + 8)2) = 0,

where Z, Jf, JT, 2) are first minors of i.

n
i-X»

Since

-1 a p

a a + 2X A

p A * + 2X

y 9 f

aZ + )8if+yiV+(X + 8)2>

c+ 2X
= 0,

= 0.

IB ^- [aZ + ^M^ yN\ \D\

and «» + y* + «* + fi^[flur + )8y + ys + 8m7J = 0 ;

.-. 2>(«' + y' + «') + a2>u?a: + ^Thoy + yi)ira = - 82>ir'

= w>[aZ + ^M^ yJV+ XD] ;

.-. a\I>vox - ir'Z] + /3 \J)wy - ii^'if ] + y \J)w% - tt^iV^

+ i>[«» + y» + «•] - X2>«r» = 0.

Frasbr — Reduction of a Quartic Surface to a Canonical Form. 73

This denotes then a system of qnadrics possessing a Jacobian
suiface.

That is, the system of quadrics passing through the nodal conic and
touching the quartic surface twice possesses a Jacobian surface, and
there are fiye such systems.

Dx - 2wl

Dtp

Dy-2u)M

D%-2wN

J>y

-\Dw

= - i>*a>»[2)«« + y* + «*) - 2u>lLx -^ My -^ N%] + KDw*^.
J containa tc* as a factor ; and the remaining surface is the quadric

«» + y» + 1* ^ + Xtt^« = 0,

And there are five, and only five, such qnadrics.

CoDsider the point which has w for its polar plane with respect to

a« + )8f^ + yw + T = 0,
where

U " Dwx - w^L

V = Bvoy - u^if ,

W « Dw% - u^N^

r « D [«* + y" + 1» - Xfi^],

2)[air + 2»] = 0, I^fivo + 2y] = 0, l]{o.w + 2i] = 0,

'f f > «» w being the coordinates of this point ; hence if, in iii, we

■abstitate for a, /3, y,

X y %

RtpectiYely, we get

0 + 2X
A
9

A
&+2X

as the condition which the coordinates must fulfil — that is, the locus
ef the point is a quadric.

^ + 2X

Ji- [«« + y» + «"- 1^ (Xi« + Jf,y + Ny%) + X|W»],

74 Proceedings of the Royal Irish Academy,

where /(Xi) - 0,

^'^ f(M) ■ I « + 2Xi h g I

h h + 2\i / m

g f tf + 2Xi n

I m n <^-Xi*

Consider

Dr = 8V + &c. . . . f{K) = - 8V + &c. ;

hence the coefficient of (2:* + y' + s*)* vanishes ; and for the same
reason the coefficient of («' + j/^ + s') tc [x, y, %] also vanishes.
The coefficient of

ic*x^ = 2S* ^ ■ ^' + 4.2*

But since /{K) = 0,

Z,» - ^^^ « 0.

.-. ^* = ui, - - 4V + &c.

Hence [2XrDr + ^-^H = (16 - 16) V + &c. ;

therefore the coefficient of tr'^ also vanishes.
The coefficient of

^^.x^^'-.'X -^'^'

but

£rM^-M,D^~ 0.
:^^ = ^ = - 2AV + &c

Hence the coefficients of to* [xy, y%, tx] all Tanish ; and hence

that is, that the squares of these five Jacobian Quadrics are connected
by a linear relation.

YKASKK^Bef/ucfion of a Quartic Surfacf* to a Canonical Form. 76
Agaiiiy consider

the coeffideiit of

(*» + jf» + ■•)»- 2,* -^^^- --1,
the ooefficient of

"*^"^' '/(a:) = -*^— ta:^ — = **'

lienoe the teims

ir[4r», y», ««, «V> *'«, y'«> y*** «'*» «V]
do not appear, aa their coefficients all vanish for the same reason.
The coefficient of

.-, ArDr - X,« - 0

^raVIBV + 4V(« 4- 3 + c)} + &c. - 4A,{4V + 4V(> + c\]-\

~ \ AK) J

ud of coane the ooefficient of teV » - i, and of tcHs* - - 0.
The ooefficient of

8X,^1C ^r8A,.J,-| 8A,[-A(c + 2X,Xa-V) + &c.]

Similarly the coefficient of u^yH » - 2/, and of tc^ » - 2y.
The coefficient of

hence this coefficient » - 21, that of ti^y » - 2m, and of t^s » - 2n.
The coefficient of

ii^ = V—^y /Z+iiiJr+nJV'+(c?-X»)i)=/(X);

^^M^Naie only quadratics in X ; hence
Vi>r = - Xr/(A^) + A j;/Z^ + wiC + n^^^ + ^Z)J = d . 8A/ + &c. ;

Proeeedittgf, of the lioyal Irish Aeadenty.

V /(\.) = 0,

Si'

/(M

= -rf;

Hence we have the equation of this qnartic surface expressed in a
canonical form

VIZ. :

where

**m ■•'-•■''•

We might consider particular cases of this quartic surface, accord-
ing as the conic becomes an ellipse, hyperbola, parabola, or circle.

In particular, the imaginary circle at infinity.

In this case, the plane to becomes the plane at infinity, and the
axes Xj y, 2 rectangular ; hence we may put, without introducing any
other peculiarity, /=:0 ^g^h, and a = ^ = y=l in the original
equation.

And the Jacobian quadrics become of the form

a;* + y» + «» '■ ~ ^— + Xw*,

▼iz., spheres, where
0=/(X)-

a + 2X 0 0 /

0 S + 2X 0 m

0 0 0 + 2X n

1 m n d - \*

The system of quadrics which have double contact with the surface
must also be spheres, since they pass through the circle ^imaginary) at
infinity, and the locus of the pole of the plane at infinity with
respect to them, that is, their centre, is

ur X y %

X a+2 0 0

y 0 & + 2X 0

% 0 0 (; + 2X

Frasbr — Seduction of a Quartic Surface to a Canonical Form, 77

which BhowB that the five quadrics helong to the same confocal

SJBteilL

iP.

and hence, in this casei the identical linear relation becomes

«»y» + «» \_{ahcfgK) {p^Y\ = 0

represents the general equation of a binomial qnartic curve haying the
points s a 0, J? a 0 ; s » 0, y « 0 for its two nodes.

« (flU? + )8y + yx) + ay = 0

denotes a conic passing through the same two points.

If it has double contact with the quartic, then it most also have
doable contact with the conic

henoe

(o« + )8y + 7f«)» + {ahefgh) (ays)* « 0 ;

(«r + jSy + yx)» + {ahefgK) («ys)» + 2X [s (<u? + )8y + ys) + «y] - X» ;

md therefore every minor of the discriminating determinant must
vanish.

Let A denote it.

• + a» A-l-X + a^ ^ + a(X + y)

* + A + /8a h^fF /+^(X + y)

y + (X+y)a /+)8(X + y) ^ + (X + y)«-X»
10 0 0

a a + a» A-fX + a)3 y + a(X + y)

/» A + X + )8a l^fF /+^(X + y)

X + y ^ + aCX + y) /+)8(X + y) (, + (X+y)*-X«

1 .a -)8' -(X + y)

a a A+X y

/8 A + A » /

1 9 f •-^'

and every minor of the latter form of A must vanish.

Proceedings of the Royal Irith Academy,

a A + X g

h + \ h f

= 0;

9 f c-X?

ii. ae + pF-i-{\ + y)C~0;

iii.

1 -a -P

a a h + X.

= 0.

/3 h + X. b

i. ifl a qnartic for X, and to each value of X we have a linear rela-
tion connecting a, p, y, given by ii., and a quadratic relation between
a, fif given by iii.

ay + 1 (cur + )8y + y8) = 0,

aG-\-pF'{'{\-\-y)C = 0;

is an equation of enveloping conic ; and its form shows that it belongs
to a system which has a Jacobian cubic curve, viz. :

s 0 y

0 s X ^ Of

Cx -20% Cy- 2F% - 2XCk

or - 2«[X(7«» ^Cxy-Fui' Oty] = 0.

Hence the system which, since it passes through two fixed points, has
a common Jacobian conic

ay- J.[i^x+^y] + Xi« = 0,

a conic which also passes through the same two points.
To each value of X, we have a corresponding conic

a(Ckr- ^t*) + /3(Cky-iT5') + Gry + XCk» == 0.

If this conic becomes a pair of lines, then

0 C aC

CO pC =0,

aC pC 2XC-'2aG-2pF

- C* l2apC - 2XC + 2aO + 2pr] « 0 ;

Fkasbr — Beduction of a Qiiaviic Surface to a Canonical Form. 79

rejecting the factor C*^

Inxt

2aQ + 2pF- 2\C + 2a/8C « 0 ;

-1

a
a

P A+X

= 0.

Eliminmting P between these two equations, we get a qnartic for a,
and to each valne of a one yalue, and one only, of p.

Hence, to each Talne of X, we have four conies of the system which
reduce to a pair of lines, that is, a conic consisting of a tangent from
each node of the binodal qnartic.

The node of this conic, t . e» the point of intersection of these two
tangents, must lie on the corresponding Jacobian. Hence, the four
Jacobians, as written aboTe, are the equations of four conies, each of
which paasee through four of the sixteen points of intersection of the
tai^ents to the binodal quartic from its nodes ; and hence, we may
infer that the anharmonic ratio of the two pencils of tangents is the
aame, since the conies pass through the nodes.

If the point xys has the line s = 0 for its polar with respect to

then

a[Ckar- W| + ^8 [ Oiy - i?lB»] + C[«y-A«»] = 0,
oCk + Cy = 0, aCk + Gc = 0 ;

y «

a A + X
X A + X h

= 0.

Hence, the locus of the poles of the line s = 0 with respect to the
coidcs of the system lies on a fixed conic ; and the above is its equation
detennined in terms of the coefficients of the quartic.

Conesponding to each value of X, we have a conic.

The Jacobian conice are

Consider

..A.

SI S^ ^M^MffB of the Royal Irish Academy.

^ ^^4t,tj«;?* o^ 4V». 2«V» «a:y* vanish since 9 is of the first
Aui. >\ \ s >^*^ *i« of the first order, and C of the second order in X,
.. K^inf: A? lW l\>urth order.
71*^ <vv4!&*'if«t of 2* is

»V vyidllcient of s^ is zero.
T)^ \»efficient of

FrO.-CrH^^Q, V /(A,) = 0,

t1t« coefficient of

and

Again consider

The coefficient of

FiiASSR — Seduction of a Quartie Surface to a Canonical Form. 81
The oodBcient of

^ ^V(M ^' f{K) •
The coefBdent of

The coefficient of

The eoefScient of
The coefficient of
Hcaice

^/(mL*^ — c. — ^^-^J

The bidrciilar qnartic might be treated in precisely the same
manner, s » 0 denoting the line at infinity in the plane, and a: + ty
written for ar, ar - iy written for y. But it can also be treated
<3ir«^y, thus —

(**+y*)' + «p* + V + ^ + 2ya? + 2/y = 0
denote* its equation.

82 Proceedings of the Boyal Irish Academy.

Iftheciiele a^-^ff* + 2ax + 2fy-¥y-0 has double contact with
it, then

hence, as before, every minor of

A =>

a + X + 4a' 4afi 2ay + y + aX

4i8o 5 + X + 4/8* 2i8y+/+i8X

2ay + y + Xa 2^ +/+ X^ c + Xy + y»

- 4a - 4^

a a + X
iS 0

X X

4 + 2 ^

0
5 + X

-(X + 2y)

/
x»

^-4

i. I a + X 0 y
0 3 + X /

^ /* X*

= 0, or -^ + r^ -<? + — = 0 ;
' fl+X i+X 4 '

1 - 4a - 4^8
a a + X 0

p 0 ^-i-X

4a«
0, or — r +

a+X i^+X

+ 1=0;

a fl + X 0
iS 0 i + X

X X

4+2 ^

2ay 2^8/ X

i. Shows that X is determined by a quartic.

ii. Shows that the centre of the enveloping circle moves on a fixed
conic, and also to each value of X we have a determinate conic, and
the conies are confocal.

Frasbr — Reduction of a Quartie Surface to a Canonical Form. 83
iiL ShowB that the circle cuts the fixed circle
, , 2a 2/ X ^

orthogonally; hnt, since

a + Xi i + Xi 4

a 4 Xfl
2^"

= 0,

2/»

Xi + X,

(a + XO(a + X«) (^ + X0(^ + X^)

hence the fixed circles are orthogonal, and as in the case of the hinodal
'iuartic the 16 points of intersection of the tangents from the circular
points to the quartie lie hy fours on these circles,

^/fe)[<-*'''*^

'*^*\

where
aud

T-o,

m (jfl + jf*y + to* + hy* + 'iffx + Ify + *,

era /(A,) -

« + A, 0 g

> + A, /

, / .-¥

The lednction is just the same

(7-(<i + x)(a + x),

-^-^(4 + X),

-©-/(a + X),

&c. . . .

= 0.

84 Proceedings of the Royal Irish Academy.

It is interestmg to show, from this point of view, that

(a + \)(4 + X)

(a + X) (4 + X)» 2'

a + Xi + X 4"(<» + X)(6 + X) '

ako

{a + X)* (^ + X)' " 2 SA • (a + A)(^ + X) " (a + X)Ci + X)'

since

/(X)=0;
hence

- 5j ^^ - (*» + y')» + «r* + V + 2^« + 2/y + r.

PROCEEDINGS

OF THE

ROYAL IRISH ACADEMY

VOLUME XXIV

SECTION B.-BIOLOGICAL, GEOLOGICAL, AND
CHEMICAL SCIENCE

DUBLIN

PITBLISHSD AT TEE ACADEMT HOUSE. 19, DAWSON STREET

SOLD AtBO BT

H0DOB8, FIGGIS, ft CO., LimTiD, 104, GBAFTON STREET
An R VILUAHS ft NORGATE, LONDON, EDINBURGH, AND OXFORD

1902-1904

The Acadbmt desire it to be understood that they are not
answerable /or any opinion^ representation of facts^ or train of
reasoning that may appear in any of the following Papers. The
Authors of the several Essays are afone responsible for their
contents.

CONTENTS

SECTION B-BIOLOGICAL, GEOLOGICAL, & CHEMICAL

SCIENCE

PAOB

Babbbtt- Hamilton (Captain O. E. H.)» B.A., F.Z.8.,

. M.B.LA. :—

^ Abstract of a Physiological Hypothesis to explain the
Winter Whitening of Mammals and Birds inhabit-
ing Snowy Countries, and the more striking
points in the Distribution of White in Vertebrates
generally, 808

An Addition to the Lint of British Boreal Mammals, . 816

Caepihtbb (Gkobge H.), B.So. Lond., M.B.I.A. : —

>-0n the Belationships between the Classes of the

Arthropoda. (Plate YI.)i .820

CoLi (Gbikviixe a. J.), M.B.I.A., F.G.S. :—

' The Intrusive Oneiss of Tirerrill and Drumafaair, . 861
On Composite Gneisses in Boylagh, West Donegal.

(Plates L-V.) 208

FlBmx (Gsobob), B.A. : see Btan (Huan).

McAbdlb (Davto) :—

A List of Irish Hepaticfld, 887

McHmnr (Alkxandeb), M.B.I.A.: —

'Report on the Ox Mountain Bocks and their probable
continuation from Galway and Mayo into Donegal,
TyronOy and Londonderry, • • • • . 871

CoRtemU

Nichols (A. B.), M.A. : — page

A List of Irish Echinodenns, 231

O'Bullt (J. P.), C.E. :—

On the Waste of the Coast of Ireland as a Factor in

Irish History, 95

Peabou (Robxbt Iildtd), B.A., B^. : —

On Types of Distribution in the Irish Flora, . 1

" Gleanings in Irish Topographical Botany, 61

Btan (Huoh), M.Am D.Sa, F.B.UJ., and Gsobgb Ebbiix,
B.A.:—
• The > Synthesis of Glycosides: Some Derivatives of

Arabinofie, 879

SoHUOT (B. F.), B.8o., PhJO. :—

. Some BodmAb on the Atlantis Problem* . . 268

DATES OF PUBLICATION

PiST 1. Pages 1 to 94. July, 1902.

ff 2. „ 95 „ 280. September, 1902.

„ 8. „ 281 „ 802. April, 1908.

„ 4. „ 808 „ 886. September, 1908.

„ 6. „ 887 „ 502. January, 1904.

PROCEEDINGS

THE ROYAL IRISH ACADEMY.

PAPERS READ BEFORE THE ACADEMY.

ON TYPES OF DISTRIBUTION IN THE IRISH FLORA.
Bt R. LLOYD PRAEGER, B.A., B.E.

[Read March 16, 1902.]

FoK the purpose of czpressing the horizontal range of flowering
I^ts in Great Britain, H. 0. Watson^ has employed eight ** Types
of Distribution," which he has named and defined as follows : —

1. British type— species widely spread through S.Hf.N. Britain.

2. English type — species chiefly seen in S. or S.M. Britain.

3. Scottish type — ^species chiefly seen in N. or N.M. Britain.
Intermediate type — species chiefly seen in Mid Britain.

4. Highland type-— species chiefly seen about the mountains.

5. Germanic type — species chiefly seen in East England.

6. Atlantic type — species chiefly seen in West England.
Local species, restricted to single or few provinces.

Watson is careful to state that in the use of the names for these
types he does not make any suggestion regarding the centre of dis-
persal or route of migration of the plant-groups which they represent ;
be uses them simply to express facts of present distribution.

Since range in latitude corresponds phytologically to range in
altitude, it will be seen that the first five of these divisions are, to.

^CfUlt BriiMHiga, i. 43 (1847), xy. 409 (1859), and Compendium of the CybeU
iriummem, 23 (1868-70).

a.I.A. YBOC., TOL. Tin., SBC. B.] B

2 Proceedings of the Royal Irish Academy.

a conBiderable extent, a grouping according to one and the same
standard — ^the latitudinal, or vertical, range of the species, whichever
term we prefer to employ. The vertical limit of plants is usually
more defined than their latitudinal limit. A small range in altitude
corresponds to a comparatively large range in latitude, and the limit
of latitudinal range is often obscured by local conditions. Thus, while
the vertical limit of a plant may often be represented by a strai^t
line, the latitudinal limit frequently resembles rather an indented
coast-line, with promontories, bays, and outiying islands. The fifth
and sixth '' Types of Distribution " are of a different character, and
represent eastern and western range in England. The focus of the
'^ Qermanic " plants is in the south-eastern coimties, of the '* Atlantic ''
group in the south-western.

In books and papers dealing with the vegetation of Ireland, whether
of the whole coimtry or of selected districts, it has been usual to
analyse the flora according to these types of distribution, which were
chosen with reference to Great Britahi only, and without reference to
Ireland.^ The distribution of plants in Ireland was not, indeed, in
Watson's time sufficientiy worked out to allow of its being ranged
alongside Great Britain. Kow that the distribution of species in this
country is at least as well known as in Great Britain, it is possible to
institute comparisons and analyses. I propose, in the first place, to
review the distribution in Ireland of Watson's Types, and from that
to pass on to the consideration of natural Types of Distribution in
Ireland as revealed by a study of the flora of this country.

The most convenient way of expressing the facts to be dealt with
is by means of a series of statistical maps, constructed according to a
imiform plan. As regards the basis of these maps, the lists of the
Watsonian plant-groups are compiled from the '' Compendium of the
Cybele Britannica," which, though now over thirty years old, is the
latest pronoimcement on the subject. In his works, each species is
referred by Watson either to a definite type, such as '^ English," or to
a qualified type, as ** English-Germanic," which signifies that the species
belongs to the former type, with tendencies towards the latter. It
should be noted in passing that these qualified tjpes approximate
nearly to each other, so that, as Watson admits, the reference of a
species to one such type or its counterpart may become arbitrary.

1 This fact was recognized in the first edition of CyheU Hihemioa by the consia-
tent use of the term **type in Great Britain," instead of merely ''tjrpe*'; an
important distinction which has not been retained in the second edition*

Praeoee— On Tf^8 of LiBtiibution in the Irish Flora. 3

Between " English- Atlantic " and '' Atlantic-English" no wide diffe-
rence exists, and it can be readily imagined that the distribution of a
species may place it between the two. Especially in sach cases, the
finding of a plant in a couple of new counties might turn the scale.
3Cany such discoyeries have been made since Watson defined the '' type
of distribution " of each British plant in 1870, yet the '' types " hsve
not been revised. Therefore, for our purposes, it will be better to use
pore types only, where possible.

The maps are constructed, according to a uniform plan, in five
depths of shading. The units of area employed are the forty county-
diTiaons of '' Irish Topographical Botany " and the standard used as a
list of the Irish flora, and its distribution, is taken from the same work,
posted ap to date. For the construction of the maps, the distribution
in the forty divisions of the component species of each group has been
talmlated. In order to balance the statistics, and maintain their
Hientiflc integrity, sub-species {i,e. those printed in italics in '' Irish
Topographical Botany") are not reckoned, nor records of doubtful
Talne (t.«. those of which the accuracy is doubted, or to which the
marks signifying *' probably introduced" or '^certainly introduced"
are applied). From the totals thus obtained for the county-divisions,
giving the number of plants of the type present in each, the lowest and
bighest figures are taken, and the intervening space divided into five
equal portions. The forty totals are grouped according to these five
portions, and the map shaded accordingly in the order : —

(1) white, (2)= (8)=|=H=, (4);li|t|i. (5) black.

An example will make the process dear. Say we find that of the
pUnt-gronp in question the maximum nxmiber of species occurring
in anyone of the county-divisions is 30, and the minimnm 11. Divid-
ing this difference into five equal portions, we get as our series : —

11 to 14, 15 to 18, 19 to 22, 23 to 26, 27 to 30
white — =1=1- i|i|i black.

It 18 to be distinctly understood that the shading of each division
reptiaents the number of plants of the group which occur in it, not
their HdrihUum in the division. For instance, in the Highland type
map, the actnal distribution of the species in many divisions would
show as little more than a few dots on the map ; instead of which
«a even shading is spread over the whole of each division according to
the number of Highland plants growing within it.

4 Proceedings of the Boyal Irish Academy.

1 . Bbhibh Ttfe : " Species widely spread through S.lf .N. Britain.'^
— ^To this type belongs the mass of our oommoiL plants. From the
definition of the type we should expect to find plants of this group
largely represented and widely spread in Ireland. According to our
standard list, the nnmber of Irish plants of purely British type is 377;
the list for Great Britain adds but a very few to this nnmber — ^namely,
one species, Avena pratensisy unknown in Ireland, and two or three
others whose claims to native rank in Ireland are doubtful or inad-
missible. If we include in the list all plants of qualified British type^
the number of Irish absentees is increased to eight, which will be
found listed in " Cybele Hibemica/' p. xlii ; most of these are of
British-EngUsh type, or haye, in other words, a southern tendency in
G^at Britain.

As examples of typical "British" plants, Watson selects the
following : —

Ahms glatinosft. Cnicas palustiis.

Betula alba. Plantago lanceolata.

Corylufl Ayellana. Polygonum ayiculare.

Xxmicera Periclyinenum. Urtica dioica.

Hedera Helix. Juncos effusuB.

Calluna Tulgaris. Gazex panicea.

Baaunculns acris. Poa annoa.

Censtium triyiale. Featuca ovina.

Trifolium repena. Anthozanthuin odoratum.

Stellaria media. Pteris Aquilina.

Lotus corniculatua. Polypodium vulgare.
BelliB perexmiB,

All of these occur in every Irish coxmty-division.

Of the distribution of the 377 typical British type plants in Ireland,
I have made a somewhat minute analysis, to discover if ^e varying
conditions of soil and climate produce any increase or diminution in
their numbers in north or south, east or west. There is no indication
of the kind. It appears that the number of species present in the
forty divisions ranges from 85 to 99 per cent, of the Irish total — a very
small amount of variation. On mapping their distribution, the result
is found to correspond so remarkably with map lY. of ''Irish Topo-
graphical Botany," which shows the extent to which the flora of each
division is at present known, that there can be no doubt that, in the
majority of cases, the absences are only apparent, and that, as a group,.
these 377 species will eventually prove to be as evenly spread in
Ireland as in Great Britain. The only portions of the country to>
which a comparison within such narrow limits can be safely applied

P&ABGBR— Oft Ti/pes of Distribution in the Itish JFIora. 8

«re fhose wbich have 1)6en practically thoroagMy explored; namely,
Keny and Ck>rk| Dnblin and Wicklow, Donegal, and the North-east.
Ab these areas are widely scattered, the figures may be worth compar-
ing, especially since the diTisions in question are all maritime, which
renders them more comparable :

Antrim,

.. 376

North Keiry, • •

. 861

Bovn,

.. 372

West Cork,

.. 369

Deny,

.. 369

Wicklow,

. 369

But Donegal, ..

.. 863

South Kerry,

. 366

West Donegal, ..

.. 362

Dublin,

. 366

EtttCoik,

.. 362

Mid Cork,

. 361

The smallest number on record is 297, in Honaghan — ^the least
▼orked of all the Irish divisions.

It should be noted, howeyer, that the whole of these British type
plants are not widely spread in Ireland. There are a few notable
exoeptiona. One, as already mentioned, is absent from this country.
A few others are very rare therein, as exemplified below, where the
first number shows in how many of the British 112 yice-counties each
species occurs, the second number in how many of the Irish 40 :

Adoxa Koschatellina,
Ulmua montana,
lleretnialia perennia,
Jun^ema eooununis,
Poa neoioralia,
FSlnlaria globulifera.

Great Britain.
91, or 81 per cei
98 „ 88
107 „ 96
77 „ 69
90

per cent.

Ireland.
1, or 2) per cent

„ 80
69 „ 63

., 27

12 „ 30

16 „ 40

6 „ 12

2. EveusH Ttpb : *^ Species chiefly seen in S. or 8.M. Britain."
^Theee are the southern plants of Qreat Britain, having their head-
quarters in the south of I^gland. They are largely lowland species
&Touiing light soils.

As typical '* "Rngliab " plants, Watson selects the following : —

nita^^^^^ catharticuf •
Uleinaniis.
Tanroa eomnranii.
Bryonia dioica.
HottoDia palnatria.
Chlora peifoHata.
oiaon Amonram.

Xiinaria Elatine.
Ranunculus parriflonu.
T<aininin Galeobdolon.
Hordeum pratenae.
AlopecuruB agreatis.
Ceterach oi&cinaruni.

Of these, six are nnknown in Ireland in the native state ; of the rest,
Ckhrm uid C^t^'o^ are the only ones which are not rare and local.

6 Proceedings of the Boyal Irish Academy.

Here again the number of the group in Ireland is bo large — dose
on 400 altogether — that we may restrict our analysis to those plants
which are of purely English type. Of such plants, 135 are included,
according to our standard list, in the Irish flora. But of these, no less
than 44, or 33 per cent., are reckoned in Ireland as possibly, probably,
or certainly introduced. Here, in fact, we come upon the home of
the large section of our vegetation which owes its presence in the
country to the operations of man — the weeds of cultivation, and light-
soil plants. And while 44 represents the number of doubtfully native
plants of this type which have established themselves in Ireland, the
number which occur in this country more or less sporadically would
largely increase this figure. For our present purpose, however, we
are concerned only with the balance of 91 species which are reckoned
indigenous in Ireland. The maximum in any county-division is 63, or
69 per cent, of the Irish list, in Dublin ; the TniniTmiTn 18, or 20 per
cent., in Monaghan. A map constructed according to the principle
laid down gives the following result : —

Fio. I.— DIstribation of ** English " plants.

This map shows clearly how the English type plants reach their
maximum along the east coast in Dublin, Wicklow, and Wexford, as
we shotdd expect them to do from considerations of position, soil and
climate. Their great abundance in Clare is a remarkable point, to

Fraegbr — On Types of Distribution in the Irish Flora. 7

which we shall presently retnm. Por the rest, excepting their
frequencj in Antrim, they decrease from S.E. to N.W., reaching their
miniminn in Sligo, Leitrim, Monaghan, and Tyrone. It may be
pointed ont that the group embraces a number of maritime plants, and
therefore the maritime divisions are necessarily slightly richer in
species than divisions situated inland.

3. Scottish Ttps : '< Species chiefly seen in N. orN. M. Britain."
*This tyjpe is the opposite of the last. With headquarters well up in
Scotland, the species range southward in diminishing numbers. They
aie the northern plants of Britain.

As characteristic examples of the Scottish type Watson cites —

Smpetenm nigram. TrientaliB eoxopflsa.

Sabus aazatilis. Ligusticuin sootioiun.

Trollius earopflBUfl. Hertenaa maritiina.
Qemiiam ■ylTatieum.

Of these, jMentalis is absent from Ireland ; of the remaining six,
three are confined to the north. This and the succeeding groups being
much smaller than the British and English types, we will call in the
foil strength of the group, whether the species be of pure or qualified
type, in order to strengthen the features indicated by their distribution.
The Scottiah type in Ireland is represented by 50 species, or less than
haU of the British total—

Scottish.

TioUins euTopflBUB. Mertexuia maritima.

Tiok latea. Salix pentandia.

Gcmiiiim ■ylyaticum. Habenaria albida.

Phmos PadoB. Potamogeton filiformiB.
CaQitriche antunmalis. nitens.

Ihoaera anglica. Scirpus nifus.

ligutienm aooticiun. Carez limoBa.

PjTola media. Featuca sylyatica.

•eeimda. Polypodium Dryopteria.

Melampynim sylTaticum. Equisetum umbroaum.
Ajnga pyrmmidalis. Tariegatam.

Lantmii i&tennadiuni.

Seottish'British.

ThaEetram minua. Oaleopais Ternoolor.

SagXBa sabalata. Carex dioioa.
Tieia sjlTatica. filifoimis.

PaniMia paluatris. Elymua europnuB.

Aatnnaria dimca. Polypodium Phegopteria.

PyraU aunor. EqaiBetum hyemale.
Piagmciila Tvlgana.

Proceedings of the Royal Irish Academy.

Rttbus aazatilifl.
Saxifraga hypnoides.
Gircea alpina.
CrepiB paladosa.

Arenaria verna*
Yicia Orobus.

Orobanche rubia.

Scottish' ffiyhland,

liobelia Dortxnanna.
Empetrum nigrum.
Salix phylicifolia.
Listera cordata.

Seottish'Intsrmediate,

Saxifraga Hirculus.
Potamogeton prolongus.

Scottish' A tlantic,

Eriocaulon septangulare.

This is a purely native group ; not one of them is under any
suspicion of introduction. Host of them are plants of thoroughly
wild ground — hillfl, heaths, glens, lakes, and bogs.

The maximum in any division is 43 (or 86 per cent.) in Antrim, the
minimum 5 (or 10 per cent.) in East Cork. Constructing our map we
get the following : —

Fio. 2.— Distribution of " Scottish " planu.

The result is striking. The Scottish type plants are concentrated
in the north, as we should expect. Thence they range down the coast
on either side : but while on the east they greatly diminish south of

Pkaeoer^ Oi» Types of Diatribution in the Irish Flora. 9

Go. Down, on the west coast they maintain their sway as far south as
dare, or even Korth Kerry. Inland everywhere l^ey are few in
number, Westmeath alox)p falling barely witiiin the third grade. The
latter fkct cannot be accounted for by the absence of saline conditions
in central Ireland, as out of the 50 species only four — Ligwticum^
Mtrientia^ Seirpue rufiu^ and Slfftnus — are maritime plants. Kor is
it due to an avoiding of the Central Plain on account of its limestone
expanses, for the limestones of Sligo, Leitrim, and Clare yield them
in abundance, and they attain their minimum in the soutii-eaBt and
south, where limestone is only very locally developed. Neither does
the distribution of hilly ground satisfactorily account for their range,
which 18 apparently due to climatic conditions as yet imperfectly
midentood.

4. HioHLAKn Ttps : " Species chiefly seen about the mountains."
—As H. C. Watson points out, the more characteristic members of this
gitmp might be better called Arctic Type, as they consist of hig^
narthem species, brought into our latitudes by the elevation of the
land into mountains. This group occupies the northern end of the
series of four latitudinal types — English, Intermediate (a small and
indefinite group), Scottish, Highland. Its headquarters are on the
b2(^ Scotch mountains and in the extreme north of that country.
The list of Highland type plants in Ireland is as follows : —

ThaHetrum alpinum. Hieracium strictum.
Suholaria aquadca. gothieum.

Diaba incana. corymbosum.

Azabia petnea. Arctoataphyloe Uva-uni.

Sflaoe acauUa. Vaocinium VitiB-Idaea.

Dryaa oetopetala. Polygonum yiyiparum.

Raima ChanuDmorua. Oxyria digyna.

AldMmilla alpina. Salix herbacea.

EpDolniiiii alnnefoUum. Juxiiperus nana.

Bednm Rhodiola. Carex paudflora.
8«xxiraga tteUaria. rigida.

niTalia. aquatilia.

Aira alpina.

oppoaitifoUa. Sealeria cenilea.

Galium boreale. Poa alpina.

Sauararaa alpina. Cryptogramme crispa.

ffiendam aeneacena. Aapidium Lonchitis.

angjieum. Aaplenium viride.

irienm. Lycopodium alpinum.

prenanthoides. Selaginella selaginoides.

crocaitam. laoetea lanistris.

10 Froceediuf/s of the Royal JmA Academy.

All of these are classed by Watson as of purely Highland type except
Suhularia and Vaccinium Vitis-Idaa^ which he ranks as Highland-
Scottish, and Sesl&ria^ which goes as Highland-Intermediate. Plants
of this group are thinly spread in Ireland, as is to be expected from
the conformation of the country ; but taking into account the charac*
ter and altitude of the mountein-groups, the total does not fall much
below what might be expected. With the vertical distribution of the
species, the present paper is not concerned; but some interesting
points become apparent from the mapping of their horizontal range.
Here the maximum is 29 (or 69 per cent.) in West Donegal, the
minimum 0 in Mid and East Cork. (Fig. 3.)

Being essentially a mountain group, it is desirable to contrast their
distribution with that of high land in Ireland— say of oyer 1000 feet
elevation. The actual distribution of land of over 1000 feet in
Ireland is shown in fig. 4.

But for purposes of comparison, it may be well to construct a
graduated map on the same principle as the floral maps (fig. 5). A
difficulty is encountered here, for the amount of high land in two of
the divisions — ^Wicklow and South Kerry — so far surpasses that which
is found elsewhere, that were an evenly graduated scale employed, the
varying elevation of the rest of the country woidd not be brought out.
We therefore employ the following scale : —

0-25 square miles over 1000 feet elevation, white
26- 60 „ „ „ =

61- 75 „ „ „ =:|i:|-

76-100 „ „ „ i|i|i

200-2-25 yj 9f „ black

For comparison, I add the distribution of ground over 2000 feet
elevation (fig. 6), according to the scale —

0 square mile Gver 2000 feet, white

up to 1 „ „ =

Ito 3 „ „ -1=1=

8 to 8 „ „ ±\i\±

20 to 25 „ „ black

If we contrast these two maps with map 8, we have the materials
for comparing the distribution of << Highland " plants in Ireland with
that of high ground. In area of high ground, whether ^the 1000 foot

FRAEOER^On Types of DUtHbution in the Irish Flora. 11

Fhl j.~Duti>biition of " Highland " plants. Pio. 4.— Actual diBtribution of land over 1000 ft.

'^ $^I>iitzibatio& of land over xooo feet.

Fio. 6.— Distribution of land over 2000 feet.

1 2 Proceedings of the Royal Irish Academy.

or 2000 foot contour line be taken, Wicklow and South Kerry far
outstrip any other portion of the country : yet both fall below the
mayimum of alpine plants, which is carried off by West Galway, and
the two divisions of Donegal. The distribution of highland ground
is in fact no criterion of the distribution of the highland flora. The
amount of high ground in western Ireland (Kerry to Donegal indusiTe)
is about the same as that in eastern Ireland: but the oollectiYe
Highland flora of the western half is double that of the eastern.
If we want to get an analogue of the distribution of the alpine flora
we will turn to the Scottish type map (fig. 2), and will at once see
many points of resemblance* The distribution of these two allied
groups is distinctly similar, the bulk of the species which compose
them inhabiting cliiefly the hilly grounds of the north and west, and
being but sparsely spi^ over the east and south-east, and also of
course over the centre.

Those "Highland" plants which occur in the east, as on the
Moume and Wicklow mountain ranges, are usually truly alpine in
habitat ; in the west a change of conditions is clearly shown by the
frequent descent of alpines to sea-level, and by the ascent of maritime
plants to high elevations (such as CoehUaria officinalis, Silent maritima^
Armeria maritima^ Phntago maritima) which are absent on the eastern
mountains.

Leaving for the present the distribution in Ireland of Watson's
latitudinal types, we must briefly consider those which are by their
definition longitudinal — ^namely, the Germanic and Atlantic types.

5. Gekmanic Type : '* Plants chiefly seen in East England." — ^This
is a special group of English type plants, segregated and separately
classed on account of their marked aggregation towards the south-east.
As Watson points out, the Cretaceous deposits lie almost exclusively
in the eastern and south-eastern provinces of England, so that the
*^ chalk plants " fall within this type.

As examples of the Germanic type Watson names

Frankenia iBdvis. PiiUoaria vulgaris.

Anemone Pulsatilla. Lactuca Scariola.

Beseda lutea. Atriplex pedunculate.

Silene conica. Aceras anthiopophoia.

noctiflora. Bpartina stricte.
Pimpinella mag;na.

Of these, only one, Pimpinella magna, is certainly native in Ireland;
of the rest, two alone. Reseda lutea and Silene noctiflora, are included
in the Irish flora, marked as doubtfully indigenous.

Praeger — On Types of Distribution in the Irish Flora. 13

Being the f artliest remoyed from Ireland as regards not only actual
dutance, but soil and climate, it is to be expected that this should be
the type least nmneronsly represented in this country, and such is the
esse. Out of 102 " Germanic" plants in England, only thirteen are
enumerated in the Irish flora, and four of these cannot be reckoned in
the certainly indigenous list The list is as follows; the extent of
lange of the members of the group in Ireland is so yariable, that after
each speciea the number of divisions in which it is known to occur is
added, in order to illustrate this feature.

Oermanie.

^Crepis bienniB, 13. Scirpus triqueter, 2.

t Unzacifolia, 18. Glyceria Borreri, 2.

Pblyganum mite, 4.

OermaniC'BritiBh.
Attngalns Hjpoglottii, 1. ^Senecio yiaooeufl, 1.

JGaUum erectum, 8. Teucrium Scordium, 7.

Hjpopitfaya mnltiflora, 6. Orchis pyramidalifl, 38.

liinoeeUa aquatica, 2. Bxomiu erectus, 9.

This is, in Ireland, distinctly a caldcole group of plants. All but
twf^^Crepte Uennis and Polygonum mite — are confined to limestone
districts or to limy sea-sands. Leaving out of account the two
''certainly introduced " species, Cr&pis biennis and Senedo viseosus, as
their range is devoid of phyto-geographical signiflcance, and giving^
the remaining two dubious natives the benefit of the doubt, the
distribution of the group works out as shown in fig. 7, next page.

Here the maximum is 8 species in Clare, the minimum 0 in Tyrone.
Our scale is 0-1, 2-3, 4-5, 6-7, 8- 9 species. The group is seen to attain
its maximum in Clare, 8.£. Galway, andDublin; while the only divisions
in which more than one species occur are certain counties in which
limestone largely predominates. This result is significant, even though,
vhen dealing with the distribution of so small a number of plants, it
is unwise to lay too great emphasis on present results. The fact ia
that, as a group, the Germanic plants have no place in the Irish flora ;
such stmgglers as have found their way here have a distinctly limestone
nnge.

Proceedings of the Royal Imh Academy.

Fio. 7. — Distribution of " Gennanic " plants.

6. Atlantic Type : " Species chiefly seen in "West England." —
This group has its headquarters in the south-west of England, and is
in this way related to the " Hibernian " and " Lusitanian " groups of
Ireland, since among them are the remnants of the old southern flora
that flourished on the lost south-western shore-line of the British Isles.
Watson's Atlantic type has other components besides these ancient
species, but it is stiU the smallest of his British plant-groups, number-
ing altogether but 62 species.

As typical examples of '< Atlantic " plants, Watson cites —

Sinapis monensis.
Matthiola sinuata.
RaphanuB maiitimus.
Sedum anglicum.
Cot jledon Umbilicus.
Bartsia visoosa.
Pingtiicula luaitanioa.

Euphorbia portlandica.
Scirpus Sayii.
Sibthorpia europsea.
Erica ciliaris.
Polycarpon tetrapb jllum.
Adiantum CapiUus-Veneria.
Cynodon Dactylon.

Of these, four are unknown in Ireland ; three are confined to the
south and west ; most of the others have a wide range in this country.

Pbaboer— On Tia>es of Distribution in the Iriah Flora. 16

Of the 62 " Atlantio" plante occurring in Britain, Ireland poflseeses
33, as follows : —

Atlantic,

MattbioJa nnuata.
B^»haniit maritimuB.
Tiola CurtiBii.
LBTBtera arborea.
Eiodium motchatum.
Carum TerticUlatuin.
Crithn*iiiii maritixiiujii*
Eubia peregrina.
Wahlenbergia hederacea.
Bartaia TtMOM.

Sibthorpia europaBa.
Euphorbia Peplia.

portlandica.
AaparaguB officinalis.
Ehynchospoia f luca.
ScirpuB Sayii.
Asplenium lanoeolatum.
Adiantum CapiUuB-Veneiis.
HymenophyUum tonbridgenfle.

Atlantic-British,
Hypericum Androoemum. Cotyledon Umbilicus.

Lastrea nmnla.

AtlantiC'Unglish,

linom angdstifolium.
Hypericum elodee.
Erodiom maritimum.
Sedom anglioum.

Pingmcala luaitanica.

Inula crithmoides.
Statioe occidentalis.
Euphorbia Paralias.

Atlantic- Scottish.

ScillaTema.

Atlantic-Highland.
Hymenophyllum unilatarale.

Atlantic-Intermediate,
Meconopsia cambrica.

Two other members of the group, S&nehicra didyma and Brmnue
»*dritiiuiij are omitted as probable introductiona into Ireland. It
▼ill it (mee be remarked that a large number of these — a full dosen —
are maritime plants. The rest are largely plants of rocks and bogs
the gnmp is characteristic of thoroughly wild ground.

Froceedinga of the Royal Irish Academy.

The maximum in any Irish oonnty-diyision is 24 (or 72 per cent,
of the Irish list) in Sonth Kerry, Vest Cork, and Vaterford, the
minimum 2 (or 6 per cent.) in Eildare, Our map works out very
prettily : —

Fio. 8.— Distribution of *' Atlantic " plants.

The group is seen to be essentially coastal — partly on account of
the plentiful sprinkling of maritime species, partly because the
remainder are largely plants of the rough country which often
accompanies the older rocks ; such country as is found in the home of
the group in Devon, Cornwall, and Wales, and in Ireland round a
great portion of the seaboard. The group also shows an increase south*
ward, and attains its full luxuriance round the shores of the southern
half of Ireland.

Before proceeding to briefly sum up the features brought out b j
the foregoing series of maps, it will be well to consider one important
factor in plant-distribution. Apart from climate, the most potent
influence affecting the flora is undoubtedly soil, and it is the
presence or absence of lime in soils that most affects the vegetation
which they support. Ireland consists, roughly speaking, of a great
plain of Carboniferous limestone occupying the centre, with more

Prabger — On Types of Distribution in the Irish Flora. 17

flerated and broken ground formed of non-calcareous rocks around
the margin. The actual distribution of limestone is shown in black
on the following map (fig. 9). Let us compare this with the distri-
bution of lime-loving and lime-avoiding plants.

Fig. 9.— Actual distribution of Carboaiferous HmMtonc.

The data are at hand. Mr. Colgan has paid much attention to
these BoQ-ielations in Ireland, and has compared his results with those
obtained in France ; in the second edition of ** Cybele Hibernica " he
'Mirat^g the calcicole and calcifuge ^ecies, using three grades
(A, B, C) for each, according to their degree of preference for a
limy soil or soil free from lime, '' A " indicating the most markoi
preference in either case.

The calcicole plants of "Cybele** range as follows: —
Caleicole A.

GcmiliiiD laddam.
Poteotilk fruticoea
Galiom lylTestre.
CarUiia Tulgarii.
Gcntiana Tenia.

&.IJk. nOC., TOL. Tin., SBC. fi.]

Calamintha officinalis.
Galeopsia Ladanum.
Orchis pyramidalis.
Ophrys apifera.
Sesleria cserulea.

ProceedingB of the Royal Irish Acadetny.

CdleicoU B.

Aquilegia yulgaiis.
Reseda Luteola.
Hypericum perforatum.
Anthyllifl Yulneraria.
Poterium Sanguisorba.
Pimpinella Sazifiaga.

magna.
Bubia peregrina.
Galium boreale.
Aapenila cynanchica.
Erigeron acre.
TuBsilago Farfara.
CaiduuB nutaxiB.
Centaurea Scabioea.
Crepis tarazacifolia.
Leontodon hirtua.

Leontodon hupiduB.
Chlora perfoliata.
Gentiaoa Amarella.
Lithospermum officinale.
Verbascum Thapsus.
Salvia Verbenaca.
Origanimi vulgare.
Opbrys muscifera.
Spiranthes autumnalis.
Juncus glaucuB.
Carez diyulsa.

glauca.
Trisetum flayeacens.
Arena pubeacens.
Adiantum CapiUuB-Veneru.
Ceteracb officinarum.

Caleieoh C.

Arabia hirsuta.
Siaymbrium AUiaria.
Viola birta.
Ceraatium arrenae.
Euonymua europaaua.
Antennaiia dioica.

Pulicaria dyaenterica.
Conyolyulua arvenaia.
Habenaria conopaea.
Carez muricata.
Featuca rigida.

Total 53. Our maximum is 50 in Clare, minimum 20 in Tyrone.
The map works out as shown in fig. 10, opposite page.

The result is somewhat unexpected. The calcicole group has its
headquarters, not in the Limestone Plain, but in the west, reaching
its maximum in Clare, S.E. Oalway, and Limerick. Thence it
follows the edge of the Hmestone northwards, so that although Weet
Oalway has nothing more than a strip of limestone along its eastern
edge, this division is high in the scale, along with N.E. Oalway and
E. Mayo. A prevalence of calcicole plants appears also in E. Cork,
Kilkenny, Eildare, and Dublin, none of which occupy the first rank
as regards area of limestone. Elsewhere the distribution of the
group is what we should expect : the minimum is reached in Ulster,
where, on the Silurian area and elsewhere, but few calcicole plants
maintain an existence. The reason for the great development of the
calcicole group in the west is not far to seek ; it lies in the occunence
of bare limestone pavements in the Burren area, in Limerick, and
around the great lakes of Corrib and Mask. It is the presence of

P&ABOEB— On Types of Distribution in the Irish Flora. 19

lire limestone rock over large areas that produces the calcicole flora
in its full development. The tough limestone drift which covers
the rock over the greater portion of the Central Plain and eastern
counties, may, as a matter of fact, have all the Hme washed out of its
sorhce layers, and yield a non-calcareous soil.

'^^^

Fig. 10. — Distributioa of calcicole plants.

To turn to the calcifuge group. The plants which show a
preference for a non-calcareous eoil are more numerous than ^those
which prefer lime; so that the species classed as calcifuge A in
" Cybele " almost equal in number the whole calcicole group and
wiU alone suffice for our purpose. They are as follows : —

Cakijuge A.

CorrdaliB daTiculata.
Tiolii palustria.
Poly^U serpyllacea*
MontU ioatana.
XUtine hexandra.
Hypericam elodes.
Bftdiola linoides.
CTtims scoparius.
XnexGallii.

Latl^TUf macTorrhizus.
Saziban stellaiia.
imionMa.

Broeera rotundifolia.

anglica.

intermedia.
Peplis Portula.
Qalium saxatile.
Gnaphalium iiliginosam.
Senecio sylvaticus.
Lobelia Dortmanna.
Jasione montana.
Wahlenbergia hederacea.
Yaccinium Vitis-Idaa.

MyrtilluB.

(72

Proceedings of the Royal Irish Academy.

Calcifuge A — continued.

Calluna Yulgaris.
Erica Tetialix.

cinerea.
Microcala filifonniB.
Digitalis purpurea.
Scutellaria minor.
Polygonum Hydropiper.
Bumex Acetosella.
Narthecium ossifragum.
JuncuB squarrosus.

Bupinus.

Potamogeton polygonilolinB.

ScirpUS C8B8pit06US.

fluitans.
Carex pilulifera.

binervis.
Deschampsia flexuoea.
NarduB stricta.
Blechnum Spicant.
Athyrium Filix-faemina.
Lastrea dilatata.
Osmunda regalia.

Total 46. The group reaches its maximum in Kerry and West
Cork with 44 species, minimum in Westmeath with 26 species.

» ^eoHnED

— ssgiii ;

?sspi

Fig. XI.— Distribution of calcifage plan

It will be noted that calcifuge plants are^ more widely spread
than calcicole. Not only is the scale somewhat higher than that
of the calcicole group, being 56 to 95 per cent, of the group, as
against 38 to 95 per cent., but the number of high percentages is
much greater, eyen though we are dealing with only the most
strongly calcifuge species. The reason is clearly to be seen in the
fact that non-calcareous soils are to be found in limestone areas, both
on account of the washing-out process referred to, and by reason of

Prabobr — On Types of Distribution in the Irish Flora. HI

aocmnalation of yegetable matter, in woods, and mncli more in bogs,
▼liich are largely developed in the Irish limestone districts. On the
other hand, no natural process is at work in this country producing
a calcareous soil in districts devoid of limestone, except on coastal
sands, where shelly acenmnlations may have a distinct effect on the
flora.

Glancing at the map, it will be seen by comparison with map 8
that, as contrasted with the distribution of limestone, the range of
the caldfuge flora is quite normal. It reaches its maximum on the
old non-calcareous rocks that stand grouped around the coast ; its
nrinimum in the Limestone Plain. The absolute minimum is reached
in Westmeath, the only division in Ireland which can be said to be
altogether under the sway of the limestone. It will be noted that in
Clare, where the calcicole flora attains its greatest development, the
caldfuge flora is also at high-water mark; but in the rich and
remarkable flora of that varied county, almost every group, whether
Eagiiah or Scottish, Atlantic or Germanic, calcicole or calcifuge,
attains or approaches its maximum!

Considering generally the series of maps showing the range in
Irehmd of the ''types" of Great Britain, it will be seen that we
hare really three topographical groups to deal with : —

(1) Ehsubh and Gebxakic, the latter a peculiar and intensified
section of the former. A southern group, often light-soil and often
ealdoole in their proclivities. The Germanic plants represent the
zoophile and thermophile element in the flora of England, and are
eoiigregated where a comparatively continental climate produces* hot
and dry summers. In Ireland these groups are concentrated along
the east and south-east coasts, where position, soil, and climate
apparently account for their predominance ; and in the Clare district,
vbere the warm dry limestone pavements probably form the attraction.

In referring the paucity of '' Germanic " plants in Ireland to the
betaking down of the Irish-English land-connection prior to that of
the Engliah-Continental, the editors of '' Cybele Hibeniica," ed. n.,
remark (p. xHii) : — ** The advance guard of aggressive species, the
British type and a large section of the English type, had time to push
westward into Ireland before its eastward land-connections were
broken down ; but the rear-guard of more slowly spreading species
ioond their westward progress checked by the^land subsidence which
created the Irish Sea. The mass of this rear-g^iard was probably
ionned of the Qermanic type plants, a group so little aggressive in

22 Proceedings of the Royal Irish Academy.

character that it seems to have been quite unable to push its way, as
a whole, across England, in the face of the more hardy settlers who
had gone before and occupied the ground." I do not altogether follow
this explanation. Bather than compare the'stream of plant-migration
to the march of an army, of which the main body does the fighting, and
the rearguard has merely to follow across ground cleared for its
progress, ought we not to choose as a simile the spread of an
empire, which enlarges its boundaries without in any way relaxing
its hold oyer the ground already won. The plant-army had to
conquer a presumably weaker flora which was in possession of the
ground ; but the rear-guaid had to oust the conquerors ! It is a fair
assumption that each successiye wave of migration was composed of
species more hardy and aggressive — better fitted for the struggle for
existence — ^than those which preceded it; otherwise it would not
advance. Then, is the ''Germanic " group composed of species '' more
slowly spreading " and ' ' so little aggressive in character " ? An exami-
nation of the list of '' Germanic " plants is not favourable to this view.
Mr. Clement Beid ^ has conveniently summarised the seed-characters
possessed by British plants which assist them in migration, and has
pointed out that capacity for migration consists' largely of the power
of a species to cross *' deserts " — a desert being an area unsuitable to
the plant : it may be water, low ground or high ground, dry soil or
wet soil, limy soil or soil free from lime. We fail to find, among
'' Germanic " plants, any characters which render them inferior to tbe
other groups in power of dispersal. Five of our Irish " Germanic '^
plants are under more or less suspicion of being recent human intro-
ductions: namely, XOdium erectum, ^Crepts hiennU^ \C, tarazacijblta^
*8eneeio viseosus^ ^Bromus ereettu. All but the fourth have spread
widely by natural means, moving freely about the crowded country,
and showing no lack either of aggressiveness or of rapidity of migra-
tion. The " Germanic " plants may have been the rear-guard of the
Post-Glacial migration which provided our islands with the bulk of
their present flora. But if so they fought their way right acrosB
Earope (where many of them have a wide distribution) against the
''British'' and "English" plants that had gone before; while the
mobile remnant that reached Ireland before the breaking-down of the
land-connection marched right across the coimtry, or round its former
margin, and still holds its ground on the very edge of the Atlantic.
The range of Watson's Germanic type in the British Isles appears

^ Origin of the BritiBh Flora, chap. iu.

PfiASOBR — On Types of Distribution in the Irish Flora. 23

to be dne largely to suitability of soil, partly to oonditionB of
climate.

(2) ScoTTiBH and Highlavi). Tbeae are the nortbem plants, the
latter an intensified group of the former. In Ireland they are concen-
tiated in the north, spreading somewhat abundantly down the western
coaaty much more sparsely along the eastern. It should be noted that
the distribution in England and Wales of these plants offers many
points of resemblance to their Irish range, though the species extend
somewhat further southward in the larger island. As in Ireland, the
group spreads far down the west coast of England, much less so down
the eastern, so that, on a rough examination, South Wales appears to
contain as many '' Scottish " plants as the Trent province. Physical
conditionB will suggest themselves in explanation of this in a manner
not applicable to Ireland, where the problem is more difficult of solu-
tion. A line drawn north-eastward from the Bristol channel to the
Wash will cut off, on the northward, most of these plants ; and this line
would appear to correspond well with one in Ireland drawn from the
Shannon mouth to Bundalk Bay.

To account for the greater abundance of alpine plants in the west
than in the eaat of Ireland, the suggestion has been made, in ''Cybelo
Hibemica," ed. n., and elsewhere, that during the Glacial Period the
mantle of ice drove these species downward to the seaboard in the milder
weaty whence, on the retreat of the ice-sheet, they colonized the western
mountains. This appears as likely a hypothesis as can be put forward.
But the similarity of the range of << Highland " and of << Scottish "
ffpedea suggests that at least some of the '' Highland " plants, which
in Ireland are not alpine in range, may have come into Ireland with
the '* Scottish " plants, many of which probably colonized this country
from the north-east. Another point to be remembered is that — ^pre-
sumably on account of greater moisture — the west of Ireland is un-
doubtedly more suited, even at low elevations, to the growth of alpine
plants than the eastern, and the " lowest limit " line, which for many
species almost touches sea-level along the west coast, may, in the east,
pass above the tops of the mountains.

(3) AzLAjmc. In England south-western, and including a consider-
able nomber of maritime plants. This is the hygrophile element of
the XngliBh flora, composed of plants which prefer the equable tempe-
ratme and abundant moisture that pertain to an insular climate. In
Ireland the group is rather southern, distributed in fair proportion
round the southern half of the littoral, but many of the species occur
round the greater part of the Irish coast.

24 Proceedings of the Royal Irish Academy.

Glancing at the maps showing the diBtribntion of calcicole and
calcifuge species, it will be seen that, while the range of the '' ScottiBh,"
'' Highland " and '^ Atlantic " plants corresponds broadly with that of
the calcifuge flora, the distribution of ''English" and ''Germanic"
species offers many points of resemblance to that of the calcicole group;
which facts we should expect to be apparent when we consider the
petrological conditions prevailing in the homes of Watson's various
" types."

So far as I can gather without an elaborate study of the distribu-
tion of the flora of Great Britain as known at present (which would
be outside the scope of the present paper), there is a greater overlap in
northern and southern forms in England than is the case with Ireland.
If we construct isophytic lines to represent the limit of the main body
of the " Scottish " and " English " floras respectively in Great Britain
and in Ireland, they will run somewhat like this : —

Fig. 12.— Isophytic lines in the British flora.
AA. Northern limit of the " English " flora, bb. Southern limit of the " Scottish" flora.

But this statement is only put forward tentatively and as a side-
issue of the subject in hand. The material for a proper study of the

Fraeokr — On Typei of Distribution in the Irish Flora. 25

question is now in existence, though much scattered, and will, I trust,
be some day brought together and analysed. It is an interesting
point.

To come now to the second portion of my paper — the question of
natural geographic plant-groups in the Irish flora. Eollowing on
Watson's lines, an essay has been made to group the native species
according to their present horizontal range, and without reference (in
the first instance) to the environmental or other cause of such distri-
bution. For this purpose, a set of maps was employed, representing
the whole Irish flora, each map showing, by means of a uniform wash
of colour, the range of one species, the data used being those given in
" Irish Topographical Botany " brought up to date. On these maps,

Fig. x3.
CoBtiinioiis nnge of a nattm plant
{Cicufa vfVvca).

Fic. 14.
Discontinuous rang^ of an introduced plant
{Sedum Telepkium).

divisions in which any plant was considered as probably or certainly
introduced were left uncoloured. The set of over eleven hundred
maps was then sorted by eye according to the distribution of the
colour on each. In this way, by making the process as mechanical as
possible, I hoped to determine the natural grouping of the plants, and
to eliminate theoretical considerations. The groups thus obtained
were then critically examined, and the claim of each member to belong
to it considered. This involved questions of relative frequency through-
out the range, and considerations relating to possible introduction in

26 Proceeditiffs of the Royal Irish Academy.

certain divisions. All species ranked as probably or certainly intro-
duced in Ireland were kept apart, as their range could throw but little
light on natural plant groups. It may be remarked that the maps
brought out very clearly the discontinuity of range which marks the
alien flora. Speaking generally, the native lowland plants are charac-
terized by a continuous range between their limits, while the alien
plants frequently exhibit a broken and discontinuous range (figs. 13,
14). Bemarkable exceptions to both rules exist, and will be men-
tioned later on.

In arranging the maps under types of distribution, one difficulty
was quick to make itself felt. This was, that in the natural flora
every gradation exists between any two types of distribution which
we may select. The difficulty was met by using Watson's plan — ^the
only possible one— of modified or intermediate types as already referred
to ; but considering the unsatisfactory nature of such fine distinctions
in a flora not yet fully worked out, the creation (by publication) of
such intermediate types is for the present withheld, and lists will be
given chiefly of those plants whose range is sufficiently characteristic
to allow of their being referred without qualification to one definite
type of distribution.

The grouping of the maps established in the first place two classes:
(a) plants which show no aggregation in any portion of the country;
and (b) plants which show an aggregation or diminution in some
portion of the country.

Class A consists of (1) universal species, i.e. species on record for all
the forty botanical divisions, and showing no marked increase or decrease
in frequency in any direction; (2) species of probably imiversal distribu-
tion, the occasional gaps on the maps being with littie doubt the result
of incomplete knowledge. To sections ( 1 ) and (2) some 360 species may
be referred, or about one-third of the Irish flora. (3) Pollowing these we
have a range of species of diminishing frequency but wide distribution,
the list extending from the border of the '' probably universal " species
down to plants which have only a few widely scattered stations in
Ireland, and which might be separately classed as of local type.
Following the nomenclature of Watson, who gave the name of British
type to all species evenly spread throughout Great Britain (though not
necessarily continuously distributed), we may define the three groups of
the above Class A as of ''Irish'* type so far as the Irish flora is
concerned; but the use of this term, except with the qualification
appended, might mislead; and I prefer to employ, in the present
paper, the term '* General type of distribution," for all species whose

Praegbr— On Tia)es of Distribution in the Irish Flora. 27

distribution shows no marked aggregation in any part of the country.
It does not appear necessary to list those common plants, over 260 in
number, which are at present known to be universal in Ireland —
inhabitants of every botanical division. As might be expected, the
plants of universal distribution in this country are almost all of
Watson's British type likewise ; it will be worth while to note the
exceptions. In the following list the '^ probably universal" species
which are not of British type are added to the ''universal" plants,
and distinguished by an asterisk.

Plants of gbvbbal msrEiBXTnoN iir I&ELA.Nn which aee not of
BninsH TTPB.

British-ISngliih^ 41 species.

Banuneuliu bulbosus. 'Veronica montana.

Barbflzea vulgaris. ^Mentha sativa.

^Reseda Luteola. *Lycopu8 europseus.

^Anoaria trinervia. ^Polygonum lapathifolium.

Halva sylvestrifl. Bumez nemorMUS.

Hyperioum tetrapterum. Euphorbia Feplus.

Trifolium dubium. *Saliz alba.

*Loloa uliginosus. ^Allium ursinum.

Oeum urbanum. Typha latifolia.

Agrimonia Eupatoria. ^Sparganium simplex.

Epilobium parviflonim. *Fotainogetoii crispus.

Ciit9a lutetiaoa. Cares remota.

*£thnn Gynapium. * sylvatica.

Caucalif Anthrisout. hirta.

Sambuem nigra. Brisa media.

Vibomum OpnluB. *Bromu8 giganteui.

Enpatorium canuabinmn. * asper.

Petatites officinalis. Nardus stricta.

Ansgallis arrensis. Scolopendrium vulgare.

Myosotis palustriB. Aspidium angulare.
^Yeraaiea polita.

British'Scottishf 5 species.

Potentina paluttris. Eriophorum vaginatum

Habenaria viridit. Botrychium Lunaria.

Sciipos enspitofus.

British-Highlandy 3 species.

Chiysotplenium oppositifolimn. Lycopodium Selago.

Ysocimiim If yrtillua.

28 Proceedings of the Royal Irish Academy,

JEnylish'British, 11 species.

*Nupliar lateum. Caljstegia sepinm.

Potentilla reptans. *ConYolyulii8 arvensis.

Epilobium hinutum. 'Junciu glaucus.

*Bidens cemua. Garez disticha.

LyBimachia vulgaris. *£qui8etu]n maximum.
*Samolu8 Valerandi.

Seottiah'British, 1 species.
^Antennaria dioica.

AtlantiC'Britiih, 1 species.
Hypericum Androflasmum.

AUantiC'English^ 1 species.
* Cotyledon Umbilicus.

English^ 7 species.

Sagina apetala. Pulicaiia dysenterica.

Euonymus europeus. Arum maculatum.

Pyrus Malus. Geterach offlcinanim.
Apium nodiflorum.

SeottiBh-Highianiy 1 species.
•Grepis paludosa.

From this analysis, we find that the ''Universal" plants of Ireland
which are not equally widespread in Great Britain are generally in
the latter island of ratlier southern range (British-English and English-
British) ; and a few are distinctiy southern (English) ; this result we
might expect from a comparison of the range of latitude of the two
islands. The most noteworthy feature of the few other species in the
lists is the abundance in Ireland of the distinctiy northern (Scottish-
Highland) Crepis paludosa,

Cnieus pratensis has an anomalous range. Though recorded from
every Irish division, it is rare in the east and increases westward,
becoming abundant in the west and north ; while in Great Britain it is
of characteristic English type, being unknown. north of Yorkshire.
Its range in Ireland is the reverse of that of most English type
plants — see fig. i. above.

In the case of a few other " Universal" plants, their distribution
over the country is not even, but it is yet not suficientiy accentuated

Prasqbr — On Types of Distribution in the Irish Flora. 29

t4) render their separation necessary. These are mostly calcicole or
caldfuge species, which increase in abundance in the centre or round
the margin according to their proclivities.

As regards the plants of General type which are not universal.
In some cases the gaps in range are with little doubt only apparent ;
this applies particularly to critical species, such as the two whose
distribution is illustrated below (figs. 15, 16).

Fic. 15. — Viola Reick^HhackiaHH,

Fig. 16 — RanuHcuius heterophyllua.

But many other cases are instances of genuine discontinuous
distribution; four good examples are illustrated below, all being
well-known and easily recognized species (figs. 17-20).

x'.^'Hypifpithyt muitifiora.

Fig. i%,*-Limana repent*

Proceedings of the Royal Irish Academy.

Fig. iq.—Siachys Beionica, Fio. ao. — Cepkalanthera enstfolia.

The following is a list of the species of General type of distribution
other than the universal species, plants of intermediate or doubtful
type being omitted; for comparison, the Watsonian type in Great
Britain is appended (in initial) to each species^ : —

Name.

Irish
Census.

Watsonian

type.

Name.

Irish
Census.

Watsonian
type.

Anemone nemorosa,

Silene Cucubalus,

Banunculus trichopbylluB, 30

Stellaria Holostea,

Drouetii,

—

Arenaria trinerria,

heterophylliw,

—

serpyllifolia,

Auricomus,

Sagina nodosa,

Nuphar luteum,

£B

Montia fontana,

Fumaria pallidiflora,

Ilex Aquifolium,

confusa,

—

Trifolium medium,

Boreei,

—

procumbens,

muralis,

—

Anthyllifl Vulneraria,

officinalis.

Lathyrua macrorrhizus,

Nasturtium eylvestre,

Fotentilla procumbens,

Cardamine flexuosa,

—

Bosa apinosissima,

Lepidium campestre,

involuta,

Beaeda Luteola,

mollis.

Viola palustris,

Cotyledon Umbilicus,

canina,

Myiiophyllum spicatum,

Beichenbachiana,

—

Callitricbe atagnalis.

—

arvensiB,

obtusangula.

—

Polygala vulgaris,

Apium inundatum,

serpyllaeea,

—

JEtbuaa Cynapium,

^ In this and following lists, since statistics are no involyed, segregates ar
included if their distribution is characteiistic.

Praeoeb — On Types of Distribution in the Irish Flora. 31

Name.

Irish
Ceaws.

Watsonian

typ«.

Name.

Irish
Census.

Watsonian
t>pe.

Sctndiz Pectcn-Veneiis,

Asperula odoiata.

Salix alba.

Taleriana Mikanii,

—

Empetrum nigrum,

Seabiosa arrenais,

Caratophyllum demenuiUi

Solida^o Virgaurea,

Neottia Nidu8-a?i8,

Bideoa eernua,

£B

Cephalantbera ensifolia,

AchiUaa Piarmica,

EpipactiB latifolia,

Seoeeio ijlTadcuB»

Orchis latifolia,

Arttitim minuf ,

Allium urBinum,

Crepis paludoea,

Juncus glaucuB,

Hieracium muronim,

Sparganium simplex,

umbellatum,

affinitf,

Erica clnerea.

Butomus umbellatus,

Hfpopithys multiflora.

Fotamogeton heteropbyllui

1,30

Lyiinwchia nemorum,

nitens,

Samoloa Yalerandi,

lucens.

M jototia yersicolor,

perfoliatua,

lithoipennuin offidnale,

obtufiifolius,

CooToWnlus airenais,

Eleocharis acicularis,

Tcronica hedenef olia,

Sciipus sylTationa,

agrettia.

Eiiophorum yaginaturo,

poUto,

latifolium.

Carex sylvatica.

Anasallia.

Tesicaria,

icutdlata,

Phleum pratense,

Hdampyniin pratezue.

Glyceria plicata.

Utzkiilaria Tulgaria,

Bromus giganteus,

miiior,

asper,

Mentha lativa.

Lolium temulentum,

Lyoopiia euiopeuB,

Kardua Btricta,

Scutellazia galericulata,

Poly podium Dryopteris,

Stacbyi Betonica,

Botrychium Lunaria,

T^mp^in liybridunit

Eouifietum maximum,

Teuerivm Scorodonia,

byemale.

Polygooum lapathif olium

, 33

If we analyse this list, making in the case of composite type plants
a certain allowance in each of the types concerned, we find that the
plants are 70 per cent. British type, 24 per cent. English, 5 per cent.
Soottiah, and less than 1 per cent, each Atlantic, Germanic and High-
land. This result calls for no remark.

Toder the general type also we can best place our common
maritime plants (see p. 39) — species which are of general occurrence
where saline conditions prevail, such as CoehUaria officinalis^ Arenaria
f*pMeSy Eryngium nutritimum, Salicomia herhacea, Trigloehin mariti-
MUM, OUfceria maritima^ AspUnxum marinum. These are about forty

32 Froceedings of the Royal Irish Academy,

in number: they are almost exclusively of British type in Great
Britain ; their distribution calls for no remark, and we need not delay
over them.

Before leaving the plants of general distribution in Ireland, it ia to
be noted that a group of species drops in here which belongs neither to
Class A treated of above nor to Class B which follows. These are
widely distributed species the feature of whose range is their absence
from some one defined area, The distribution of these is not sufficiently
even to allow of their being placed in Class A» nor is it sufficiently
restricted to permit of their inclusion in any portion of Class B. The
thinning out or absence of a plant as shown in these instances is of
high interest, and of as great phyto-geographical importance as
the restriction of another to the same area. It will presently
appear that the two phenomena sometimes go hand-in-hand, and I
shall further refer to ranges such as the above-mentioned after the
converse case of plants characteristic of the same areas has been
discussed. It may be pointed out that even if such gaps in range ai^
eventually fiUed up by the discoveiy of a few stations in the blank
counties, the result will be to obscure rather than to do away with an
interesting feature of their distribution, for they are in any case much
rarer in these areas than in the rest of Ireland.

We now pass to the second and more important of the two large
classes into which the set of maps naturally divides itself — ^plants
which show an aggregation in some portion of the country.

It is to be remembered that, as compared with Great Britain,
Ireland is small, with a more restricted range both in latitude and in
longitude, and in altitude as well. It is also of more even shape, being
roughly elliptical in outline, and possesses less variety of surface and
climate. It is not surprising, therefore, that the flora of its various
portions displays a reduced diversity; in other words, that tlie
number of species of strongly marked local range is not large.
Nevertheless, some definite features of distribution came out clearly
as the maps were studied. The first strong character displayed is a
tendency towards a central or marginal distribution, a peculiarity not
found to any marked degree in the flora of Great Britain, and resulting
from the physical features of the country. The non-calcareous rocks
and the mountain-groups lie around the edge of the island, and here
is concentrated the flora pertaining to such conditions; while the low-
lying limestone plain with its numerous bogs, marshes, and lakes, is
the head-quarters of a difPerent set of species. By referring a plant
to the Central type of distribution, then, we signify that it is found

Prabgbr — On Types of Dktrihution in the Irish Flora. 33

chiefly in the Central Plain. 11711116 plants of this type often extend
to the margin of the island in the east and west, they show a marked
lestrietion of range towards the north and south.

The area of the Central type of distribution may be defined as being
limited by a line joining the Shannon month witii Wateiford on the
south, and a line joining Sligo Bay with Dundalk Bay on the north,
while in its most characteristic form it docs not touch either the
etstem or western margin of the island. The circle on fig. 21 approxi-
mately defines its ideal boundary.

The Marginal type, which is generally speaking the conyerse of
this, hardly requires definition, as its name is sufficiently descriptive.
The plants which belong to it are characterized by a tolerably eyen
though frequently discontinuous range through those divisions which
lie around the margin of the island, and by an avoiding of the Limestone
Plain* The negative character of avoidance of the Central Plain is the
most striking feature of this type of distribution ; and the ring which
mariu the range of the constituent species frequentiy thickens consider-
ably in the north and south, where the coast-line lies far from the
edge of the plain. The area of the Marginal type of distribution may
be described as lying outside the circle on fig. 21.

A number of the rarer and more interesting plants of Ireland are
more or less marginal in distribution (being rare in the Central Plain),
but are restricted to limited areas ; while many others show a general
increase towards the north, south, east, or west of the island. As
r^aids these, the strongest phytological boundary which developed
itBeU is one which corresponds with the curves evolved from a consider-
ation of the range in Ireland of the northern and southern plants of
Great Britain (see fig. 12) ; and this boundary can be best localized by
drawing a line from Oalway Bay on the west to Dundalk Bay on the
east The need of a dividing line between eastern and western plants
slso became clear ; and the most natural boundary appeared to be a
line passing through the cities of Londonderry and Cork — a division
which corresponds with the partition into eastern and western already
«ttployed in " Irish Topographical Botany."

The central circle and these two intersecting lines, then, define six
types of distribution which I believe are founded on the actual range
of plants in the country. The names most convenientiy employed for
the "types" will be

2. Central. 5. Mumonian.

3. Marginal. 6. Lagenian.
4* TJltonian. 7. Connacian.

a-LA. raoc, VOL. vui., sac. b.] D

Proceedings of the Royal Irish Academy.

the last four being named after fhe four proTinces of Ireland, in wbicb
each tjrpe respectiyely reaches its maximunii namely : TJltonian ty^
in Antrimi Mumonian type in East Cork, Lagenian typein Dubliiii
Connacian type in West Galway,

Fio. tx.— Boundaries of the areat of the Irish Types ot Distribation.

We have now to consider in detail the six types of distribution
above defined. In the Ibts of species which follow, only the more
characteristic plants of each type are mentioned. No attempt is
made, considering the still incomplete state of our knowledge, to

Pbabgbb — On Types of Distribution in the Irish Flora. 35

refer €My Irish plant to some type or combination of types. It is
quite possible that some of the plants which appear in the lists
below may, in the course of time, receiye an extension of range that
will place them outside of the type to which they are at present
referred. On the other hand, future confirmation of the present
limits of range may allow other species to be definitely referred to
one or other of the '' types " which are at present undassed. After
the name of each species, its Irish census is given, and its type in
Great Britain is added for comparative purposes.

2. Cehteal Type. — Thirty-eight species, with an average range
of 15 divisions per species, fall into this group. The distribution of
four characteristic plants of this type is shown in figs. 22 to 25.

Fio. tt,r-5UUaria fabuiru*

Fio. %i,^Orchu Mono,

F*G. x^—Sium Uti/olsmm,

Fig. t$,''Andromoda Polifolia,

Proceedings of the Boyal Irish Academy,

Thalictnim flATuniy
fianimciiliis ciiciiiAtiiBy
Caltha ndkani,
Stdlaria ptlastrif,

LathyruB palustrisy
Poterium Sangoiaorbft,
Mjriop^jllum TerticiUataiii,
Siam latifoUnm,
Cornas sangninea,
Galium uliginoeum,
Ezigeron acre.
Inula BaHoina,
Carlina Tulgarifl,
Centauiea Scabioaa,
Crepifl tanxacifolia,
Tragopogon pratenaifl,
Andromeda Polifolia,
Pyrola rotundifolia.

23 E

13 EG
2 —

19 EB
19 E
11 E
21 E
25 E

14 EG
13 E
13 BE

17 E

2 absent

28 EB

21 BE

18 G

22 BE

24 IS
1 SG

Gentiana Amarella,
Tencrinm Scordium,
Betnla Tenrooosa,
Orchis Mono,
Ophiys apifeiay

muscifeiay
Juncus obtusiflonis,
Lemna polyibizay
Sagittaria sagittif olia,
Potamogeton plantagineus,
Carex paradoxa,

PMndo-cyperus,
Equisetom Tariegatum,
Chara decmacantha,

pdlyacantha,

diyt^qdata,

tomentosa,
Toljpella glomerata,
Nitella tenuissima.

29 BE

7 6S
17 -
20 S
26 £G

8 £6
26

8
21

£
S
£
23 EB

I
22
15
21
22

6
11

LI
£

An examinatioii of this list shows a characteristic compositioii.
Eleven of the thirty-eight are
agnatic; species nine more are
marsh plants. Ten helong to
pastures and dry ground; two
are hog plants. Eight are
calcicole, and none calcifuge,
according to the standard of
"Cyhele Hibemica." All are
lowland; none are charac-
teristic of the uplands or
highlands. None are woodland
species. If we construct a sta-
tistical map according to the plan ^p^^^^^rT
already used, the distribution ^^^ffi
of the group comes out clearly cC^ ^
(fig. 26).' Here the minimum r''"^^ ' "
is 1 in South Kerry and We^t zf^^
Donegal, the maximum 35
(or 92 per cent, of the group; f.o. a6.-Dirtrib«tion of - Centr^ » pu„ts.

It ™*T ^ pointed out that the relatiye Talue of the depths of shading id
really higher than expiessed hy the numerical equiyalents, since species Reoenll^
thin out towards the limits of their nmge, and reach their maximum abundanc^
about the centre of their areas of distribution ; whereas on these maps a unif onx
Talue is awarded to each plant throughout its whole ninge.

Fraboer — On TypeB of Distribufiofi in the Liah Flora. 37

in Westmeafhy followed by 31 in S.E. Ghdway. It will be noted
that Antrim fnimBbes a snfficient nmnber of Bpecdes to raise it
abore the minimiiTn grade ; this is the effect of the Lough Neagh
flon, which indndes a number of stragglers from the waters and
manhes of the Central Plain, such as Thaltetrum JIavum, Bhamnut
uihgrtieuij E. Frangukiy LathyruB palustrisy Sagittaria iagittifolia.
The prolongation of certain species into Mid and East Cork appears
to be dne to the extension of the limestone into those divisions.
AnaljBed according to their British distribution, this '' Central " plant-
gronp is strikingly southern. Of 29 whose types are given by Watson,
IS, or nearly two-thirds, are of English or Germanic type ; seven
more have '' English " tendencies, while the only northern plants
ve one ''Scottish" species, JEquUetum varUgatum^ and two bog
plants of Scottish tendencies, Andromeda Polifolia and Pyrola rotund*-

3. MAMssMkii Ttfb. — Ezdnsive of maritime plants (of w:hich more
anon), the Marginal type is somewhat more nnmerons in species
than the CentraL Forty-six plants are listed below; they have
tt STenge range of 16 divisions per species.

SabohriaaquAtiea,

Sanisnrea alpina,

Hieraciom anglicnm,

Ccnidam tetnmdmiD,

Schmidtii,

MBudflMBdnmif

gothiciun,

arreoae,

cMum,

—

SHioacfliate,

£B

Lobelia Dortmanna,

rabttkte.

Batiaebeumdia,

Centunculiu minimus,

£B

flyperieiimelodet.

Myosotis collina,

Endina BiMehatam,

Staohya aryensis,

Trifotioffl striatuiii.

Galeopaia Tonioolor,

tirtom.

Sderanthua otti^^h*,

inphnoBf

Salixherbaoea,

TiekijrlTiiica,

Zaiinichelliapalil8tria(aggr.)> 20

'k'uwiuaodoraU,

Carex dioica,

fiuifngmitelUrit,

rigida,

Sddion Rhodiola,

Milium effuBum,

CtUitridie bamnlata,

Lyoopodinm alpinum,

Cowamticfflatum,

Isoetea lacuatris,

^■t^vmiam.

Pilularia globulifera.

<^*H>kaUiias7h«ticiim,

Chara canescens,

—

Aatkenk noUIia,

Nitella traualuoena,

—

Proceedings of the Royal Irish Academy.

The distribution of four characteristic examples of this type i»
giyen in figs. 27 to 30.

Fig. ifj.— Hypericum •lodes.

Fio. tH.'-LcMia Dorimatma.

ViQ. 99,—Pinguicula lutiianic€t.

Fio. lo^^NHella iranslucene.

It will at once be noticed that this group, of homogeneous dis-
tribution, is composed of heterogeneous elements, the result of the
yaried conditions which the marginal area provides. The leading
sections are alpine plants, brought in by the numerous mountain
groups ; calcifuge plants, rejoicing in the absence of limestone ; and
xerophytes, for wldch the sands of the coast are an attraction. No
less than 23, or one half of the group, are characteristically highland
or upland species (though only fiye of these do not occasionally

Praeoer — On Types of Distribution in the Irish Flora. 39

descend to near sea-level). Thirteen more favonr light or sandj
kSIb, Nine are water plants, but marsh and bog plants are few.
Seren are typical calcifuge species, one only {Cerastium arvense) is
sli^tly caldcole. Comparing their distribution in Great Britain, a
corresponding diversity is apparent* Eight are of pure Highland type,
one of Scottish, seven more have Highland or Scottish tendencies.
8ix are of Eng^sh type, two of Atlantic ; the remainder are British
or composite*

Por our map we have a mazimmn of 39 species (or 88 per cent, of
the group) in Antrim, a minimum of 0 in Longford* (^ The avoidance
by the group of the low-lying Central Plain and limestone areas
oomes out clearly.

Fic. 3X.— Distribution of '* Marginal " plants.

Haritime plants of general distribution round the coast are not
daased as of Marginal type, as their proper place seems to be rather
in the General type. In cases of restricted range, they are placed
under the type to which are referred other species of similar distri-

4. ITLTOiriAir Ttpe. — Into this group comes much of the northern
ekment of the Irish flora. The list given below numbers forty-five
species, which might be swelled by the addition of numerous Hawk-

Proceedings of the Royal Irish Academy.

weed segregates. The restricted range of plants of this and the follow-
ing groups (arising from the more restricted area by which the types
are defined) as compared with that of the " Central " and '* Marginal"
plants, is evident from the fact that the average number of divisions
per species drops to 6, as compared with 15 and 16 in the two preced-
ing types.

FlO. it.—Saxifraga oppotiiifolia.

Fio. ii.—Circaa aipina.

Fig. 34. — Cieuia vifvsa.

Fio. 3S.^PofamogeioM filiformit.

The distribution of four selected examples of this type is shown in
figs. 32 to 35. All of these are comparatively wide-ranging species,
but many plants of the type have a quite limited range, mostly with

Prabobr — On Types of Distribution in the Irish Flora. 41

Antrim as focus. The peculiar and defined range of Cieuta is note-
worthy, and without a parallel.

iUniinealas finitsiis.

Hieraciuxn atriotuniy

TroOini enropmUy

corTmbosiun,

Cudimine amazAy

auiatuzn.

—

Btrbana intermedia.

—

crocatiim,^

Teeidalia nudicaolif ,

HypocluBris glabra,

Saeiie acaalla,

Yaccimum VitU-Idsea,

iireiaiiium BjlTatifiimi,

Pyrola media,

pimtenae.

minor.

Phmu Badiu,

secunda,

Melampynim sylyaticum.

Bonbibeniiea,

Polygonum miniu,

Sudlnga oppodtifolk,

mite,

aisoidesy

Balix pentandra.

CaOitridie antuinnalisy

phylicifolia.

I^Oobiiim angoatifolitim,

nigricans.

Oiroaa alpmA,

Potamogeton flliformis,

Cieuta Tiroaa,

Garez pauciflora.

lignitieiim aeoticam.

elongata.

Galinm Cradata,

Buxbaumii,

•—

Adoza Moaehatelliiia,

irrigua,

—

Antiiim nemoroeuin,

—

GalamagrostiB atriota,

Hieiadiim laaiophyUum,

—

Cryptogramme crispa,

aigenteunit

—

Bquiaetum pratenae.

The character of this group is shown by the fact that some SO out
of the 45 are hill Jor mountain species, though very few of these are
timfimii to high elevations. Four are water plants; five frequent
manhes, and four peat bogs. One is a miaritime species. Only six
affect dry or cultivated soils. The group is characterized by an absence
of either calcicole or calcifuge plants. Analyzed according to the
British types, the Highland and Scottish species largely predominate ;
out of 38 classed by Watson, nine are of Highland type, ten of Scottish;
six more have Scottish tendencies. Only five are Ihiglish. Only one,
Aiout^ is of British type.

Per the construction of our statistical map, we have a maTimum
of 37 speciea (or 82 per cent, of the group) in Antrim, a minimum of 0 in
Tarious southern divisions. The northern grouping of the species comes
oat dearly.

> And a number of other Hawkweeds of more reatrioted range.

Proceedings of the Royal Irish Academy.

Fig. 36.— Distribution of " Ultonian " plants.

5. MmcoHiAv Ttpe. — This is the largest of the six groups, being
approached in nnmbeis bj the TiageniaTi type alone. Sixty-six
species are listed below ; these have an average range of 8 diTisioiiB
per species.

The range of four selected examples of Mnmonian plants is shown
in figs. 37 to 40, and these are characteristic of well-marked sub-types.

Fto. n.—RoMUMCulus Lenormandt,

Fio. 1%,—PinguicuIa grand(/hra.

Praeoer — On Types of Distribution in the Imh Flora. 49

In y^-^ampanula TrackeUum,

Fio. ^o,^Chlora perfoliata.

Yi%k 87 {Sanuneului Zenormandi) exemplifies the manner in which
TariotiB sonthem plants range np the east coast. As further examplea
of this peciiliarity of distribution may be mentioned

Lmnm angastifolitiitt.
SythrBa pulchella.
Wahlenbergia hederacea.

Orobanche major.
Salvia Yerbenaca.
Jiinciu aoutuB.

Kg. 38 {Finguieula grandiflord) shows the well-known Kerry-
Cork type, of which several of the famous Lusitanian group are note-
worthy examples. Other plants of this group are

Bom micranthA.
fiazifrEga Oenm.
ArbutuB Unedo.
Mierocala filifonnis.
Sibthoipia europna.

Allium Scorodopraaum.
Cares punctata.
Aaplenium lanceolatum.
Nitella Nordstedtiana.

Fig. 39 {Campanula Traeheliutn) illustrates the peculiar Barrow
Ttlley range whidi is shared by this species and CoUhieum auiumnaU —
a well-marked and very rare type of distribution.

Fig. 40 {Chlora perfoUata) exemplifies the more wide-ranging
Mumonian plants. Chlora might, indeed, be called a Central Plain
species extending southward, rather than a southern plant extending
northwazd ; it is strongly calcicole. Zeontodon hispidus is a parallel
case, and has precisely the same northern boundary. The line joining
Dublin and Killala marks, indeed, the northern limit of several of the
mate widespread of the Mumonian plants.

Froceedingg of the Royal Irish Academy.

EonanculuB tripartitos,

Microcala filiformiB,

Lenormmidi,

ErytbrBBa pulchella,

parviflonifl,

CynogloBsum officinale.

Olaucium flayum.

Antiirbiaum Orontium,

Hatthiola mnuata,

Sibtborpia europ»a,

BraBsica nigra,

Orobancbe major,

Lepidium latifolium,

Pinguicula grandiflora.

absent

Viola hirta,

Calamintha officinalis.

lutea,

TbymuB ChamssdryB,

—

liDum anguBtifoliam,

Salvia Yerbenaca,

Geranium pusillum,

Scutellaria minor.

rotandifolium»

Cbenopodium rubrum,

coliunbinum,

Atriplex portulaooidea,

OmithopuB perpuflilluB,

Bumex maritimuB,

Erodium maiitimum,

pulcber.

Trif olium filif orme,

Eupborbia Peplia,

Mercurialifl annua.

Alohemilla alpina,

SpirantbeB autumnalis.

Boaa miczantha.

Allium Scorodoprasum,

Sazifraga Geum,

abBent

Colcbicum autumnale,

dedpiensy

JunouB aoutuB,

Foeniculum officinale,

(Enantlie pimpinelloideB,

flabellatuB,

—

Bubia peregrina,

ScirpuB paxTuluB,

—

Dipsacus BylyestriB,

Carez muricata.

Diotia candidiBsima,

divulfla.

Hieiacium hypocheroidea,

-—

axillaria.

Leontodon hL^idoB,

punctata,

—

Campanula Traohelium,

Festuca Myuroa,

Wahlenbergia hederacea,

BromuB erectuB,

G£

Arbutua Unedo,

absent

Lig:uBtrum Tulgare,

Agropyron pungens.

—

Chlora perfoliata,

Asplenium lanceolatum,

The chief character of this large group as regards habitat lies in
the fact that more than half of them are plants of pastures, light soils,
and dry places. Four are water plants, and five marsh plants ; none
are characteristic of peat bogs. Eleven are maritime species. Only
eight are plants of the uplands or mountains. Analyzed according to
their distribution in Great Britain, they are a markedly southern and
western assemblage. Of 57 classed by Watson, 27 are of English type,
7 Atlantic, and 10 more combinations of English with Atlantic or
Germanic. Viola lutea is the only Scottish species, and one other.
Allium SeorodoproBum^ has Scottish tendencies. The Highland type
is entirely absent except for AlchemiUa alpina^ and the pure !6iitiah
type is unrepresented.

Prabger — On Types of Distribution in the Irish Flora. 45

In tormiiigthe statUtical map, the maxinmm is 39, or 60 per cent,
of the group, in East Cork, doselj followed by 38 in West Cork : the
mininiTiTn is 0 in varions northern counties. The manner in which the
•onthem plants run np the east coast, and avoid the Central Plain^
comes out very clearly.

Fio. 41.— Distribution of " Mamonian " plants.

6. LAonoAir Type. — Under this head forty-nine species are placed,
with an average range of eight diyisions per species.

The distribution of four chosen examples is shown in figs. 42 to 45.
Fig. 42 {Lepidium hirium) shows the eastern range at its maximum,
ilg. 43 {Trifolium gUmeraium) at its minimum. Pig, 44 {Cynoglotsum
•fcinals) illustrates a very characteristic Lagenian range, while in
fig. 45 {Seilla vema) a more northerly trend is illustrated. The] most
diaiacteristic of the Lagenian plants is the group of light soil or
land plants which are spread along the coast between Wexford and
Louth, such as

Sisymlniiim Iiio.
Medicago fjlyettris.
Tiifolinm subterraneum.

^omeretuiD.

•cabmm.

Trigoaella omithopodioides.
Scnecio eruoifolius.
Featuca uniglumis.
Equisetum Moorei.

Proceedings of the Royal Irish Academy.

Fio. ^,^Lepidium hirium.

Fio. \i^-~Trifolium glonuratum.

Fig. ^^.'-Cynoglossum officinaU, Fio. 45.— ^Iri/fa vema.

The list of Lagenian plants is as follows : —

ThaHctrum dimeiiBe,

—

Trigonella ornithopodioides.

Sisymbrium Irio»

Trif olium tcabrum,

Lepidium hirtum,

glomeiatum,

Thlaspi anrenBe,

tubteiraneum,

Elatine Hydropiper,

Yicia lathyroides,

Hypericum hinutum,

Sazifraga granulata.

Malya mosohata,

£B

Epilobium roseum,

Medicago sylyestris,

—

Chsdrophyllum temulum,

Pbaeobr — On Types of Distribution in the Irish Flora. 47

Inthriicai Tuls&ris,

SciIlaYema,

Seneck) erodf oliuB»

Lemna gibba.

Cirdniis crispuBy

Zannicliellia polycarpa,

—

Pidii echioides.

Garex dioica.

Hiendum boraale*

paludosa.

LjeopoB anrenuSy

Poa palufltrisy

—

CyBOgloMiim officinale,

Olyceria aquatica.

lithoipennum axrense.

Borrerii

Echium Tolgare,

Festuca imiglumis,

Scrophulam umbroea.

Hordenm RAcaliniiiTiy

Cihmtnfha Adnos,

murinumy

Gtleqpds Ladanumy

Equiaetum Moorei,

absent

Larniam allnmiy

Chaia conniyena.

—

Galeobdolon,

Tolypella prolif era.

—

Atriplez fazinosa.

intrioatay

—

Saliz triandia.

Nitella gracilis,

—

Aspuagus officiiialiBy

ThiB Eastern group is, relative to the other types, zerophytic.
Serenteen of the 49 species are inhabitants of sea-sands or sandy
soil, 12 more of dry banks
or lig^t cultiTated ground.
Seven are water plants, 5
frequent marshes, 3 more are
onially found by river sides.
One, Lepidium hirtum, is dis-
tineilj calcifuge, but on the
other hand Galeopsis Lada-
mum is strongly caldcole, and
aereral others also prefer a
hmy Bon. None are moun-
tain plants, or even hill
phmts, excepting perhaps
Hi&raeium horedU.

The distribution in Great
Britain of the group is chiefly
••English," 12 out of 40
speciee classed by Watson
being of that type. SeiUa
uma (Atlantic-Scottish) is
the only species in which the northern element makes its appearance.
One is of Atlantic type, one of Germanic : the rest mostiy intermediate
Itetween Knglish and British: in other words, slightiy southern in
range. As compared with most of the groups, there is an increase
of British type, no less than 5 being purely British.

Fio. 46.— Distribution of

Proceedings of the Boyal Irish Academy.

For our map, the TnaxinrnTn is 38 species, or 80 per cent., in
Dublin; the mininnini 0 in West Mayo (fig. 46).

7. CoNNAdAif Type. — This and the Mumonian group are the most
numerous in species of the six Irish '' types." Sixty-three species are
listed below ; these have an average range of six divisions per species.

Here belong plants exhibiting a general western trend in their
range ; such as Taxw haceata (fig. 47) and the more restricted
Adiantum CapiUus- Veneris (fig. 48). In this section of the group
may be placed nearly one-half of tlie plants listed below.

Fio. ^.--Adiantum Ca^iL'ut'Vtturu.

Fio. ^<^.^Hah«naria intacia.

Fio. $o,^Erica mtdiUrranea.

The remainder include two very distinct sub-types, which may
be called respectively the Burren type and the Connemara type.
The former is represented in fig. 49 {Hahenaria intacta\ the latter

Frasobb— Oif Tffpea ofDMribution in the Irish Flora. 49

in fig. 60 {jR^ meditmwMa).
groaptie

HeUantfaemiiiii Tineale.
AitngBluB Hypoglottis.
BpinM FQipendola.
FdteotalLi frntieoM.
Oettiana Tenia.

The leading memlwn of the Buiren

TiimoBftllft aqoatiea.
Ajuga pTTamidalis.
Epipactis atro-rubexiB.
Habenaria intaota.
Potamogeton lanoedUtuB.

The gronp reaches its waxiTniiTn on the hare limestone hiUs of Barren,
whenee it spreads in diminishing nomhers oyer the pavements or
*' cngs" oi GaLway and Mayo. It is essentially a oalcioole group.

In marked conlzast to this stands the adjoining oaloif age Connemava
group. In its restricted sense, it numbers but three species — ^the
remarkable ''Luaitanian" heaths, JBrieamediierraneaj E.Maehaii^ and
JkUoeiapoU/oUa; but a few more widely-ranging species, such as the
equally interesting NaiMflexUii^ may fairly be classed with those.

The list of the Connacian type is as follows : —

Thifietnun alpinum,

eollinnm.

Anbiidliata,

Dnbabcana,

BfHinthfunnm tizimIb,

gattatum.

Viola itagmiia.

Aienuia ciliata,

absent

Gcnminm tangninenm.

^mSsr^

OB
EG

PotentiDa Initiooea,

Saafaan niTalis,

absent

Btember^

absent

BnMoa intennedia,

Galium boreale.

■yWcstre,

AiperaU eynanchicay

Hioaciiiia iricum.

EiicaMaekaii,

absent

BMditenmnea,

absent

DtbMciap^ifolia,

absent

ArctoiUphyUM Uva-nni,

Geatiaiia Tenia,

l^iDQwUa >n***Hfii.

absent

BmoaTiaeoM, "

Ajjppyiamidali.,

Fdygooum Tinpanun,

Ozjna digyna.

Euphorbia biberna,

amygdaloides,
Juniperus communis,

nana
Taxus baocata,
Epipactis atro-nibens,
Habenaria intacta,
Sisjninchium angustifolium,
Simethis bicolor.
Allium Scbienoprasum,
Juncns tenuis,
Potamogeton Eirkii,

lanceolatus,
Naias flezilis,
Eriocaulon septangulare,
Scirpus triqueter,
Bbynchospora f usca,
Carez Boenniughausiana,

trinenpis,
Descbampsia alpina,

disoolor,
Sesleria csBrulea,
Pea alpina,
Tricbomanes radicans,
Adiantum CapiUus-Yeneris,
Asplenium Tiride,
Aspidium Lonobitis,
Eqmsetum traobyodon,
Lyoopodium inundatum,
Isoetes ecbinospora,
Nitella Nordstedtiana,

absent

—

absent

—

a.i.a. rmoc., tol nn., sac. b.]

50 Proceedings of the Royal Irish Academy.

Of this assemblage, the largest oonstitueiit groups are mountain
plants, which number 16, and plants of dry places — ^in most cases
limestone pavements — ^which are 19 in number. Of the mountain
plants, onlj half are confined to the higher grounds ; the others often
occur in great quantities at low levels. Bog plants are well repre-
sented by six species ; marsh plants number four, water plants seven.
Only one, Seirpus triqueUry is maritime in habitat. The group has
distinct proclivities for limestone, seven of the species being recognized
calcicole plants, while two are calcifuge. No less than ten of the
species are absent from Great Britain. The remainder are quite hetero-
geneous as regards their distribution in the sister island : four are
English, two Litermediate, one Scottish, four Atlantic, two Oermanic,
one local ; far the largest group is formed by the Highland type (here
continually lowland), which numbers thirteen species. Of British type
there is one definite and one doubtful example (namely, Juniperw
communis and Oeranium sanguineum respectively).

For our map the mftTiTnnin is 37 species, or 60 per cent, of the listi
in West Ghdway ; the wn'TiiTTmni o in half a dozen eastern counties.

Fio. 5x.— Dittribation of" Connadan " plants.

It may be useful to set forth in tabular form the characteristicB of
the six plant-groups which have been described.

n
o

B
H

Twi

*9iinraLio{)

•oi^OTnv

-puv[q9iQ

•ij«WOog

*o;vipeTiuo;nj

'H*U^"a.

'VPHff

pTEV[qSiQ

■pmidx^

*;no9tMS

*d9njppi3

'dlooppiQ

'f^oaiXjppin

-ffpooiipin

•*)g ;wa

•V^H

•w»»ili

^ «o o«

^ ^ eo CO

CO rH C4 ^

^ 09 vH

w*
Mt

e«

^ vH 1^ C4

«D ^ C4

s s

e I

Jr2

52 Proceedings 0/ the Royal Inah Academy.

The figures, especially as regards habitat, axe of course merely
approziixiate, as exact data in such matters are not possible ; neTflT-
theless, the table brings out in a yery striking manner the widely
divergent characteristics of the yarious Types of Distribution, and fits
in well with the maps previously given, showing the distribution of
plants according to soil and elevation, and also according to type in
Ghreat Britain.

Notice has already been taken of the fact that certain plants of
wide distribution in Ireland are nevertheless characterized by an
absence from definite areas. This interesting point must now engage
our attention for a few minutes. One of the most marked ranges of
this kind is illustrated in fig. 52 {Drosera angliea\ which shows an
absence from those south-eastern counties where the Lagenian flora
reaches its maximum. The following species show a similar absence
from or rarity in the south-east : —

Nymphsea alba. H jrica Gale.

Bubiis Bazatilis. Scirpus pauciflorus.

Hyriophyllum alterniflorum. Bhynohospora alba.
Pinguicula yulgaris.

— while in the case of Pamastia palustris and Selaginella telayinouUi,
the centre of the '* absent " area lies further south ; most of tiiese are
imiversal over the rest of Ireland, while in Great Britain they are aa a
group ^'British" with a "Scottish" tendency. It will be noticed that
in the main they are plants of lowland boggy places, and such ground
reaches its minimimi in these divisions ; nevertheless some further
reason appears necessary to explain their absence. They might be
classed as Anti-Lagenian rather than as Pan-Connacian, since they
do not exhibit any marked increase westward.

Fig. sz.'-Dmem anglica. Fio. ^i^—CSnaMtke Pkgliandrium,

Pkabobb^Oh Type%^ of mkttihuiim isk the Irish Flora. &3

Another type of absence appears in fig; 53 ( OShanihe Phsttandrium),
consisting of a dying out along tho monntain-iim of Ireland, especially
in the west. Seven species —

Kattortiam pafaistie. Kumex Hydrolapathom.

amphibinm. Elodea canadenBis.

Simii angustifoliam. Lemna triflulca.
CBsantHe Phellandrium.

— exhibit this feature coDspicnonsly, and others to a less degree ; all
lie practically uniyersal^ elsewhere. This is quite a homogeneous
gn>np, inhabitants of lowland marshes and ditches, with a strong
" English " tendency in Great Britain.

A third group is illustrated in fig. 54 ( Vteia angmii/oUa\ with
vhichmAy be classed

ffitymbriiim AQiaiia. Yaleiianella olitoria.

Ulttc OaOii.

Hke the character is a marked absence from the proyince of Connaught
—the middle part of the Connacian district. These plants have little
in common, SiBymhrium being caldcole, UUx calcifuge, the others
neutral ; and they show similar diversity in other respects, the only
pomt of agreement being a preference for dry situations.

Fms. 54.— ^mm amgutt^folia» Fio. 55.— X^m/m^ kirtut.

i fiy " nniyenal " I meoa piMont in all dnriaioxu.

54 Proceedings of the Royal Irish Academy.

A better-marked group is fhat whose featiire is raritj in or absence
from Ulster, especially central Ulster (fig. 55). We may cite

Papayer EhsBas. Scrophulaiia aquatica.

Brasflica alba. Origanum yulgare.

Sazifraga tridactylitea Parietaria officinalis.

GaiduuB pjonocepbaluB. Orohia pyramidalis.

Leontodon hirtua. Festuca rigida.
Yerbascum ThapsoB.

The proclivities of this " Anti-Ultonian " group are lowland, light soil,
caldcole, and '' English" ; characters which reach their miniTnum in
the Ulster flora. The sandy soils by the sea enable most of these
species to creep northward round the Ulster coast, but inland they
are rare or absent.

Lastly, there is a group of " Anti-Central " species, too widely
spread to be cited as characteristic " Marginal " plants. One of the
best marked is figured (fig. 56) in (Enanthe erocala^ and the following
resemble it in range : —

Hypericum hnmifusum. Hyosotia repens.

Filago germanica. Carex IsBvigata.

Fig. siS,—(EHamtAe crocafa.

These are all plants of somewhat marginal type, and increase on
the hills and non-calcareous rocks.

One small but well-marked type of distribution which, though in
a broad sense " Central," does not typically fall into any of the seyen
types already defined, deserves mention. The plants composing it

Fbabobb — On TypeB of Diatribution in the Irish Fhra. 55

show a diagonal range acroBs Ireland from the north-east towards the
9oath«we8t, haying a marked absence in the north-western and sonth-
eastem areas. An example (fig. 57, Sydroeharis Monus-rana) will
illustrate this type ; and to show the similarity of range of the gronp,
a statistical map is added (fig. 58) according to the usual plan, show-
ing the aggregate range of the following members : —

Biaunenlos ciitinatafl.
Ltth jrna palostrii.
Hydrocliaris Monus-rans.
Sagittaris aagittifolia.

EleochariB acioularia.
Carex acuta.
Laatrea Thelypteris.

Fto. ^j,~~//jdroeAans Mortufrana, Fio. 58.— Moantain-folding of Ireland.

On the map I haye added the main lines of the old '' Caledonian "
and ''Hercynian " folding of Ireland, as demonstrated by Prof. Cole/
since this, I belieye, is the key to the peculiar range of these species.
It will be noticed that they are without exception marsh and water
plants, and they follow the great central trough of the island, spread-
ing oyer the basins of the Shannon, Erne, Boyne, and Bann, which
lie in the synclinal area, but ayoiding the great anticlines of Leinsterf
Mayo and Donegal, including the riyer-system of the south-east ; and
are absent eyen from the great western lake-system of which Lough
Cozxib is the predominant member.

Lastly, as to the distribution of plants which are probably or certainly
introduced in Ireland. As before stated, the aliens are generally
marked by a discontinuous range. A large number are widely spread;

> KnowUdge, April, 1898.

Proeeedmgs of the Royal Irish Academy.

hut cOien fall in with Yarioiu Tjrpes of Distribation, and as Hub b n»
donbt in most casee the effect not of chance, bnt of soil or climate, it
will be worth classifying them. LeaTing ont of account species which
have a restricted range, sach as Seneeio tpiaUduB and Straiioi&8 tMi&ti
the aHens of well-marked range ran as follows : —

Central-^

AxenaxiatenuifoliB,

E Matricaria diseoidea
Marginal—

—

LyclmiB Githago,
Inula Helenium,
Silybum Marianum,
Centanrea Cyanus,
Giohorium Intybus,
CuBcuta Epithymum,

B Cuaouta Trifolii,
E P Lydum barbanim,

B Plantago media,
E Btomus aecalinua,
E

UUoniat^—

EB
BE

MynfaiB odoiata,
AnehuBa aemperrirent,

I ? Veronica peregrina,
£

Ifumonian —

Senebiera didyma,
Yaleiianella Auricula,
Picrifl liieracioideB,
lixiaria Elatine,
minor,

E MaiTubium yulgaie
E Humulua Lupulus,
E Narciflsus biflorus,
E Leucojum sostirum.

E
E
B

6£

Diaba muraliB
Siflymbrium Sophia,
Medicago maoulata,
Lactuca muralis.

— Campanula lapunculoidea,
BE Ballota nigia,
£ Acorus Galamusi
£

Connaeian —

LI
£

Allium Babingtonii, —

It will be seen that the more successful aliens, other than those of
general distribution, are grouped round the margin of the island,
especially in the south and east. They reach their minimnm in the
centre, north, and west. Their distribution, in fact, coincides with
that of the '^ English " plants (fig. 1), to which type belong, as will
be seen from the analysis appended to the list, 16 out of 26 classed by

P&ABOSR— Oft Types of Distribution in the Irish Flora. 57

Watson. Thifl Mumonian orLagenian range of the aliens is of interest,
foftliere can be no doubt that it is the result of conditions of climate and
toiL In support of this yiew, one remarkable instance maj be cited«
Clorer seed, imported from England, is sown widely in Ireland;
ofidal information supplied to me is to the effect that no more doyer
is sown in the south and east than in other parts of the country.
With the dorer come the seeds of the parasite Orohanehe minora a
plant of English type, not a native of Ireland, and unknown therein
until some forty years ago. The plant is now an established and
spreading colonist, and its present range coincides in a striking degree
with that ol the group to which it belongs — the light soil English

type plants. In the central portions of its range — Wexford paiticu-
Urly — ^it is now abundant and permanent. This further emphasizes
the floial peculiarities of the south-eastern portion of Ireland, which
bire already been demonstrated both from the presence and absence
themn of certain groups of species. The great Leinster anticline
is an important factor in Irish plant distribution, and a phyto-
logieal boundary of marked character is formed by the line where its
uplands sink into the Central Plain, and by the prolongation of that
tine northwards and southwards.

Of the seren Types of Distribution proposed in this paper, five have
ti^ analogues in the types which Watson instituted for Great Britain.
In both series we haye a General type, and a Northern, Southern,
Bait«n, and Western type. The General, Northern, and Southern
types of Great Britain and of Ireland in a wide sense correspond in their

58 Proceedings of the Roi/al Irish Academy.

composition. The Eastern group of Ireland is seen to be esaentiallj
sonthem in Gh!eat Britain, while the Eastern group of Great Britain is
practically absent from Ireland ; neyertheless the two correspond in
character, representing in each case the nndens of the thennophile and
xerophile elements of the flora — ^in England a much more intensified
group than in Ireland. The Western plants of the two islands also
exhibit a wide diyersity in range, those of Oreat Britain being
Southern and Marginal in Ireland, while those of Ireland are not
to anj extent Western in Oreat Britain, and include besides a
number of species absent from the sister island. But here again
the two have affinities, both being hygrophile and frigofuge in
character. The two remaining Irish types, the Central and Mar-
ginal, have no analogues in Great Britain. The former consists largely
of "English" species, the latter chiefly of "British" plants which
do not penetrate into the Limestone Plain.

It will be observed that no type corresponding to Watson's Mid-
land type is proposed for Ireland. Plants of this kind form inlreland
a much less distinct group than in Great Britain, being largely reduced
in numbers, and not nearly so montane in habitat. Moreoyer, they
haye not any so definite head-quarters as, in Great Britain, they find in
the Highlands of Scotiand. In Ireland, plants of Highland type are
distributed almost equally between the Marginal, TJltonian, and Con-
nacian areas. I^one belong to the Lagenian and only one to the Mumo-
nian, although in those districts occur the largest areas of high eleyation,
as well as the loftiest summits, in the country. The actual alpine flora
of Ireland is extremely limited. Taking the 42 Irish plants belonging
to Watson's Highland type, we find that one-third of them descend in
Ireland to sea-leyel. Sixty per cent, may be found at eleyations of
500 feet or less. Fully one-half of the group fiourish at these low
eleyations in places where alpine ground — say oyer 1000 or 1600 feet —
does not adjoin, so that their occurrence means not merely the washing
down of seeds from their natural high-leyel habitats. Only 30 per
cent, keep aboye the thousand-foot contour line, only 5 per cent, aboye
the 2000-foot line. In fact, Watson's Highland plants cannot be defined
in Ireland as a group " chiefiy seen about the mountains." They are
chiefiy seen in certain hill-regions, but the presence of eyen high
mountains does not necessarily inyolye their appearance. The species
seen about the mountains are largely British type plants, with a
yariable admixture of " Highland " species, and certain local groups —
in the north " Scottish " plants, in the south often " Lusitanian." It
does not seem desirable to attempt to construct out of these hetero-

PsABOBB — On Types of Distribution in the Irish Flora. 59

genonB materials any group of Irish plants '^ cliieflj seen about the
mountains.'* The peculiar range of the Irish high-level plants — often
absent from lofty mountains in the east and present on hills of less
eleration or on low grounds in the west — ^is best brought out by treat-
ing them with the other plants of similar distribution.

80 much for the facts. The causes which lead or haye led to the
distribution of the flora as we now find it are difficult to determine.
The effect produced by the distribution of lime, and of open light
soils, is fiurly dear ; but climatic effects are not so easily dealt with.
As regards temperature, some of the characteristic plants of Connacian
type are without doubt frigofuge — in other words, their chief need in
onr climate ia a sufficiently high winter temperature ; and in fig. 60,
which shows the isotherms of the coldest month of the year in Ireland
(January), parallels between isophytic and isothermal lines may easily

Jitt*

Fig. 60.— JaaiufT ttothemu.

Fio. 6x. — ^Jttly iaothenns.

be drawn from among the plants of the south and west. Fig. 61 like-
wise, showing the isotherms of the warmest month (July), suggests
that a number of the south-eastern species may be thermophiles —
plants for which the most pressing need is a high summer temperature
for the ripening of fruit. Questions of rainfall probably effect but little

00 Proeeeding% iff the Royal Irish Academy.

the dktribatioii of plants in Indflnd, mnce (fig. 62) then ia evoryiriifire
a anfflciency. Bnt the limit of the Connanian type (fig. 51) will be

Fig. 6a.— Annual tainfall.

aeen to correspond with that of the wet west^of-the-Shannon district
of Ireland, while the driest area is inclnded in that which nuirks the
range of the Lagenian and most of the English type species. There
is eyen a dry area around Gfalway Bay which no doubt helps to
produce the remarkable aggregation of *' English " and " Qermanic "
species in north Clare. But the facts are not yet brought together, nor
the observations made which will enable us to determine how far the
present distribution of plants is effected by climatic causes. Nor is
this the only direction in which work is required. We can neyer hope
to understand our phyto-geography till its problems have been attacked
by the historical method. Yet the history of the Irish flora is still an
absolutely unworked field. The records lie buried below our peat
bogs and superficial deposits, and their elucidation will furnish evidence
of the highest importance. No branch of Irish botany has more
pressing claims on the field botanist than this.

[ 61 ]

II-

GLEANINGS IN lEISH TOPOGRAPHICAL BOTANY.
By ROBERT LLOYD PRAEGER, B.A., B.E.

[Read Ma&ch 16th, 1902.]

Of the material contributed or collected for '< Iriflli Topographical
Botany," a quantity of gleanings remain after the crop has been
gathered in. These consist mainly of two sorts— notes of varieties
and hybrids, and notes of aliens. Except where the information
rejecting species could be amplified by including records of their
varieties, or where alien species could be admitted as naturalized,
neither of these classes of records was included in the book referred
to. In now publishing a selection of these notes, I haye kept '* Cybele
Hibeznica" before me, and have aimed mainly at giving such records
u aie [supplementary to the information therein contained. Though
the bulk of the notes which follow are unpublished previously, I have
not hesitated to include records scattered through inaccessible publica-
tiona ; and have sometimes given, in condensed form, all information
rekting to a plant, whether published or unpublished, usually indi-
cating what matter is original. The notes axe arranged under county .
diTirioDs, and records are quoted according to the rules adopted in
*' Irish Topographical Botany." References to published papers are
giTen by means of the numbers prefixed to them in the same work ;
which is followed in other details of arrangement also.

As regards the records for which I am responsible. The heavy
field work of 1896-1900 gave little opportunity for the study dt
critical plants; but in some genera, notably AtehemiUa^ JEuphrasia^
and Ckaray large collections were made, and the distribution of segre-
gates and varieties in Ireland to a great extent ascertained. My best
thanks are due to the several critical botanists who devoted much
Ubour to the inmiiTig of the material collected — ^Mr. Arthur Bennett,
the late Yroi. A. W. Bennett, Mr. G. C. Druce, Messrs. H. & J. Groves,
^r. E. F. Linton, Rev. £. S. Marshall, Rev. W. Moyle Rogers, and
3b. tnienck Townsend.

62 Proceedings of the Royal Irish Academy.

Bannnoiiliis peltatiu Fr., yar. dongatns F. Schultz.

38 Down Canal near Scanra (Lett)— V. B. E. C. '94-6.

Also in 32— Bee Irish Top. Bot.

S. aoris, L., yar.^Boroanns (Jord.).
In 12, 16, and 26, '95^-6— £. 8. MarshaU.

yar. Stereni (Andrz.).

39 Antrim " Glenarm, &c.," '99— Druce 28^.

Calfha radicana Forst., yar. setlandioa Beeby.

28 Westmth. £. S. Marahall refers (J41) £. F. Linton's plant from

Brittas Lake {Cyh., p. 10) to this.

tFnmaria Vaillantii Loisel.

8 Limrck. Limerick '97 [among casuals] (G. Fogerty) — Hfrh.

L.F.C. !

Hastortinm amphibinm x sylrestre « H. barbarosdes Tansch.
6 Waterfd. Cappoquin '99— P.

29 Leitrim Carrick-on-Shannon '99 — ^P.

Mr. Bennett is not positiye of the detennination, but both paients
are now known to occnr on the Shannon and Blackwater, and theie
seems little reason to doubt the determination.

H. officinale L., yar. mierophyllum (Beichb.).
In 12 Wexford, and 16 W. Galway— E. S. Marshall.

Barbarea vulgaris E. Br., yar. aronata (Eeichb.).
1 Kerry S. ♦Bossbehy '00— Scully.
89 Antrim Drum Bridge '94 — ^Stewart.

Arabia hirsnta Scop., yar. glabrata Syme.

9 Clare Inishmore '95 — P.

♦Alyssum oalyoinum L.
Divisions 1, 4, 5, 19, 20, 21, 40.

*A. maritimum L.
5 Cork E. Queenstown '90— Phillips, and '98— Mrs. Peisse !

Praeobb — Qkaninga in Irish Topographical Botany. 63

Sifymbriom officinale Scop., var. leiooarpnm DC.
8 Cork V. Glandore '96 (J. Groves)— E. A. Phillips.

♦Erynmom orientale E. Sr.
DivisioiiB 1, 3, 4, 5, 8, 12, 22, 39.

^Biinias orientalis L.
Divisions 6, 12, 21— P.

^Camelina satiya Crantz.
Divisions 4, 12, 16, 20, 21, 23, 27, 31, 32, 38, 39, 40.

^Lepidinm Braba L.
Divisions 5, 8, 11, 12, 21, 87, 88, 89.
ippears to be establishing itself at Limerick — P.

*L. mderale L.
Divisions 8, 5, 8, 12, 21.

fPolygala caloarea F. Schultz.

Specimens collected near Taam are doabtfnlly referred to this form
by Prof. A. W. Bennett, while others from Devil's Bit (N. Tippenuy)
and Atblone (Eoscommon), collected by myself, and Kilrea (Derry) by
Xh. Leebody, are marked by the same authority " approaching
M^cM^M." P. eaicarea being as yet unrecorded from Ireland, the
ooennenoe of these intermediate forms is interesting.

*Silene Armeria L.
Divisions 9, 12, 21, 37, 39, 40.

8. Cnenbalns WibeL, var. pubenila Syme.

13 Carlow Aghade *99— P.
15 Galir. SE. Garryland '00— P.

I believe frequent in Ireland, but I did not note localities.

Cerastium glomeratum Thuill., var. apetalum Dum.
3 C«k W. Timoleague '97— Phillips.

C. triviale Link, var. holosteoides Fr.
6 Watofd. By the Blackwater below Cappoquin '99— P.

64 Proceedings of the Boyal Irish Acadetny.

Arenaria serpyllifolia L., var. leptoolados (Gubb.).

3, 4, 6 Cork Frequent— PhillipB.

6 Waterfd. Dunganran '99, Carrickbeg '00— P.

7 Tipp. 8. Clonmel'OO — ^Phillips. Cahir and Slievenaman '00— P.

8 Limrck. Limerick '99 — SomerviUe. L. Gnr 'OO^P.

10 Tipp. N. Nenagh '00— Phillips. BaUingairy '00— P.

11 Kilkny. Ballyragget '99, Fiddown bridge '00— P.

13 Carlow By Barrow above Bonis '99 — ^P.

14 Queen's liountrath '97, Maryborough '96— P.

15 Galw. SE. Coole '00— P.

18 King's Birr '00, Banagher '98— P.

19 Kildare Nuiney'97— P.

20 Wicklow Kilmacannoge '94 — ^P.

21 Dublin Sutton '94— P.

31 Louth Soldier's Point '96— P.

Also in 23, 26, 35, 37, 38, 39— see CyheU II.

var. Lloydii (Jord.).

12 Wexford Camsore and Gorey districts '97 — ^E. 8. Marshall.

Stellaxia nmbrosa Opiz.
5 Cork E. Near Castletownroche '00— R. A. Phillips.

Rev. E. F. Linton writes of the specimens : — " 8. umirosa and
S. media major seem to run into one another, and this may be regarded
as 8. umhrosa with bluntly tubercular fruit or as var. major with
glabrous inflorescence ; i . tf., there are connecting links which seem to
abolish 8. umhrosa as a species^ and then we have two contiguous van.
at times barely separable — 8, media var. umbrosa^ 8. media var.

;br."

SteUaria umhrosa is unknown in Ireland hitherto.

Spergula arvenais L., var. vulgaris (Boenn.).
Divisions 1, 2, 3, 4, 5, 6, 11, 12, 16, 20, 21, 38.

var. sativa (Boenn.).

Divisions 1, 3, 4, 5, 6, 11, 15, 16, 21, 27, 85, 36.
Apparently the two forms are about equally common.

Hontia fontana L., var. minor (Gmel.).
Diviaions all, ^cept 8, 1818,

Prabobr— 67feant/i^« im Irish Topographical Botany. 65

yar. riyularis (Gmel.).

Diviaioiifl 4, 6, 7, 10, 13, 19, 20, 21, 26, 27, 29, 30, 38, 89, 40.
Chiefly about the mountaiiiB.

Geranium Bobertiaiiiuii L., var. modestnm (Jord.).
6 Wateifd. Bungarvaii '82 (Britten and Nicholson)— B.E.C. '82.

yar. pnrpareani Font,
2 Kerry N. Lower Lake of Killamey '01 — G. C. Druce.

Ononis repens L., var. horrida Lange.

5 Cork £. Youghal '00 : frequent near the sea — Phillips.

6 Waterfd. Bunmahon and Knockmahon '82 — Hart j8^.
21 Dublin St. Doulagh's '96— P. Rare.

38 Down Aidglass (Waddell)— S. & P. 874,

*Kedicag^£alcata L

5 Cork £. Queenstown Junction 1894-1900, Tivoli '96— Phillips.
8 limrck. Corbally '98 (Bentley)— jy^jr^. L. F. C. ! Old quarry
at Limerick ! '00— R. D. O'Brien.

♦Kelilotns alba Desr.
Diyisions 1, 4, 5, 8, 9, 12, 21, 22, 34, 38, 39.

Lotos comicnlatns L., var. crassifolins Persoon:

12 Wexfoid Near Wexford '96— E. S. MarshaU.
38 Down Newcastle '98— G. C. Druce.

And in 39, 40— see Cyheh II.

Vicia tetrasperma Moench.
5 Cork K. Ballyvodock '00— R. A. PhiUips.

V. angnstifolia Roth., yar. Bobartii Koch,

4CQEk]Gd Coachfoid '97— R. A. PhiUips.

Y. craooa L., yar. inoana ThuilL

16 Gtlw. W. Clonbur '95— M. & S. 5^5.
23V«tmth. NW. end of L. Owel '95— Leyinge,
26 Mayo E. Clonbur '95— M. & S. 5^5.

»• L A. nuCKUf TOL. Tm., BBC. B.] F

Proceedings of the Boyal Irish Academy.

Lathyrus macrorrhiziu Wimm., yar. teniufoliiu Iteicli. fil.
5 CorkE.

10 Tipp. N.
16 King's
19 Eildare

Glanmire '00— K. A. PhiUips.
*PrnnTU domestica L.

Nenagh '00— R. A. PHllipe.
Clonad Wood '96— P.
Carbury '96— P.

Snbiu foBcns x incnrratiui.
23 Westmth. Crooked Wood '95 — H. C. Levinge, andjoj.
S. mncronatuB x pyramidalis.
L. Corrib near Cong '95— M. & S. 5^5.

S. corylifoliuB x roflticanuB.
Malahide '93— P.

S. oorylifoliuB x lencostachys.
Maryborough '93— P.

S. corylifoliuB x osutu.
Howth Junction '93 — P.
Knock Ross — Leyinge '94 484.

Potentilla suberecta Zimm. » P. procombens x nlvestriB.

10 Tipp. N. Near Cloughjordan '00— P.

14 Queen's Base of Arderin '97 — P.

16 Galw. W. About Clonbur '95— M. & S. 5^5.

26 Mayo E. Clonbur '95— M. & S. 5^5.

29 Leitrim Ballinamore '00, Lurganboy '99 — P.

33 Ferman, Newtownbutler '49 (Dr. Mathew) — Marshall 555.

36 Tyrone Omagh '96— Miss Knowles.

38 Down Newtownbreda '49 (Mateer) — Marshall 555, and Pur-

chas, Jwrn, BoL xxxi., 374.

39 Antrim Glenshesk '93 — Shoolbred Sjy.

Oeumintermediom Ehrh. = 0. riyale x nrbantuu.
18 King's Clonad Wood '96— P. .
20 Wicklow West of Baltinglass '99— P.
26 Mayo E. Ballinrobe '91— Mrs. Persse !
29 Leitrim Annaghearly Lake '99 — P.

40 L'derry Limayiidy '95 — B.N.P.C. Garvagh — Miss Snowies.

Also in 14, 15, 19, 21, 24, 33, 37, 39— see Cybele II.

26 Mayo E.

21 Dublin

14 Queens

21 Dublin
23 Webtmth.

TuAEOWi—Okaninffs in Irish Topogrt^hieal Botany. 67

Alchemilla vulgariB L,

The paper on the distribution of Alehmnilla segregates in Ireland,
by Bev. E. F. Linton in Jmtm. Bot. and Iriih Nat,^ April, 1900,
summarizes our knowledge up to that date. As Mr. Linton has since
named for me a large batch of gatherings, and as he has not in many
instances given localities in his paper, I give in full all the information
I have.

A. pratensiB Schmidt.

9 Clare Co. Clare {Hwrl. R. P. Murray)— Linton '00 50/.

17 Oalw. NE. Annaghdown '00— P.

18 King's Tullamore and Clara '99— P.

22 Meath Ballivor '00, Hill of Down, Oldcastle— P.

23 Westmth. Knock Eyon '99, Moate, Hare L, Coosan L.— P.

24 Longfd. CasUerea '00, Ballymahon, Killashee — ^P.

25 Boscomn. Lough Key '97— P.

27 Mayo W. Pontoon and CrossmoUna '00— P.

28 Sligo Ballysadare '00, Lough Key— P.

29 Leitrim Lough Melvin and Lurganboy '99 — P.

80 Cayan Lough Gowna '00, Lough Sheelin — P.

81 Louth Kavensdale '00— P.

83 Ferman. Lower L. Macnean '00, Castle Coole— P.

84 Dongl. E. Brown Hall '00— P.

86 Tyrone Lough Muck '97, Cookstown— Miss Knowles.
88 Down Kear Holywood '85— P.

39 Antrim White Park Bay '97, Dunloy— P. Belfast (Stewart)
— Linton jor.

A. alpeitria Schmidt.

8 Cork W. Gurtavehy '00, Skibbereen '89— E. A. Phillips.

10 !Rpp. N. Youghal Bay '99— P.
16Galw. W. Eecess'94— P.

18 Kings AboTe Kinnitty '00— P.

26 Mayo £. Near Claremoms '00— P.

27 Mayo W. Castlebar '96— E. S. Marshall.

28 Sligo Mullaghmore '00, Keishcorran, Lough Gill — P.

29 leitrim Glenade and Lough Gill '99— P.

80 CaTan Ballyoonnell '00 — A. Somerville.

81 Louth Carlingford Mountain '00-»P.

83 Fennan. Florencecourt '00,*arragh Creagh,. L« Mcdvin— P.

Proceedings of the Royal Irish Academy.

35 Dongl. V. Lough Salt {Eerh, Brit. Mus.)— Linton jaz.

36 Tyrone Omagh '97 — MIbb Knowles.

38 Down Saintfield '96 : common (Waddell)— W.B.E.C.'96-7.

39 Antrim Dunseyerick '97 — ^P, J^equent.

A. fllicanlis Buser.

4 Cork Mid Dripsey '89— R. A. Phillips.

5 Cork E. Fermoy '50 (T. Chandlee)— JT^jtJ. S. & A. M.

6 Waterfd. Cappoquin '99— P.

7 Tipp. S. Fethard '97, Lough Muskry '00— P.

8 Limrck. Adare '99^^omemlle. Capantimore— JJ^rft. L.F.C.

10 Tipp. N. Ballingarry '00 : common — P.

11 Kilkny. Kilmacow '00 — Miss Home. Ballyragget — P.

13 Cariow Goresbridge and Ballintemple '99 — P.

14 Queen's Grantstown '98, Arderin '97— P.

15 Galw. SE. Chevy Chase '00 : common— P.

16 Galw. W. Moycullen, Bolbeg Ferry, '99— P. Clonbur.

17 Galw. NE. Knockmae '99, Dunmore, Barbersfort, &c. — P.

18 King's Einnitty '00, Shannon Harbour — P.

19 Eildare Ballymore '00, near Baltinglass — P.

21 Dublin Kilteman '94— P.

22 Meath Balliyor and Slieve Bregh '00— P.

23 Westmth. Knock Eyon and Lough Iron '99 — P.

24 Longford Longford '98 : common — P.

25 Roscomn. Eockville '99, Slieve Bane, Mote Park— P.

26 Mayo E. Ballinrobe — Mrs. Persse !

27 Mayo W. Mweelrea '82— Hart 380.

29 Leitrim Binn Lough and Carrick-on-Shannon '99 — P.

31 Louth Kearney's Cross '97 — P.

32 Monaghn. Drumreaske '00 — A. Somerville.
34 Dongl. E. Lag '98— H. C. Hart.

36 Tyrone Omagh '97 — ^Miss Knowles.

37 Armagh Tynan Abbey '92— P.

38 Down Scrabo HiU '87— P.

39 Antrim Cave Hill '98 — G. C. Druce. Common.

40 L'derry Benevenagh— S. A. Stewart. Frequent.

Eosa tomentosa Sm., var. scabriusoula (Sm.).

29 Leitrim Lurganboy '99 — ^P.
Also in 20, 33, 39— see CyheU !!•

T^A^ann^Oleanings in Irish Topographical Botany. 69

5 Ck>rk £.
7 Tipp. S.

11 Kilkny.

12 Wexford
16 Galw.W.

18 King's
26 Mayo E.
84 Dongl. E.

38 Down

39 Antrim

35 Dongl. W.
9 Clare

3 CorkW.

5 Cork E.
14 Queen's
16 Galw. W.
23 Westmth.
26 Kayo E.
35 Dongl. W.

38 Down

39 Antrim

40 L'derry.

38 Down
23 Westmth.

7 Tipp. 8.

19 Kildare

37 Armagh

38 Down

39 Antrim
iOUderry.

S. canina L., var. Intetiana Leman.

Mitchelstown '97 : frequent — ^£. A. Phillips.

Fethard '00— R. A. Phillips.

Kilkenny '00— R. A. PhilHps.

Near Wexford '96— E. 8. Marshall.

Clonbur '95— E. 8. Marshall.

Edenderry '96— P.

Clonbur '95— E. 8. Marshall.

Near Killygordon — Fl. Donegal.

Saintaeld '95— C. H. Waddell.

Common '93 — W. A. 8hoolbred.

yar. sphierica (Gren.).
Ardara — Fl. Donegal.

rar. sentioosa (Ach.).
Ballyvaughan '00— R. A. Phillips.

Tar. dnmalia (Bechst.).

Skibberecn '96 : frequent — Phillips.

Near Mitchelstown '97 : frequent^Phillips.

Abbeyleix '00— Phillips.

Clonbur '95— E. 8. Marshall.

Clonave *95 — Linton 50J.

Clonbur '96— E. 8. Marshall.

Oweebaira estuary — Fl. Donegal*

Saintfield '95— C. H. Waddell.

Glenarm '99 — Druce 28^* Common.

L. Neagh '99 — Druce ^^5. Eglinton — ^Mrs, Leebody.

f . vertieillacaniha (M6rat.)
Saintaeld '94— C. H. Waddell.

Tar. urbica (Leman).
Knock Body '95 — ^Linton 50J.

Tar. dumentorum (Thuill.).
Fethard '00— R. A. PhilHps.
Co. Kildare— Cy^. II.
Near Lough GiUy '98— Druce 28^.
KiUowen (8tewart)— iSif^/. Fl. NE.
Mazetown (8tewart)— /^upp/. Fl. NE.
Toomebridge '98— Druce 28^^

70 Proceedings of the JSopal Imh Academy ^

Tar. arratica Baker, f . casia Sm.
35 Dongl. W. Egliah Biver (E. J. Hanbury )—/'/. Donegal.

The above being the first attempt to show the distribution of
12. canina forms in Ireland, I have given all the reliable records of
which I have knowledge.

S. glanca Yill., var. coriifolia (Fr.).

39 Antrim Cave Hill '98— Druce 28s.

40 L'derry. Lough Neagh '98— Druce 28s.

CratsBg^ Ozyacantha L., var. ozyacanthoides (Thuill.).
Divisions 34, 35, 38. Kot distinguished elsewhere.

var. monogyna (Jacq.)
Divisions 12, 21, 23, 34, 35, 38. Probably common.

Pynu Halus L., var. acerba DC.
Divisions all except 2^ S, 4, 9, 12, Slf, S5, 36, 40.

var. mitis Wallr.
Divisions all except i, 2, 3, 4, 6, 9, 23, 25, 27, 34,-35, 40.

^Bryonia dioica Jacq.

22. Heath Thicket by the Boyne at Oldbridge '01— Miss R.
Smith!

Apinm nodiflomm Beichb. fil., var. oereatum Bab.

12 Wexford Near Wexford '96— E. S. Marshall.

16 Galw. W. South side of Lough Mask '95— M. & S. j^s-

21 Dublin Ireland's Eye '94— P. 736.

23 "Westmth. L. Owel and L. Derevaragh '95 — Linton 505.
35 Dongl. W. Bamelton and Xincashla — Fl. Donegal.

A. innndatnm Beichb. fil., var. Uoorei Syme.

17 Galw. NE. Biver Clare near Tuam '99— P.

22 Meath Navan '00— P.

34 Dongl. E. North-west of Ballyshannon — Fl, Donegal.

FrequentaboutDownpatrick and in the Bann basin, in 37, 38, 39, 40.
A remarkable variety.

Prabger — Oleaninga in Irish Topographical Botany. 71

Sambnens nigra L.^ var. ladniata L.

15 Galw. SE. Qort '00— R. A. HiiUips.

Oaliam palnstre L., yar. Witheringii (Sm.).

Divisiona 6, 8, 11, 12, 16, 21, 23, 25, 26, 28, 29, 30, 34, 87, 88, 89.

Of fifteen Irish gatherings of 0. palustre suhmitted to !ReY. £. S.
Manhall, only one is referred (and that douhtfuUy) to the type,
winch appears to he yery rare in Ireland.

YalerianeUa olitoria Poll., var. lasiocarpa Reichh.

12 Wexford Common hetween Greenore and Camsore '97 — E. S.
Marshall.

Matriearia inodora L., var. salina Bah.

5 CoA E. Ballycotton '96— R. A. Phillips.
12 Wexford Near Wexford '96— E. S. Marshall.

16 Galw. W. Salthill '00— R. A. Phillips. Roundstone— 50^.

20 Wicklow Kilcoole '94— P.

21 Dnhlin Howth '94— P.

39 Antrim Giant's Canseway '93 — W. A. Shoolhred.

Artemiiia ▼ulgaris L., var. coarotata Porcell.

16 Galw. W.
26 Mayo E.

I Ahout Clonhnr '95— E. S. Marshall.

Seneoio vnlgaria L., var. radiatna Eoch.

6 Waterfd. Bunmore East '01 — ^Mrs. Persse !
8 Limrdk. EUmallock '00— R. A. Phillips.

S. Jaeobsea L., var. flosonloBOs (Jord.).

6 Waterfd. Tramore '99— P,

10 Tipp. II. Dromineer '00 — R. A. Phillips.

16 Galw. W. Knocknagoneen *99— P.

22 Meath Korth of Laytown '96— P.

28 EOigo Strandhill '97— P.

34 Dongl. E. Bancrana (J. Hnnter) — Hart 40^.

38 Down Groomsport ! '86 — Stewart.

Also in 2, 9i| 1^, 81, 35— see CyheU II.
Frequent on coast sandhills : very rare inland.

72 Proceedings of the Royal Irish Academy.

S. i^naticuB L., var. pennatifiduB Gren. et Godr.

16 Galw. W. Maam and Clonbur '95— M. & S. S4S'
26 Mayo E. Cong '95— M. & S.5^5.

Cnicns palnstris x pratemdB.

Plants found in several parts of Ireland conyinced me that tliis
hybrid is of not infrequent occurrence. In 1900, in a damp pasture
north-west of Claremorris, a large colony of plants was found which
exhibited every grade from C. palustris to C, praiensis, both of
which were present in abundance. Of specimens midway in the
series from this locality Mr. Arthur Bennett writes : — "19. 12. '00. I
have to-day compared the Carduus specimens at the British Museum.
Among the British specimens there is nothing so extreme as your
specimens. The nearest is an Irish specimen. Watson's Forsteri is
an anglieum [= pratemis] with its tenuity of the leaves retained—
yours has more the harslmess of palustris. In the general collection
at the British Museum there is nothing like it ; but when I showed
Mr. S. Moore the specimens he said "I think a|hybrid." Clearly,
your specimens retain to the heads more the characters of pratensis
than palustris. Did C, tub&rosus occur in Ireland I should have been
much inclined to name one of your specimens (i,e. 3. 7. '00 N. of
Claremorris) C. palustris x tuherosus. You will say — well, then,
what do you name the specimens after all ? I answer, though with
some doubt — a hybrid as you make them, probably C. pratensis
X palustris, but more extreme than any I have seen of English
specimens so named." The conditions under which these plants
occurred do away in my mind with any doubts which might hang
around the few dried specimens submitted; and I group with this
gathering other plants obtained in Queen's County and Carlow. As
regards certain previous records of this hybrid, no valid reason
appears for excluding them. Eev. E. S. Marshall confirms (5^/)
Mr. Levinge's Westmeath record {484), and also Rev. E. F. Linton {in
litt,) ; and if Mr. Levingc, a discriminating observer, is correct about
his Westmeath plant, he is probably also correct about his Clare one.
I think, therefore, that in recording the undermentioned stations for
C, pratensis x palustris we are on tolerably safe ground : —

9 Clare Lisdoonvama '92 — H. C. Levinge.

13 Carlow Below St. Mullins '99— P.

14 Queen's Mountrath '97— P.

Prabobr — Okaning% in Lnsh Topographical Botany, 73

28 Westmth. Lough Owel— H. C. Levinge 484,

26 Mayo E. North-west of Claremorris '00— P.

40 L'derry. Oarvagh and Jackson Hall (D. Moore) — Cyh. I.

C. arvenns Hoffm., var. mitis Koch.

3 Cork V. Glengarifl '90— Druce 284.
40 L'derry. Toomebridge '98 — 0. C. Druce. Aghadowey '95
(WaddeU)— W. B, E. C. '95-6.

yar. horridiu (Adam).

1 Keny 8. Eenmare '90 — ^Druce 284.
11 Kilkny. Ferrybank '95— M. & S. ^45-

Tar. setOBUS Bess.

27 Mayo W. Gortnaiaby '00— P.
34&35Dongl. Frequent— see Fl. Donegal,
38 Down Lambeg '00 — Davies 2^2.

40 L'derry. Between Kilrea and Garvagh '98 — Stewart and Miss
Enowles.

^Lactnca virosa L.

8 Limrck. Waste ground at limerick ! '00 — K. D. O'Brien.
Holding its own in two stations here.

nmoaenm oflcinale Web., yar. erythrospermnm (Andrz.).

5 Cork E. Near Youghal '00— R. A. Phillips.

6 Watrld. Tramore '99— R. A. Phillips.
21 Dublin Portmamock '01— P.

yar. palustre (DC).

8 Limrck. Near Limerick '99 — A. Somerrille.

9 Clare Near Ballyvaughan '99 (Playlair)— W.B.E.C. '99-'00.

26 SITe'^ } ^^^ ^^ ^^^^ ^^®^ '95— M. & S. S45'

29 Leitrim Drumcoura Lough '00— A. Somendlle.

30 Caran Ballyconnell '00 — A. Somerrille.
32 Monaghn. Mullyash Hill '00— A. Somenrille.

35 Dongl. W. Fanet, &c. : north only — Fl. Donegal.

Also in 1, 4, 19, 20, 21, 28, 36, 38, 39— see CyheU II.

74 ProceedingB of the Royal Irish Academy.

var. ndum (Jord.).
12 Wexford Camsore '97— E. S. Marshall.
16 Oalw. W. Near Clonbur '96— E. S. Marshall.
^3 Westmth. Knock Drin '95 — Levinge. Caatietown — ^Marshall.

26 Mayo E. Near Cong— Marshall '99 ^41.

27 Mayo W. Mallaranny '99— E. S. MarshalL

Also in 1 6-^see Cyh$U II.

Sonchiu arventiB L., var. angnttifolia Mey.

26 Mayo E. S.E. end of Lough Mask '95— M. & S. 5^5.

Tragopogon pratenBe L., var. minus (Mill.).

Diyisions 7, 8, 15, 18. I believe this is the usual Iriah form, but
information is lacking.

Jasione montana L., var. major £och.^
16 Galw. V. Mount Gable '95— M. & S. j^j.

Statice auriculsBfoUa Yahl., var. intermedia Syme.
21 Dublin Howth, Killiney, &c. (Hart)— i^. Donegal.

EryihrflBa Centaureum Pers., var. capitata Eoch.
3 Cork W. Baltimore '96 : frequent on coast — ^PhiUips.
9 Clare Lahinch '01— Miss E. Armitage.

27 Mayo W. Mallaranny '99— E. 8. Marshall.

Symphytum officinale L., var. patens (8ibth.).

3 Cork W. Skibbereen '97 : frequent— Phillips.

4 Cork Mid. Frequent— Phillips '00.

1 1 Eilkny. Graiguenamanagh '00 — Phillips.
15 Galw. SE. Portumna '00— Phillips.

*S. tuberosum L.
Divisions 4, 5, 21, 22, 36, 38.

*Borago officinalis L.
Divisions 3, 4, 5, 6, 8, 12, 14, 21, 22, 26, 38, 39, 40.

^Antirrhinum miyus L.
Divisions 4, 5, 7, 11, 21, 25, 31, 35, 38, 39.

*Linaria purpurea L.
Divisions 4, 11, 12, 21.

'PRABOVSt-^OJeanings in Irish TopoffrapMcal'Botant/. 75

Seroplmluia aqnatioa L., var.einerea Bum.

27 Mayo W. Newport '99— E. S. Marshall.

And in 16, 23, 2&;-6ee Cyhele 11.

Veronica Anagallia L., var. anag^alliformifl Bor.

12 Wexford Wexford *96— E. S. Marsliall.

16 Galw. W. Clonbur and Cong frequent '95— M. & 8. 5^5.
23 Westmth. Knock Brin and Scraw bog '95 — ^Levinge.

26 Mayo E. Cong frequent '95— M. & S. 5^5.

27 Mayo W. Mallaranny district frequent '99— Marshall.

Euphrasia officinalis L.

Except for a few gatherings chiefly by English visitors, Irish Eye«
brights have as yet, except for my own collecting, been almost
nnworked. I am under a deep debt of gratitude to Mr. F. Townsend,
who has identifled all the plants which are recorded in my name
below. As no attempt has been made except in Mr^ Townsend's
Mono^aph (1897) to give the Irish distribution of the Syebrights, I
bare included all reliable records of which I have cognisance.

)E. ftricta Host.

6 Wateifd. BaHyscanlan Lough '99— P.

8 limrck. Mullagh '00— P,

10 Tipp. N. Cloughjordan '00— P.

15 Galw. 8E. :Rinville House '99— P.

17 Galw. NE. Menlo and Billower Lough '99— P.

18 King's Clonmacnoise '99— P.
27 Mayo W. Ballina '00— P.

E. borealis Wettst.

12 Wexford Between Oreenore and Churchtown '97 — Marshall jjg.

16 Galw. W. Oughterard '99— P. Connemaia '5» (F. Edrk)— <^90.

19 Eildare South of Kildare '97— P.
22 Meath Laytown '96— P.

24 Longfd. Common by Lough Bee '00 — P.

27 Mayo W. Ballina, Eosserk, Berreen, Crossmolina, '00 — P.

33 Fennan. Lower Lough Macnean '00 — ^P.

Proeeedinga of the Royal Irish Aoademy.

7 Tipp. 8.

8 Limrck*

10 Tipp. N.

11 KilVny.

13 Carlow

14 Queen's

15 Galw.SE.

16 Galw. W.

17 Galw. NE,

18 King's

19 Eildare

21 Dublin

22 Meath

24 Longford

25 Eoscomn.

26 Mayo E.

27 MayoW.

28 SHgo

29 Leitrim

30 Cavan

31 Louth

36 Tyrone

38 Down

39 Antrim

40 L'derry

E. brevipila Bumat et Oremli.

Thurles '98— P.

Askeaton '00— P.

L. Ouma '99, Keeper Hill, Devil's Bit, Cloughjoidaii

Portunma — P.
Tory Hill '99, Urlingford '98— P.
Borris *98, Bagenalstown '97— P.
Abbeyleix, '98, Mountrath '97— P.
Gairylandand Chevy Chase '00 — P.
Oughterard and Gentian Hill '99— P.
Tuam, KeekiU, Oranmore, Menlo '99 ; Clonbrock—P.
Clanmacnoise '99, Edenderry '96— P.
Bathangan '98, Bathmore, Leizlip, Kilcock, Kibneage,

Carbury — P.
Near Tallaght '97— P.
Laytown and Enfield '96— P.
Castlerea '00— P.

Slieve Bane '00, Arigna, Corkip Lough, Athlone— ?.
FW. of Claremorris '00— P.
Mallaranny district common '99 — ^Marshall. Ballina

district common '00 — P.
Carrowkee Hill '97, Lough Gara, Inismurray — P.
Ballinamore '00, Garadice L., Glenade, Binn L., L.

GiU— P.
Mount Nugent '96— P.
Ballymascanlan '00, Bavensdale, Ardee, Togher,

Lurgan Green — P.
Mullaghcam and Omagh '96 — ^Miss Knowles.
Greypoint '85— P.

Glenarm and Cave Hill '98— G. C. Druce.
Toomebridge '98— G. C. Druce.

f. subglahra.

10 Tipp. N. Cloughjordan '00— P.
26 Mayo E. East of Foxford '00— P.

f. eglandulosa.
17 Galw. NE. Menlo '99— P.

Pkabobr— Gfeiniwjr* in IHsh Topographieal Botany. 77

f. subefflandulosa.
10 Tipp. N. Devfl'e Bit '98— P.
19 Kildare Aboye Baihmore '98— P.

E. brevipila x Eosfkoviana.

27 Kayo W. Lough Conn south of Derreen '00 — P.

The eame hybrid is doubtfully named from 11, 15, 23, 31.

E. nemorosa Mart.

10 Tipp. K. Cloughjordan '00— P.
15 Galw. Sfi. Lough Derg '96 (N. Colgan)— A. Bennett.
23 Westmth, Lac Lean [Lough Lene ?] — see Townsend Sgo.
M Fennan. Florencecourt '00 — P.

Also recorded from " Mayo " in Wettstein's Monograph.

f. tetraquetra,

9 Clare Murrough '95 (N. Colgan)— A. Bennett.

22 Meath Oldcastle (W. S. Millar)— Townsend 8go,

E. curta Fries, yar. glabresoens Wettst.

9 Clare Lahinch '01— Miss E. Armitage.
17 Galw. NE. Enockmae '00— P.

38 Down Newcastle '98—0. C. Druce.

33 Antrim Lough Neagh '98— G. C. Druce.

E. occidentalis Wettst.

5 Cork £. Poorhead [» Power Head] '95 (Phillips)— Townsend

8go.

6 Wateifd. Tramoro '99— P.

W Down Templemore [« ToUymore] Park '98— G. C. Druce.

E. gracilis Fries.

7 Tipp. 8. Fethard and near Cashel '98— P.

10 Tipp. N. Between Deyil's Bit and Ballyhoul '98— P.

11 dkny. Inistioge '98— P.

17 Gdw. HE, Annaghdown '00, Keekill '99— P.

22 Meath OldcasUe '96— P.

23 Wcsfcnith. Athlone '98 — ^P. L. Dereyaragh — Levinge and Qroyes
2a Sligo BaUysadare '00-P.

Rroaedingn of ihe Royal Irish Academy.

29 Leitrim Lough Melvin '99— P.
32 Monaghn. Creeve Lough '00 — P.
39 Antrim Near Ballintoy '93— W. A. Shoolhred.

A few records are withheld,' as the determinations were made prior
to the recognition of so many forms in these countries.

E. Bfottioa Wettst.

7 Tipp, 8. Thurles '98— P.

13 Carlow South of Carlow '98— P.
16 Galw. W. Boss Lake '99— P.

19 Kildare Hill above Rathm<»e '98— P.

25 Roscomn. Athlone '98 — P.

26 Mayo E. East of Foxford '98— P.

27 Mayo W. Mallaranny '99 (E. 8. Marshall).— A. Bennett. Ballina,

Bathroeen L., frequent by Lough Conn '00 — ^P.
31 Louth Carlingford Mountain '00 — P.

E. Sostkoyiaaa Hayne.

1 Kerry S. Dingle '53 (D. Oliyer)— Townsend 8go.
6 Waterfd. Dungarvan and Kilmacthomas '99 — P.

8 Limrck. Doon '00 — P.

10 Tipp. N. Dromineer '99, Devil's Bit '98— P.

11 Eilkny. Thomastown, Inistioge, Barleeagh Wood, '98 — P.

13 Carlow Killedmond River '99— P.

14 Queen's Rathdowney '98, Emo '96— P.

15 Galw. 8E. Marble Hill '97, Chevy Chase '00— P.

20 Wicklow Scalp '72— JST^rJ. G. Pim.

22 Meath Oldcastle '96— P.

23 Westmth. Moate '99, Athlone — P- Bog of Lynn — Levinge.
25 Roscomn. Arigna '00 — P.

28 Sligo Ballysadare '00— P.

36 Tyrone Baronscourt '96 — Miss Enowles.

E. Salisburgenflis Fiink.
Divisions 8, 9, 15, 17, 26, 29— see CyheU and IrUh Top. Bot.

E. SalisburgensiB x brevipila.

18 King's Lough Goura '98— P.

** JE'. Salishuryensis x hrevipih^ I doubt not." — T. Townsend.
Mr. Townsend favours me vrith a- formal descriptioB of this . se-w

Praboer — Oleanings in Iriah Topographical Botany. 79

hybrid, which, he remarksy is an exceedlDgly interesting plant. He
is ftnzioiu to see further specimens, and I hope any botanist visiting
King's CoTinty will not lose the opportunity of searching for it. The
finding of the parent E, Salishirgensis in King's County would alone
be a yaluable discoyery.

Of fourteen •* species " of Eyebright — the " species " seem as well
founded as those of the £ubi — known as British, Ireland possesses
ten. Of these, two, JS. hrevipila and JE. 'JRostkoviana,^ appear to
constitute the bulk of the Irish Eyebright flora. JE, hrevipila is very
widespread in Europe, but in Great Britain appears to increase north-
ward; while E. jRostkoviana, a plant of middle Europe, in Great
Britain increases southward. For the rest, £. horealis, JE. gracilis^ and
E. $eottica are northern in their range, though not so in Ireland,
r. nemoroia is southern in Great Britain ; JE". oecidentalis appears to be
a very local plant occurring in N.W. France and Britain, and £!.
itrieta, a lowland mid-Europe form, seems to be a rare plant in these
cuuntries. The well-known JE. Salishurgenais, a northern and alpine
plant, is as yet unkncrwn in Great Britain, though ranging widely
along the west coast of Ireland. Of the four British forms not yet
found in Ireland, J?, latifolia and E, Foulamsis are high northern, and
i*. Kermri a limestone plant of limited distribution. With the
exception of E. Salishurgmsis /none of the Irish Eyebrights show a
marked range either horizontally or vertically, though the species
differ much in abundance.

BhinanthuB Crista-galli L., Yar..fjallax W. & G.

J 6 Galw. W. Between Clonbur and Mt. Gable '95— M. & S. 5^5.
23 Westmth, Lough Owel '95 (W. R. Linton)— -B. E. C. '95.
39 Antrim Near Toomebridge '98 — Druce ^Sj.

Tax. S. Btenophyllus Schur.
25 Roeconm. KiltecTan*97— T. A. P. Mapother.
Rhirmtthus forms are almost unworked in Ireland.

Bartsia Odontites Huds.

Tar. wma is known from divisions 5, 8, 11, 12, 26 s var. serotina
trom 5, 12, 21, 39. Further information is needed as to their distri-
bution in Ireland.

80 Proceedings of the Royal Irish Academy

^Ueiitha alopecnroides L.

38 Down Dmmy Water Bridge '90 — ^P. This is the M. roiunix-

Jolia of Hart j<9/, and of Stewart & Praeger <?7j.

^TL longifolia Huds.

5 Cork E. Dodge's Glen (J. Sullivan)— Allin ig.
1 1 Kilkny. Callan '00— E. A. Phillips.
15 Galw. SE. Gort '00— K. A. Phillips.

Also in 1, 3, 21— see CyheU II.

*Tt. yiridis L.
Divisions 4, 5, 18, 23, 34, 38.

U. hirsnta Huds., yar. subglabra (Baker).
36 Tyrone By Lough Neagh '96 — Miss Knowles.

U. sativa L., var. palndosa (Sole).

15 Galw. SE. Woodford Biver '85— Linton ^02.

18 King's TuUamore and Clara '99— P.

23 Westmth. L. Derevaragh '95 (W. R. Linton)— B. E. C. '95.

24 Longfd. Near Ballymahon '00— P.

Also in 2 and 40 — see CyheU II.

var. Bubglabra Baker.
14 Queen's Graigue '99— P.

var. rubra (Smith).
Divisions 4, 12, 13, 38.

var. graoilifl (Smith), f. oardiaoa.
7 Tipp. S. Fethard '00— R. A. Phillips.

M. verticillata L. =» arrensis x aquatioa.
40 L'derry Toomehridge '98— G. C. Druce.

K. Pulegiom L., var. erecta Syme.
1 Kerry S. Gallarus '87 (A. 'L&Y)—Herh. Glasnevin

StachyB palustris x sylvatica.
4 Cork Mid Blarney (R. MiUs)— AUin ig.

11 Kilkny. BaUyiagget '98— P.

12 Wexford, Courtown '94— Mrs. Tattow !

Prasgbr — Okaninga in Irish Topographical Botany. 81

13 Carlow Borrifl '98— P.

19 Eldaie Leixlip '96— P.

20 Vicklow Glen of the Downs '93— P.

22 Meath Athboy '00, Maynooth, Beauparc, Oldcastle — P.

24 Longfd. Newtowncaahel '00— P.
31 Louth Togher '96— P.

85 Dongl. W. Near QrweedLorer—Fl, Donegal.

86 Tyrone Newtownstewart '96 — Miss Knowles.

Also in 5, 21, 37, 38, 39, 40 — see Cghele II.

Usually nearer palutiris than iyhatiea, and cannot then he referred
to S. amhigua Smith.

Plantago CoronopnB L., yar. pygmssa Lange.
27 Mayo W. MaUaranny '99— E. S. Marshall.

«Chenopodiimi mnrale L.
DiTidons 3, 5, 8, 21, 22, 39 ; recently seen in all except the first two.

Polygonum ConyolvalnB^L., yar. subalatnm Y. Hall.

16 Galw. W. Oughteraid '95— M. & S. 5^5.

31 Louth Mouth of the Boyne '96— P.

3S Down Saintfield '93, Magheralin— Waddell 8g6.

P. avicnlare L., yar. arenaatrum Syme.
Diyisions 6, 21, 27, 38.

yar. littorale (Link.).
Divisions 21, 22, 31, 38.
P. Persicaria L., yar. glandnlosa Y. Bosch.
38 Down Ballynahinch '86— P.

var. incaniutt auct,
88 Down Warrenpoint town reseryoir '90 — P. This is the P.

lapatki/olium of Stewart & Praeger S^j.

Euphorbia Cyparassias L.
19 Kildare Curragh '97— P.
24 Longfd. Saint's Island '99— Miss B. Smith !
In wild ground in these stations.

K- K A. nOC., TOL. Tin., IBC. B.] O

82 Proceedings of the Royal Irish Academy.

Betnla pubescexu Ehrli., yar. denudata Oren. et Godr.

8 Limrck. Thomfields bog '01 — Mas E. Armitage !
20 Wicklow Altadore '93— P.

Salix triandra x fragilis = 8. deoipieni Hoffm.
32 Monaghn. Lough Ayaghon '00 — ^P.

8. pentandra x fragilis » 8. onspidata Scholtz

16 Galw. W. *Maam '99— Marsliall 5^/.

19 Kildare fKilcuUen '97— P.

26 Mayo E. *Cong '99— Marsliall 5^/.
A rare hybrid.

8. aurita x cinerea >s 8. lutescens A. Xem.
3 Cork W. Inchigeela '97— R. A. Phillips.
7 Tipp. S. Between Fethard and Cashel '98— P.
16 Galw. W. Maam '95— M. & S. s^j.

8. nigricans x anrita = 8. coriacea Forbes.
23 WcBtmth. Knock Drin neighbourhood '95 — ^Linton 50J.

8. vinunalis x Capress « 8. 8mithiana Willd.

Divisions all, exetpi 2, 9, 12, 15, 15, 17, 26, 27, 31.

S. viminalis x cinerea.

7 Tipp. S. Between Cashel and Fethard '98— P.

20 Wicklow Shillelagh '99— P.

22 Meath Kildalkey '00— P.

8. Timinalifl x anrita « 8. fruticosa Doell.
1 1 Kilkny. By the Nore above Ballyragget '99 — ^P.

8. viminalis x Caprea.

6 Waterfd. Ballyscanlan Lake '99 — P.

18 King's Clara '99— P.

23 Westmth. Quarry bog '95 — ^Levinge. Near Athlone '99 — P.
25 EoBcomn. Bockville '99 — ^P.

*8. porpnrea x yiminalis » S. rubra Huds.
I have gathered this in divisions 8, 26, 28, 29.

Praboer — Gleanings in Irish Topographical Botany. 83

Popnlus tremnla L., Tar. glabra Syme.

8 Limrck. f Adare '9.9 — ^A. Someryille.
16 Galw. W. Clonbur '95— E. S. Marshall.

Iris Fseud-aconu L., Tar. acoriformiB (Boreau).

27 Majo W. Mallaraimy district abundant '99 — Marshall.
Appazently the usual Irish form.

Allium vineale L., Tar. oompactum (ThuilL).

8 Limrck. Limerick '99 — A. Somerrille.

Juneus effosufl x glaucus » J. difhunu Hoppe.
7 Tipp. 8. Carrick-on-Suir '00— R. A. Phillips.
Also in Dublin only.

Luzula ereota Desr.

f. umbellata in 5, 10, 15, 18, 29, 86, 37, 38, 39!

1 eongesta in 1, 5, 6, 8, '10, 15, 16, 18, 29, 38, 39, 40.

Both are probably uniTersal.

Arum maculatum L.
The spotted-leaTed form, from which the species deriTCS its name,
is ipparently Tcry rare in Ireland. The only notes I haTO are : —

15 Galw. 8E. Portumna '99— Mrs. Joyce. Dunsandle '99— P.
21 DuUin Glencullen '02— G. H. Pethybridge I

Typha latifolia L., Tar. media Syme.

9 Clare Tnishmaan '90 — Nowers and Wells dji.

Bparganium ramosum Huds., Tar. miorooarpum Neum.

16 Galw. V. Maam and Clonbur frequent '96— M. & 8. 5^5.
26 Mayo E. Frequent at Clonbur '96 — ^M. & 8. S4S'

8B Down Loughanisland (Waddell)— 8. & P. 874.

Fotamogeton crispus x obtosifoliuB » P. Bennettii Fryer.
37 Armagh Canal below Caledon '92— P. See Irish ML, n., 182.
This hybrid was described by Alfred Fryer in 1895 {Joum. Bot,
«™ii, p. 1, tab. 348) from Grangemouth, Stirlingshire, whidh still
I the only other British station.

84 Proceedings of the Boyal Irish Academy.

P. luceiLB L., yar. aonminatus Fr.
28 Westmth. Lough Derevaragh '92 — H. C. Leyinge.

P. pnflilliiB L., Tar. teniiiBsimiiB Koch.

12 Wexford Near Wexford '96— Marshall 5^7.

34 Dongl. E. Doagh Island '98— H. C. Hart.

37 Armagh Armagh '76 (S. A. Stewart) — Eerh. A. Bennett.

38 Down Saintfield '00 (Waddell)— W.B.E.C. 1900-1.

Zostera marina L., yar. angfostifolia Fr.

12 Wexfoid Wexford Harbour '96— E. S. Marshall.
15 Galw. 8E. Kinyarra '00— P.
lOGalw. W. Gentian HiU '00— P.

21 Dublin Malahide inlet '99— P. Baldoyle— JEri. 8. & A. ¥.

22 Meath Nanny Biyer '96— P.
27 Mayo W. Near Killala '00— P.

31 Louth Boynemouth '96— P.

34 Dongl. E. Trawbreaga Bay '98— H. C. Hart.

38 Down IStrangford Lough, common — P.

Carex teretiuBCula Good., yar. Ehrhartiana (Hoppe).
17 Galw. NE. Near Clonbrock '96— P.

Also in 34 and 35— see Fl. Donegal.

C. Ooodenovii J. Gay, yar. elatior (Lange), f . angostifolia.

39 Antrim Harbour Island '98— G. C. Druce.

yar. junoella Fr.

5 Cork E. Toughal '96— R. A. Phillips.

7 Tipp. 8. Thurles '00— R. A. Phillips.
15 Galw. 8E. Woodford '00— R. A. Phillips.
17 Galw. NE. Donamon '97— P.

32 Monaghn. Bessmount and elsewhere '00 — Waddell.

34 Dongl. E. Near Ballyshannon — Fl. Donegal.

35 Dongl. W. Milf ord Lake and Glenties- ^. Donegal.

Also in 1, 4, 12, 16, 23, 26, 39— see CyheU II.

C. extensa Good., yar. ptunila Anders.
12 Wexford Wexford Harbour '96— Marshall 5J7.
27 Mayo W. Mallaranny '99— E. 8. Marshall.

P&AB6BB — Oleaninga in Irish Topographical Botany. 85

C. flava L., yar. (Ederi Retz.

Hr. Marshall finds this in 12, 16, 26, 27~in all the localities he
has worked, so it is probably common in Ireland.

var. minor Towns.

17 Oalw. NE. Ellower Lough '99— P.

27 Mayo W. liallaranny district abundant '99 — Marshall.

40 L'dcrry Lough Beg '98— G. C. Druce.

C. flava X fdlva.

3 Cork W. Gurtavehy Lake '00— R. A. Phillips.

leOalw. W. EossL. '99— P. NW. side of L. Corrib—M. & 8.5^5.

26 Mayo £. S. side of L. Mask '95— M. & 8. 5^5.

37 Mayo W. Newport and west of Castlebar '99— Marshall.

37 Armagh Derryadd Lough '92— P.

Agrostis alba L., yar. stolonifera (L.).
DiTisionB 6, 11, 13, 16, 17, 22, 31 — P. No doubt common.

yar. maritima Mey.

3 Cork W. Baltimore '96 : common on coast — Phillips.

21 Dublin Portmamock '90 — Druce 284.

22 Meath Laytown '95— P.

yar. coarctata (Hoffm.)

39 Antiim Islands in Lough Neagh '98 — 0. C. Druce.

40 L'deny Common near Lough Beg '98 — Q. Q. Druce.

A. Yulgarifl With., yar. pumila (L.). ^

Diyisions 1, 3, 4, 5, 8, 12, 13, 14, 20, 21, 24, 26, 27, 29, 33, 36,
38, 39, 40. Merely a state induced by staryation.

Arrhenafhemm avenaceum Beauy., yar. nodosum Beichb.

S Umrck. Thomfields '01— Miss E. Armitage.

23 Westmth. Knock Drin — Leyinge 484,
36 Tyrone Omagh '96 — Miss Knowles.

Phngiiiitet commiinia Trin., vir, nigricans Gren. et Godr.
23 Wsrtmth. Lough Owel '95— H. C. Leyinge.

86 Proceedings of the Royal Irish Academy.

Koderia cristata Pens., var. graeiliB (Bor.)-

26 MavoE I South of Lough Mask '96 — ^M. & S. 5-^5.

Sederia Oflemlea Aid., Tar. luteoalba Opiz.
Diyisioiis 9, 15, 17 : apparently frequent where the type occurs.

Poa pratensifl L., var. subcamlea (Sm.).

12 Wexford Rosslare '97— E. S. Marshall.

21 Dublin North BuU '00— R. A. PhilHps.

23 Westmth. Bog of Lynn '95— F. C. Levinge.

27 Mayo W. Mallaranny '99— E. 8. Marshall.

Also 34, 35 — see Fl. Donegal,

P. trivialis L., yar. Koeleri (DC).

12 Wexford Rosslare '97— E. 8. Marshall.
27 Mayo W. Mallaranny '99— E. 8. Marshall.
29 Leitrim Glenade '99— P.

Tar. glabra Doell.
12 Wexford South of Greenore '97 — ^Marshall 5J9.

Festaca ovina L., yar. eapillata Hackel.

3 CorkW. Glengariff '90— Druce -?<^^.
16 Galw. W. Maam abundant '95 — M. & 8. j^/j.

F. elatior L., Tar. pratensis Huds.

DiTisions 8, 11, 16, 18, 19, 20, 21, 24, 26, 28, 29, 33, 34, 36, 36,
38, 39, 40. Probably in all diyisions.

F. elatior x Lolium perenne.

18 King's Pallas Lough '00— R
31 Louth Boyne mouth '96 — ^P.

34 Dongl. E. Innishowen Head and Guldaff (Dickie} — Fl. JDoneyal.
Also 13, 17, 23, 39— see Cyheh II.

BromuB giganteuirL., Tar. triflonu Syme.
DiTisions 36, 38, 39. Frequent?.

Praeoer — OUanings in Imh Topographical Botany. 87

Agropyron repeiu Beauv., yar. barbatnm Duyal-Jouye.

11 Kilkny. Ballinlaw Perry '99— P.

21 Dublin Portmamock '94— P.

31 Louth Queensborough '96 — P.

36 Tyrone Omagh '96— Miss Knowles.

*Lolinm perenne L., yar. mnltiflonun (Lam.).
36 Tyrone Strabane '96— Miss Knowles.

*yar. italienm (Braun).
Diyisions 8, 12, 32.

Ceterach offleinanun Willd., yar. orenatnm Milde.

Diriaions 4, 6, 6, 7, 9, 10, 11, 13, 16, 17, 18, 20, 22, 24, 25, 26,
27, 29, 38, 39. Generally occurs wbereyer tiie plant grows strongly.

Aspleninm Adiantnm-nignun L., var. acntnm (Bory).

5 Cork E. Ifear Whitegate '00— R. A. Phillips.
1 1 Kilkny. SnowhiU '99— P.

Also in 1, 2, 3, 4, 8, 16, 20, 26, 27, 38— see CyleU II.

Laatrea Tiliz-mas Presl., yar. affinia Bab.

Krittons 6, 7, 8, 10, 11, 12, 13, 15, 16, 17, 18, 19, 20, 21, 22, 28,
24,25, 26, 27, 28, 29, 30, 31, 32, 33, 37, 38, 89, 40.

yar. paleacea Moore.

DiTirions 1, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21,
22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 37, 38, 39, 40. Both
are, no doubt, nniyersal.

yar. abbreviata Bab.

The only locality whence I haye seen undoubted ''Lastreapro-
pinqna" is the Moume Mountains. Mr. Hart has recorded it from
Btindon and Mount Leinster {38 1)^ and states that this is the usual
fona on the upper part of the Donegal mountains {FL Donegal).

L. dilatata Presl., yar. oollina Bab.
I Kerry 8. Brandon '87 (A. Ley)— B.E.C. '87.

88 Proceedings of the Royal Irish Academy,

Polypodinm vnlgare L., var. serratnm Willd.

Diyisions 4, 5, 8, 9, 11, 23, 28, 39. Not anfrequent in limeBtone
districts when the plant is growing strongly. It develops into var.
semilaeerum and that again into Tar. eambrteum.

Eqmsetnm limosnm Sm., var. flnviatile (L.).

Divisions aU except 3, 4> 5> ^, -?-?, ^0, 32^ 34, 35. The usual form
in shade or in running water.

£. palustre L., var. polystachyum auet.

8 Limrck. Thomfields '01 — Miss E. Armitage!
15 Galw. 8E. Marble HiU '97— P.

Merely a sport, produced by exuberant growth or more often by
injury to the axis.

var. nudum Newm.
29 Leitnm Lough Melvin '99— P.

A sport, consisting of an absence of branching, the result of exposed
habitat.

CH AivAC£iE«

So much material was obtained during my five years' field-work,
that I have in most cases supplemented my notes with brief references
to additional records, as given in Messrs. Groves' papers, &c., so as to
make the following a complete account of the distribution of Charaoee
in Ireland so far as present information goes. In Irish Top, JB<d.
the records given under each species include those of its varieties.
In the following lists I have separated typical forms from varieties.
My warm acknowledgments are due to Messrs. H. and J. Groves for
naming or confirming the very large number of the plants listed below,
with which my name is associated.

Chara fragilis Dcsv.
Type— Divisions all, except S9, 35, 40.

var. barbata Gant.

1 Kerry N". Near Yentry '94— D. McArdle.

9 Clare Killaloe '96— Colgan 2ig,

Pbabgbb — Okanings in Irish Topographical Botany. 89

11 Elkenny TTrlinglord '98— P.

14 Queen's Bathdowney '98— P.

15 Galw. SE. Loughrea '98 : widespread and frequent — P.

17 Qalw. NE. Clonbrock '96— P.

18 King's Banagher '98, Shannon Harbour, Geashill — P.

19 Kildare Bathangan '98, south of Kildare '97 — P.

20 Wicklow Muirough '95 (D. Moore) — Groves j^q.

22 Meath Kear Ardee bog '97— P.

23 Westmth. Loughanstown — Leyinge^<S^. Mullingar — Groves j^9.

24 Longfd. Canal at Killashee '98— P.

25 Eoscomn. Bockville '99— P.

26 Mayo E. Kilmovee '99— P. About Cong '95— M. & S. 5^5.

27 Mayo W. Lough Conn near Derreen '00 — P.
31 Louth Killencoole '96— P.

38 Down Holywood hills '91, Clandeboye, Craigauntlet — ^P.

39 Antrim Portmore L. ['46] (Thompson), Tardree Hill (Grainger)

— S. & P. S74.

var. oapillaoea Cobs. & G.

1 Kerry S. Cloonee Lough '98— P.

2 Kerry N. Long Range '87 (Scully)— Groves j^q.

16 Galw. W. Galway W. '75 (More), Renvyle — Groves j^j, j^g,

21 Dublin Howth '94— P.

29 Leitrim Glenade Lake '99— P.

31 Louih Soldier's Point '96— P.

32 Monaghn, L. Naglack '01— Bullock- Webster. Creeve L. '00— P.

33 Fennan. Three miles N. of Enniskillen '81-2 — Barrington 108.
38 Down Glasdrumman '98— Davies. Holywood '85— P.

var. HedwigiiKuetz.
32 Monaghn. Lough l^aglack '01— O. B. Bullock- Webster.
Abo in divisionB 16, 21/ 26, 28, 38 — see Ghroves j^g.

var. delioatnla Braun.

3 CoA W. Inchigeela '97-r-B. A. PhiUips.
14 Queen's Bathdowney '98, Maryborough '96 — P.
16 Galw. W. Oughterard,Moycullen,'99— P. Bccess ( Linton)— j^9.
n Galw. NE. KiUower Lough '99— P.

a. 1* A. PROC., vol., VIII.,

SEC. B.]

Proceedings of the Royal Irish Academy.

21 Dublin Clondalkin '94— P.

22 Meath Oldcastle '96— P.

23 Westmth. Lough Owel '01— G. R. Bullock- Webster.
26 Eoscomn. Lough Ree at Galey '97 — P.

26 Mayo E. Lough Corrib near Cong '95 — M. & S. ^4^,

27 Mayo W. Lough Cullin near Pontoon *00 — P.

28 SHgo Lough Gill '99— P.

29 Leitrim Glenade Lough, Annaghearly L., L. Melvin, '99— P.

81 Louth Braganstown bog '97 — P.

82 Monaghn. Greaghlone Lough '01— G. R. Bullock- Webster.

33 Ferman. Blaney Bay on Upper L. Eme '81-2 — Barrington loS,

84 Dongl. E. Bundrowes Riyer — Fl. Donegal,

35 Dongl. W. Tullyconnell L. '98 — Hart ^05. KinnyL. L.Sessiagb.

36 Tyrone Favour Royal '96 (Mrs. Leebody) — J. Groves.

37 Armagh Common in Lough Keagh '92 — P.

38 Down Lough Neagh '98— J. F. Davies.

39 Antrim Lough Beg '94 — P. Frequent in Lough l^eagh.

C. eonnivens Braun.
12 Wexford Lagoon north of Wexford Harbour '97 — E. S. Marsliall.

C. aspera Willd.
Type— Divisions all, except 6, 7, 8, 13, 25, 28, 31, 32, 3If, 36.

var. oapillata Braun.
21 Dublin Blanchardstown '89 (Scully) — Groves j^^.

var. ourta Braun.
15 Galw. SE. Portumna demesne '81 — B. "King^d^.

var. subinermis Kuetz.

3 Cork W. Near Lough Hyne '96—1. Groves & R. A. Phillips.

12 Wexford Lagoon north of Wexford Harbour '96 — ^Marshall 55 7,

15 Galw. SE. Kilcolgan '99— P.

28 Sligo Lough Gill River *84--B. & V. 112.

29 Leitrim Glenade Lough '84— B. & V. 112.
37 Armagh Lough Neagh '92 — P.

89 Antrim L. Neagh '84 (Lett)— Groves j^j. Rathlin ; Ram's I
40 L'derry R. Bann below Coleraine '94 — P.

'PsLA^OER^Okaninga in Irish Topographical Botany. 91

var. lacnBtris H. & J. Groves.
38 Down Longh Neagh '98— J. H. Davies.

Also in 16, 27, 36, 37, 89 — see Groves j^^. So far as at present
known, confined to three lakes, Lough Keagh (where it is abundant).
Lough Cullin, and a lake at Boundstone.

sub-sp. desmacantha H. & J. Groves.

32 Monaghn. Near L. Naglack '01— G. R. Bullock -Webster.

Also in divisions 7, 8, 10, 11, 14, 15, 16, 17, 18, 19, 21, 23, 24,
25, 26, 27, 28, 29, 30, 33, 38— see Irish Top. Bat.

C. polyaoantha Braun.
32 Mooaghn. Carrickmacross and L. Naglack '01 — Bullock-'Webster.

Also in divisions 5, 7, 8, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21,
22, 23, 25, 26, 27, 28, 32, 37— see IrUh Top. Bat.

C. oontraria Euetz.

Type— Divisions 1, 10, 13, 14, 15, 17, 18, 19, 21, 22, 23, 24, 25,
27, 29, 30, 32, 35, 37, 38, 40.

var. hispidnla Braun.

Divisions 16, 28, 26— see Groves j^g. Common in the Westmeath
lakes: elsewhere at Cong only.

C. denndata Braun.
23 Wcstmth. Brittas Lake '94— Levinge ^Ss-

C. tomentosa L.

Divisions 10, 15, 23, 24, 25— see Irish Top, Bot. Westmeath
Iskes and Shannon only.

C. hispida L.

Type— Divisions aU, except 3, 6, 12, 26, 29, 30, 33, 34, 36, 39.

var. macracantha Braun.
Divisions 19 [not 21], 20 — see Groves j./c^.

Proceedings of the Boyal Irish Academy.

Tar. rudiB Brann.

7 Tipp. S. Clonmel '97— P.

8 Limrck. Adare'99 — Somerville. L.Our, CiUTaghChase/OO— P.

9 Clare Iniahmore '91 (Stewart) — Groves ^4g.

10 Tipp. N. BaUingarry, aoughjordan, Youghal Bay, '00— P.

11 Kilkny. River Nore above Ballyragget '99— P.

14 Qneen's Qraigae '99, Erkina marslies, Portarlington, &c. — P.

16 Galw. BE. Loughrea and Portumna '98— P.

16 Galw. W. Clonbur '96— E. S. MarshalL

17 Galw. NE. Barbersfort '99 : very frequent— P.

18 King's Canal at Edenderry '96— P.

19 Eildare Levitstown '99, Monasterevan, Maynooth, &c. — P.

21 Dublin Royal Canal at Lucan '94— P.

22 Meath Lough Crew and Enfield '96— P.

23 Westmth. MuUingar '99 — P. Knock Drin, Brittas L. — ^Levinge.

24 Longfd. Ballymahon '00, Killashee, Priest's Island — ^P.

25 RoBComn. Corkip L., R. Suck, '99 — ^P. Common in L. Ree.

26 Mayo E. Claremorris '00 — ^P. L. Mask— M. & S. 5^5.

27 Mayo W. Derreen '00— P. Ballina '91— A. Somerville.

28 Sligo I5allysadare '00, Rosses Point — P. Frequent in east
SOjCavan ■»«. Lougb Sheelin '96— P.

32 Monagbn. Carrickmacross district f. '01 — G. R. Bullock- Webster.

87 Armagh About Armagh and Loughgall '92 — P.

38 Down Money Lake '91— P. Loughanisland— -^. NJS,

C. vxQgariB L.

Type— Divisions all, except 5, 6, 26, 5If, 36.

var. longibraoteata Kuetz.

2 Kerry N. Blennerville '88 (Scully)— Groves j^g.

5 Cork E. Midleton '72 (T. Allin)— Groves j^^.

7 Tipp. S. Carrick-on-Suir '00--^P.

9 Clare Co. Clare '95 (N. Colgan) — Groves J55.

11 Kilkny. Granny '98— P.

14 Queen's Mountrath '97— P.

15 Galw. SE. Kinvarra '00, Portumna '98— P.

17 Galw. NE. Barbersfort and Ballyloughaun '99— P.

19 Kildaie Hills above Rathmore '98, Carbuiy '96— P.

20 Wicklow Base of Great Sugaiioaf '94— P.

Prabgbr — OleaningB in Irish Topographical Botany, 93

21 Dublin Ireland's Eye '95— P. Sworda (D. Moore)— j^^.

22 Meath OldcasUe '96— P.

2S Westrnth. Lough Iron '99— P. Frequent

25 Boscomn. BockviUe '99— P.

26 Mayo E. Kilmovee '99— P.

28 SHgo Bosses Point '97— P. Glencar L. '84— B. & V. ii2,

29 Leitrim Lough Melvin '99— P.

31 Louth Oreenore '00, Castlehellingham— P. Dundalk— j^^.

32 Monaghn. CamckmacrossandMoynaltyL.'Ol — ^Bullock-Webster.

33 Feiman. Belleek '00— P.

88 Down Kircubbin '90— P. Near Belfast (Stewart)- j^9.

39 Antiim Springfield '57 (Hind) — Groves ^^g.

yar. papillata Wallr.

I Kerry 8. Waterville '89 ( Scully )^Groves j^g.

8 Limrck. Limerick '99 — A. Somerville.

11 Kilkny. Thomastown '98— P.

12 Wexford North side of Wexford Harbour '96— Marshall 5J7.

13 Carlow Bagenalstown '99— P.

14 Queen's Near Farmhill '90 (Scully)— GroYes 34g,

15 Galw. SE. Near Portumna '97— P.

27 Mayo W. BaUina '00— P.

C. oaneaoeni Loisel.

DiTisions 1, 12, 17 — see Iri^h Top. Bot

Yor the Irish species of the genera TolypeUa and NiUlla^ I have
no infonnation additional to that given in Irish Top. Bot, except the
feOowing: —

Hitella tenuissima Euetz.

23 Weatmth. Lough Owel '01— G. B. Bullock-Webster.

H. muoronata Euetz.

32 Mmiaghn. Camckmacross, Lough Naglack, and abundant in
Moynalty Lough '01— G. B. Bullock-Webster.
An interesting addition to the Irish flora.

B. I. A. raoc., TOL. Tin., 8BO. B.] I

94 Proceedings of the Boyai Irish Academy.

H. flezilis Ag., yar. cnuua Braun.

24 Longfd. Lough Gowna '00— P.

yar. nidifica Wallm.

25 Eoscomn. Lough Allen at Arigna '00 — P.

30 Cayan Lough Sillan '01— G. R. Bullock-Webster.

32 Monaghn. Annaghmakerig Lough '00 — P.

A very rare yariety, its only other British record being Marlee
Loch, East Perth (Sturrock, 1882).

I haye included in these notes records up to the Spring of 1902.
The unequal nature of the records shows how much work is still required
among Irish critical plants. An excellent example of the interesting
results which will still reward careful field work in the country is
shown by Key. G. R. Bullock- Webster's discoyeries in the Charaaa
of MonaghaUy of which he has giyen an account in Irish Naturaliti,
yoL xi., pp. 141-146, 1902.

OK THE WASTE OF THE COAST OF IRELAND AS A FACTOR
IN IRISH HISTORY. By J. P. O'REILLY, CE.

[Bead DBcnuont 9, 1901.]

HATnre proposed to myself the exaininatioii of certain points relatiye
to the forms and stmeture of some of the ancient monaments of
Ireland, I was led on to the study of the past and present physical
geography of the conntry, as being intimately connected with its
history, and, therefore, with that of the peoples to whom certain of
these monaments have been attributed. Modem historians show their
strong appreciation of this connexion, by the care they take to illustrate
by maps and drawings the localities or places wherein or whereat have
taken place the eyents which they treat of, as also in pointing out the
changes which haye occurred in the localities since the period con-
sidered by them in their narration. That this is no easy task has
been shown by Sir Charles Lyell in his '* Principles of Geology/'
Tol. L, p. 252, where he says: —

*' To thoee whose attention has never been called to the former
changes in the Earth's surface which geology reveals to us the position
of land and sea appear fixed and stable. It might not seem to have
onde^one any material alteration since the earliest times of History ;
but when we inquire into the subject more closely we become convinced
that there is annually some small variation in the geography of the
globe. In every century the land is in some places raised and in
(^theis depressed in level, and so likewise is the bed of the sea. By
these and other ceaseless changes the configuration of the Earth's sur-
het has been remodelled, again and again, since it was the habita-
tion of organic beings ; and the bed of the ocean has been lifted up to
the heif^t of some of the loftiest mountains ; the result is in general
new insignificant, if we consider how slightly the highest mountain
ehaiDs cause our globe to differ from a perfect sphere. Chimborazo,
thoogh it rises to more that 21,000 feet above the sea, would be
lepreeented on a globe of about 6 feet diameter, by a grain of sand
rhal less than -^Vth of an inch in diameter."

a.UA. rmoc., vol. zxiv., sbc. b.] IT

96 Proceedings of the Boyal Irish Academy.

Indeed the great difficidty which meets the conBcientionB writer of
History, at every step, is that of placing the reader in a position to
realize as fully as possible the conditions under which the erents he
narrates have taken place, and that for the particular period he may
be treating of. Eyen correct maps can only represent the present state
of the country or ground considered, since accurate surveys may be
said to date from the nineteenth century only. As to the geographical
delineation of countries previously to that time, one has only to examine
the maps of the eighteenth and seyenteenth centuries to become aware
of their insufficiencies and defects howeyer yaluable they may have
been at the time or may be still historically. As to restorations, the
remarks of Sir Charles Lyell in this respect are well worthy of dtatioo.
*^ The difficulty," he says, ''or rather the impossibility of restoring the
geography of the globe, as it may have existed at any former period^
esx>ecially a remote one, consists in this, that one can only point out
where part of the sea has been turned into land, and we are almost
always unable to determine what land may haye become sea " (Lyell'6
" Principles of Geology," yol. i., p. 265).

Aa regards Great Britain and Ireland the splendid maps of the
Ordnance Survey give us the correct representation of these countries
as they are at the present period, and furnish therefore a thoroug^y
<)omplete and reliable standard by which to judge of the changes that
time may bring, or by which to work out what may have been the
geography of these countries in former times.

It might be supposed that but few changes can have taken place
in the outline or general character of these islands during historical
time, and that any such changes, if of any magnitude, would be found
recorded in some document or historical work. That many records of
such changes exist is certain, but that all have been noticed or reooxded
is very doubtful. The changes attributable to atmospheric erosion
during historic times, are probably on the whole not very great, and
have been more or less approximatiyely estimated by geologists. Those,
however, which have been due to the action of the ocean on the coast
lines are in certain respects more important and more easily observable.
Few great storms from the west, north-west or south-west fail to leave
their mark on the coasts somewhere or other; and the steady continuous
beat of the Atlantic waves on the rocky headlands and coast lines
works their disintegration and removal, slowly it may be, but most
effectually. As an example of this action on the east coast of Ireland
may be pointed out the coast line between Killiney and Bray, as also
that between Bray Head and Greystones, along which considerable

O'Bbillt— O/i the Waste of the Coast oflreland, 8fc. 97

extents of the coast are formed by the drift, the sur&ce of which slopes
down towards the sea, and seems to have met it at some former period.
Fonning at present a cliff of more or less altitude, it is being steadily
€aten into by tbe tidal action and waves to such an extent and so
npidly, as to have given cause to the withdrawal of the railway Une
farther inland during the last quarter of the past century, more par-
ticnkrly along the part lying between Bray Head and Oreystones ;
and to have rendered necessary frequent if not incessant, and therefore
Terj costly, defence and embankment work by the Railway Company for
the safety of the line (quite lately, January 22nd, 1901, the Chairman
of the Company estimated this cost at £18,000). It is to be regretted
that such inroada of the sea along our coasts are not more carefully
noted, measured, and mapped for future record and information as
their total amount must in time become very appreciable.

Beodes this steady corrosion of the coasts, especially on those most
txposed to the Atlantic waves and storms, are to be counted with, the
alow alterations of level which have been noticed in Great Britain, if
not in Ireland, during historic time. Hence, it may be inferred that
nnleas land be reformed proportionally to the waste along our coasts
arising out of the tidal and wave action and erosion, the superficial
area of these countries must be slowly decreasing, and presuming that
the same causes have been in action during the past, this area must in
former times have been greater than it is at present. This decrease of
mperfidal extent of land is thus noticed in a criticism of '* The Reclama-
tion of Land from Tidal Waters," by Alex. Beaseley, M.nrBT.c.B. (1900),
irhich appeared in Nature, vol. 62 (19th July, 1900), p. 266 : —

" The area of this country is gradually diminishing by the con-
tinoal waste that is going on all round the coast. On the Yorkshire
<:oa6t it is estimated that two miles have disappeared since the Roman
occopatian, and more modem records show that towns and villages
ikare disappeared with their houses and churches, and in some cases
the whole parish has been washed away. Along the Norfolk coast the
only record of several villages is ' washed away by the sea,' and on
the Seotish coast churches and houses have fallen down the cliffs, on
whiehare to be seen the bones formerly deposited in a vanishing church-
f arl On the south coast, although the chalk cli& at the east end of
the Englifih Channel are subject to continual falls and slips, more care
has been taken to protect them, but along the clay cliffs of Dorsetshire
lU waste is continuous ; here 20 acres slipped down seaward in one
sight from the cliffs at Axminster.

" On the west coast the nets of the fishermen are said to become

98 Proceedings of the Royal Irish Academy.

occasionally entangled in the mins of houses and buildings boned is
the eea some distance from the coast of BlackpooL"

These considerations would apply still more strongly to the islands
which border the western coast of Ireland, and which may be presumed
to hare been in former times larger and more important than they are
at present, as also more numerous. Such greater extent^ number, and
importance of them in former times, would enter as Actors into the
question of the adrent of man in these countries and of the various
eolonixationB which it gave rise to. It is mainly to geology, and in
part to tradition and history, that we must have recourse for evidence
of these alterations in the coast line of the country, more particularly
as regards former extent. With the aid of the Admiralty maps which
furnish reliable data as regards soundings, and which, by ^e aid of the
contour lines in depth, which these soundings enable us to draw, can
be indicated the probable extent and nature of the changes, which
have taken place in the coast lines of the country as the result of
immersion or emersion and general action of the sea. From this point
of view the subject has been very fully and lucidly treated by Professor
Boyd Dawkins in his work '^ Early Man in Britain and his place in the
Tertiary Period," published in 1880, and from which work the follow-
ing exti^cts are taken : —

He opens (p. 3) with the remark: ''The continuity between
Geology, Prehistoric Archseology, and History is so direct that it is
impossible to picture early man in this country without using the
results of all these sciences."

(p. 6.) — ^He states: "Before our ancestors were in Europe, and
before our country was an island, there were PalsBolithio tribes in
Britain, ignorant of the use of polished stone and of metals, without
domestic animals, living solely by the chase, fishing, and fowling. Of
these, the older or ' river drift men ' have left evidence that they
wandered over the greater part of western and southern Europe, over
North Africa, Asia Minor, and over the whole of India, while the
newer or ' cave men ' have been traced over a large part of Europe.'*

(p. 9). — He gives his reasons for starting on his inquiry with the
commencement of the Tertiary Period as foUows : " In the Tertiary
Period each life group is so closely linked to that which went before
and followed after, that there is no break of sufficient importance to
be used as a starting-point in our special inquiry into the Andent
History of Man. We shall therefore be compelled to treat in outUne
the principal changes which took place in this country from tlie
beidnning of the Tertiary Period down to the time when man finst

O'Rrilly— Oil the Wmie of the Coast of Ireland, Sfe. 99

appears on the stage, and to see how they are rehited to the vaiTing
CQaditions of life on the Continent.''

He then gives the following classification of the stages of the
Tertiary Pmod : —

CharaeterUiicM,
I. Eocene ; or that in which the Mammalia now on the W • ;«„ «,^^»«. a
EarA w««.pr«.nted.byalUed fonn. belonging to ^a«TJt
exittmg orders and families. ; «**«.!» o»B«i.

II. Miocene ; in which the alliance between living and \ jy^tj^ ^g_ _^
ibesil mammals is more close than before. ) ^^ ^

in. Pliocene ; in which living species of animals appear. Living species.

IV. Pleistocene; in which living species are more abundant ) Living species abun-
than the extinct. Man appears. J dant. Man appears.

Y. Piehistoiic ; in which domestic animals and cultivated \ Man abundant. Do-
fruits appear, and man has multiplied exceedingly > mestic animals.
on the Earth. J Cultivated fruits.

VI. Historic ; in which the events are recorded in history. HiBtorical record.

(p. 14.) — He says : *' The invasion of Europe by the Placental
MftfnmftlR is the great event which is the natural starting-point for our
inquiry into the ancient history of man, since the conditions by which
he was surrounded on his arrival in Europe form part of a continuous
sequence of changes from that remote period down to the present day.'*
(p. 18.) — ^He gives a sketch map (fig. 3) of the geography of north-
western Europe in the Eocene Age, and having given the reasons which
enabled him to give its outline he says :

(p. 23.) — *' From these considerations (zoological, botanical, and
geological), Eocene Britain (and Ireland) may be taken to have formed
part of a great continent extending north and west to America by way
of Iceland and Greenland, while to the north-east it was continuous
with Norway and Spitzbergen. It extended also to the south-west
across what is now the Channel to join the western parts of France.
This great north-western continent or ' Northern Atlantis ' as it may
be tenned, existed through the Eocene and Miocene Ages, offering a
neaas of free migration for plants and animals, and it was not finally
broken up by submergence until the beginning of the Pleistocene Age."
(p. 43.)— As regards the continuity with North America he states :
"The researches of Professor Heer into the forest vegetation of the
Continent, Britain, Iceland, Spitzbergen, Greenland, and Grinnell-
land prove that the whole of this portion of the Earth's surface was
^ land in the early Tertiary Period, offering free means of migration
to plants and animals from the Polar regions into America on the one

100 Proceedings of the Royal Irish Academy.

hand, and into Europe on the other. The 500 fathom line indicates
the prohable coast line during both Eocene and Miocene, and the rapid
increase of depth in the Atlantic to its west would allow of a consider-
able depression taking place without altering in any important degree
the position of the sea margin.

(p. 44.) — ** Professor Heer places his * Atlantis' to the south-west
of the line represented in the map (fig. 6), but the enormous depth
of the North Atlantic renders it yery improbable that there was dry
land in that region at a time, geologically speaking, so recent as the
Miocene Age " (reference will be again made to this remark further on).

'' The principal mountains in the British Isles were in their present
positions in the Miocene Age, but were considerably higher, (probably)
double what they are now."

As bearing on the fact of the former extension of the land to
America and towards the north-east of Europe it is of interest to cite
the following note on the " Report of Messrs. Newton and Teall on the
Lava sheets of Franz-Josef Land " from Nature^ vol. 57 (Nov. '97 to
April '98), p. 324 : '* The immense lava sheets that cover an area of
some 200,000 square miles in the Deccan of India have been looked
upon as the greatest examples of Vulcanism in the world, but an even
more extensive outpouring of similar material must formerly have been
evident in the northern hemisphere if we can accept the conclusions
reached by Messrs. Newton and Teall from a study of the geological
collections made in Franz-Josef Land by the Jackson-Harmsworth
Expedition (see Quart. Journal Geolog. Soc. December, 1897). That
Archipelago is formed of the fragments of an ancient basalt plateau
which must have stretched far beyond its present limits. Similar
igneous rocks are found in Spitzbergen, Jan Mayen, Iceland, Green-
land, the Faroes, the Hebrides, and north Ireland ; and the authors are
inclined to regard all these areas as the isolated fragments of a formerly
continuous land area, the greatest part of which has sunk to form the
northern portion of the North Atlantic Ocean. The period of this
outpouring was probably the end of the Cretaceous and beginning of
Tertiary times. The period seems to have been distinguished by similar
occurrences in other parts of the world, for the great lava flows of the
Deccan and of Abyssinia are of the same age.'' '* In Auvergne, in the
Miocene Period, the volcanoes burst through the granitic and gneissose
plateau of central France " (Geikie, " Text-book of Geology" (1893),
p. 203).

(p. 66.) — ^Boyd Dawkins says : ** There is no proof of the presence
of man in Europe during the Miocene Age."

O'Rbilly— On the Waste of the Coast of Ireland, 8fc. 101

In this respect it may be worth citing the following from an article
in the Nineteenth Century ^ ** On the Cradle of the Human Eace," by
Samuel Waddington : —

" Others see reason to beHeye that there is Httle or no doubt that
the human race has existed on the face of the Earth for more than one
million or eyen two million years.

" Darwin, it will be remembered, was of opinion that man may have
existed in the Eocene Period; while Mr. Wallace holds {Nineteenth
Ceniwy, 1 887) that he certainly did exist in that period. Professor
Huxley also appears to have held this view, for he observes that the
first traces of the primordial stock whence man has proceeded need no
longer be sought by those who entertain any form of doctrine of pro-
gressiTe development in the newest Tertiaries, but that they may be
looked for in an epoch more distant from the age of the Mephas primi-
feniue (Mammoth) than this is from us."

" The remoteness of the date," observes Sir John Evans, "at which
the Pakeolithic Period had its beginning almost transcends our power
of imagination"; and Professor Eatzel in his '* History of Mankind,"
states that a regular workshop for the manufacture of chert flakes
which was discovered on the banks of the Mississipi in Minnesota, dates
from the Intergladal Era, and that hunters chased the long-extinct
beasts of the Drift Age in Mexico and in Argentina.

He (Mr. Waddington) asks, " But how long ago is it since the
commencement of the Eocene Period ? " and taking into consideration
the statements of Lord Kelvin as to the probable time since the
solidification of the Earth, he says : ** The date of the beginning of the
Eocene Period cannot therefore be estimated at less than four million
jean before the present time.

** When the great Mastodon, now in the British Museum, was
found by Dr. Kock in the Ossage Valley, Missouri, a number of stone
arrowheads and charcoal were found near it, and one of the arrow-
betids lay underneath the thigh-bone of the Mastodon, and in contact
with it. The animal was found, it will be remembered, at a depth of
20 feet under several alternate layers of loam, gravel, clay, and peat,
with a forest of old trees on the surface."

{p. 72.) — ^Boyd Dawkins sketches out the Geography of Britain
rand Ireland) in the Pliocene Age (fig. 10), and says : ** The North
Sea, which was small in the Miocene age, and did not touch our
present coast line, was now gradually enlarged at the expense of the
land, and ultimately a direct communication was made witti the Arctic
Sea, by the sinking of the land, extending from the Scandinavian

1 02 Proceedings of the Royal IrUh Academy.

Mountains to the British Isles, to Iceland and Qieenland on the oof
hand, and to Spitzbergen on the other."

This depression by which the Arctic became continuous with, the
North Sea, caused it also to become connected with the Atlantic.

(p. 73.) — He says : '* The Atiantic line at that time may be taken
to be marked by the steep slopes passing downwards from the 100
fathom line, to the 300 or 400 fathom line, which imply that a land
barrier was in that position for a Tery long period.

''It would make very little difference in the map of Pliocene
Britain (and Ireland) if we were to take the western coast line to be
marked by the 300 fathom instead of the 100 fathom line."

(p. 75.) — He shows the evidence of icebergs at that period'off the
coast of Great Britain, and the submergence of the tract of land
uniting Ireland with the continent of Miocene Europe, by which
currents of cold water from the Polar regions obtained free access to the
Korth Sea of the Pliocene Age, from which they had been before shut
out by a barrier of land.

(p. 93.) — He states that, as evidence stands at present, the
Geological record is silent as to man's appearance in Europe in the
Pliocene age.

(p. 94.) — Speaking of the Pleistocene Age, he says : " New
Mammals now appear belonging for the most part to living species.
Their remains were associated with human implements in such a
manner as to show that man was a member of the fauna which
characterises the Pleistocene Period of this quarter of the world."

(p. 110.) — Describing the great geographical and climatal changes
of the period, he says : '' Britain must also have formed part of
the mainland. Ireland must also have been united to Britain, to
have allowed of the groups of animals (mentioned by him) finding
their way so far to the west. The elevation above the present
sea-level necessary to account for this distribution of the animalfi, is
not less than 600 feet or 100 fathoms" (fig. 24). The Straits
of Gibraltar could not have been in existence when the African
elephant ranged as far north as Madrid, and the Caffir cat, African
Lynx, and spotted Hysna sought their prey in the Iberian
Peninsula.

(p. 112.) — He says: "From these considerations, it is evident
that Pleistocene Europe must be looked upon as intimately connected
with Africa on the south, and with Asia on the east, and that it
offered no barriers to the migration of Asiatic and African •«iT>%y\^
as far to the west as Britain and Ireland."

O'Bbilly— On the Waste of the Coast of Ireland, Sfc. 103

(p. 113). — ^He Bays: '' From the distribution of the Pleistocene
Mammals, we may infer that the climate was severe in the north and
warm in the south; while in the middle zone, comprising France,
Germany, and the greater part of Britain, the winters were cold and
the summers warm, as in Middle Asia and North America. There
were secular changes of climate in Pleistocene Europe, and while
the cold was at its maximum, the Arctic animals arriyed at their
southern limit ; and while it was at its miniimim, the spotted Hyeena,
and the Hippopotamus, and other southern animals roamed to their
northern limit."

The climatal and geographical changes which then took place in
Britain were marked by Glacial phenomena, which are summarized as
follows : —

(p. 115.) — " (1) The first Glactation was a period of elevation."
(p. 116.) — ''The ice at that time was sufficiently thick to have
overridden Schihallion in Perthshire, at a height of 6500 feet, and the
hills of Galway and Mayo at 2000 feet."

(p. 117.) — " (2) Tlie Iceherga — A period of depression.
''(3) The depression continued. The glaciers disappeared, and
.the sea beat upon an archipelago of islands, which gradually sank
beneath the sea to a depth of from 2300 feet below their present
level on the flank of Snowdon, to 1200 feet at Yale Eoyal, on the
road between Buxton and Macclesfield, and to about 1400 feet in
Scotland.

" (4) A reversion to a severe climate.

** (5) Period of elevation. The climate becoming temperate, there
followed an upward movement of the land, until the Upper Boulder Clay
became dry land, and Britain and Ireland became part of the mainland
of Europe as represented in the map (fig. 32). The climate was
leM severe than in the preceding period, and was gradually again
becoming temperate,

*' As the Upper Boulder Clay deposited on the sea bottom became
lifted up, it was gradually covered by forests of yew, Scotch fir,
oak, ash, and alder, in which the Pleistocene Mammalia found ample
food in the eastern and midland counties."

As regards the Glacial Period in Great Britain and Ireland, it may
be well to cite here the opinion of the eminent geologist. Professor
I^pworth, given in his ** Intermediate Text-book of Geology," 1899.

^p. 385.) — " That the glacial conditions of Britain and Western
Kuiope were accompanied by a certain amount of depression is
generally acknowledged, but whether that depression was excessive

104 Proceedings of the Royal Irish Academy,

and general, and was broadly coincident with the Middle Qlacial times,
or whether it was relatiyely insignificant, repeated, and local, is yet
a matter of dispute."

(p. 125.) — ^Boyd Dawkins says : ** In all probability the geogra-
phical conditions of Britain and Ireland at the time of the River
Drift (Pleistocene) Hunters, were identical with those of the Late
Pleistocene (fig. 32, p. 150), when our country formed part of the
Continent.

(p. 148.) — *' The remains of the late Pleistocene animals lie
scattered oyer a large area in Britain, and it is necessary to conclude,
from their presence, that our country formed part of the mainland of
Europe at that time."

(p. 150.) — " The Ghjography of Great Britain (and Ireland) in the
late Pleistocene Age is indicated by the map (fig. 32, p. 150)" ; and
he says : —

(p. 151.) — " It may be concluded that Britain (and Ireland) stood
at least 600 feet aboye its present leyel, and so that the Scyem R.
united its waters with the rivers of the south of Ireland."

(p. 152.) — He then defines the Range of the late Pleistocene
Mammals over Britain and Ireland: — '' In Ireland the Mammoth has
been found in the Counties of Cavan, Ghdway, Antrim, and Waterf ord,
and in the Shandon cave near Dungarvan, in the last of these counties
along with the grizzly bear, wolf, fox, horse, stag, and alpine hare.
This irregularity in the distribution of the animal remains is intimately
connected with the geographical and climatal changes which were
going on in the obscure and complicated portion of the late Pleistocene
Age, known as the Glacial Upoeh^ and it is highly probable that all the
Irish Mammalia mentioned aboye are pre-Glacial."

(p. 153.) — ** We must further realise that all the climatal and geogra-
phical changes known as Glacial, happened while the Late Pleistocene
Mammalia were living in the regions not covered by glaciers, or
overwhelmed by the sea, and that they wandered to and fro, as the
barriers to their migration were altered."

(p. 169.)— He says : " The Palaeolithic Hunter of Mid and Late
Pleistocene River deposits in Europe belongs to a fauna which
arrived in Britain before the lowering of the temperature produced
glaciers and icebergs in our country. He may therefore be vie'wed
as hemg prohahlg pre-Glacial. When the temperature was lowest, he
probably retreated southwards, and returned northwards as it grew
warmer, precisely in the same manner as the Mammalia on which he
depended for food."

O'Beillt— Ow the Waste of the Coast of Ireland, 8fc. 106

(p. 171.) — •* It may therefore be concluded, ho says, that man wa&
probably pie-Oladal and Glacial in Europe, but was certainly post-
Glacial in the area of the North Thames."

It may be worth while to cite here the views put forward by
Lapworth, in the work already cited on the Glacial Period.

(p. 352.) — " During the last half century, abundant evidences have
been obtained of the existence of man as far back as the final stages
of the Glacial period ; and the glacial and post-Glacial formations
have consequently been separated off from the Tertiary, and erected
into a distinct series by themselves, which has been termed the
Qttatemmy, while the period of Geological time which they represent,
has been denominated Anthropa%oie.*^

(p. 378.)—" l!he recent deposits were formerly referred to as the
BumoHy as it was supposed that they alone afforded evidence of the
existence of man, but the discoveries of late years, have made it clear
that man existed in Pleistocene times, at any rate, in the later stage
of the epoch, if not throughout the whole."

As bearing on this question, it may be of use to cite here the
opinions of the eminent French geologist, de Lapparent, as stated in
the last edition of his ** Trait6 de G6ologie," vol iii., 1900, in his
critical review of the general characteristics of the Tertiary Era.

(p. 1632.) — He considers the question of the Establishment and

Viassttudes of the great Glacier Sy and says : — " It would seem that

independently of a first or primary Pliocene phase, there were, as well

in America as in Europe, two other great Pleistocene phases of

extension of the glaciers. These phases were separated by intervals

of ^e, during which the climate was at least as favourable as at

present, and the surface of the land became clear of ice, even into the

very hearts of the mountain valleys. The greatest extension took

place anteriorly to the development of the Palaeolithic civilization

▼hich made its appearance during inter-Glacial Periods, when the

£Upha$ primigenius commenced to associate with the Elephas antiquus.

"The first mentioned species, accompanied by a fauna, on the

▼hole of a colder climate than the previous one alone, was in existence

doling the succeeding extension to which followed, even if it did

not accompany in part, the deposit of the great LoSs, At that time a

period of dry cold supervened to interrupt the active flow of the

rirers. Man then took refuge in the caves and under rock-shelters,

wbilst in the meantime became developed in our part of Europe, first

tlie Eqwdit and then the Reindeer^ an animal known to dread fogs

whilst supporting easily dry cold.

106 Proceedings of the Bayai Iriah Acadeniy.

*' Before the end of the * Mjigd«l«nian CiTilicafion ' diaracterutic
of the period, the humidity of the atmosphine reappeared, without
however being accompanied by any earth movonentt anffidentlj
marked to have giTen rise to renewed energy in the eroaiTe action of
the water-connes.

'^The Beindeer was almost completely driven ncMrth, and the
Keolithic dvilization came into existence everywhere. It is probable
however, that the phenomena were more complicated, and that in
addition to the great Glacial Periods already indicated, it wonld be
proper to add intermediate phases, since every day's experience more
strongly causes it to be recognised that the reLations between the
morainic deposits is more or less complicated."

(p. 1634.) — ^He says: ''It is therefore logical to admit that
successive movements of emersion of the continents which took place
suddenly (par saccade), marked the phases of the activity of the
water-courses."

Discussing the prohainlity of a rapid transition from the age of tk$
Reindeer to that of the Turfboge^ he says : '* Whatever may have
been the exact course of events in our part of Europe, the transition
from the regimen of great water-courses, to that of the reindeer,
must have taken place rapidly. Otherwise, the rivers which at first
carried only coarse gravel, would have little by little filled up their
principal channels with silt, as has been justly remarked by Belgrand.
On the contrary, the principal channels which during the Pleistocene
Epoch hardly sufficed to carry the river floods, must have been dried
up suddenly, thus laying bare the horizontal surface of gravel over
which a thin stream continued to meander. Consequently, when
later on, a sufficiently humid regimen reappeared, it was the peat
which undertook to fill up the main water-courses, wherever the
permeability of the slopes secured for the rivers a regimen exempt
from violent floods. With the bogs commences the actually existing
regimen. During the period of dry cold, the fauna of the Mammals
was that of the Siberian steppes. The humidity of the age of the
bogs by favouring the development of timber, determined the
incoming of a forest fauna. The temperature henceforward under-
goes but slight vicissitudes, and with the exception of some alterna-
tions of invasion and retreat of the sea in the Elemish regions, the
contours of the continents have become fixed, and the story of
succeeding events belongs rather to Archseology and to History than
to Geology."

At p. 247 of his work already cited Boyd Dawkins says : '* The

O'Reilly— On the Waste of the Coast of Ireland, 8fc. 107

^S'Histwie Period covers all the events which took place between the
Pleistocene Age on the one hand, and the beginning of history on the other. ^^

(p. 248,)— He sajB : ** The Fre-Historic Period is separated from
the PleiBtocene by a long interral, during which not only great
changes in the zoology of Great Britain (and Ireland) took place, but
also corresponding changes in the geography.

'' At the dose of the Pleistocene Age (fig. 32), the valleys which
imited Britain to North France, Germany, and Scandanayia, as well
as to Ireland, were gradually depressed beneath sea level, and the
North Sea, the British Channel, the Irish Sea, and the Western
Atlantic Coast line generally, became very much as we find it now
'see fig. 95, p. 254). An examination, however, of the submerged
forests and peat bogs, proves that the downward movement had not
ceased tmtH a late period in the Neolithic Age,*'

(p. 250.) — He shows that, *' In West Somerset and at Minehead,
we may infer that man was living in this region during the time that
a dense forest overshadowed a large portion of what is now the British
Channel, and before the deposit of tibe blue fresh-water clay, and the
marine silt, at a time not later than that marked by the layer of peat
ot vegetable soil in which the prostrate trees are embedded.''

(p. 251.) — ''The submerged forests are merely scraps spared by
the waves of an ancient growth of oak, ash, and yew, extending in
Somersetshiie underneath the peat and alluvium, and joining the
great morasses of Glastonbury, Sedgemoor, and Athelney, in which
Xeolithic implements have been met with by Mr. Stradling.

'' In Torbay as well as in North Devon and Somersetshire, man
was in possession of the country when the land stretched farther out
to sea than at the present time. In this particular case (Torbay),
Mr. Pengelly estimates the submergence to have been not less than
40 feet, since the forest was alive."

Similar proofs of submergence are to be met with on our coasts^
wherever the land dips gently under the water-line.

(p. 253.) — He says : *' It is worthy of remark that the enormous,
trunks of the trees prove that the Scotch firs, oaks, yews, willows, and
hirehesy of which the forest was in these places mainly composed,
most have grown at some distance from the ancient coast line, since
the westerly winds sweeping over Lancashire from the Atlantic at the
present time prevent the free growth of vegetation on every unpro-
tected spot on the coast. The prevalent winds, however, are proved to
have been very much the same, since then as now, by the position of the
trees, which lie prostrate, with their heads pointing towards the east.

108 Proceedings of the Royal Irish Academy.

'^ The depression of land throughout Great Britain and Ireland
since the trees flourished could not have been less than from 30 to 40
feet. The ten-fathom line, therefore, considered by Sir W de la
B^che to be roughly the boundary of the land at that time, may
be taken to represent the sea margin (of that time) with tolerable
accuracy."

(p. 254.) — " This forest growth is proved to belong to the KeoUthic
diyision of the Fre-Historic Period by the presence of animals originally
domestic, and introduced by the Neolithic tribes, the Celtic short-
horn, and the sheep or goat, as well as by the absence of the Fleisto-
cene Mammals."

(p. 256.) — He cites Dr. Jas. Geikie's work on the geography and
climate of North Britain : — " When these buried trees darkened the
now bleak islands (Orkneys and Shetland) with their greenery, the
land stood at a higher leyel and the neighbouring ocean at a greater
distance. To have permitted this strong forest growth we are com-
pelled to admit a former elevation of the land, and a corresponding
retreat of the ocean.

'^ The same inference may be drawn from the facts diBclosed by
the mosses of Ireland and England. On the coasts of France and
Holland peat dips under the sea and along these bleak maritime
regions of Norway, where now-a-days the pine tree will hardly grow,
we find peat mosses, which contain the remains of full-grown trees,
such as are only to be met with in districts much further removed
from the influence of the sea." (See ** Great Ice Age," c. xxvi.)

(p. 263.) — ^He says : '* Such changes in the Mammalia and in tiie
geography of Great Britain (and Ireland), in the interval separating
the Fleistocene from the Fre-Historic Period, eauld not have taken place
in a short time ; and when we reflect that comparatively little change
has taken place in this country during the last 2000 yean, it is
obvious that the one period is separated from the other by a lapse of
many centuries, of how many we cannot tell."

(p. 265.) — He says : " It may be concluded that the former period
was, beyond calculation, longer than the latter."

(p. 482.) — ^He says: '* Britain, at the beginning of the historic
period, differed considerably from the Britain of to-day, althougih
there is no reason to suppose that any vertical movements have altered
the relation of sea to land. The dash of the waves for the last 1900
years has destroyed large tracts of land, where clifis are composed of
soft and incoherent materials. The inroads of the sea on the south
coast have been so great, in some places, such as Fevensey and Fayham,

O'Rbilly— 0» the WasU of the Coast oj Ireland, 8fc. 109

in Snasex, that it is by no means improbable that the Isle of Wight
may have been imited at low-water to the adjoining coast during the
Roman occupation. (It was an island in the days of Claudius).'^

(p. 483.)— He says : ** The rainfall, at the beginning of the Historic
Period in Britain, must have been greater than it is now, because of
the large extent of forest and morass. The surface of the country was
densely covered with trees."

(This relatively greater rainfall may be taken as implying, amongst
other causes, a relatively greater height of the mountain parts, in the
interior, since such greater height would necessarily favour a greater
amount of condensation, and, consequently, of rainfall).

These many extracts from Boyd Dawkin's work, and from the
other authors mentioned, show us not merely the former varied geo-
graphical conditions of Great Britain and Ireland, relatively to the
Continent and to one another, but also allow us to appreciate the
immense interval of time that must have elapsed since the commence-
ment of the Tertiary Period, and consequently how very small the
(listorical Period must appear in comparison therewith, and there-
fore how valuable all the data that can be collected either in the
form of traditions, or as observations and historical records relative
thereto. The early traditions regarding this country, which appear in
O'Flaherty'B ** Ogygia " and in the ** Annals of the Four Masters,"
merit, in this respect, careful and considerate attention.

Thus the commencing lines, '* The age of the world to this year of
the Deluge, 2242 : Eifty days before the Deluge, Csesair came to
Ireland wiUi fifty girls and three men ; Bith, Ladhra, and Fintain,
their names."

This passage, which is fully commented on in the notes to O'Dono-
van'B edition, from the purely scholarly and literary point of view, is
c^iaUe of assuming another aspect if taken in connexion with the
series of submergencies of lands and islands, which formed part of the
great northern continent, or group of islands, considered in the Eeport
of IfesBrs. Newton and Teall, ''On the Lava Sheets of Franz-Josef
Land," already referred to. Ireland was evidently affected by the
i^ries of volcanic movements, which seem to have lasted from the
Miocene Period onward, and of which series Iceland is still an im-
portant and active centre. It may be, therefore, that the '' Deluge "
referred to in the Annals represents the echo of a tradition from
Pre-Historic times of one of these sudden and catastrophic volcanic
movements (such as that of Krakatoa in 1883) which affected Ire-
land and gave rise to a remarkable depression which, if sudden and

110 Proceedings of the Royal Irish Academy.

accompanied by a series of tidal waves from the ocean, might well be
tantamount to a deluge. One has only to read of the recorded appear-
ance and destructive effects of such waves in modem times to recogniBe
how truthfully they are described as " Deluges," particularly by the
survivors who have witnessed the catastrophe and suffered from it. The
details given in the Annals as to the places of their deaths and inter-
ments rather tends to prove that the '^ Deluge " was not such as to
have prevented people from surviving and living on the island, which
so far favours the supposition of a sea- wave or subsidence with sea-
wave. Too much stress cannot be laid on this view of the question,
since it bears a certain relation to the submersion of the Island of
" Atlantis," as mentioned by Plato, and the two together might be
taken as connecting links in that chain of events implied by the break-
ing-up and submersion of different parts of that Great Lava Plateau
spoken of in Messrs. I^ewton and Teall's report. It is further interest-
ing to note that a series of modem archaeological discoveries, resulting
from the excavations so successfully and scientifically carried out in
Egypt and in Asia Minor, as also in Mesopotamia, and now being
actively pursued in many other quarters of the East, have resulted in
pushing back the record of time so, that already dates of 7000 b.c.
are spoken of, and we may well foresee that further researches will
in not many years hence push the antiquity of human records back to
10,000 B.C. There even appears in the Setenttfie American Supple-
ment of January 26th, 1901, p. 20960, an article entitled ** Archaeology
in the Past Century," by Prof. W. M. Flinders Petrie, d.c.l., ix.d., of
University College, London, in which he says : —

" We, therefore, have passed now at the end of this century to a
far wider view of man's history, and classify his earlier ages in Europe
thus: —

( 1 st). Eolithic — Eudest massive flints from deposits 600 feet up.
(2nd). Palaeolithic — Massive flints from gravels 200 feet up and

less ; {AchuUen).

(3rd). „ Cave-dwellers, flints like the preceding and

flakes ; {Ifcust^ien).

(4th). „ Cave-dwellers, flints well worked and

finely shaped; {Solutrien).
(5th). ,, Cave-dwellers, abundant bone- working and

drawing; {MagdaUnien).

(6th). Neolithic — Polished flint working; pastoral and agri>

cultural man.

O'Rbilly— On the Waste of the Coast of Ireland, 8fc. Ill

<* What time those periods cover nothing yet proves. The date of
4000 s.c. for man's appearance, with which helief onr century started,
has been pnshed back by one discovery after another. Estimates of
from lOyOOO to 200,000 years have been given from various possible
clues. In Egypt an exposure of 7000 or more years only gives a
faint brown tint to flints, lying side by side with Palaeolithic flints that
are black with age. I incline to think that 100,000 B.C. for the rise
of the (2nd) class and 10,000 b.c. for the rise of the (6th) class will be
a moderate estimate."

Thus the period of time stated to have elapsed since the submersion
of the Atlantis Island, as mentioned by the Egyptian Priest to Solon,
according to Plato's narrative, was 8000 years before his time ; this
up to the present, has been treated as fabulous and as throwing discredit
on the statement, but it now becomes not merely credible, but harmo-
niBes with the date which Flinders Petrie assigns for the (6th) period
above-mentioned of Human History, that is about 10,300 years ago.
It is of importance to point this out, since the submersion of the
Atlantis Island may, as already stated, be but one of a series of volcanic
and seiflonic movements in the Great Lava plateau of I^orth Western
Europe, having been marked by immersions of parts of the plateau,
the formation of islands, the further immersion and destruction of
these with accompaniment of great tidal waves comparable in their
destructiveness to deluges as already stated.

In connexion with this question there is room for citing the work
of Sir Jos. Prestwich, '^ On certain phenomena belonging to the close
of the last geological period, and their bearing upon the tradition
of the Flood *' (1895). He says (p. 72) : " In any case these tentative
estimates, are in accordance with the conclusion I have arrived at on
other grounds, that the Glacial (including the post-Glacial) Period,
together with PalsBolithic man came within 10,000 to 12,000 years of
our time."

Assuming, as argued by Boyd Dawkins, and as indicated by his
map (fig. 32), that at the close of the Pleistocene Age, Great Britain and
Ireland were still in connexion by land one with the other, and with
the continent of Europe, and that subsequently a series of depressions
intervened which resulted in the isolation of these countries, it is
reasonable to accept that these changes took place relatively slowly
and aaccessively, and that they were contemporaneous with changes
in the Atlantic coast line, probably in connexion with the volcanic
phenomena of the Icelandic, Greenlandic, and Franz-Josef group, all

K.I.A. PBGC., VOL. ZXTT., SVC. B.] L

112 Proceedings of the Boyal Irish Academp.

which had for result the present general coast outline of these
countries.

It may be asked would the 100 fathoms line as supposed by Bojd
Dawkins truly or approzimatively represent this Atlantic coast line
at the end of the Pleistocene time ? and are there not grounds for
admitting the existence of a more extended western land and of
adjacent islands, which could haye afforded the shelter necessary for
the forests which covered these countries prior to the growth of the
bogs.

This query is in some degree met by the results pointed out in
an article by Dr. Keusch in ** Naturen," cited in Nature, voL 63
(Dec. 13th, 1900), p. 160, in which he calls attention "to the
changes of level that have taken place in Iceland in recent
geological times viz. since the Ice Age." He says : " In 1896 the
Danish Ingolf Expedition investigated the sea bottom between Jan
Mayen and Iceland. In examining the dredged material Herr A.
S. Jansen made the observation that almost everywhere on the bottom
of the deep ocean lie shells of dead Molluscs of well-known shallow-
water forms, side by side with deep-water forms. It was very remark-
able to dredge up from depths of 500 fathoms to 1300 fathoms
Yoldia Arctica which now lives at Spitzbergen, and in the Kara Sea
at depths from 5 to 100 fathoms. Dr. Eeusch suggests that these
remains of Arctic life forms cannot have been carried there by drift-
ing ice, but that the sea bottom, in comparatively recent times during
the Ice Age, must have been much nearer the sea level than now. At
that time the Arctic shallow- water forms must have lived there ' m
situ ' ; then a sinking of the sea bottom has taken place which can be
estimated at not less than about 2500 metres (about 1355 fathoms).
It is easy to see that these results of the Danish naturalist have an
important bearing upon the phenomena of the Ice Age." It is evident
that from these results, there may be presumed a much greater exten-
sion of the Western European plateau and of its resulting islands
than Boyd Dawkins was prepared to admit. Whatever the chain of
events was that gave rise to the depression of the land, it is reasonable
to assume that the causes thereof were more active on the Atlantic
side of Ireland than on the eastern coast, and that the surface of land
affected thereby was more extensive. The breaking up of the land and
accompanying island groups during the Pre-Historic Period necessarily
occupied a considerable time, and allows us to admit that from the
coast of Spain, northwards to Ireland, and even farther, a great
number of islands, more or less inhabited, and of greater or less extent.

O'Reilly— 0» the Waste of the Coast of Ireland, Sfc. 113

existed during a certain time, and had not finally been worn away and
submerged in the Atlantic before the dawn of Irish traditional history.
Thus we Bee room for the tradition of the Island of Brazil (with
reference to which a very interesting paper was published in the
Royal Dublin Society's Proceedings, N. S., n. 1880, p. 173) and for
a Bockall far more prominent and extended than the present island,
which now can hardly be approached, so bold and precipitous are its
coasts.

As to the probable existence of other islands in the North Atlantic,
we can only arriye at a conclusion relatiye thereto by the aid of a
bathymetric chart of that ocean (that is, a chart coloured according to
the relative depths by means of contour lines of equal depth), and
keeping in mind the remarkable conclusions arrived at, in Dr. Beusch*s
article already cited.

As regards the coast of Ireland in general, and the changes they
have undergone during past ages, the only data that can at present be
availed of, are the records of the various Geological Surveys made of
them, and of the parts of the coasts of Great Britain which axe as
fully exposed to the action of the Atlantic waves and storms as are the
Irish coasts, as well as of the adjacent island groups, the Hebrides,
the Faroe, Orkneys, and Shetland groups. Certain descriptions of the
maritime counties both in Ireland and in Great Britain also furnish
observations and have been availed of. These records have up to the
present not been brought together and presented in a collected form,
and in the present paper it is proposed to so present them as a basis for
a more complete recension of all the data bearing on the question of
the wear of the Irish coasts. As in regard to many parts of the Irish
coast, the data are meagre, if not entirely wanting, while for much of
the coasts of Scotland, Wales, and England such data are available in
greater or less sufficiency, it has seemed reasonable to employ these
data when concerning parts of the coasts which are directly exposed to
the Atlantic Ocean ; since it is evident that whatever has been the
destmctive action of the waves and storms upon these, it cannot be
supposed to have been less on those parts of the coasts of Ireland
which are more directly and more immediately exposed to the full
action of the Western Ocean storms. Hence the indications existing
as regards the wear on the Scotch and Cornish coasts can to a certain
extent make up for the meagreness or absence of details as regards
the western, north-western, and south-western coasts of Ireland.

Sir Charles Lyell's" Principles of Geology" (1872) supply some
very valuable information in these respects, offering excellent terms of

114 Proceedings of the Royal Irish Academy.

eomparison for an appreciation of the wear that these coasts hare
imdergone, and consequently merit detailed citation. Under the
heading ^^ Destroying and Tramporting Power of Currents*^ and sub-
heading *' Action of the Sea on the British Coast ^ Shetland Islandt,
^<?.," he says (p. 607, vol. i.) : " The northmost group of the British
Isles, the Shetlands, are composed of a great variety of rocks, in-
cluding granite, gneiss, mica schists, serpentine, greenstonea, and
many others, with some secondary rocks, chiefly sandstones and
conglomerates. These islands are exposed continually to the uncon-
trolled violence of Hie Atlantic, and no land intervenes hetween the
western shores and America. The prevalence therefore of strong
westerly gales causes the waves to be sometimes driven with iireeistihle
force upon the coast, while there is also a current setting from the
north. The spray of the sea aids the decomposition of the rocks and
prepares them to be breached by the mechanical force of the waves.
Steep cliffs are hollowed out into deep caves and lofty arches ; and
almost every promontory ends in a cluster of rocks, imitating the forms
of columns, pinnacles, and obelisks."

(p. 609.) — **In some of the Shetland Islands, as on the west of
Mickle Roe, dykes or veins of soft granite have mouldered away, while
the matrix in which they are enclosed, beiog of the same substance
but of a firmer texture, have remained unaltered. Thus long narrow
ravines, sometimes 20 feet wide, are laid open, and often give access to
the waves."

After describing some huge cavernous apertures, into which
the sea flows for 260 feet at Loeness, Dr. Hibbert, writing in
1822, enumerates the other ravages of the ocean: ''But the most
sublime scenes are where a mural pile of porphyry, escaping the process
of disintegration that is devastating the coast, appears to have been
left as a sort of rampart against the inroads of the ocean. The Atlantic,
when provoked by wintry gales, batters against it with all the force of
real artillery, the waves having in their repeated assaults forced them-
selves an entrance. This breach, named the Grind of Naver (fig. 47)
is widened every winter by the overwhelming surge that, finding a
passage through it, separates large stones from its sides, and forces them
to a distance of not less than 1 80 feet. In two or three spots the frag-
ments which have been detached are brought together in immenfie
heaps, that appear as an accumulation of cubical masses, the product of
some quarry" (Hibbert, '' Description of the Shetlands." Edin., 1822).

'' There are localities in Shetland in which rocks of almost e very-
variety of mineral composition are suffering disintegration. Thus the

O'Bbilly— On the Waste of the Coast of Ireland, 8fc. 115

sea makes great inroads on the day-slates of Fitfel Head, on the
serpentine of Yord Hill, in Fetlar, and on the mica schists of the Bay
of Trieste, on the east coast of the same island, which decomposes into
angular blocks. The Quartz Bock, on the east of Walls, and the
gneiss and mica schist of Garthness suffer the same fate."

(p. 511.) — ^Lyell says, under the heading ^'Destruction of Islands** :
'^ Such devastation cannot be incessantly committed for thousands of
years without dividing islands, until they become at last mere clusters
of rocks, the last shreds of masses once continuous. To this state
many appear to have been reduced, and innumerable fantastic forms
are assumed by rocks adjoining the islands, to which the name
* Drongs ' is applied, as it is to those of similar shape in Peive. • The
granite rocks (fig. 48) between Papa Stour and WiUswick Kess afford
an example ; a still more singular cluster of rocks is seen to the south
of Hillswick Kess (fig. 49), which presents a variety of forms as
viewed from different points, and has often been likened to a small
fleet of vessels with spread sails. Midway, between the groups of
Shetland and Orkneys, is Fair Island, said to be composed of sandstone,
with high perpendicular cUffs. The current runs with such velocity
that during a calm, when there is no swell, the rocks on its shores
are white with the foam of the sea driven against them.

** The Orkneys, if carefully examined, would probably illustrate
our present topic as much as the Shetland group. The north-west
promontory of Sanda, one of these islands, has been cut off in modem
times by the sea, so that it became what is now called Start Island,
where a lighthouse was erected in 1807, since which the new strait
has grown wider."

As regards the Orkneys, the following is taken from the Ordnance
Guutieer of Scotland^ under that head : —

'' Except in the Pentland Firth, where the depth of the sea reaches
40 fathoms, the water in the straits between the islands and their
immediate neighbourhood is nowhere deeper than 20 &ithoms ; a rise
of 120 feet in the sea-bottom would unite the whole group, except
Sivona and the Pentland Skerries, into one mass of land, which would
^ separated from the mainland of Scotland by a strait of from 2 to 3
nules broad, where the Pentland Firth is. If these sounds are, how-
€^r, of moderate depth, their number and the broken and winding
outline of the coast are evidences of the hard struggle that constantly
takes place between the land and the Atlantic surge."

'* The intricate, indented coast-line, worn into creeks, and caves, and
oreihanging cliffs — the crags, and Skerries, and sea stocks, once a part

116 Proceedings of the Royal Irish Acadefny,

of the solid land, but now isolated among the breakers— the huge pile
of fragments that lie on the beach, or haye been heaped up far above
the tidal-mark — tell only too plainly how yain is the resistance even
of the hardest rocks to the onward march of the ocean. The rate of
waste along some parts of those islands is so rapid as to be distinctly
appreciable within a human lifetime. Thus the start-point of Sanda
was found by Mr. Stevenson, in 1816, to be an islMid every flood
tide ; yet, even within the memory of some old people then alive, it
had formed one continuous tract of firm groimd. Nay, it appears that
during the ten years previous to 1816 the Channel had been worn
down at least 2 feet.

** Some few years back (about 1874), when the Channel fleet were
in the north, they attempted to pass to the westward through Westray
Pirth, in the teeth of a strong spring flood ; but all the Queen's horse-
power and all the Queen's men could not do it, and they had to turn tail.

'* Short storms of great violence are not the worst, being surpassed
by the long continuance of an ordinary gale, and during great storms
the devastation and ruin is very great. During a peculiarly severe
storm, in 1862, in Stomna (in Caithness), in the Fentland Firth, the
sea swept right over the north end of the island, lodged fragments of
wreckage, stones, seaweed, &c., on the top, 200 feet above ordinary
sea level, and then rushed in torrents across the island, tearing up the
ground and rocks in their course towards the opposite side. The
heaviest rains and the most prevalent and strongest winds are from the
south-west and, south-east."

As the west coast of Ireland is largely made up of the same classes
of rocks as those forming the Hebrides, the Shetlands, and the Orkneys,
and is more fully and directly exposed to the force of the Atlantic
waves, it is reasonable to assume that all that has been herein stated
as to the destructive force of the ocean on these islands holds good,
even more strongly, as regards the western coast of Ireland — ^the
** Wild "West," as it has been called — and we may admit that wear and
waste is going on there incessantly, even although we have no obser-
vations in support thereof.

As regards the coast of Britain, from the coast of "Wales south-
wards, more has been observed and noted, and the resulting wear
recorded would tend to show what must have been the waste along the
south-western and southern coasts of Ireland, even although we had
no records regarding them.

There is a very interesting article, by D. Mackintosh, Esq., f.o.6.,
in the QtMrt, Joum» of the Geolog. Soe. of London, vol. xxiv., 1868^

O'Bbilly— On the Waste of the Coast of Ireland, 8fc. 117

p. 279, '* On the mode and extent of the encroachment of the sea on
Home parts of the shores of the Bristol Channel." He says : '< In a
paper read hef ore this Society, Noyemher 8, 1865 (see vol. xzii., p. 1),
Vi. Qoodwin Austin brought forward very satisfactory reasons for
amdading that the area of the Bristol Channel was dry land during
the (now submarine) Porest Era, and that it must afterwards have
snbsided to a depth of at least 1 20 feet, as a submerged land is now
found at that depth under the sea level. Whatever relative changes
in lerel the land and sea may have subsequently undergone, it is
obnons that the general tendency of the " waves " and " ground sea,*'
or'' Atlantic drift," which is sensibly felt beyond Watchet (18 miles
▼est of Bridgewater), has been to destroy the contour of the gradually
lisbig shores by wearing them back into difCs. As a consequence, the
extent of the encroachment since the forest area went down may in
some localities be approximately ascertained."

He then gives a section of the coast-line near Watchet, and ex-
amines the relation of the difb to the exposed shore, and says : "It
vill be obvious that the sea has recently had no small share in the
deoodation of the Bristol Channel, whatever may have been the cause
of the original excavation."

As evidence of the ** recent rate of encroachment, he says : "I
learned from a very old fisherman at Watchet, whose veracity no one
•eemed to doubt, and whose statements concerning the encroachments
of the sea were directly or indirectly corroborated by others, that not
Qore than 150 years ago a brewery, belonging to a Mr. Davies, stood
at a distance of at least 200 yards from the present diff, east of
Watchet harbour, and that the rocks under its site are still recognised.
There was likewise a village (or hamlet ?) called Easenton, to which
the fisherman's great-grandfather was in the habit of going for a mug
of beer, the rite of the furthest east part of which is now about one-
fonrth of a mile from the coast."

He adda a note: ''I found the following record among the
documents of a solicitor of Williton : — ' North of Eacloze, a part of
Watchet in 1662, a bam and other buildings, with orchard and garden
beyond; in 1751, all gone to sea.' To the west of Watchet the sea is
encioaching on a high ridge and imdermining large blocks of sandstone,
interwoven with alabaster, which it carries away entirely, or scatters
and piles in a strange confusion. The configuration of the sea-bed,
asder and for some distance from the clilEs, very much resembles the
nneven ground at the base of many inland escarpments."

(p. 281.)—'' JBneroaehmsntM on Westan-super-mare. — The sea is

118 Proceedings of the Royal Irish Academy.

conyerting slopes into difCs, where it is not silting up flat aieas from
Brean Down to a considerable distance northwards. Near Weston
the sea is forming a line of cliffs on the north-western side of
Weorle Hill. At Bembeck cove its encroachments hare disclosed,
or rather nearly destroyed, the last remnants of a genuine raised
beach.''

Professor John Ehys, m.a., of Oxford, in his '^ Celtic Folklore oi
Wales and Isle of Man," has some interesting legends as regards the
waste on the Welsh coast, in vol. ii., p. 401, under the heading,
** Triumphs of [the Water WorW^ He says : " More than once tn
the last chapter was the subject of submersions and cataclysms brought
before the reader, and it may be convenient to enumerate here che
most remarkable cases, not to mention that one of my informants aad
something to say (p. 219, vol. i.) of the submergence of Caer Arianrlod,
a rock now visible only at low- water between Celynnog Fawr and
Dinas Dintte, on the coast of Aif on ; but, to put it briefly, it is ui
ancient belief in the principality that its lakes generally have swallowed
up habitations of men.'

(p. 403.) — " Perhaps it is best to begin with historical even*j,
namely, those implied in the encroachment of the sea and the sand, «n
the coast of Glamorganshire, from Mumbles in Gower to the mouth :>f
the Ogmore, below Bridgend. It is believed that formerly the shor^
of Swansea Bay were from three to four miles further out than tie
present strand, and the oyster-dredges point to that part of the baj,
which they call the " Green Grounds," while trawlers, hovering ova*
these sunken meadows of the Grove Islands, declare that they can
sometimes see the foundations of the ancient homesteads, overwhelmed
by a terrific storm which raged some three centuries ago. The old
people sometimes talk of an extensive forest, called Coed Ariam
(* Silver Wood'), stretching from the foreshore of the Mumbles to
Kenfig Burroughs, and there is a tradition of a long lost bridle-path
used by many generations of Mansels, Mowbrays, and Talbots, from
Penrice Castle to Margam Abbey. All this is said to be corroborated
by the fishing up, every now and then, in Swansea Bay, of stags'
antlers, elk horns, those of the wild ox, and wild boars' tusks, together
with the remains of other ancient tenants of the submerged forest.
Various references in the registers of Swansea and Averavon mark
successive stages in the advance of the desolation from the latter part
of the fifteenth century down. Among others, a great sandstonn is
mentioned which overwhelmed the borough of Cynffig, or Kinfig, and
encroached on the coast generally ; the series of catastrophes seem to

O'Reilly— 0« the Waste of the Coast of Ireland, Sfc. 119

have calminated in an inundatioii, caused by a terrible tidal wave, in
the early part of 1607.

" To return to Kenfig : what remains of that old town is
near the sea, and it ia on all eddes surrounded by hillocks of finely
powdered sand, and flanked by ridges of the same fringing the
coast. The ruins of several old buildings, half buried in the sand,
peep oat of the ground, and in the immediate neighbourhood is Kenfig
Pool, which is said to have a circumference of nearly two miles. When
the pool formed itself I have not been able to discoTor."

(p. 404.) — " On this coast is another piece of water, namely,
Crymlin or Crumlin Fool, now locally called ' the Bog.' It lies on
Lord Jersey's estate, at a distance of about one mile east of the mouth
of the Tawe, and about quarter of a mile from high- water mark, from
which it is separated by a strip of ground known as Crymlin Burrows.
The story about this pool, also, is that it covers a town buried be-
neath the waters. An article of the South Wales Daily News, of
February 16th, 1899, says of Crymlin : * It is said by the old people
that on the site of this bog once stood the old town of Swansea, and
that, in clear and calm weather, the chimneys, and even the church
steeple, could be seen in the bottom of the lake.' The lake was at one
time much larger than at present.*'

(p. 416.)—** The writer of an article in the Monthly Packet for
1859 gives a sketch of the story of the country overflowed by the
oeighbouring portion of Cardigan Bay, mentioning, that once on a time
there were great cities on the banks of the Dovey and the Disynni.
'Cities with marble wharfs,' the author says, *bu8y factories, and
c:. rchea, whose towers resounded with beautiful peals and chimes of
^u]J.' The author goes on to say, that Mausna is the name of the city
<'Q the Dovey ; its eastern suburb was at the sandbank now called
Borth, 'its western stretched far out into the sea.' The name Berth
«t..Qdfl for * Y. Borth,' i.e. * the Harbour,' "

Passing from the south of Wales and the Bristol Channel to the
peninsula of Devon and Cornwall, which is beaten by the waves of
the Atlantic in all their force, we find in Sir Henry de la B^che's
Report on the Geology of Cornwall, Devon, and West Somerset (1839)
aveiy interesting chapter on the '* Action of the Sea on the Coast,'^
from which the following citations are made : —

p. 435.) — ** As about 472 miles of coast, exclusive of estuaries and
miiior irregularities, are in the district under consideration exposed to
the action of the sea, considerable facilities are afforded for the study
of this action, more especially as the rocks brought within its influence

120 Proceedings of the Royal Irish Academy,

are very difPerent, and the conditionB under wldch they arc exposed
are also yariable. We soon perceive that the streams of the tide, to
the movement of which so much abrading power has been attributed,
have very little influence upon this coast ; that it is chiefly in those
places where the tides have little strength, but where the softer rocks
prevail, and the exposure to the prevalent winds, and hence to
breakers, is considerable, that the chief loss of land by the action of
the sea is greatest. In fact, the tides rarely run beyond one or two
miles per hour, except round the headlands, which are nearly all com-
posed of hard rocks, the softer parts of the coast having been hollowed
out by the breakers, during the lapse of ages, into creeks, coves, and
bays, so as to be removed from the main stream of the tide."

It would be difficult to form a correct idea of the geological
time during which this coast has taken its present form, when we
perceive so many hard rocks worked into creeks and coasts, and learn
as indeed from their aspect we would expect, that no appreciable
change has been observed in them during the memory of man, we can
readily believe that the present condition of this coast is due to no
ordinary lapse of time as reckoned by him.

(p. 436.) — '* The hard quartzose and trappean rocks of Trevose
Head, the greenstone and trappean rocks of Pentire Point, near
Padstow, the hard slates of St. Agnes Head, the compact sandstones
and hard slates of Godrevy Head, the greenstone of St. Ives Point,
the greenstone and hardened schistose rocks of Gurnards Head, and
the granite of the Land's End — may be readily supposed incapable of
being appreciably wasted by the action of the streams of tide which
pass over them. In like manner the granites of many other points in
the Land's End district.''

(p. 437.) — " A very short experience of the destructive effect of
breakers will be sufficient to afford evidence of the form which a
coast must take according to the variable manner in which it may be
exposed to them ; so that after the lapse of ages any given coast will
readily show, from the wearing away of the softer rocks into creeks,
coves, and bays — ^the harder being gradually left to protrude as points
and headlands — that it has been scooped out according to the unequal
resistances of the rocks on the one hand, and the variable power of
the breakers on the other, due allowance being made for the original
form of the land, and the indentations produced by the entrance of
the sea, at its high-water level, into valleys, producing estuaries."

(p. 438.) — " It rarely happens that breakers do not fall on the
western part of the coast, even in the calmest weather, undulations

O'Reilly— Q» the Waste of the Coast of Ireland, 8fc. 121

from the Atlantic, prodnoed by gales of wind on some part of it, not
too distant, to allow the wayes so caused entirely to subside before
they meet the land, rolling in upon the shores, and often breaking
with a heayy crash on them, causing, probably, as much abrasion as
the wares at any other time. These are technically known as ' ground
swells,' as they tear up the beaches exposed to them, hold abundance
of pebbles or sand, as the case may be, in mechanical suspension, and
eren seem, as it were, to rise from the bottom of the sea, hurling
the mechanically-suspended substances upon the beach or against the
cliffs with a heavy grinding noise, frequently beard far inland. As
these ' ground ewells ' very often roll in from the westward, the
coast from Morte Point to the Land's End is much exposed to it,
paiticalarly towards the latter place. When, as it often does, the
Atlantic or ground-swell rolls from the south-westward, a large
portion of the southern coast, otherwise protected, is exposed to it ;
generally the formidable breakers caused by the swell, even in calm
weather, do not extend beyond the Prawle and Start-Points."

(p. 439.) — ** The ordinary breakers are well known to be the
crash of the waves produced by winds blowing on the coast, and
according to the exposure of the coast to open sea, other things being
eqiul, are their magnitude and destructive powers.

'* In many situations common atmospheric influences so combine
with the action of the breakers to produce the destruction of the
cliffs, that it may be difficult to say whether the loss of land may not
be more due to the one than the other ; in most places, however, the
breakers nearly cause tbe whole loss, leaving isolated rocks to show,
t'> a certain extent, the destruction they have caused. The cliffs,
^rom Trevoee Head to new quay, may be selected as affording a
good example of the destruction of a coast by the action of heavy
breakers."

(p. 440.)—" The rocks between Teignmouth and Lyme Regis suffer
much loss from the action of the breakers upon them to an extent
that, if the latter possessed the average force of those which wear
away the coast last mentioned, very considerable inroads would be
ni*de upon them, and the bay would be much enlarged northwards in
the course of a few thousand years. Independently, however, of the
loss by landslips, the Lias cliffs near Lyme Regis are readily seen to
^ washed away by the breakers, as may easily be observed between
Ckannoutb and that town, as also to the eastward of it. Consider-
•hk waste of this coast has thus been occasioned within the memory
of persons now living — ^a waste first recorded, we believe, by De Luc,

122 Proceedings of the Royal Irish Academy.

and one which is still proceeding *at a considerable rate, the fall of
the clifEs being frequent, particularly in bad weather. Much loss
of coast is sustained near Sidmouth, particularly in the direction
of High Peake, the small green-sand cap on the top of which will
disappear at no distant date, geologically speaking, and be onlj
known to hare existed from the geological maps now constructed of
that part of the country."

(p. 441.) — ''The isolated Ked Sandstone and Conglomexate rocks
on several parts of the coast, the evidence of its former boundaries
between Sidmouth and Teignmouth, are often worn away by the
breakers in a manner well illustrating the unequal resistance offered
by drfferent portions of them. One of these isolated maases, named
' the Chit Eock,' which long rose above a ledge on the west of Sid-
mouth, was a few years since upset, in consequence of a central part
of it having been finally worn so thin that the upper portion was
knocked off by the breakers in a gale of wind."

Near Watchet, East Quantockshead and Iiittle Sto1(e, the lias
red marl and sandstone of the coast suffer much annual loss — not so
considerable a loss as they would sustain if exposed to the action of
heavier breakers than now reach them, even to such as now batter
and wear away similar rocks near Lyme Begis and Sidmouth, bat
sufficiently to become well marked.

The minor effects of the breakers are easily seen on every part
of the coast, the harder rocks resisting their action, while tbe
softer are worn into caves, creeks, and coves, of every variety of
form.

(p. 442.) — '^ The sea, by its action upon rocks of unequal hard-
ness and the fall of some compact portions of them, or of la]^
indurated nodules contained within them, often raises a barrier
against itself, and the lower portions of cliffs become protected for a
time ; beyond that they would remain otherwise firm, even in some
cases producing points of land composed of these blocks or more in-
durated masses of rock. {Note). — Indeed it may be said, on this
head, that beaches generally, more especially shingle beaches, are
only the harder part of abraded cliffs reduced to somewhat smaller
dimensioDs).''

(p. 443.) — " Having thus briefly adverted to the destructive action
of the breakers on this coast, we should notice the protection afforded
by the common beaches thrown up in front of low lands.'* (The
author then enters into considerations regarding the formations and
influences of these beaches and the formation of dunes.)

O'Rbilly— On the Waste of the Coast of Ireland, 8fc. 123

Tinder the heading of ** Cornwall,*^ " Lewis' Topographicfd Diction-
aij of England " saye (p. 516) : '' The sea has considerably encroached
nponthe coast within ^e last sixty years (1771 to 1831), in the
Irandreds of Stratton and Lesnewth, especially near Bude Harbonr,
where the waves are rapidly wasting the sandhills."

As a considerable extent of the north-western and western coasts
of France are fully exposed to the action of the Atlantic storms and
hreaken, and consequently to waste, it will be of interest to note
whst is said lelatiTe thereto in the *' Dictionnaire de Qeographie
UniTeneUe de Tivian de St. Martin " ; and for that purpose the coasts
of the diiferent departments exposed to the Atlantic action will be
eottfidered in their succession from east to west — that is, Calyados,
Handle, Ille-et-Yilaine, Cotes du Nord, Finisterre, Morbihan, Lmre
ln^., and Yendfe.

Speaking of the *' Configuration physique " of the Department of
CalTadoB, the Dictionary says : ** Les c6tes du d6partement o£Erent un
dereloppement de 120 kil.-carr^B formant une courbe rentrante peu
iinnense, bord6e de falaises dont la mer ronge insensiblement le pied,
et d'nn difficile acc^s k cause des nombreux rochers, d6biis d'un rivage,
qne les souTenirs historiques constatent s'6tre plus ayanc6 en roer
antiefois qu'aujourd'hui. Les rochers de Cakados ont donne leur
noo an d^partement ; ce mot de Calyados, est le corruption de I'Es-
P<^ol SahadoTf et ' le Salvador ' 6tait un yaisseau de I'inyincible
^nnada qui se brisa sur cet 6oueil."

Speaking of the coasts, the Dictionary says : '^ Cette cote a subi
<Iea revolutions dont les vestiges sent visibles sous les eaux et dans
rint^eur des teires. Le plateau du Calvados n'est pas autre chose
^Q^la hase d'nn prolongement des falaises du Bessin qui a 6t6 rase par
la mer. On ne saurait gu^re chercher ailleurs que dans ces falaises
^^^trnites, la source des attemssements qui ont combl6 les anciens
goifea de TOme et de la Dives.

*' La foT^t de Hautefeuille, disent encore les traditions locales,
omhragait au commencement du XVI"* siSde la large lisi^re sur
^aelle s'^pandent aujourd'hui les marges audessus de BemiSres et
de Langrane. De nombreuses et puissantes racines s'enfoncent en
eflet dans les fissures des rochers mis 4 nu. Les commissaries du Car-
^^ de Richelieu trouv^rent, k d6faut de la foret, un petit port 4
^cnuires ; la Seniles y debouchait en s'inflechissant cL TE., et de vastes
B^^nis s'^tendaient k Pouest jusqu'i Anelles 4 12 km, de distance.

124 Proceedings of the Royal Irish Academy.

Deptds la mer a d6Tor6 le port et lea maiaiB ; elle a racoorci le conn de
la Seulles de 3 km., et il ne reste plus du Mvre de 1640, qu'xme eerie
de bas fonds o\i la retraite de la iiiar6e laisse de long;ae8 flaqnee d'eao.
II est du reste permis de yoir iin indice de transformations bien plus
yastes, dans les vestiges de retranchements romains qui, de Revien i
Tailleville et k St. Aubin enveloppent Courseulles. La cbarrue met
souyent 4 d^couvert dans leur yate enceinte des briques, des fragments
de poterie antique, et des m^dailles. H est peu probable que les
Romains se fussent si f ortement install^s sur ce point s'ils n'avaient eu
qu'une insignifiante station nayale k prot6ger, et Tandenne confignn-
tion du riyage donnait sans doute 4 leur Stablissement militaire des
raisons d'etre qui n'existent plus. La cote est encore rongee par le
flot ; mais, 4 mesure que les dentelures s'en emoussent, elle donne moins
de priseaux courants.

The same Dictionary, speaking of the English Channel (the
^'Manche"), says: ''La triple action des m6t6ore8, des courants et
des yagues de maree, continue de ronger incessamment les liyages da
grand d6troit. Batz, Traigoz, les sept ties, Br6hat, sont les restes
d'un littoral disparu. Au N. les Scilly Is., le cap Land's end, le cap
Lizard, se dressent dans une mer toujours agitee par le flotet le jusant.
Les m6t6ores fissurent en haut les rochers que la mer sape en bas, et
d'6normes blocs s'6croulent daps les flots et y forment des ^cueils.
De quelques uns de ces 6cueils Pimagination populaire a fait les ' Aimed
Knights ' (les cheyaliers armee), d6fenseurs des continents ; mais i
leur. tour ils cedent 4 la pression des vagues et s'engloutiront tout k fait
k Vest des granits r^sistants de Treguier, les cotes ont 6t6 fortement
entam6e8 par le flot, et des p6ninsules comme Herviant, Yerderlet,
Cezembre, sont deyenu des iles.

<* Dans le sud de Tile de Jersey des roches et des greves, que ser-
yaient de fondement 4 des terres disparues, s*etendent 4 mar6e basse.
jusqu'4 3 k./metres du riyage.

''Les falaises de Normandie compos6es de mat^riaux beaucoup
moins durs que les promontoires de la Manche occidentales sont plus
facilement entam^es. Le recul des falaises de la Seine Inf^rerieure et
du Calyados est d'enyiron 25 4 30 c/metres en moyenne par an. Sur
les cotes de P Angleterre situees en face, Perosion est plus rapide encore.
La masse totals des roehsrs que la Manehs Orientals hrise ehaqus anti^e
est ivaliUe A etwiron 10 millions de m.j cubes} En 1862 pendant une

* Mar^chal, ** Ann. des ponts et chauss^es."

O'Rbilly— 0« the Waste of the Coast of Ireland, 8fc. 125

tempete les rochers de la Heve se sont eboules sur une ^paisseur de
15 metres. Banfi les ages ant^rieurs I'Angleterre etait rattach6e 4 la
tvrre ferme, par un iethme k jonction qui a 6te graduellement rompu
par le choc et la pression des vagues. Plus k Test au Pas de Calais,
la falaise du cap Gzis-nez r6ciile en uiojenne de 25 metres par si^cle
A ToQest de Dover, la f alaiae de Shakespeare d'apr^s Mr. Beete Jukes
a recule de 2 k. /metres depuis le temps de Jules GsBsar.

'* Hais le flot qui d^truit, 6difie aussi, et k Textremit^ orientale de
la Ifianche on pent constater un des examples les plus frappants de
rinflnence des marges sur la forme des rivages. ' Une plaine basse et
nurecageose dite, Bomney Marsh,* qui se termine k la pointe de
Bimgeness s*est formSe par I'arrSt du courant de mar6e qui vient de la
¥er da Nord. Le flot de I'Atlantique arrache k la base des falaises
crajenaes de Hastings des d6bris siliceux qui ne pouvant passer la
pomte de rencontre des mar6e8, s'arretent le long du Bomney Marsh
et le prolongent continuellement en mer. La pointe Dungeness
s'accroit d'environ 0* 50 par an.' Ailleurs la mer a procM6 par
enraaement, en d6posant des debris d'algues et d'animalcules m616s
aa sable et k Pargile, et c'est ainsi qu'elle a fait avancer le profil
(lea riTages dans le golfe de Carentan, k la racine de la p^ninsule
dc Cotentin."

The same Dictionary speaks as follows of the coast of the Depart-
ment of " nie et Vilaine " :—

"leplus 41ey6 (plissement de terrain) de tons, Haute-Foret, n'a que
255 m. ; dominant les sources de TAff, affluent de TOult, il se dresse
pm des fronti^res du Morbihan, dans la foret de Paimpont, dont les
6070 hect. sont un faible debris de I'antique et c61dbre f or^t de £roc6-
liaode. Toutes les coUines du pays, tons ses plateaux, tout cela fut
jadia une foret immense, qui de plus en plus s'6£Cace.

'*De ces bois sans fin faisait partie, du temps de Eomains, le Scissia-
MRumttt, la foret de Scissey, envahie brusquement par la mer en 709
<^t 00, et deTenue alors une grive mar^cageuse qu'on a des86ch6e, et
remplacee par une plaine humide de 15,000 hect. appartenant k 23 com-
mimea. C'est le marais de Dol, entre le massif de Dol et celui de St.
Malo ; < Petite Hollande,' ayant ses dignes parfois 6br6ch6es par les
aasanta des flots, ses canaux, ses moulins 4 vent, ses marais, ses brumes
<?nae«. CTeet en per^ant, pendant les grandes mar6es d*6quinoxe, un
cordon littoral allant des caps de OranviUe au Grouin de Cancale
qae la mer recouTrit les terres basses du Seissiaeum n&mu$, Ce fut la
T>las grande perturbation qui se soit produite sur la cote de la France.
La digue qui protege le marais de Dol domine d'un m^tre et demi les

126 Proceedings of the Royal Irish Academy.

pluB hautes marges d'6qninoxe. De I'embouchure du Ginonlt d la
pointe de Ohateaa Bicheux la riyage reete plat, mais k cette pointe
commence la famease cote de Bretagne, I'une des pluB dechiqnet6es,
des pluB sauTages, des plus sombres, deB pluB orageuses dn monde
entier. Le littoral n'est pas moins d6chiqaet6 qu'^ Test et les terribles
dcueils dont la mer est parsem^e, le rendent tr^s dang^reax pendant
leB mauvaifi temps."

Speaking of the Departement de Finisterre, this Dictionary Bays: —
'* Comme I'indique son nom, le Finisterre fonne au sein de I'Ocean
Payant garde dn Continent Armoricain, les deux points de Baz et de
St. Mathieu, celle-ci granitique, celle-lst gneissique, a'ayancent dans
les flots et Bontiennent vaillamment leurs premiers assauts; des roches
de Bcbiste, moins r^sistantes, ont c6d4 sous I'effort de la T&gue foriense,
et y a crens^ la rade de Brest et la baie de Douamenez. Fouett^es 4
fois par joor aux deux marges montantes et aux denx jnsants, second
BouTent par des yagnes de tempetes, les extremitSs de la peninsule eont
assaillies par ces terribles agents de destructioD. De lit tontes ces
petites lies violemment B6par6es de la terre ferme, ces pans de mur, a
demi-ens6vilis sous les sables, et ces traditions de yilles d6triiite8 qui
temoignent partont des progr^s de .POc^an. A I'aide des jalons de
granit 6pars sur les bas fonds, on pent, avec quelqne certitude fixer k
25 km. le minimum des enyahissements de TAtlantique depuis I'anti-
quit6. Dans la baie de la ForSt, sur les cotes de Penmarch, dn Bas,
du Conquet des constructions et des • troncs d'arbres temoignent des
pertes r6centes qu'a fait ici la terre ferme."

Describing the Department of the Loire Inf6rieure the same
Dictionary says : '^ Le ' silhn de Bretagne ' est 61ey4 presque par-
tout de 60" el 80" ; — seulement k 91" de hauteur supreme pr^s du
Temple de Bretagne. Entre Sayeney et Port Chateau il commande
Us Briires, marais souyent inondes, prairies tourbeuses, dont la plus
yaste, est la Grande Brihre^ entre Loire au S. la Yilaine an N.
et les hauteurs de Quemande qui la s^parent de I'Atlantiqne i
rOuest. Longue de 15 km., large de 10 km., elle a bien 8000 hect.
k I'altitude moycnno de 3". Ancienne forSt mouill6e, pleine encore
de troncs noircis par un long s6jour dans le tourbe, on la voit tour &
tour et suiyant la saison, nappe sans profondeur, oii Pon chasse \e»
oiseaux d'eau, prairie oO pait le mouton, et d'oii leB Bri^rons tirent par
milliers de tonnes la tourbe-— entre la grande Bridre et les ' bogs '
d'Irlande, il n'y a point de difP6rence — des for^ts qui croissaient
autrefois sur le sol enyahi pai les to^urbes, ont 6t6 ^touff^es, et les
arbres tomb6s pour la plupart, dans la direction du yent principal^

O'Bbillt— On the Waste of the Coast of Ireland, 8fc. 127

•e montrent encore paiiaitement con8ery68 soas la couche k demi
carboniB^ des spbaignes; lenr boiB devenu aussi noir, aussi dur,
que l'6Une, est assez bien con8ery6 pour ^tre d6bit6 et transf orm6 en
parquet

As regards the D^artment de YendSe, the Dictionary eays : —
*' L4 dans oe que nous nommons maintenaut la Baie de Bourgneuf
entre la c6te de Erance et Noirmoutier, elle a depos^, depose et
deposera, des alluvions jusqu'd. comblement, deyant le continent
toajonrs accru, i I'abri de la roche, et du sable de I'ile, qui s'41^Te
en briselame contre les yagues du grand large. II n'y apr6sentement
ki qa'one seule terre en mer, Noirmontier, mais quand ce rem-
blaiement commen^a, le rivage regardait trois lies, Noirmoutiers,
an K.W., Betz, rocher de sohiste au 8.E., et entre les deux,
rUe de Monte. Ce que rOc6an laissa tomber ici, ce qui y torn e
encoro, c'est la mine des caps bretons, la yase de la Loire, les menus
fragments des caps de Noirmoutiers, et quelques boues des ruisseaux
da HTsge. De plus le sol s'exhausse, du moins on le croit. Tout
eDdigaement k part, la France a gagne durant les deux demidres
nicies quelques 700 hectares dans la Baie de Bourgnouf, ainsi appel6e
de la yille de Bourgneuf en Betz (Loire Inf.) jadis riyeraine, tandis
que 2 k.m. de plaine basse et de marais salants, la s^parent du flot ft
cette heure ; de m^me Beauyoir-sur-mer est k 4 km. de la mer."

If it were merely wished to establish the general fact of the
wasting action of the sea on the coasts of Great Britain and Ireland
many more examples of such wasting action could be cited from
Lyell's " Principles of Geology," particularly the excellent details as
regards the north-east of Scotland and the east and south-east of
EngUnd. The main object in making the citations already giyen, has
been to allow of a more just appreciation of the action of the Atlantio
waves on the coast of Ireland, and of the consequent waste which
must haye been, and is incessantly going on day by day, although
onobseryed and unrecorded for the most part. Turning, therefore,
to the examination of the coasts of Ireland, that of the eastern side
will be considered first, commencing with the coast-line of Wexford.

Th^NatumtH Oauiteer (1868) says of the coast: **The coast is
generally low and shingly from Kilmichael Point, in the north-east, to
Wexford Harbour, a distance of nearly 30 miles, and is skirted along
the entire line by a series of sand-banks marked at their northern
extremity by the Arklow ship-light,"

Dr. Joyce, in Philips' ** Atlas of Ireland" (1833), says of the

ma.^ PBOO., xxiy., bbo. b.] ^

128 Proceedings of the Royal Irish Academy.

^' Coa»i4ine " : " The coast is low, and for the most part sandy, inter
rapted in a few phioes hy fringes of rock ; it is unbroken from Kfl-
michaft] Point to the Bayen Point ; hnt from this to Wateiford
Harbour it is much indented by inlets."

In the Proceedings, Roy. Ir. Academy, Series ii., Science, vol. iu.
(1877-83 )y Mr. George H. Kinahan published a Paper on Sea-beaches,
especially those of Wexford and Wicklow, of which the following are
extracts : —

(p. 191.) — " During the time I haye been engaged on the Geolo-
gical Surrey of Ireland (oyer twenty years) I haye had, when stationed
in maritime districts, fayourable opportunities of obserying the sea-
beaches. This has been especially so during the last six years
while I haye been engaged in examining the counties of Wicklow and
Wexford, and in these years the obeeryations made were both numerooB
and minute."

(p. 192.) — *' The western Salteecuiient runs north-east to Kihnore
Pier, where it turns westward, and forms the ' counter tide ' that
meets the Hook current at the Keragh Island. At the meeting of
these two currents a shoal has accumulated. Under ordinary dream-
stances the current from Hook carries the beach with it only to the
neighbourhood of Keragh, as proyed by the fact that the stones from
Hook promontory are rarely found beyond Keragh. The 'counter
tide ' west of Kilmore carries the beach north-west along Ballyteige
Bay, and during the last 40 years (since the Ordnance Map was made)
has lengthened the Ballyteige sand-hiUs more than 200 feet."

(p. 193.) — <* The Cahore shingle beach is about 3 miles long, and
is Imrgely composed of fragments of the Greenore and Camsore rocks ;
with these there are others from the cliffs along the Blaokwater
coast. Opposite Gourtown (north of Gahore) is the ' nodal^ or ' hinge-
line,' of the tides on the south portion of the Irish Sea, where the
rise is least and the current greatest. The refuse from the shipping
at Gourtown Harbour, such as bits of brick, tile, slate, coal, &c, are
principally stranded along the beach a few miles soutii-west of Kil-
michael Point. On this beach, Greenore and Gamsore rock fragments
are not unoonunon ; but in the two small bays to the north of the
Point the grayel and shingle is made up almost solely of the local
rocks, many of the fragments being more or lees angular. The dSbris
from the shipping at Arklow is principally beached on the strand
south-west of Mizen Head.

(p. 195.)— *':n^ effecU of the Wind Wawti The waves of this
elass that act on this coast are of two kinds, yix., ' ground swells,' or

O'Rbilly— 0« the Waste of the Coast of Ireland, 8fc. 129

wavei generated by storms in the Atlantic or the Channel, and the
waves directly due to the winds blowing on the coast. Their effects
aTD either to pile up and fill the beaches or to cut them out. If they
strike the beach at a right angle, they fill it up, forming ' fulls ' or
' itorm beaches, while, if they are running in a more or less opposite
direction to the flow-tide, they cut out the beach."

(p. 197.)—'' Consequently, during the last 40 years, the coast-
line between Lady's Island lake and Kilmore has been considerably
denuded away, especially in the vicinity of St. Patrick's Bridge.
These beaches, during the continuous east and north-east winds of
the spring of 1876, changed from their ordinary grayel into " fulls "
of shingle.

'* In the North Bay all winds seem to ' cut out,' this being due to
the complication of the tidal currents, the beaches rarely being full
( xcept in the summer and autumn, when there are no winds. On
iccount of the great cutting out along the beach, the marginal cUlb
hare been yasUy denuded within the last 40 years. The ' Cahore
Single beach,' at the north of the bay, is fullest during south and
south-east gales, while it is cut out by winds from the north-east,
and by ' ground swells.' After south and south-west gales it is often
smothered up with fine sand blown from the adjoining accumulations
of ^lian Drift.

** A little to the north of Cahore Point is Poulduff Pier, with the
l^aehes accumulated since it was erected, while farther north are the
piers and other works at Courtown.

"On the coast-line, south (fig. 4, pi. 6) and north (fig. 5, pi. 6)
of Kilmichael Point, there has been considerable denudation of the
.«and-hills since the Ordnance Survey was made (i.^. 40 years ago).

'* In the first-mentioned localities oyer 37 acres have been carried
away by gales from the south-east. Here there is an exceedingly
swift tidal current to the north-north-east, which, under ordinary
^^^fcnnutances, carries all the beach with it, and leayes no protection
between the sea and the sand-hills ; consequently, under these circum-
stances, during south-east gales, the wind-wayes have full power on
the latter, which they then rapidly denude away.

" (aVo<«.)— (This is a most remarkable place, as in recent years the
Mnd-hiUs at one time seem to be forming and at other times wasting
^▼ay. Some of the old men can point out the extensions of the sand-
hills prior to the Ordnance Survey, and the roads that used to lead to
thf-m, which now end at stifE cliffs; while one old man, in June,
I'^TS, pointed out in a diff that had only been uncovered the previous

130 Proceedutgs of the Royal Irish Academy.

winter, an old quarry that must have been worked with inm took
prior to the accumulation of the sand-hillB that existed there when
the Ordnance Survey was made).

" Northward of Kilmichael Point, in the bay at the mearing of tiie
eonntieB Wicklow and Wexford, the denudation of the sand-hills hu
also been considerable within the last forty years (fig. 4, pi. 6), more
than 20 acres in the townland of Gloon Lower and Upper having been
swept away."

(p. 199.) — " It ought to be specially pointed out, that the storms
which cut out the beaches may not be the same as those which denude
away the marginal difis. A small storm, when the strand is empty,
may do great damage to the coast-line.

<' {Not0.) — After very wet seasons great falls of cliffs often take
place. The natives will often tell you that so many yards are going
yearly, and, in proof of this assertion, will point to the waste of the
previous winter, they supposing the same happens every year. The
greatest falls occur at the highest cliffs, on which account the greatest
waste is supposed to be taking place in those localities; but, after
careful calculation, I find this not to be the case. None of the high
eliffs reach an average waste of 0*75 feet per annum, and generally
the loss is less than 0*50 feet per annum, while in places the low
cliffs have been denuded away as much as 2*5 feet per annum. The
greatest denudation on the whole line of coast between Hook and
Dalkey is at the low cliff near St. Patrick's Bridge, Kilmore.

''^ Extraordinary high tides, unaccompanied with wind, seem to
do little or no damage on an open seaboard. In March, 1867,^ there
was a remarkable high tide on the coast of Galway, the traces of
which were scarcely perceptible along the open coast, even on tiie
sand-hills ; but in the land-locked bay? it did considerable damage to
the piers and sea-walls.

•* On January 3rd, 1877, there was on the east coast a very high
tide, which along the Wicklow coast was accompanied by a very
moderate wind. This did considerable damage to the Dublin and
Wicklow Railway between Greystones and Wicklow ; not so much
by the direct force of the waves as by their height, they flowing over
the line, and the overflow cutting into the land side of the embankment,
thus gradually eating out the beaches. Between Newcastle and
Wicklow Chemical Works it encroached in many places, as mueli as
8 yards into the Morrough (anglice^ the plain)."

^ 1667. 2. Sunspot minimum.

O'Rbilly— On the JTaate of the Coast qf Ireland, Sfc. 131

(p. 202.)—" Some of the big waves or * roUew ' that visit the
coast, on rare oecasionB are due to earthquakes."

(p. 203.)—" At the Kish Bank, off Dublin Bay, an attempt was
made to erect a light-house on screw piles ; but it was given up, as
the flanges of the piles were broken by large blocks in the accumula-
tion of sand."

(p. 207.) — " On no coast are groynes so necessary as that now
nnder consideration, especially in parts of Wexford and Dublin where
valuable land is yearly disappearing ; yet they have been erected
only in isolated spots."

In the Memoir of the Geological Survey of Ireland, accompanying
iheetaNos. 158 and 159, including the district around Enniscorthy,
County Wexford, by G. H. Kinahan, m.b.i.a. (1882) the following
obserrations occur : —

{p. 32.) — " Some of the newest accumulations are the estuarine
wiikmds of the north intake in the Wexford lagoon. These mudlands are
described in the published memoir to accompany sheets 169, 170, &c.
To the north-east of this intake, at Curracloe and Ballinesker, both
inade and outside the jSolian Drift, which is the northern end of the
HaTen Spit, is deep peat. That outside is cut when the tide is out,
and carried up above high- water mark, to be dried and made into
torf. Outside the marginal ^olian Drift hills in the Bam Channel,
peat has been dredged at the 4 fathom line."

(p. 34.) — " In Ballynaclash, about i mile south-west of the mouth
ot the Blackwater there is the following section : —

"(Section No. 4.)

feet

8. Sdl,

2 "

7. CUy,

6. Peat,

from 1 inch to

6. Blue Glaj,

from 7 inches to

1-6

4. Peat,

from 1 inch to

1-26

S. Pebblj Clay, very irregakr, as it is \

fillmg what

seems to be a water f

. 8

excavation in theaasociated sand, i

. .

2. HanureSand,

;

1. Clayey aiadaloid Drift, . .

44-76

At this place the denudation is excessive, the cliff being altogether
changed since Mr. Wyley made a sketch of it about thirty years ago
(UoO-52); while since 1840 a strip of land about 175 feet wide has
disappeared or at a rate of oyer 4*25 feet per annum.

From here south-west to Ballinesker (1*75 mile) the drift is

132 Proceedings of the Royal Irish Academy.

alternations of sand and marl, within places, Glacialoid Drift ; also to
the south of Ballyvalloo, the previously mentioned rib of Boulder
Clay Drift (fig. 5). The denudation of the cliff is considerable, rarely
less than 2 feet per annum ; and in some places, as in the vicinity of
Ballyvalloo House, it is over 3 feet.

In the Memoir to sheets 169, 170, 180, and 181 of the
Oeological Survey of Ireland by Mr. G. H. Kinahan, h.r.i.a. (1871'),
p. 14, the author states : —

(p. 15.) — ''In connexion with this area, the denudation of the
coast line should be described, as a considerable waste of land yearly
takes place. The Ordnance maps of the country were published
about 1 840 or thirty-five years ago. {Note, — The cliffs were examined
in 1875.) If, therefore, the present coast line is mapped, an estimate
can be made of the annual waste of the drift cliffs. While examining
these cliffs, it was observed that the effects of denudation varied
according to the nature of the drift. The Glacial Drift, in general, best
resisted the encroachment of the sea ; but in a few places it has been
considerably wasted. In such places the sea undermined the bottom
of the cliffs, causing great falls, principally due to their own weight.
Joints open in the marl during the dry weather; if these are
perpendicular or oblique to the line of the cliff the water percolates
through them, and the cliffs remain more or less perpendicular ; but if
they are parallel or nearly so to the line of cliff, they fill with water,
causing great slips, which masses, after coming under the influence
of the waves of the sea, are dissolved and wasted away. Such clifik
gave way in mass, but the slips are so extensive, that it takes years
before the sea can remove the d6bris, thus giving time to a protecting
slope to form. If marl is interstratified with sand, gravel or other
drift, the waste of the cliff is usually very rapid."

To the south-west, the Baginbun promontory is margined with
rocks, and very little denudation is apparent, except in the black
shales at Petit's Bay, between Gamivan and Baginbun Heads. To
the east of Bannow Bay, for about 3 miles, the sea cliffs, in general,
have a rock foundation, over which the drift may be glacial, aqneous,
or meteoric. In a few places the rocks have protected the cUffs, but
in many places they have been eaten away. In the townland of
Bannow, to the south of the old church, over 60 feet in depth have
been cut away since 1840, or at a rate of 1*71 feet per annum. In
the vicinity of Kiln Bay, and at the east margin of the townland,
there has also been considerable waste, respectively, of about 2 and

O'Bbilly— 0» the JFaaU of the Coast of Ireland, iSfc. 13»

2-5 feet per annmn. Further eastward, at the south-west of the
townland of Haggard, the waste is nearly 1*5 foot per annum, hut in
the east portion of the townland, also in those of Blackhall and Loftus
Acre, it has been excessive, in places reaching 3 feet per annum.

''Further eastward, in the townland of Ballymadder, the denuda-
tion is less, being about 1*5 feet per annum. This is also the average
in the western portion of Cullenstown. In these places, at the base of
the cliff in the aqueous drift, are recent sandstones or conglomerates
that resist the sea action, and thus preserve the accumulation resting
on them, while at the east of the townland, opposite the end of
Ballyteige Warren, there has been great waste, over 3 feet per annum.
The BaUjteige Warren is an irregular ridge of JSolian Drift. Its
outer margin seems to have been more or less cut away since 1840,
whilst its western end has grown more than 400 feet in length. At
the east of Grossfamoge Point, to the eastward of Ballyteige Warren,
ihingle has accumulated west of Eilmore Quay ; but a little farther
eastward, north and north-east of St. Patrick's Bridge in the
townland of Nemestown, there has been considerable waste, in some
places as much as 200 feet since 1840, or over 5 feet per annum ; at
this place there seems to have been the maximum denudation on the
wnth coast.

''Farther east-north-east, in the townlands of Ballygrangans,
Bastardstown, and Ballyhealy (Wexford -V^), the waste is about 1*4
per annum, and in Bingbaun and Ballagh about 1 foot.

"Between Tacumshin and Lady's Island lakes, the denudation
of the coast line has been from 5 to 50 feet in the last thirty-seven
years, while the bank enclosing the latter lagoon (Lady's Island lake)
bas been pushed inland more than 60 feet. Eastward, in Burrow,
Wexford (Wexford Sheet 53), the coast has been cut away at about a
nte of *75 foot per annum, the denudation ceasing suddenly as we
approach Gamsore Point.

"At Gamsore Point and from that northwards to the old coast-
g:uard station at St. Helen's, there has been very little general
denudation of the coast, although the strand margin for the most part
is drift In a few places, the denudation is more or less considerable.

" From the old coast-guard station of St. Helen's to Greenore, and
^romthat eastward and northward to Rosslare coast-guard station,
<^on8iderable denudation has taken place. Here the cliffs are high
ud formed of marl. They therefore nearly invariably come down
in slides, the d6bris of which must be removed by the sea before
another slide takes place, and but for this the waste would be much

134 Proceedings of the Royal Irish Academy •

more rapid. Between St. Helen's and Qreenore Point, the waste hu
been 1 foot per annum. Between Greenore and Ballygeary pier,
-25 foot, and between the pier and Rosetown about -50 foot. The
denudation here being greatest in the townland of Kill of the Sea,
•75 foot.

(p. 17.) — " The new pier of Ballygeary exemplifies how easily, by
a well regulated system of groynes, the denudation of this coaflt
could be arrested. The pier was commenced in the spring of 1873,
and in 1875 a large accumulation of sand had collected along the
coast to the south-east, but especially in the vicinity of the pier
where now (1878) sand-dunes are forming. This accumulation fonns
a foreshore that is gradually stopping the denudation of the cliff.
The cliJS to the westward of the pier also now suffers leas from
denudation.

''From RoBslare coast-guard station (Eosehill) to Ballinesker
(Wexford V^), are the banks (Bosslare and the Baven) that enclose
Uie lagoon of Wexford Harbour. They are composed of .Solian Drift,
and have been considerably altered, especially Bosslare since 1840.
Opposite White House, at the land or south end of Kosalarey the
coast line has moved westward or inland over 100 feet (3 feet per
annum). Opposite Bosslare House, about 75 feet (2 feet per annum),
and a mile and a quarter farther north, where the denudation is
greatest, about 203 feet or 5*74 per annum. From this point north-
wards, the denudation decreases to the Bull's Perch, where it is
50 feet (1*45 per annum), but to the north thereof it again rapidly
increases, being at 170 yards north of that point, 150 feet (about
4 feet per annum), the banks in places being breached, and the sea
passing through it during gales from the south-east. Further north
the bank originally had a very irregular outline, but now it has been
considerably added to inside, while outside in places it has been cut
away. The length of the bank has also increased northward.
When it was examined in 1 876, the Dogger Bank, off the mouth of
Wexford Harbour, was of considerable size, and in part an island.
This, however, has quite changed in the last two years, the island
having disappeared, and consequently the form of the northern portion
of the Bosslare has also changed."

(p. 18.) — " The changes in the forms of the Bosslare and Haven
Banks are in a great measure due to the intaking of the north and
south midlands in Wexford Harbour, as now the outflow of the
water is much less than formerly ; consequently the ' Flow tide '
wave current from the south changed the form of the Dogger Bank

O'Bbilly— 0« the Waste of the Coast of Ireland, Sfc. 136

from a masdye ahoal to the aouth of the entrance, into a long narrow
bank that oyerlapped the mouth, and even the end of the Baven Bank,
and changed altogether the tidal currenta and the drift of the sand.

(p. 32.) — ^^TomKaggard Distriet. — The sectiona along thia coaat have
greatlj changed since they were examined thirty years ago (1858)
by Wilson. They are even much changed since the above records
were noted in 1873. Sometimes when 'the beaches are full,' none
of the base of the cliffs can be seen, as occurred in September, 1878."
(p. 36.) — << Raiom andRoulare Burrows. — ^There is a tradition that
the ancient entrance into Wexford estuary, was to the north, in the
Tidmty at Curracloe. Thia is not at all improbable, but it must have
been in very ancient times to allow for the great depth of peat now
accomuktedat Curracloe.

(p. 36.) — " The hollows occupied by Wexford estuary are very
ancient. The present outline for the most part was induced by the
25 feet sea, but since that time, while the land was lower than at
present, and aubsequentiy the shore lines underwent various modi-
fications. The surface area of the estuary, however, has changed
considerably since the time of the 25 feet sea-beach. Subsequently
to the time of the 25 feet beach, the land was at least 30 feet
higher than at present to allow the peat to grow. After it had
Again sunk, the aand-bars seem to have formed, but farther seaward
than at present, as otherwise the peat would not be found under,
flul to seaward of them. Other submerged peaty and lagoon deposits
have been recorded south-west of Qreenore Point, in the neighbour-
hood of Ballytrent and St. Margaret's, where there are sites of small
lagoons margined by sand-ridges. Under and outside the latter,
being peaty accumulations. Of the latter locality, Mr. Wyly re-
cords ' Bog with trees ; exposed between high- and low-water marks
(rf spring tide.' "

(p. 47.) — ^*K%lnwre district — ^Further eastward the cliffs are low,
ui] hare in general ^olian Drift above, and gravelly Glacialoid Drift
helow. The sections given are very different to those recorded by
Mr. Wyly when examining the coast about thirty years ago (1858),
ts the marine denudation during the intervening years has been
^xcemre. Prior to leaving the drift, it may be mentioned that, aa
in all the Glacialoid Drift between Crossgamoge Point andTacumshin
Lake, fragments of shells and flints may be found, more especially
near the top and the bottom of the accumulations according as the
gravels and sands lie under or over it. They are also numerous in
the inlying patches and layers of sand.

136 Proceedinga of the Rcyal Iriah Acadetny.

(p. 48.)—'' 7%e Saltee Islands.— These islands lie to the south o!
the mainland, and are connected with it by a bar or ridge that u
partially submerged and partially tidal, the latter portion being
called St. Patrick's Bridge. On the east of St. Patrick's Bridge an
some large blocks, the residue of the drift that has been cut away by
the sea, the largest being called ' St. Patrick's boat,' from a legend
connected with it. It is a remarkably large erratic."

(p. 60.) — ''The bar of JBolian Drift that separates the Lady's
Island lagoon from the open sea is, in places, swept oyer daring
storms and high tides. Outside the bar, during storms, the sea tears
up large pieces of sandy, clay peat, similar to that at the bottom of
the lake. On the east coast, between Crossfintan Point and Cama
House, there is a low ridge of ^oHan Drift, while north of Caina
House is a submarine peat extending below low-water mark."

Wieklow Coast. — Memoir Geological Survey of Ireland, 121 and
130 (1869), by J. Beete Jukes, ma., p.k.8., and G. Y. Du Noyer,
M.B.I.A. No particulars given.

Memoir Geological Survey of Ireland, 138 and 139 (1888), B. J.
Cruise, m.b.i.a.

(p. 22.) — *^ Raised Sea-Coast. — All along the coast of the counties
of Wieklow and Wexford, the evidences of a recent rise in the
sea bed and adjoining coast are remarkably clear. These consist in
the occurrence of terraces and flats formed of silt, sand, and shelly
gravel, rising a few feet above the high-water line, and bounded inland
by cliffs or abrupt banks (according to the nature of the nuterial and
form of the ground), which originally constituted the sea margin at
the time when the terraces and flats were submerged. The actual
extent of rise of the coast and sea-bed is uncertain ; but the old sea-
bed generally lies from six to twelve feet above the highest tides,
The level is often increased by hillocks or dunes of blown sand which
have been thrown up by the winds, as is the case in Brittas Bay,
Arklow Bay, and other protected inlets. In the district contained in
Sheet 139, examples of the raised coast are to be observed in the
bays lying between the headlands, and in a direction from north to
south they occur —

''(1) In Brittas Bay, between the Castle Rock of Ballynacarrig

(Sheet 130) and Mizcn Head.
** (2) Between Mizen Head and the coast cliffs of Kilbride.

(yBBiLLi— On the Waste of the Coast of Ireland, 8fc. 137

*' (8) In Arklow Bay, both to the north and south of the Ovoca

Biyer.
" (4) In the bay south of Arklow Head (see fig. 3) ; and
" (5 and 6) In the bays north and south of Eilmichael Point.

*'The old sea margin is in these cases generally yery clearly defined
by lumks from 10 to 30 feet high, formed either of marl or slaty
strata, from the base of which the low terrace stretches seaward as
Isr |s the abrupt descent which forms the margin of the existing sea-
shore. The old sea-bed is now covered either by sand-dunes, or,
▼here these are absent, is green with coarse grass and other land
plants."

DMin Coast. — ^Memoir Geological Survey of Ireland, Sheets 91
and 92 (1871), by Edward HuU, m.a., f.r.s., f.g.s., and R. J. Cruise,
ir.E.i.A.

(p. 42.) — " Raised Beaches, — A raised beach is seen in detached
places along the shore from Balbriggan to Lowther Lodge. North
of .the lodge it stands from 5 to 8 feet above high-water mark; and
shells which had in most cases lost their colour, and were generally in
a fragmentary state, were found therein. (List of same by Mr.Baily).

DMin Coast. — ^Memoir Geological Survey of Ireland, Sheets 102
and 112 (1861).

(p. 50.) — ** Drift (Lambay Island). — This deposit exists merely on
the low ground which extends along the western margin of the island,
and in no respects differs from the ordinary brown drift gravel clay,
containing boulders and rounded lumps of the Carboniferous limestone
▼hich is so common over Ireland. Here, however, as we would
naturally expect, this clay contains a large percentage of rounded
fragments of the local porphyries, but nothing to stamp it as a local
deposit. It is, however, evident that, at the period of its depositioD,
Lambay Island formed a part of the mainland. This clay does not
eitend more than 100 feet, if so much, up the flanks of the more
elevated portion of the island, and beneath it the rock surfaces are
rounded, smoothed, and scratched, the directions of the striee being
iHvth-weat and south-east."

(p. 66.) — " The Drift.— The drift over the whole of this county
coDsists of two deposits. The first a black or brown gravelly
calcareous clay, containing a large amount of rolled limestone
fngments of various sizes as the lowest deposit; and the second,
loose sand and gravel, consisting ' principally of limestone pebbles.

lS8r Proceedifiga of the Royal Irish Academy.

though fragments deriyed from all sorts of Lower Silurian^ and each
kinds of rock, form a large percentage of the whole. Pebbles of
granite are not uncommon in this gravel, and chalk flints and
pebbles are also sometimes present. The lowest of these deposits is
found oyer the northern portion of the district under reyiew, but it
terminates at elevations varying from 300 to 400 feet above the sea."

(p. 67.) — <' Shennick's Island, opposite Skerries, affords an interest-
ing proof of the extreme age of this drift clay. It measures ( 1 860) 575
yards in length, from north-west to south-east, by about 150 in width,
and is formed of thin gravelly clay, which the sea has now abruptly
escarped on the north-west of the island to the depth of 46 feet. On
the opposite shore, south of Skerries, the same deposit is also escarped
by the sea, to the depth of 41 feet, the distance between the two
being in one place dose on three-quarters of a mile. This channel
has, therefore, been cut by the sea long subsequent to the deposition
of this clay, which, no doubt, represents the remains of what was
once a very large extent of land stretching into the Irish Sea. The
same fact, just noticed, has been mentioned in connexion with Lambay
Island, which is two miles and a-half from the nearest point of the
mainland, the deepest part of the channel being over five fathoms.
The east face of Howth, also, affords us another proof of the existence
of land having extended here far into the sea. On the top of the
cliffs, from Foxhole to the north of Lough Levin on the south, a
distance of 600 yards, we found this brown gravelly clay, containing
numerous limestone pebbles, plastered against the rocks, and termi-
nating at an elevation of about 100 feet above the sea, having a main
width of only 70 yards."

Enough has been said to prove the great antiquity of this deposit
by the amazing amount of denudation which has taken place since its
formation.

On the shore, one quarter of a mile west of Malahide, there is a
layer of gravelly clay, three feet thick and six feet above high-water
mark, containing recent shells and fragments of granite, chalk, and
flint. At low-water mark there is exposed on the beach east of
Malahide, blue marly clay, containing the dead shells of a species of
Pholas, &c.

Memoir of Geological Survey of Ireland, to sheets 121 and 130,
portions of Counties of Wicklow and Dublin. J. Beete Jukes, x.a.,
P.B.8., and G. V. DuNoyer (1869).

(p. 46.) — " The area described lies wholly in the County Wicklow^

O'Rbilly— Of* the Waste of the Coast of Ireland, 8fc. 139

except a bidaII part along its northern border, which belongs ta
the County Dublin. Along the seabeach, between Bray and Grey-
stones, there are low difis of marl, with limestone and other pebbles
and fragments of shells, over which is a deposit of sand and grayel."

(It is in these cliffs that the action of the sea, already referred
to, has been so marked, undercutting them and causing the cliff face
to fall away, and be subsequently removed by the tidal action. As
already mentioned, the waste has been so considerable that the railway
line as originally laid down, at a slight distance from the face of th&
diif, has been so endangered by the approach of the cliff face, that
the line had to be withdrawn inland at some considerable expense,
while costly works of undeipinning and strengthening the base by
stockades and groyns has been going on up to quite lately. The same
remarks hold good as regards the stretch between Bray river and
Ballybrack).

Memoir of Geological Survey of Ireland, sheets 91 and 92 (1891).
The area included in the sheets embraces portions of the Counties of
Heath, Louth, and a small tract of the County Dublin (no available
particulars given).

Memoir of Geological Survey of Ireland, Sheets 81 and 82 (1871).
These maps embrace the greater part of the County Louth, showing
fouteen or fifteen miles of its coast-line (no particulars given).

Memoir of Geological Survey of Ireland, sheets 60, 61, and part of
71 (1881). The District described in the Memoir lies wholly in the
Coanty Down.

(p. 20.) — " Raised beaches. — There are numerous indications around
the coast that the land has been raised in recent times. These indica-
turns occur in the form of terraces, consisting of stratified sands and
grarelB, often containing marine shells of the species now inhabiting
the neighbouring seas, with possibly a few forms which may have
disappeared. These terraces were clearly old sea-beds, and they have
sinoe been raised into land-surfaces beyond the reach of the highest
tides. Such terraces are found skirting the northern shores of Dundrum
Bay, partially covered and concealed by sand-hills, and extending to
the foot of the high ground at Newcastle. They again appear, form-
ing a very narrow strip along the coast at Annalong, where they have
been subjected to the wasting effects of the waves ; but on both sides
of the entrance to Carlingford Lough, at Soldiers Point and Greenore

140 Proceedings of the Royal Irish Academy,

Point, they f onn considerable tracts of level land and may be con-
veniently examined. Of these terraces, which may be probably called
' raised beaches/ there appear to be two ; the lower, rising from 3 to
to 7 feet above high-water mark of spring-tides ; the second, from
10 to 15 feet above the same datum. Mr. Trail has described these
terraces as they occur at Greenore, where, in the stratified gravels of
which they are formed, there are bands of oyster shells, together
with shells of the genera M ytilis, Fecten, Natica, littorina, &c. At
Killowen, near Rostrevor, similar shells were found in shingle, 10 feet
above high- water mark. The two terraces on the opposite shore are
similar in formation.

'' On the lower terrace, that of about 10 to 15 feet above high-water
mark, is built the town of Warrenpoint, together with the old keep
of Narrowwater, on the estuary of the Newry river.

'* At Annalong Harbour the terrace is at 40 feet elevation above
Ordnance datum, or a little over 22 feet above high-water line ; this
is, therefore, the upper terrace. The terrace bordering the coast near
Dundrum is referable to the first or lower level."

(p. 21.) — ''Other remains of raised beaches are to be found at
intervals along the shores of Carlingford Lough to Warrenpoint from
10 to 12 feet above the water-line.

" In addition to the raised beaches, clear indications of terraces,
formed out of the drift deposits are to be observed at several levels,
viz. at those of 50, 75, and 150 feet. These are often more easily to
be recognised when viewed at some little distance than when standing
upon them. The terraces of this class are of more ancient date than
those described above, and are probably referable to the period when
the land was emerging from the sea, towards the close of the Oladal
Period, the terraces having been formed during long pauses."

Memoir of Geological Survey of Ireland, Sheets 37, 38, and part
of 29 (1871).

The area described in this Memoir lies altogether in the County
Down.

(p. 42.) — Raised Beaches, — Skirting the shores of Belfast Lough,
between Hollywood and Donaghadee, we find a deposit of marine sand
gravel, the maximum elevation of which is about 20 feet above the level
of high water. In this deposit, artificially formed, flint flakes were dis-
<30vered some time back, of which Mr. G. Y . Du Noyer, in a communica-
tion addressed to the Secretary of the Royal Geological Society ol
Ireland, thus writes : ' I may remark that when these singular flakes were

O'Reilly— 0#» the Waste of the Coast of Ireland, 8fc. 141

first diBOOTered in the district round Carrickf ergus, abont fiye years ago,
their mechanical origin was questioned. Indeed I myself thought at
first that they were due to the crushing by natural causes (the weight
of the basalt) of the flint nodules, forming the original drifts oyer the
atmospherically eroded surfaces of the chalk. The chippings around
the edges of the flakes can, howeyer, only be accounted for by
artificial means, as they afford dear eyidence of design in their forms
and mode of occurrence. Subsequent examination clearly showed me
that eyery flake, no matter how rude its form, or how sharp its edge,
exhibited at one end a flat surface, transyerse to the longest axes of
the flake, and from this surface a blow was giyen at a point on it,
which caused a flake to come off from the original nodule, and this
fiake below the point of concussion, exhibited a conchoidal fracture
and a * bulb of concussion,' features which could only be formed by,
and were the result of, * an intelligent blow.* And further on he
says : ' The conclusions which my present information on this subject
leads me to aixiye at with regard to the origin and explanation of the
mode of formation of these flint flakes are these : During the period
of formation of our present raised sea-beaches, the men of that period
resorted to the out-crop of the chalk for flint nodules, from which to
manufacture their mallets, hand-axes, kniyes, rude spear- and arrow-
heads, and other implements, and these are the refeeta of that
manufacture during an unknown period, the localisation of the raw
material conducing to the localisation of the worked implements, lost
or rejected, and which was then coyered by the sea, but which is now
^ land skirting the the present coast line.' These flakes are
generally found close to the upper surface of the drift grayel, but at
Ballyholme Bay near Bangor, they occur at a depth of from 6 to 8
feet from the surface, in stratifled sand and grayel. On the beach
under the cliff there is a submerged bog, with stems and roots of trees
▼iaihle at low water.

Worked flint-flakes are also found on Beagh Island, in Strangford
LoQgh, in a raised beach on the north of the island. About 1 mile
north of Ballywalter, near Ballyferris Point, is a raised beach
consisting of stratifled sands and shells. It is about 3 feet aboye
hig^-water mark."

Memoir Geological Suryey of Ireland (Sheets 49, 50, and part of
61), 1871. The district considered is situated wholly in the County
Down and along its eastern shore.

(p. 11.)—'' The two islands of the ' North and South Bocks ' lie

142 Proceedings of the Royal Irish Academy.

off the eastern shore at a distance of about 2 miles, and are about the
same distance apart. This outer coast has a very shallow sea-board,
with numerous low sunken rocks, and being much exposed is
dangerous to navigation. These islands are the most eastern limits of
land comprised in the Irish survey.

''Along the western margin of Strangford Lough, the largest
islands are to be found, viz., Islands, Taggart, More, Mahee, and
Beagh, while many smaller ones are thickly dispersed all round.
Extensive sandy and slob-lands occur surrounding these, the islands
themselves being composed chiefly of rounded drift-hills, tailing off
down to the water's edge. Some of these, on their exposed side, end
abruptly in a steep escarpment, sometimes on the northern, and at
others on the southern side. On none of these does rock in situ occur,
while on those along the eastern side, the margins of the islands are
formed of solid rock, with a central covering of drift. Strangford
Lough is connected with the sea by a narrow strait about 5 miles
long, and from half to three-quarters of a mile in average width."

(p. 12.)— " Near its exit occurs Eock Angus and several isolated
rocks ('Pladdies') of a dangerous nature, and upon the bar (on
which is always deep water) at times, a terrible sea breaks.
Throughout the strait, a wide and deep channel exists, attaining a
depth 26 fathoms off the Cloghy rocks, and 35 fathoms (or 216
feet) between Portancarlagh and Ballyhenry Bay. This strait has
mostly rocky shores, and is kept clear by the scouring action of the
tides, which here run with a very rapid current, at about 5 to 7J
knots an hour for ordinary tides, and up to 9 knots for some spring
tides. In parts, the passage is contracted to comparatively small
dimensions, the narrowest being between Isle-o' -Valla and Rue Point
on Bankmore Hill, where it is only 1700 feet wide. This latter side
being an obstruction of a projecting drift hill, is gradually wearing
away, thus tending to widen the channel of this place. A little
south of this, between Black Islands and Qowland Bocks, at low water,
the passage is reduced to only 1000 feet, with a depth of 15 fathoms.
Here there is a whirlpool. About half a mile to the southward ocean
also a series of whirlpools of considerable size, where there is a depth
of 26 fathoms, whose influence is felt for upwards of half a mile, and
which are called "the Routen Wheels." Here a bad sea always
prevails, and small vessels even hesitate to pass through them except
at slack water. The roaring of these breakers is often heard for many
miles distant. It is probably to some irregular or peculiar conforma-
tion of the bottom, with the rapid current flowing over it, that these

O'Bbillt— Oft the Waste of the Coast of Ireland, 8fc. 143

aie to be attributed. The width of the channel opposite Strangford to
the aoothem end of the village of Portaferry, is a little under 1800
feet, or about i mile, and in part is 15 fathoms deep. The difference
in time of high water at the bar and at the northern extremity of the
LoQgh ie nearly two hours."

The author gives an estimate of the quantity of water which
panes through this strait in filling and emptying the Lough at each
tide

(p. 13.) — '* I may further remark, with regard to Strangford Lough,
tbat we have evidences which would lead us to suppose that it differs
very materially in its present state from what it originally was, that
its very existence is probably due to its having been a *' Geological
basin" of limestone, of which traces are still to be found in the
narrow skirting thereof at Gaatle-Espic, but which has all been
Kmoved by denudation and ' atmospheric solution,' and that instead
of, as at present, being a lough connected with the sea by a strait, it
was originally a fresh-water lake. (Note. — History does not state
as much, but an old tradition seems to exist that such was formerly
tie case)."

Memoir Geological Survey of Lreland (Sheets 21, 28, and 29),
(1876). The area forms a portion of the great volcanic region of the
Coimty Antrim bordering on the eastern coast.

(p. 2\,)—^^ F%int impUmmts, — Flint-flakes, celts, cores, &c., are
found over the entire district, not only on that portion where the
ci^ is subjacent, but also on the high grounds occupied by the
Wdt. In many places the fragmentary chips are very abundant,
sndi as on the dialk outcrop in Drain's Bog, as well as along the
OQtcrop south of Glenarm, pointing out the locality where these
implements were made.

Vemoir Geological Survey of Ireland (Sheet 14), (1886). The
district presents some of the most striking features in the scenery of
County Antrim, and no one visiting it can fail to be struck vnth its
fine headlands and deep glens opening out upon the sea-coast. The
district included in the northern half of Sheet 14 extends from the
River Bush, near Armoy on the west, to the sea-coast on the east at
Coshendnn^ and southward to Bed Bay.

(p. 9.)—" The coast line is, in general, bold and precipitous. North
«i Coahendall, south towards Glenariff, it is rather low and un-
dulating, and exhibits some fine examples of the old sea caves of the

m.I.A. TBOC.y TOL. XXIT., BBO. B.] i^

144 Proceedings of the Royal Irish Academy.

25 feet raised beach. They are to he seen at Bed Bay Tunnel, and
south of it excavated out of the New Bed Sandstone — ^the principal
one being called ' Nanny's Cave.' A little south-east of Cushendon,
there are also some very fine and extensive sea-caves occurring in the
conglomerates of the Old Bed Sandstone. The raised beach itself
ranges from the 25 feet to the 40 feet contour, and is well marked
along the coast at several places between Cushendun and Glenariff,
forming a slight escarpment or cliff of drift and rock along its course.
The southern half of the sheet, has for its eastern boundary the shore
extending from Bed Bay to Glenarm Bay."

Memoir Geological Survey of Ireland (sheets 7 and 8), 1888. The
district described extends along the coast of Antrim and Derry ; it also
includes the Island of Bathlin.

(p. 7.) — "The shore line from Portrush to Fair Head, east of
Ballycastle, composed for the most part of cliffs formed of successiye
tiers of basalt resting on chalk, is generally bold, often inaccessible,
more especially in the neighbourhood of Bengore Head, which rises to
a height of 367 feet above the waters of the sea, where the celebrated
Giant's Causeway makes an interesting geological feature. Striking
as is the appearance of Bengore Head, it is completely surpassed by
that of Benmore or Fair Head, rising 636 feet or nearly double the
height of the former. This latter has a cap composed of a massive
sheet of dolerite which, on the sea face, is broken up into great poly-
gonal monoliths over 250 feet in length. At the base of this lofty
cliff broken columns of basalt are confusedly strewn over the slopes to
the waters edge, covering the underlying Carboniferous beds.

" The peninsula of Portrush lies in the extreme north-west comer
of the county Antrim, and is fenced on its western side by perpendi-
cular cliffs composed of a sheet of dolerite some 70 feet in thickness.
The most westerly promontory is called Bamore Head. At a distance
of half a mile north-east from Bamore Head, a chain of islands,
sixteen in number, called * the Skerries ' commences, and extends in
an easterly direction for about a mile and a half, forming a natural
breakwater to the north Atlantic waves, which even in comparatively
calm weather may often be observed breaking over the seaward faces
of the Skerries and throwing the spray high into the air.

''The coast line at and west of Portrush consists chiefly of cliffs of
basalt and dolerite, bounded at the base by a narrow uneven margin oi
the same rocks, indented by numerous small irregular creeks and bays.
The surface at the top of these cliffs stands generally at a height of

O'Rbilly— On the Waste of the Coast of Ireland, Sfc. 145

25 to 40 feet above the sea, reaching, however, about 100 feet in the
Ticioifj of Island Doo. This island is one of a number of outliers
belonging to the basalt, which lie at a short distance out from the
shore, some being separated from the mainland merely by narrow necks
of water at full tide. The rocks are often traversed by fissures,
some o{ them being open up to the surface so as to form < puffing
holes' or ' blow holes,' through which the air, accompanied by spray
IB projected with more or less force by the waves rushing into the
cavities below. One of these occurs at filackcastle rock dose to
Portstewart, the fissure here communicating with a cave which runs
southward between the rock itself and the mainland. Another is
found on an island south-west of Blackrock. Of the few caves that
occur on this part of the coast, none are of large dimensions. Close
to the ruins of Ballyreagh Castle, an opening of this nature runs in for
about 18 yardsy being 7 yards wide at the entrance and at most 7 feet
in height."

(p. 9 ) — <' Bathlin Island, — ^Although so close to the mainland,
lUthlin Island is very difficult of access, owing to want of proper
harbour accommodation and the liability to dangerous seas due to
tidal currents between the island and the mainland. The northern
coast face, consiBting of tabular and columnar basalt resting on chalk,
ia formed of bold, often inaccessible cliffs, between 300 and 400 feet
in height, while along the opposite side of the island the sheets slope
towards the south, and along their seaward faces, show a similar
sapeipodtion of the basaltic and Cretaceous beds."

(p. 20.) — " The next westward outcrop of the chalk is to be found
at low water on either side of Dunluce Castle ; whence the rock gradu-
ally rises and forms a cliff at the * White Eocks ' of about 150 feet.
This cliff 80 seen from the sea presents a varied aspect owing to the
nomerous caverns and fantastic forms into which the rock has been
(wed by the erosion of the sea, a process which is still going

(p. 21.) — ** £atUm Island, — The Chalk formation is the foundation
Tock of the Island of TJathlin : the thickness of the formation is about
220 feet, and is the greatest of the chalk either in Eathlin or on the
mainland.

'* Good sections of this division of the basalt (Lower Basalt) are
exposed to view in the steep sea-cliff faces between Ballycastle and
hallintoy, in which latter direction it thins out to not more than a
hnndred feet or so; while in the vicinity of Ballycastle, the mass
f^snnot be leas than 350 feet thick."

146 Proceedings of the Royal Irish Academy.

(p. 24.) — ^^ Upper Basalt. — Although most of the district is capped
with sheets of Upper Basalt, the only good sections are to be found in
the almost inaccessible clifEs east and west of Bengore Head. Hera
they form successive tiers of columns varying in size and sometimeB
in position. These tiers indicate successive outflows of lava. Tlie
most remarkable of them is the lowest of the series which caps the
iron ore deposits, and forms the Giant's Causeway."

(p. 27.) — " Ratldin Island, — The Tertiary volcanic rocks here are
well represented, and present very much the same section as they do
about Bengore Head on the mainland. The best sections are seen from
the sea, in the cliffs on the north side from Bull Point to the lighthouse.

^' At Doonpoint there occurs a causeway in the Lower Basalt, and
the longitudinal section shows vertically columnar basalt having fan-
shaped and radiating columns of smaller dimensions blended into it
from the top, showing that the two sheets amalgamated before cooling.
Dr. Haughton notes of the rocks at Doon as follows : ' The curvature
of some of the pillars is continuous through 90^, and they pass from
the vertical to the horizontal position, exhibiting, however, a tendency
to break at the point of greatest flexure, which has caused most of
them to be broken off by the action of the sea.' "

(p. 28.)—" The Dolerite of Fair Head is probably of the same
age as that at Portrush, but does not weather so rapidly. The Fair
Head sheet is remarkable for its enormous thickness, presenting, as it
does, an unbroken columnar face to the sea, near the ' Qtrej Man's
Path ' of 260 feet.

" At the base of the basaltic cliff at Fair Head an intrusive sheet of
columnar basalt 70 feet thick occurs ; and in its extension it is met
with at Drunmakill Point, to the south, where the colunms are
scattered in all directions."

Mr, Symes regards the Dolerite of Fair Head as possibly the
latest volcanic protrusion in the county Antrim,

(p. 31.) — *^ Volcanic Vents. — One of the most remarkable volcanic
vents in the county Antrim is situated at the well-known island of
Carrick-a-raide and the adjacent coast, a fine view^ of which can be
had from the celebrated * swinging bridge.' This old neck cannot
be less than from 1000 to 1200 feet in diameter, and is filled up with
massive, coarse, and tough grayish volcanic agglomerate, enclosing
large irregular masses and smaller fragments of basalt, basalt bombs
of all sizes, and chalk pieces occasionally."

(p. 34.) — ^^Peat Bogs and Alluvial Flats. — ^A deposit of peat projects
from beneath the blown sand, and follows the slope of the strand for

O'Bbilly— 0» (he Waste of the Coast of Ireland, 8fe. 147

a short distance seaward, at the point where the county boundaiy
terminates a little south of Portmsh. It is partly covered at high
water, daring which, in rough weather, masses of it are torn away
and carried out to sea. The peat contains hazel nuts, portions of
small branches, leaves, and the elytra of beetles. The presence of
this peat, within range of the sea-action, may indicate a subsidence of
the land within very recent times. Similar instances are to be
ebserred at various points of the Irish coast.

^^Rm$d Beaehei, — These occur at Portrush, Port Ballintroe, Rathlin
Island, and the Giant's Causeway. Professor Hull has recognised
the raised beaches of the county Antrim as the representatives of the
* 29 feet beach ' of the opposite coast of Scotland."

(p. 35.) — "The shores of White Park Bay and the coast-line
north-west of Ballintoy exhibit examples of the raised beaches and
their associated old sea-caves and sea-stacks (see fig. 7).

" Prehistoric remains have been found in abundance on the raised
beach platform of White Park Bay, consisting of worked flints, stone
hammers, com crushers, fire hearths, pottery, etc., and the bones of
varioTu ftnimala ; also ' kitchcu middens ' of shells and ashes.

^^Cwerru, — The Chalk at theWhite Eocks is penetrated by numerous
cayes at different elevations, but none of them extending any great
distance. Under Dunluce Castle a cave runs through the entire rock ;
this is probably artificial, and could easily have been excavated, owing
to the spheroidal nature of the rock. At the Giant's Causeway are
two caves in the Lower Basalt at the sea-level ; and in Eathlin Island
foor have been noted by Mr. Andrews, the lengths varying from
150 to 250 feet.

(p. 37.) — <( Among the sandhills in the town of Portrush, a gale
recently exposed a Prehistoric hearth, in which were pieces of pottery,
numerous flakes, and cores of flint, and a few bones. The flakes are
remarkable for their freshness ; wherever else found the majority of
them are porcelainized or weathered.

Memoir, Geological Survey of Ireland (Sheets 12 and 6) (1886).

(p. 5.) — "The district described lies entirely in the county
Londonderry, except a small area to the north-west of Lough Foyle,
which belongs to the county Donegal. The greater part of this
district have been described by General Portlock in his geological
report on Londonderry and parts of Tyrone and Fermanagh (1843).

''Lough Foyle occupies a considerable area in the central and
western parts of this district, dividing the portion in the county

148 Proceedings of the Royal Irish Academy.

Donegal from that in the county Londonderry. The former, wbicli is
very small, is part of the peninsula of Inishowen.

" The county Deny portion of the district is characterized by the
extensive plains that horder Lough Foyle ; the lowest of which is a
raised beach, bounded on the sea-side by large tracts of reclaimed
land or intakes.''

(p. 6.) — " The high ground in the east of sheet 12, with its
continuation in sheet 6, is underlaid by the basalt which forms the
great Tertiary plateau of the counties Antrim and Derry. The
boundary of l^s volcanic area is here, as in other parts of the district,
often characterized by bold, precipitous cliffs, which, towards the
north, assume magnificent proportions, and are accompanied by huge
landslips of comparatively recent date. The rugged masses thus torn
away, rise sometimes in sharp pinnacled forms in front of the steep
fac« of solid rock, and are separated from it by a gap, strewn with
blocks that have fallen in large numbers on either side.

" The bold outline, which thus denotes the boundary between the
Secondary and Tertiary formations, passes northwards alld eastwards,
with a wide sweep into sheet 6, accompanied by a gradual descent
towards the coast. Here, at the east of Umbra, it consists of a
steeply-receding cliff of chalk and basalt, about 500 feet in height,
supporting a mass of boulder day, which stands piled up against the
escarpment for a height of nearly 300 feet."

(p. 7.) — " East of Umbra the boundary of the basalt follows the
coast-line for a short distance, bending out below the sea, within a
mile of DownhiU ; while beyond this point, and as far eastward as
the locality just named, nearly perpendicular crags, composed of
basalt, with some beds of ash, overhang the shore, skirted for half
their height by an accumulation of blown sand, resting on boulder
day, and in some parts barely upholding ponderous semi-detached
masses of rock, which seem ready to fall from their position.

** From Downhill eastward to Castlerock, the cliffs directly overlook
the sea, having a more irregular and rugged outline, owing to the
constant and frequently violent action of the waves, which are at
the present day hollowing out caves in the basalt. Similar openings,
standing at a somewhat higher level, and dating back to the period of
the adjacent raised beach, occur in the chalk between Downhill and
Umbra ; whilst, in some instances, as at Backaunaboe (' the tether-
stake '), a little east of Downhill railway station, the conditions seem
to point to a continuous drilling action carried on from that day to
the present. The above name is given to a sharp sea-stack, composed

O'Rbilly— Oft the Waste of the Coast 0/ Ireland, 8fc. 149

of amydaloidal basalt, standing out from the difP at the western end
of the temple tunnel — a remnant of the northern wall of a spacious
cave, the eastern portion of which is still to be seen penetrating the
rock for a short distance (see flg. 1)."

(p. 8.) — '^ The raised beach, traces of which exist at yarious places
around the coasts of Derry, Antrim, and Donegal, and which represent
the 25 feet terrace of the western coast of Scotland, here extends
inland as far as the margin of drift composing the sloping ground west
of the basalt escarpment."

(p. 26.) — ^^ Raised Beaeh.— An extensive raised beach, at an
ayerage height of about 25 feet, fringes the southern and eastern
shores of Lough Foyle, extending to Bellarena and Magilligan, where
it has a width of from two to four miles. It is also seen on the
north-western shore in Donegal, where, owing to the nature of the
ground, it is much narrower, being only a quarter of a mile wide at
Qnigley's Point.

" The Bobbers' Cave and the Pipers' Cave, which penetrate the
Chalk at about one or two hundred yards, respectively, east of the
stream that joins the sea a little east of Umbra, in Sheet 6, standing
at about the 25 feet contour Une, belong, no doubt, to the period of
this raised beach. These openings were occasionally within reach of
the waves, during the prevalence of storms, till the construction of the
railway presented a barrier; and the floors are now strewn with
rolled blocks and pebbles of basalt and broken shells. In a place
where the basalt is laid bare among the sand dunes, 400 yards east
of Castlerock, at about the level of the 25 feet contour line, the
surface of the rock bears clear evidence of the rounding action of the
sea, and the crevices are filled with sand and shell fragments — ^pro-
bably remnants of an old beach."

Memoir of Geological Survey of Ireland, Sheets 1,2,5, 6, and 1 1
(in part), (1889).

(p. 9.) — '* The district described forms a remarkable promontory,
bounded on either side by Lough Swilly and Lough Foyle, and jutting
out fen into the Atlantic Ocean, where it terminates in the cliffs of
Malin Head. Though not actually an island, as its name indicates,
being connected with the mainland by a neck of alluvial soil, yet the
name is not without significance, as pointing to the inference that
within the historic, or at least traditionary, period it may have been
really an island, at least during high tides. As a physical fact, the

150 Proceedings of the Boyal Irish Academy,

nazTow neck by which the promontory is united to the mainland,
thongh about 8 miles from shore to shore, is formed of an old sea^bed,
which has been elevated into land, certainly in very recent times,
and in all probability within the period during which Ireland was
inhabited by Celtic tribes. It corresponds with the well-recognised
' 25 feet raised beach ' of our northern coast. This narrow strip,
along which the railway from Londonderry to Buncrana is carried,
has an average elevation of from 20 to 25 feet above Ordnance datum,
and only for a short distance, near Pennybum, is the level materially
exceeded, the ground rising to 50 or 54 feet above Ordnance datum,
or 41 to 45 feet above mean level."

(p. 10.) — ''But the island of Inishowen, thus constituted, seems
to have been itself a double island, owing to the existence of a second
narrow strait by which it was crossed at the period above referred to.
Between GuldaiS Bay, on the east, and Trawbreaga Bay, on the west,
there stretches a low neck of alluvial land, deeply covered with peat ;
and, during the period of depression, this was overflowed by tidal
waters, as the old sea-bed, consisting of sand, silt, and gravel, well
seen in the neighbourhood, underlies the peat, which has grown over
the surface since its elevation into land. The highest level of this
alluvial tract is 50 feet above Ordnance datum, or 12 feet above high-
water of ordinary tides, and of this, probably, 10 or 12 feet consists
of peat. At its western end this strait conmiunicated with the ocean
both to the north and south of Doag^ Island, which is at present
connected with the mainland by a bar and sand-dunes forming the
shore of Pollan Bay.

" (Note.) — Dr. Sigerson (Proc. Roy. Ir. Acad., 2nd ser., vol L,
p. 212, ei 90q.) has adduced historical evidence in confirmation of
the statement that Inishowen was an island, not only within the
period of human habitation, but within that of history. In the mape
of the Escheated Counties of Ireland (1609), of which facsimile copies
were taken at the Ordnance Office, Southampton, in 1861, a strip of
water is shown connecting the Eoyle and the Swilly loughs across to
the north of the ' City of Derrie,' just where the raised sea-bed
occurs. Another strip of water is shown, stretching from the * Lake
of Loughfoile,' near Saint Johnstown, to the inlet of the Swilly, near
Castle Hill. Derry itself stood on an island before the last elevation
of the land as a strip of water, recently a morass, bounded the lull on
which the old city is built, on the west. Sigerson quotes passages
from the < Annals of the Four Masters,' of the dates ▲.». 1211 and
1010, in which the name island is applied to the present promontory ;

O'Bkiixy— 0» the Waste of the Coast of Ireland, Sfc. 161

thns, in the latter case, the quotation runs: — 'a.d. 1010, (Engus
O'Lappan, Lord of Cinel Enda, was slain by Ginel Eoghain of the
Island,' ue. Inishowen. Thus historical evidence concurs with that
deriyed from an inspection of the physical conditions, that Inishowen
was actually an island up to within very recent times. The raised
beach referred to is in reality a representative of that of Kilroot and
Lame, containing numerous worked flints, and of the 25-30 feet
raised beach of Scotland, in which seyeral canoes and other works of
hmnan art have been found." (See J. Geikie, *' Great Ice Age,"
p. 311. &c.)

(p. 10.)—*' The promontory of Inishowen, as now constituted, is
exceedingly hilly, and consists largely of rocky ground, covered by
heath and mountain bog. Its culminating point is Slieve Snaight
{- *Snow Mountain '), a quartzite mountain, which rises from the centre
of the promontory to an altitude of 2019 feet above Ordnance datum.
The most prominent feature is the grand quartzite ridge of Raghten
More, which traverses the western portion of East Inishowen between
l^nnree Head and Pollan Bay, and reaches an elevation of 1655 feet.
Thns, although of no very great elevation, this mountain ridge, owing
to its position as rising abruptly from the Atlantic, and breaking off
along its western slopes in a naked wall of quartzite, conveys to the
nuod an impression of massiveness which is not altogether dependent
on its altitude."

(p. 11.) — "The coast-line of Inishowen is generally rocky and
predpitoas, except along the margin of Lough Foyle, and the inlets
through which the principal streams make their escape into the
ocean. The northern coast is particularly bold, the cliffs often rising
to heights of 500 or 600 feet, and at ' the Pounds,' north of Glengad
Head, to a height of 802 feet above Ordnance datum. Malin Head
iH- l)i although the most projecting point of the coast is compar-
atively low (125 feet) ; but Dunaff Head, at the entrance of Lough
Swilly, presents to the Atlantic waves a bold wall of granite and
qoartzite of over 600 feet in height."

(p. 13.) — *^ Islands, — Several islands rising from the Atlantic lie
^t some distance off the coast of Inishowen. The largest of these is
InifthtrahuU (flg. 3), a rocky mass, nearly a mile across from east to
vest, formed chiefly of gneiss. Some dangerous rocks, called ' The
Tor Bocks ' (fig. 2), rise above the surface a mile north of Inishtra-
hull ( s < Island of the big strand ' ).

'' ( (Kote.) — InishtrahttU means the ' Island of the Big Strand,'
/amA a Island ; ^ra » strand, and h-ull, an old and uncommon Celtic

152 Proceedings of the Royal Irish Academy.

word for big or large.* There is actually no very big strand at the
present day, as I am informed by Mr. Cruise, who geologically soi*
Teyed the island ; but there is a raised beach extending right aci06B
the western side of the island, now about 50 feet above Ordnanoe
datum, or 30 feet above high- water mark, and it is not improbable
that at the time the island received its name this may have been a Ing
strund in the ordinary sense of the word.)

'* The Garvan Islands form another group of rocky islets, three in
number, rising about a mile from the coast of Malin Head, and formed
of quartzite, while another small islet, called Glashedy Island, lies
a mile off the coast in the bay between the prominences of Duiuf
Head and Malin Head. This island is formed of quartzite."

(p. 22.) — " At Glengad Head crumpled micaceous schists forma
vertical cliff, 200 feet high ; sections in similar beds are also freelj
exposed along the coast line to their junction with the quartzite."

(p. 26.) — " Maville District — This tract of country is as wild and
rugged as almost any part of Donegal, the coast line being bold and
precipitous, except on the south-east along the shore of Lough Foyle.
On the north the sea-cliffs reach a height of 400 and 500 feet, and are
in but few places accessible except by boat."

(p. 33.) — **For some three miles south-west of Moville, a line of
gravel cliffs at an average height of about 50 feet borders Lou^
Foyle, and similar though smaller deposits may be observed on the
shores of Lough Swilly.

'* Raised Beaches, — An extensive raised beach, probably the repre-
sentative of the 25 feet beach of Scotland, borders the alluvial plain
south-east of Inch Island, continuing to the south-west along the
valley between Carowen and Burt, opening into the Blanket Nook,
while to the east it occupies the valley that extends from Bumfoot, in
a south-easterly direction to Pennybum, the average height observed
being 32 feet. At Farland Point, south of Inch Island, and along the
coast of Lough Swilly, south-westwards, portions of a raised beach at
the same elevation remain. Small portions of a raised beach are also
seen on the southern and eastern shores of Inch Island, and a more
extensive one stretches along the shore of Lough Swilly for about a
mile and a-half north of Fahan to Buncrana."

(p. 34.) — ** On Inishtrahull a fine example of a 50 feet raised
beach occurs in the centre of the island. The lightkeeper informed me

1 «• Hulk " is a word still used among country people in the north of IreUnd to
mean a big lazy fellow.

O'Ebilly— 0« the Waste of the Coast of Ireland, 8fc. 153

that in the year previous to my visit, 1885, during a gale from the
north, this beach was covered with water for over two hours.

'' On the mainland from Malin Coastguard Station a raised beach
extends for two miles to the south, being sometimes one mile wide.
This beach was at one time covered with bog, which is now nearly all
cut away. At Malin Watch Tower there are fine examples of the
25, 50, and 75 feet raised beaches. Along the shore to Malin Head
numerous patches of the 50 feet beach may be observed between the
rocks. The most important and extensive, however, of these raised
beaches is that which stretches from GuldafE to Tranbreaga Bag. Its
average height is about 50 feet, and most of its surface is covered with
bog, which is being rapidily cut away. Another extensive raised
beach stretches from Tullaghan Bay to Leenan Bay, the bog that
formerly covered this beach being nearly entirely removed. As
pointed out by Professor Hull, both these raised beaches are of a
comparatively recent date.

"Along the south coast, between Inishowen Head and Moville, the
50 feet raised beach occurs in several places. It consists of sand and
gravel, and is best seen between Greencastle and Inishowen Head.
SheUfl of existing species are common throughout the deposit. At
Tremore, Kinnoge, and Olennagiveny bogs, the 25 feet beach is
represented, and contains shells at each place."

(p. 34.) — " Kitchen Middens. — Associated with the raised beach,
moimds and accumulations of shells occasionally occur, which must
be regarded as of human origin, inasmuch as flint flakes, fragments of
bone, and burnt wood are often found in them. They were observed
on the shore north and north-west of Ballymoney, in the Garowen
district, at Fairland Point, at Inch Island, near the old castle of
the South, and at Inch Eoad railway station."

Memoir Geological Survey of Ireland, Sheets 3, 4, 5 (in part), 9,
10, 11 (in part), 15 and 16 (1891).

The region described in this memoir includes all the tract lying
between Lough Swilly on the east, Gweebarra Bay on the west, and
tile Atlantic coast, which connects these two inlets along the north.
It is the most mountainous portion of Donegal, and from its centre
rises the culminating height of the north of Ireland, the twin-peaked
Sirigal, which attains an elevation of 2462 feet above the surface of
the ocean. Its coast line is indented to a remarkable extent, and
slong the west is broken up into numerous rocky islands.

(p. 8.) — ** Errigal is certainly the most perfect pyramidical
mountain in Ireland, perhaps in the British Islands, and is a conspicuous

154 Proceedings of the Royal Irish Academy,

object far out at sea from the decks of sliips approacliing the north
coast of Ireland from the Atlantic."

(p. 14.) — ** Sea Loughs, — The coast between Longh Swilly and
Oweebarra Bay is deeply indented by several bays and sea loughs, of
which Milroy Bay and Sheep Haven are the most important. The
latter is a broad bay, the entrance to which is bounded by the bold
cliffs of Horn Head on the west, and by the less elevated coast of
Doagh on the east."

(p. 15.) — ** Islands, — In addition to Horn Head, which is an island,
there are several islands lying off the coast of north-western Donegal
requiring special notice. Of these, Tory and Aran Islands are the
most important.

** Tory Island, in ancient writings Toirinis and Torach, * the Island
of Towers,' is remarkably distinct when viewed from the mainland
between Dunfanahig and Cross Beads. It lies at a distance of about
eight miles from the coast of Horn Head, the bottom of the sea
descending to 24 fathoms. Along the north-east, the island presents
a bold front of naked rock towards the Ocean ; but on the south side
the wide bay of Camusmore affords shelter and anchorage for ships
and fishing boats. The western shore is shelving, and is lined by a
remarkable shingle beach forming a natural breakwater, and giving
evidence from its extent of the force and sweep of the Atlantic billows
when impelled by the prevalent westerly winds.

'^Aran Island, — This is a large island separated from the mainland
by a sound about half a mile across.

" The numerous rocky islets lying off the coast between Gweebarra
Bay and the Bloody Foreland are all formed of granite."

(p. 17). — " As regards the occurrence of pebbles of granite, &c.
in the quartzite, and limestone, it is believed that they have been
derived from rocks older than the granite (not improbably of Arcluean
Age), and now submerged beneath the waters of the Atlantic."

(p. 72.) — " Bloody Foreland District and adjoining Islands.'^

(p. 73.) — '* Rugged cliffs, sometimes 100 feet in height, bound
Gola Island and ITmfin Island, especially on the west. These are
penetrated by sharp fissures, hollowed out along the joint planes, so
that, in some cases, one side overhangs the sea at an angle of 70^ or
80^. Natural arches also occur, as at Scoltydoogan, north of Gola
Island, where a small inlet, entered by a narrow gully, communicates
with the sea by means of an arched passage about 40 yards in length,
at a depth of 70 feet below the surface of the ground. At
Scoltaglassan, nearly half a mile east of Torglass Island, a nanow

O'BwLLy— On the Waste of the Coast of Ireland, Sfc. 155

inlet between steep walls of granite, has been hollowed out by the
sea along a line of fissure, now partly filled with broken rock and
bonlder clay. This break, which is accompanied with little or no
dislocation, seems to run across the western part of the island,
appealing again at Scoltnalinga north of AUagh Island. A small
arch occurs in the prolongation of the same line east of Tomamullane,
foiming a gully with an overhanging eastern waU 50 feet high, that
on the west, reaching 80 or 90 feet."

(p. 74.)—*' Umfin Island has an exceedingly irregular outline, and
b bounded by steep rugged cliffs, pierced by caves and natural arches.
At about 50 yards from the extreme western point, one of the arches
runs north-north-east, along the lines of jointing quite through the
promontory. It has a length of 60 or 70 yards, and the opening forma
t conspicuous feature as viewed from Oola Island. A cave 70 yards
in length cuts through the northern part of the island.'*

(p. 74.) — " The main portion of the granite on Inishbogin (Sheet 3)
is coarsely crystalline, and it is sometimes largely porphyritic. The
jnnction with the schists is clearly traceable across the highly
^dated surfaces of the latter on the north coast of the island. At
Illanamarve, the shore line is broken by deep narrow inlets, one of
which is spanned by a natural arch, the apex being formed of a band
of fine-grained schistose granite, 4 yards wide."

(p. 75.) — '• Ro»guil Diitriet. — The northern portions of this
promontory are mainly composed of granite, which has been intruded
amongst Uie metamorphic rocks. It has been intruded amongst
qnaitzites generally, and along the coast of Doagh Bay, breaks across
them in numerous dykes and off-shoots, which are visible in the coast
cliffs.

(p. 84.) — " Qtoeedore and Aranmore Districts. — Numerous dykes
of felstone penetrate the granite and metamorphic rocks, chiefly in
the rugged area south of Inishfree Bay and in the western half of
Aian Island. Variations of colour and character are frequently
noticeable in the same dyke ; and the felstone forming the dykes of
this and other localities, in this portion of the district, weathers
rapidly into cream-coloured or light brown kaolinized rock. North
of Kincaslongh, the dykes consist usually of dark-brown rock, with
pink felspar crystals porphyritically developed, dark mica, and occa«
BonaUy blebe of free quartz. The trend here is mostly northerly.
Cares and precipitous inlets mark the point of the coast line where
rach dykes exist, as at Scalpnadinga and Illion in Aran Island, and at
the northern extremity of Cniit, near Owey Island, a dyke of coarse-

1 56 Proceedings of the Royal Irish Academy.

grained gray porphyritic f elstone 1 5 feet wide, has originated a care
with a blowing hole at its inner extremity. The erosion within has
been so extended as to form a natural bridge, so that the roof of the
former cave forms a natural bridge over the narrow inlet to the spac*
within as represented in the wood-cut (fig. 17)."

(p. 87.) — ^^ Faf%ad District {Diorite andEpidiorite). — Commencing at
the northern part of Eanad Head, we find an irregular mass of dark-
bluish evenly crystalline diorite, forming a small boss. South of this
boss at Bonnaweelmore, another large mass of finely crystalline diorite
extends from the shore inland, apparently bedded in the quartzites,
seen in clifp section. Along this coast there are several fine sea-
stacks formed of these rocks, notably that of Stookmore, or Brown
George, inside Swilly Beg."

(p. 88.) — ** From Illanmore to Lee Point, numerous dykes and
irregular masses of diorite occur, sometimes following the bedding of
the rocks, and at others crossing it. They are all of a dark greenish
colour with varying degrees of crystallization. At the ' Seven
Arches' several dykes of diorite occur, weathering freely, but the
' Arches ' are formed by the weathering of the quartzites along joint
planes and planes of bedding."

(p. 89.)—" From Ardbune Point in Caffard Bay, a well-defined
diorite dyke averaging 400 feet in width, can be traced almost con-
tinuously in a south-east direction, from the eastern to the western
shores of Mulroy waters. It forms the elevated peak called Cashel
Fort, 496 feet, a striking feature in the landscape."

(p. 93.) — ** Horn Head. — The diorite of Horn Head occupies a
considerable area of the more rugged portions of the promontory.
Sections in the cliffs, which from the coast line show that this rock
has been to a large extent intruded between the beds of quartzite
among which it can be traced in dark bands, varying in thiclmess and
confornimg in general direction to the lines of outcrop. In cases
where it has crossed the bedding horizontally, or nearly so, and the
upper portion of the quartzite has been denuded away, the diorite
appears in section as a cap resting on the truncated portion below.
South of Traglish Point the diorite ranges up among the quartzite beds
to the top of the cliff, a height of about 600 feet in a mass, 60 or 70 feet
in width."

(p. 97.) — " Bloody Foreland District — A very conspicuous dyke of
columnar dolerite 4 yards wide cuts through the schists west of
Curran's Point (Sheet 9), where it is shifted for a distance of 25 yaids
by a fault. It appears again to the north-west, penetrating both the

O'BiiLLY— On the JFasts of the Coast of Ireland, 8fc. 157

schist and a tongue of granite, and forming the wall of a chasm about
20 feet deep, at an inclination of 15° from the vertical."

(p. 98.)—" A dyke of fine dolerite reaching 4 yards in width,
passes by Lough Aninyer on the mainland north of Gh^eedore Bay.
It is in some parts porphyritic with large crystals of felspar."

*' Two separate rocks, rising from the sea, in nearly the same line
and distinguished by their dark colour, are prolongations of this dyke.

** Twry Island, — Tory Island in the Atlantic Ocean, is situated due
north of the coast of county Donegal. It is about eight miles distant
from the foreland, and lies a little to the east of north from that head-
land. The distance from the boat slip at Magheraroarty is nine miles,
and from Dunfanaghy, around Horn Head to the same part of the
I^d, about 16 miles. The Islands of Inishbofin, Inishdooey and
luishbeglie between Tory Island and the mainland, the farthest north,
viz. Inifihbeg, being about half-way across.

'* The Island of Tory is a narrow strip of rocky land lying with
its longest diameter of nearly three miles in a north west and south-
fast direction."

(p. 99.) — " At its widest part, viz. at the north west extremity,
It is less than a mile in width, and at the narrowest part, i,e, just east
of Westown, or about the middle of the Island, it is only one-fifth of
& mile across, its average breadth being about half a-mile.

** The Datives always speak of it as Tor-i, and this would appear
to be the most explicit manner of spelling, to be in consonance with
the derivation of the name.

"This island was anciently the stronghold of the Formorian
pirates, whose chief was Balor, ' and two of the tower-like rocks on the
side of Tory are still called " Baler's Casde " and " Baler's Prison "'
Of the former of these, there remains but the site. As it was situated
at a very narrow isthmus which is the only passage from the main
portion of the island on to the Doon peninsula (where peat is cut for
foel), it is probable that the islanders have removed the building.
The highest point on the island, viz. Doon Balor, 282 feet above sea
ierel, is situated at the north end of this peninsula and in the cliffs
further south, was * Baler's Fort.' Around Westown are several not
too well preserved monuments of ancient worship^ — as St. Columbkille*8
Church, St. John's altar and cross, another altar, a grave. Temple
'Uroresher, or Church of the Seven, and a Qoigtheach, or round tower.
This latter is the only one that is well-marked and preserved."

(p. 100.) — "The outline of the island as seen from the south,
although very much broken and irregular, presents a general inclination

1 58 Proceedings of the Royal Irish Academy.

from the west to the east. To the south and south-east of the light-
house at the north-west extremity, the land is very little ndsed
above the sea level, to which it runs down by a gradual slope, being
at its junction with the shingle beach which forms the south-west
part of the shores, but 26 to 30 feet above the sea. The whole of the
southern and south-west half of the island is comparatively flat and
lowland, and rises with somewhat rapid slopes to the north and north-
east terminating in precipitous clifEs, from 100 to 280 feet in height
The north and north-east coast line, in marked contrast to the south
and south-east shore line, which is but slightly indented, is more
irregular, being carved into a multiplicity of minor headlands and
points and many varieties of inlet, creek, and cave.

** Owing to the irregularities in the sea-clifEs, the coast presents the
appearance of numerous tors or isolated crags, standing up as if occur-
ring in the centre of the island.

^^ Tory Peak (see fig. 21) is a most prominent feature in the outline
of the island, appearing like a great tower standing about the centre
of the island. It is in fact a partially detached sea-stack on the north-
east coast, near the east end of the island. A wide bay is cut far into
the land south of this peak at Scoltshoarsa ; and the land about East
Town, lying very low, permits nearly the whole of this huge mass of
rock to be seen from the south. Similarly Tormore or the great Tor,
which occurs at the north extremity of the Doon peninsula and which,
in common with the latter, runs out to north-east at nearly right
angles to the main island, appears as a massive tower, at the eastern
extremity of the island, the whole of the ' Anvil,' as the ridge is
called, being foreshortened into one mass. The whole of the Doon^
which rises rapidly from the sea level at Port Doon to nearly 300 feet,
also is foreshortened into a craggy or torlike mass."

(p. 102.) — ** From the extremity of Doon Balor, a fine view of the
whole island is obtained, and on a dear day nearly all the northern
clifb are seen from this position, with the lighthouse distinctly promi-
nent at the western extremity of the island (fig. 22).

" Granite, — The greater part of the island consists of granite
which varies greatly in texture, being at one part a massive com-
pact durable rock, and at others, where porphyritic, much more
decomposable than in the former case. At the north-western
extremity at Toradardeen, the rock occurs in irregular amorphous
masses. Along the sea-cli£Ps, the rock appears as a highly porphy-
ritic gray granite with the weathered surface thickly studded with
crystals of orthoclase. The shore is much indented.

O'Rbillt— On the Waste of the Coast of Ireland, 8fc. 159

** Eaftt of the promontory of Ardlaheen, the rock still maintainB
its ooanely eryetalHiie character. A set of very distinct, nearly
Yertical cross-joints, rnnning south-west occurs here, traversing the
reddish and grey gneissose granite. The direction of these points
nearly coincides with that of the promontories and inlets in this
locaHty."

(p. 104.)^'' Along the south-east face of Meggart Headland a
basalt dyke about 2 feet thick occurs, and thins out before reaching
the top of the difP. A large basalt dyke, 10 or 12 feet wide, occurs to
the n<n^ of this, and is continued into the north-west face of Morard
Sead, where it is weathered out and forms a cave.

*At PoUabraher ('Wolfs hole'?) the sea has cut through the
small headland, forming a natural arch, and inland occurs the cave
Lagirehy (or Ram's caye), which is a round or oval shaped pit, like a
qnany hole, and has been formed along vertical joints running east
n(»1h-eaBt. It communicates with the sea by an underground passage,
and is cut out of the granite, which has slipped in and been carried
away as shingle. Torbanny, which is a small sea-stack, rises to the
eastward.

'' Along the western side of the island, a very well-marked shingle
bank forma the margin. It consists of rounded blocks of quartz, a
few of homblendic and augitic rocks, and the rest of gray gi*anite.

*' To seaward of this bank a fringe of rocks but little elevated above
the sea level, consists uniformly of grey and reddish granite."

(p. 105.) — "The quartzite in Port Doon dips generally to the
eastward at angles of 20° to 30°. It is intersected by numerous
vertical joints, running south 30° east, along which the island rock,
Torahaor, which stands as a sentinel at the entrance of this little
oatoial harbour, has been cut off from the mainland."

(p. 106.) — *' Northward, from Port Doon, the ground rapidly rises^
snd the rock in the cliffs consists of white tabular and flaggy quartzite,
dipping east 10° north at 20°. The coast-line is most irregular and
deeply carved into bays, together with headlands and numerous sea-
stacb. These cliffs end in a remarkable narrow ridge of rocks named
the ' Anvil/ which terminates at its northern extremity in Tormore.
A channel has been cut through by the sea at the southern end, where
this ridge ia connected with the mainland by a natural arch.

*' Along the western shores of the Doon the rocks^preserve the same

general easterly dip, and exhibit various sea-stacks or tors ; a rather

picturesque one being Tomaweelan, which stands at the entrance to

Portacballa Bay. In the southern cliffs of this bay and on the shore

m.i.A. pROc., VOL. xxnr.., sbc. b.] 0

160 Proceedings of the Royal Irish Academy.

occurs a confiiderable deposit of pipe-day, thoagli of a very impure
sandy nature. It is at least 12 feet wide, extends to the sunniit of
the cliff, and is nearly vertical."

(p. 109.) — ''Bloody Foreland District, ^'Deep deposits of bonlder
clay, often covered by several feet deep of peat, are seen in section in
the cliffs of Bloody Foreland, and the coast lying to the east. The
face of the cliffs is often cut by narrow gullies, worn by small
streams, which become swollen in rainy seasons. Large blocks, fallen
from the boulder clay, sometimes strew the shore, one of which, cloee
to Meenlaragh, composed of granite was found to measure about
15 feet by 12 feet by 9 feet."

(p. 110.) — "Deposits of peat occur at several points along the
coast, which lie wholly or in part below high-water level, such as
north of the Gweebarra R. near Dooey Hill, and south of the estoary
at the Black Strand."

(p. 111.) — "A peat deposit, evidently grown in situ^ has been
observed below high- water mark on the ocean side of Inishfree Island
on its south side. Fragments of peat, washed up by the waves when
more than usually large, strew the Leabgarrow strand on the east
side of Aranmore, and the peasantry speak of the possibility, not very
loDg since, of crossing to Rutland Island at low water.

" This seems, therefore, strong evidence for a recent submeigence
of the land to some extent. Mr. Harte in his Paper ( Joum. Boy.
Dub. Soc., Dublin, vol. i., pp. 25-27) speaks of ' forests that are
under the sea ' which * may be very extensive.' "

(p. 112.) — "Remains of peat bogs, now covered by the tide,
frequently occur along the coast, at Ballyness Bay near Falcanagh,
Oortahork, and Ards Point. This submergence constitutes proof of
depression of the land during a recent period.

" Dungloe District. — Large accumulations of drifted sand are to
be met with at several points along the coast, frequently forming
dunes, as at the mouth of the Qweebarra River. The drifting sand
has been disastrous to a village which formerly constituted an impor-
tant fishing-station on the east side of Rutland Island. The sand now
almost covers the desolated habitations.

" ((Note.) — Mentioned by the late Lord Qeorge Hill in the second
part of his elaborate * Hints to Donegal Tourists ' (1846-7). Rutland,
his Lordship stated, was a green island until forty years before the
date of his publication (1806-7), was then a military station with
*good houses' and * quite a gay place.' ")

(p. 113.) — "In Skull Island (Inishcoole) human remains lie

O'Rbilly— 0» the Waste of the Coast qf Ireland, 8fc. 161

entombed in blown sand, as noticed by the late Mrs. Craik (in an
< Unknown Country ' : an illustrated account of a tour in Donegal).

*' Fanad Dittriet, — ^Examples of raised beaches are found all round
the northern coast from Ballymastocker Bay, on Lough Swilly, to
Mark's Point at the ' Narrows ' on Mulroy waters.

''At Ballymastocker Bay a fine example may be seen of an old
flea-diff, 1000 feet inland from the present tidal flow. It is semi-
circular, and about 50 feet high ; the space between it and the sea is
filled with sands which, in places, are becoming cemented together.
ETidence of a raised beach can also be seen at Sessiagh Bay, and to
the south at Doaghmore Strand."

Memoir, Geological Survey of Ireland, Sheets 22, 23, 30, and 31
(in part).

(p. 7.) — ''The district described in this memoir comprises the
south- westerly portion of the county Donegal, lying to the north of
Donegal Bay, and stretching along its western margin into the
Atlantic Ocean. Along its coast-line it is indented by numerous bays
with interrening headlands."

(p. 10.) — " Sea-Cliffs and Headlands. — The promontory of Banagh
generally presents a bold and rock-bound coast to the Atlantic, deeply
indented with bays and gullies, and often rising in cliffs several
hundred feet above the surface of the ocean. Along the southern
shore of Loughros Beg Bay, the quartzite cliffs rise from the ocean in
a steeply sloping wall, 500 feet in height ; and some distance further
west, under the summit of Slieve Tooey, a nearly vertical wall of the
same rock descends a thousand feet from its edge to the surface of the
waters. At the head of Tormore Bay, still further west, the cliffs are
almost equally lofty and steep ; and all along the coast to Glen Head
they bre^ off with faces several hundred feet in height. Glen Head
is a remarkable cliff, almost vertical, with a descent of 600 feet,
surmounted by a tower, built as a watch-tower in the time of the
Spanish Armada. The long ridge of quartzite, which bounds the
valley of GlenoolumbkiUe on the south, here breaks off abruptly ;
and along with the cliffs terminating at Doon Point, encloses a little
bay, at the head of which are masses of shingle, piled up by the
powerful Atlantic waves when impelled by westerly winds. The
force of these waves must be sometimes prodigious; but their
destructive effects on the quartzite rock, which is naturally brittle,
are somewhat l^sened by the occurrence of intrusive sheets and dykes
of qpidiorite, which help to bind together, as with bands of iron, the

162 . Proceedings of the Royal Irish Academy,

masseB of natural masonry of which the coast is constructed. }bM
More headland, fonned of tough schistose rocks, though not very lofty,
seems to haye heen able to resist the force of the waves better Uuat
the clifb of quartzite, as it projects much further out into the ocean,
than the adjoining parts formed of the latter rock; and the same
statement holds good with reference to the promontory of Malin Beg
to the southward. This headland is, howeyer, somewhat protected by
the group of islands formed of felstone porphyry, of which Bathlm
O'Bime Island is the largest. All along this part of the coast the
rooks are fissured, faulted, and thrown into numerous sharp folds.

'' From EoBsarrell Point, as we proceed southwards, the coast
cliffs retreat inwards, and gradually become more and more lofty and
precipitous till they culminate in that magnificent wall of natnnd
masonry, which descends from the summit of Slieve League to the
surface of the ocean, through a height of nearly 2000 feet, and
stretches in an unbroken sweep from north-west to south-east through
a distance of about three miles. The greatest eleyation of thi&
stupendous cliff occurs immediately below the summit of the mountain,
which reaches a height of 1972 feet, though here the actual cliff
is only 1650 feet in depth, the remaining part consisting of slopes;
and from this the crest gradually descends in either direction, till at
Bunglass Bay, near the southern extremity, the cliflis about 1000 feet
in height. The cliff is formed of successiye courses of quartsite and
schist, yariously coloured, yellow, red, and gray, with a gentle dip
southward, or rather towards the south-east, along the northern and
central part of the escarpment, but becoming highly inclined and even
vertical, where the bay sweeps round to the west at Bunglass, where
it is surmounted by the cliff called ' The Eagle's Nest.' "

(p. 11.) — ''At the base of the vertical cliffs of Bunglass Bay a
shelving sliingle beach slopes downwards into deep water; and, stand-
ing on the edge of the cliff, you look down into the clear green waters
of the ocean from an elevation of 800 feet, and again upwards to the
cliffs above, rising to a similar height. This great sea-wall of meta-
morphosed strata has attained itiS present dimensions, both in length
and altitude, by the gradual undermining of the base, where the surf
is always breaking, and against which, during storms, the waves
beat with terrific force, as exposed to the full sweep of the AtLantie
waters. It would appear from the position of the summit of the
mountain immediately over the cliff, and from the direction of the
contour lines, as shown in the map, that the cliff has now reached its
maximum of elevation. When the work of excavation has proceeded

O'Rbilly— Ort the Waste of the Coast of Ireland, 8fc. 163

for some distance fnrtbei*, the height of the cliff will gradually leasen,
owing to the fall of the gFoiind inwards. Among the coast cliffs of
Ireland, and perhaps of the British Isles, there is none which reaches
in loftiness that which presents its face to the Atlantic along the
western flank of Slieye League, and which forms a hreakwater not
unworthy of the great ocean which washes its hase.

''This coast, indeed, from Carrigan Head to Donegal, consists
of a succession of deep hays, with proportionately long intervening
headlands," «

(p. 12.) — " East of the promontory of Camtullagh lies MoSwyne's
Bay, separated from Inyer Bay hy the long and narrow promontory
tenninating in St. John's Point, which, owing to its form and length,
is the most remarkable headland of Doneg^ Bay. Measured from
St John's Point to the Tillage of Dunkineely, this promontory is
orer 6 miles in length, with an average breadth of half a mile."

(p. 36.) — *' There is a large sea-cave directly beneath the highest
point of SHeve League."

Memoir Geological Survey of Ireland, Sheets 31 (in part) and
32 (1891). The district described lies in the south-western border
of the great tract of metamorphic rocks which stretch northwards
into the highlands of Derry and Donegal. (No available particulars
given.)

Memoir Oeological Survey of Ireland, Sheets 42 and 43 (1885).
The eastern and larger portion of the ground described belongs to the
Co. Leitrim.

(p. 9.) — ** The lowland belt stretches through the district as an
nndnkting or boggy tract between the central mountains and the sea,
most usually presenting a low line of cliffs and islets, or a sandy
foTcshore, to tiie full force of the Atlantic breakers.

" That pert of the Atlantic off the coast has not any great depth in
the vicinity of the land, nor does it seem to present any such abrupt
inegularitiea in the form of its bed as diversify the shape of the
ground under description. Such depths as 8, 15, and 25 fathoms, are
marked on the Admiralty Chart, within distances from the coast-line,
which, taken inland, would show differences in elevation equal in
amount to considerably more than 200 fathoms. Drumcliff Bay is
only one or two fathoms deep ; and even out in the wildest part of
Donegal Bay, mid-way between the Teelin and Ardboline headlands,
the depth given (31 fathoms) is less than the height above sea-level of

164 ProeeedingB of the Royal Irish Academy.

the broadest part of the Ardboline or Lissadill promontory at the foot
of Benbulben. Four miles from the northern shore of this promontoij
land re-appears in the small flat island of Inishmnrray and adjacent
rocks."

(p. 27.) — ^^ Raised Beaches, — Portions of the coast of Dmmciiff
Bay, not far from Carney, are marked on the map as raised beaches;
these containing oysters, clams, periwinkles, &c., are now fonr or six
or seven feet aboye high-water mark."

Memoir Geolog. Survey of Ireland, Sheet 55 (1885). A certain
portion of the description refers to the arms of the sea, Ballysadare^
and Sligo Bays. (No available particulars given.)

Memoir Oeolog. Survey of Ireland, Sheet 54 and south-west part
of 42(1880). The district described lies almost altogether in the
Co. Sligo. (No available particulars given.)

Memoir Geolog. Survey of Ireland, Sheets 39, 40, 51, 52, and
northern portion of 62 (1881). The area described occupies the
north-western portion of the County Mayo. It is bounded on the
north and west by the Atlantic Ocean, and on the south by Blacksod
and Tullaghan Bays."

(p. 7.) — ^^North-eastern Portion of District, — The tract of country
to be described is that which extends from Benmore Head, west of
Bunatrahir Bay, to the old coast-guard station at Porturlin.

" The physical features of this district are characterized by a bold
and precipitous coast-line. To the mighty roll of the waters of the
Atlantic must be attributed the varying features of headland and bay,
precipice and shingle ridge, island and gorge, which give this coast its
great interest to the geologist. The ocean waves, acting along lines
of weakness and displacement, those of fissures, jointage planes
or dykes, interesting rocks of different degrees of hardness, and in
various stratigraphical positions, have carved out the diverse features
of the coast-line, as we now And them. Eastward the greatest eleva-
tion attained by the cliffs is at Keady Point, where it reaches 352
feet from this on either side, the shore-line, while still precipitous,
gradually diminishes in elevation. Between Glengloss Point and
Belderg Sarbour the cliffs range up to 189 feet high, and are sharply
indented along lines of fault or of fissure. To the westward they
again increase considerably in altitude, attaining 640 feet at Benwee
Geevraun Point, and in continuation of Glinsk mountain to about 850

O'Bbilly— Ofi the JFaate of the Coast of Ireland, 8fc. 165

feet, although they are not so steep. Opposite the island of Illan-
master the difb are 790 feet high, and Very precipitous. From thence
to Portorlin they range in varying heights of 400 to 600 feet, with
islands detached from the several headlands and fissure gorges cut into
the mainland. Further westward the cUfb attain to still higher
eleyations."

(p. 8.) — ** The Stags of Broad Haven form a group of four islanda
rifling from the surface of the Atlantic, at a distance of upwards of a
mile and a-half from the north-west coast of Mayo. One of these
rises to an elevation of 316 feet, another to 312 feet, the next to
256 feet, and the lowest to 243 feet. That nearest the coast is
domeshaped, while the most northern of the group is pointed. They
consist of schistose rocks, and form a favourite retreat for sea-hirds."

(p. 9.) — ''The long north and south peninsula, locally called the
' Ifullet,' which is separated from the mainland by Blacksod Bay,
and an arm of Broad Haven, contains no very considerable elevations^
the highest (434 feet) being towards Erris Head on the north, whilst
the remainder of the peninsula is low lying and gently undulating
ground. The coastline from Bossport, along Broad Haven to Erris
Head and thenoe to Annagb Head, presents a continuous line of bold
precipitous clifEs, generally inaccessible, the remaining part being low,
gravelly, sandy, and rocky beaches."

(p. 17.) — ''Along the shore a little to the eastward, between
Nyxanag^ and Claddaghnahowna, at the base of the cliffs 160 feet
high, the metamorphic rocks are visible at about the sea level,
dipping apparently 20^, 30^ east-north-east beneath the Carboniferous
sandstones which overlie them unconformably, and dip from 5^ to 10^
only. By the breaking away of the sandstones, owing to the sea
action, the schists are revealed at low tides, and in the face of the
cliff adjacent, their broken uneven surface is again visible underneath
the sandstones and shales."

(p. 18.) — " At the head of the small bay, between Keady Point and
Benadeneen, into which the waves roll majestically, the inclined
face of rock is laid bare by the removal of the outer portion up to
a line of fault, which bears east-south-east, with an inclination
northward at 25^ from the vertical, the southern side remaining
intact."

(p. 22.) — ^^ Peat ho^s. — ^Peat bogs are numerous and extensive
over the whole country, particularly in the southern portion, where
they are of vast extent and great thickness, and along the shores of
Blacksod Bay and Broad Haven, extend even below low-water mark.

166 ProceedmgB of the Eoyal Irish Academy.

These extensiye low-lying bogs contain numerooB tranks and loots of
large forest trees, such as the oak, fir, &c., as well as the stems, leaTes,
and fruit of the hazel, and other stunted varieties, which proye the
existence at one time of large tracts of forest in the country, althoo^
it is now quite destitute of timber."

(p. 27.) — ** Following the shore line from Belderg Harbour,
westward. Horse Island is met with. It is almost completely
separated from the mainland by a deep and narrow gorge, due
probably to a line of fault or fissure.

** About one mile and a half further westward (of Benwee Geevraun
headland) a most interesting locality occurs, where Glinsk Mountain
abuts upon the seaboard near the townland boundary of GHnsk and
Laghtmurragha. The cliffs attain a height of about 900 feet, and
are broken into steep slopes and precipices; they are partially
accessible from the top by a winding [path, or from the bottom by
landing from a boat."

(p. 31 .) — ' *' A more remarkable^fault fissure and dyke occurs between
the islands of Illanmaster 353 feet high, and the headland adjacent
which is 790 feet high. This narrow deft is cut down to the sea
level, with almost perpendicular sides, through which there formerly
was an open water passage, but now a ridge of shingle has
accumulated in it. Traces of the dyke are visible at low water. On
the opposite side of the small islet, this fissure is further continued
east 25^ south, as a chasm into the flanks of Glinsk Mountain, the
open sides of which attain to a perpendicular height of 300 feet."

(p. 31.) — '' The most remarkable of those fissures and dykes occurs
adjacent to the Island of Torduff. A cleft has been formed along a
line of fault into the mainland, with perpendicular sides up to 400 feet
in height, and scarcely 10 feet apart in some places, but widening out
at the top. Seaward it is prolonged in a rather remarkable manner ;
first cutting off by an open chasm with vertical walls, one island from
its adjacent headland, then another— Torduff — ^from its headland,
which is over 500 feet high, then Illaunakanoge from its headland,
Fohemasmeel, almost 550 feet in elevation, and further westwaid, but
with wider interval, the Island of Carrickduff from its adjacent
headland. The view looking down this cliff with its four pairs of
opposing perpendicular headlands on either side is almost unique.

Memoir Geological Survey of Ireland (Sheet 63 and north half of
74), 1880. The district described is one of the wildest and most
inaccessible in Ireland, but is not devoid of many features of

O'Reilly— On the Waste of the Coast of Ireland, Sfc. 167

geological interest. It lies wholly in the County Mayo. It is bounded
on the south by Clew Bay, and on the west by part of Curraun, Achill,
and parte! the Blacksod Bay.

(p. 17.) — ** On the shores of TuUaghan Bay below high-water
mark, nnmerous large trunks and roots of trees are to be met with
resting in the bog, showing the existence of extensive forest in the
locality at one time. At present the whole country is quite destitute
of timber of any kind."

Memoir of Geological Survey of Ireland (Sheet 62 and northern
partof Sheet 73).

(p. 7.) — " The area described embraces a small portion of the coast
ot west Mayo, from Bally croy to Blacksod Bay. As also the Islands of
Achill, Achlll Beg, Clare, Innishbiggle, Innishgalloon, Duvillaun,
Leamarch, &c.

** The whole of the Island of Achill, and a considerable portion of
the adjacent mainland, is of extremely mountainous character. The
Island of Achill, whose extreme dimensions from .east to west are
15 miles, and from north to south 11 miles, is separated from the
"^inland by Achill Sound, a channel of the average width of from
1^ to three-quarters of a mile, narrowing to about 300 yards at
fiallsmouth, the Ferry, and Darby's Point, and spreading out into a
large expanse of wat^ north of the feny between Achill and Bally-
<^7, containing the Islands of Innishbiggle, Annagh, &c., and
sweeping away to the eastward and southward, joins Ballycragher
Bay, forming with the waters of Clew Bay at Mulranny, a peninsula
of that large tract of country which lies to the west of the village, of
iriuch Curraon Hill, 1715 feet high is a conspicuous feature.

*^ Proceeding to the west, or west 20^ north, from the ferry at the
Soond, the ground rapidly rises within a distance of 4^ miles as the
crow flies, till it attains west of the village of Mweelin, an elevation of
lo30 ieet, forming there a ridge or tableland running due south for
nearly 3 milesy and terminating at a height of 818 feet in Dooega
Head. Its western margin descends with a nearly vertical descent of
900 feet into the sea, forming the precipitous and picturesque cliffs of
Minaun* To the north of this, and east of the Protestant colony, the
ground again rises 698 feet above the sea level, whilst inmiediately
to the west abruptly rises the mountain of Slievemore, attaining a
bi-ight of 2204 feet within a horizontal distance of one mile. This
mountain gradually slopes to the west, and at a distance of 1^ miles
from its apex» terminates at Ooghnaboo, in sea cliffs 80 feet high. Its

168 Proceedings of the Royal Irish Academy.

north-east face is broken by a precipitous rift or chasm, which extends
to within a few hundred feet of its summit, and runs down nearly to
the sea cliff. From this out to the extreme west point of the island,
viz. Achill Head, the surface of the country consists of only high hills
and elevated boggy plateaux, culminating in the mountain called
Croaghaun, whose summit towers nearly vertically over the Atlantic,
at a height of 2192 feet. This mountain is sheared off by an
enormous precipice of nearly 2000 feet, running from top to bottom of
its north-west face, and forming an almost perpendicular wall to the
sea."

(p. 9.) — ^^Achillheg Island, — One and a half miles long by one
mile wide is an elevated tract of land, lying about half a mile from
the main island. Three hundred feet up the sharply inclined flank of
the south-east face of this high ground are found large perched blocb
of red sandstone, more than a ton in weight, in a condition of unstable
equilibrium. Achillbeg is bisected by a broad, sandy, east and west
cut, or passage, running parallel to the passage that divides the two
islands, and the three eastern and western valleys, occurring at
intervals of two miles each, going north on the mainland. The above
passage is nearly on a level with the sea, which has evidently swept
through it ; its direction coincides with that of the joint planes and of
the numerous eastern and western faults."

(p. 12.) — ** Starting from the north-east end of the island, Tii.
Kidge Point; the ridge is due to the hard siliceous nature of the
schists, backed up on the least weathered side, by a strip of still
harder quartzite. On the west side of this the sea has encroached
along the line of strike, leaving exposures of the harder portions of
rock here and there. Korth of Doogort the coast-line is most irregular,
the rock being soft and easily decomposed, and also cut by numerous
faults. The bold sea-cliffs at the base of Slievemore, standing out in
a semilune, are composed of a haid quartz-schist. From Dirk to
Annagh the coast-line is recessed at right angles to itself. Here the
rocks are not less hard, but we have a sudden change of dip along a
northern and southern fault at Dirk, and the change in the outline is
probably due to jointing, by which the rocks are much cut up. Thi»
increased excavation is due also to the reversed dip. From this point
the seaboard projects outwards until it terminates on the north-west
point of Gubroenacoragh, composed of hard quartzose schists. The
flanking cliffs on the north and west coasts of this tableland, being o!
felspathic or micaceous schists, have been more rapidly cut away
along the parallel jointings, which running north-north-west at

O'Reilly— 0» the Waste of the Coast of Ireland, 8fc. 169

intenrals of two feet, cut the rock into slabs. In the Croaghann cliffs
the dip changes from south-south-east to north-north-east or north-
easty and continues thus out to A chill Head, the master joints running
parallel to the line of cliffs. Along the shores on the south, Moyteoge
Head presents a harrier of quartzite to the agents of denudation, thus
sheltering the inlet or baj of Eeem from the south-westerly gales.
The hay is formed parallel to a line of fault. Further east the head-
land of Guhalennaum More stands forth in the comparative impenetra-
bility of quartzite, being carved out along the lines of jointing — the
indentations of Dooagh and Keel are cut in along the line of strike."

(p. 13.) — " On the east side of Keel Bay, in the Minaun Cliffs, the
strike bends more and more to the north-east, the jointing continuing
approximately at right angles to the dip. From Doonty Eighter, the
Bouthem point of the quartzite cliffs of Minaun, the coast-line, as a
whole, trends to the south-east, the schists becoming softer, more
micaceous, steatic, and chloritic as we go south, and this portion of
the coast being more exposed to the south-western gale has run at the
southern end of the island away to a point. It is very noticeable that
the coast-line throughout is approximately either at a very high angle
or at right angles to the direction of the dip.

'* Achill Sound itself appears to have been formed by a gradual
sabsdence of the land, the direction of the coast-line being approxi-
matively parallel to the line of the strike of the rocks, viz. north-
north-east and east-south-east, and frequentiy coinciding with that of
the major joints. On both the eastern and western shores also, the
bog is found ronning down on the beach, and forming banks at the
level of high tide."

(p. 14.) — '' At OoghreUeyranneU there is a cavern cut in along the
joint lines called the < Seal Cave.' "

(p. 15.)— "^om 8addk B$ad to Ackill JZkni.— Very fibrous, and
luf^y felspathic, coarsely crystalline gneissose schists. Cliffs dan-
geroQB, and almost inaccessible."

(p. 17.) — '' At the east side of Keel Bay in the Minaun or
'Cathedral Cliffs,' the sea having excavated passages through pro-
jecting points along the joint planes, we find the coast-line and hcad-
Lind of hard, flaggy, and tabular quartzites, with occasional bands of
argillaceous and other schists. We here notice in all the more
weathered parts, «. 6, those cut into the cliff, that the direction of
the dip is at right angles to that of the most weathered face of the
diff."

(p. 18.) — "In Achillbeg Island. The coast-line here is very

170 Proceedings of the Royal Irish Academy.

irregular, being cut into numerous inlets, among rapidly weathering
rufity-looking schists. In the guts the sea is rapidly catting its way
between the foliation planes.

Memoir, Geological Survey of Ireland (Sheets 73 and 74 (in part),
13 and 84) (1876).

The country described includes a part of the county Mayo with a
small portion of the county Galway. This tract is bounded on the
west by the Atlantic, and on the north by the southern shore of Clew
Bay, and on the south by Killary Bay.

(p. 12.) — ^' In the Atlantic, off the mainland, are some ialandB,
the largest being Clare Island, which is about four and a-half miles
long from east to west, while its greatest breadth is not more than
two and a-half . To the west of this island are the steep cUlb of Knock-
more, which rise directly from the ocean to the height of 1520 feet
Next in importance are Inishbofin, Inishark, Inishturk, and Cahir.
Most of those islands would appear to be the peaks of submerged
ridges. Cahir and Inisturk lying in a line with the ridge that extends
from the Mweelrea Mountains, towards the north-west, north of
Loughs Cunnel and GlencuUin, and the yalley of the Owennadomaun
to Cross Lough; while Inishbofin and Inishark may be either on
the ridge that forms the Rinvyle Promontory or on the continuation
of the ridge forming the promontory called Cleggan Head, both of
which are included in the district to the south."

(p. 14.) — " Clare Island, — On the eastern side of this ragged and
wild island is the only landing-stage, which is afforded by a smooth
beach. Between the western and eastern coasts the island is trayersed
by several ridges of moderate elevation, culminating in that of Knock-
more, which, as already stated, presents a bold and steep face to the
Atlantic. The northern portion of the island presents a very rugged
appearance due to the unequal denudation of strata formed of different
materials."

(p. 14.) — ** Islands in Westport Bay, — These islands are remark-
able for their uniformity both in shape and composition, while their
summits never rise to more than about 100 feet above the level of the
sea.

''Their form is apparently connected with the direction of the
original glaciation of the district ; but on the western side of each —
save that of Inishgort, which is protected by Dorinish — marine action
is making a preceptible change, leaving perpendicular cliffs, while
towards the east the ground slopes to the water's edge."

O'Bbillt— On the Waste of the Coast of Ireland, Sfc. 171

(p. 27.) — ** Since the great Olaoial Period, but probably while
^aden existed in at least some of the sheltered mountain valleys, the
lea rose at least 350 feet higher than it is at present, its waters chang-
ing the featoies of all the valleys that came under their influence
while at the same time and subsequently atmospheric waste modified
the higher portions of the country."

(p. 39.)—-" InUhark, Iniehgcrt, Iniehkinnymore, and Inishhinny'
%.--The first is the principal island in this group. It is bleak
and wild, rocky towards the west and north-west, while there is an
envelope of drift on the eastern slopes. At the north-west shore
$re lliji and almost perpendicular cliffs.

*' hiihhofin has a general east and west strike. On an average it
ia four miles long and two wide ; but the north and south coasts are
indented with bays and at Lough Bofin from sea to sea, is not ^^^^ a
mile. The island consists of five hills, namely: — Weitquarter, its
greatest hei^t being 292 feet ; it forms a promontory nearly separate
from the rest of the island, being connected by the previously men-
tioned low isthmus in which Lough Bofin is situated : — MiddUquarier^
hi^Jiest peak, 288 feet ; Cloonamore, the north east hiU, having a height
of 167 feet ; Xnoekj the hill east of the harbour, 271 feet ; and Inuh-
^OR, 143 feet, which is a tidal island, and separated from the other
hilla during high water."

(p. 42.) — " A little north-west of Lough Bofin are north 70° west
dykes, which apparently are portions of the gabbro dykes just men-
tioned. Further north-west and north-east of Bunnamillen Bay, are
maanve dykes of melaphyre ; apparently portions of the same dykes,
hat separated from one another, and shifted by faults. These have
leathered considerably and formed the deep marked fissures called
' Boher-na-collig ' (Old hag*s path)."

(p. 42.) — *' Inuhtwrh lies about five miles north east of Bofin, and
eif^t miles from the mainland. The surface is undulating ; there are
lour marked peaks : the north-west or signal tower hiU, 629 feet ; the
north-east, 428 feet ; the south-east, 588 feet ; and the south-west 240
ieet The east and south-east coast is low ; while on the west and
noth-west are considerable difis, some of which are nearly perpen-
dicolar and vary from 200 to 400 feet in height."

(p. 43.) — *' JFVuhill and Govern Islands. — These rocky islets lie
2nm one to two miles from the mainland and are composed of very
fel^Mithie, massiTe, puiplish and greenish grits and sandstones, often
pebbly and mnch cut up by quartz strings."

(p. 73.) — '* The numerous islands in Clew Bay are saddle-backed

172 Proceedings of the Royal Irish Academy.

hillfi of boulder day restmg on limestone. These islands are peculiaiij
shaped, being generally oval, withtheirlongest axes runningin aneastand
west direction, the most western of them having their face on the sea side
cut away by the encroachments of the sea, and presenting vertical difis."

Memoir Geological Survey of Ireland (Sheets 93 and 94 and adjoin-
ing portions of Sheets 83, 84, and 103), 1878. The area described is
bounded on the west by the Atlantic Ocean, on the north by tk
Atlantic and Killery Harbour, on the south by the Ocean.

^p. 8.) — *' The Atlantic Ocean, which bounds so much of this ana,
indents its coast by fiords, bays, and creeks, some of which are o!
considerable size and length. The largest and most marked of the
fiords are Eillery Harbour and Streamstown Bay. The first is over
nine miles long, and seldom, except near the east termination, over
half a mile across ; while Streamstown Bay is nearly five miles long,
and for the most part only a few hundred yards wide."

(p. 11.) — *^ The central ridge ends towards the west at Inishtuik,
in a height of 120 feet, while the northern branch immediately west of
Aughrustbeg Lough is only 78 feet high ; nevertheless farther west,
and apparently one of the partly submerged peaks of this ridge, is
Ardillaun or High Island, with an altitude of 208 feet.

*' Bordering the Atlantic Ocean on the north, and partly parallel
with the northern branch of the east and west ridge just now described,
is a low range of hills that towards the north-west ends in the cape
called Cleggan Head. North-west of this is Inishbofin, and possibly it
may be part of this ridge ; however, more probably it is part of a more
northern ridge (Revyte Promontory)."

(p. 13.) — " Besides themountcdns now described there are isolated
hills forming conspicuous and striking objects, such as Letteimore
(* the hig slope^^ 1172), forming the promontory north of Ballynakill
Harbour; Lackairea, Hhe tangled flags^^ 1307 feet above the sea,
standing over and abruptly rising from Maum Bay, and in a quarter of
a mile gaining a height of 1279 feet)."

(p. 14.) — '* Islands. — Lying off the west and south coasts, also in
some of the bays, are large and small islands, and sea rocks. ( Oarrig
and Carrygeen), The islands off the north part of the west coast are
apparently peaks of the submerged parts of ridges as previously
suggested, being situated on lines having a similar bearing to the
parts of the ridges on the mainland, t.^. nearly east and west Iiiie&.

Southward of Mannin Bay and off the south coast, other arrange-
ments seem to exist ; and the islands, if they are parts of ridges, belong

0*Rkilly— 0/1 the Waste of the Coast of Ireland, Sfc. 173

to sjBtems having other bearings. South-west of Mannin are Chapel,
Duck, and other small islands. These extend in a north-east and south-
west direction, and may be parts of a ridge that runs from Knock, south
of the mouth of Mannin Bay, to Shiprock ; while the islands that lie
farther westward (Inishdugga, Inishkeeragh, Illaun-na-neid, or Slyne
Head, &c.) may be parts of a second parallel low ridge. Off the south
coast, the islands are scattered about irregularly; still, howerer,
tbey may also possibly be peaks of submerged ridges, as Inishlackan,
Ulammacroagh More and Beg, Croagh-na-keela, and the Carriggeens or
flea-rocks, three miles further to the south-west, lie in a line which is
pAnliei to the ridges or lines of islands just described ; while Mace
Head, St. Maodara Island, and the sea-rocks called the Sherds, lie in
i second nearly parallel line."

(p. 53.) — ■« Sufh Island. — This is the most westerly land in the
couoty Galway, and is bounded on all sides by high clifb, which are
for the most part perpendicular or nearly so."

(p. 54.) — *• Friar* s Island is also wild, rugged, and very inacessi-
ble. To the north-west in the granite, occur systems of east-and-west
and nearly north-and-south vertical joints, wldch produce a columnar
aspect when viewed from the west or north-west.

"The Boathem side of Gru^gh is glaciated, grooved, and etched,
the bearing of the ice varying from north 70° west to east and west."
(p. 58.) — " Islands off the Rinvyle Promontory ; Inishbroon, off Kin-
dle Point ; lUaunananima, IVeaghtllaun North ; Crump Island, — The
exposed portions of the homblendic rock weather freely into a rusty
fcrown crust."

(p. 89.) — ^* Islands off the South-west and South Coasts ; Inish-
^W^f Illaunaminara, Lyal More and Beg^ Inishkeeragh^ and lUaunane,
vith numerous smaller Islands and Tidal Rocks, — This group of islands
lies to the south-west of Mannin Bay, between it and Slyne Head,
many of them being joined together at low water.

*^ lUaunamid, or Slyne Mead, Chapel Island, Ducklslamd, Doonna-
vsul^ IllaunaUama, and their associated earrigs and carrigeens. These
islands and rocks form the south-west extremity of the Co. Qalway.
C^rrickfin, Horse Island, Strawheaeh Island, Carrioknure, IHaunpollna-
^ffci, lUaunurra, and the adjoining seacocks, — These islands and rocks
lie in and to the south-east of the bay, that is situated east-north-east
of Slyne Head."

'p. 90.) — '• To the south-east of Bunowen Bay, in Crompaun Bay,
ud aouth of Ballyconneely Bay, there are numerous rocks and smidl
;«land8.

174 Proceedings of the Royal Irish Academy,

** Inishlaekan, lUaunnaeroayh More and Beg^ and CroaghnahesU.^
These islands lie off the mainland south and sonth-west of Roundstone
Bay.

*' Mile Roeks and Sheds. — These isolated sea-rocks lie to the south-
west and south of Groaghnakeela, and are inaccessihle except in calm
fine weather."

Memoir Qeolog. Survey of Ireland, Sheets 104 and 113, and adjoin-
ing parts of Sheets 103 and 122 (1871). The area contained within
the limits of Sheet 113 is, for the most part, occupied hy the Atlantic
Ocean and the entrance to GFalway Bay. The two northern islands of
Aran are situated towards the western margin, and immediately to the
south of it, in Sheet 122, is Inisheer, the extreme south-easterly
island of the Aran group. Parts of the islands of Lettermullan and
Gorumna, with the portions of the mainland that lie east and west of
Gashla Bay, occur inside the northern margin.

(p. 7.) — " The Aran Islands, at the mouth of Qalway Bay, He in a
north-west and south-east direction, heing about 16 miles long from
Carrickemonmaodonagha, the north-west point of lUaun-eragh (the
western of the Brannock Islands), to Trawkeera Point, the eastern
extremity of Inisheer. They consist of three large islands — ^Inish-
more, Inishmaan, and Inisheer — with four small islands off the north-
west point of Inishmore, called the Brannock Islands (p. 8) ; and on
its east coast, at the entrance of Eilleany Bay, Illaunatee or Straw
Island ; the last named being joined to the island by a sand-bank
during low water. Connected with the Aran Islands there are vpry
few detached rocks, besides these to the north-west, which are included
under the general name of Brannock Islands, only three occurring off
Inishmore, called Island-a-reefa, Craghalmon, and Garrickmonaghan ;
and one, a spring-tide rock, called Finnes, half a mile from the shore of
Inisheer.

'' Sounds, — ^North of Inishmore, between it and lar-Connaught, is
the Iforth Sound, about 5^ miles wide. Between Inishmore and
Inishmaan is Gregory's Sound, from 1 to H miles wide, while Inish-
maan is separated from Inisheer by the Foul Sound, which is 1^ mile?
wide between the nearest points, and Inisheer, from the barony of
Burren, Co. Clare, by the South Sound, about 4 miles across.*'

(p. 9.) — " Mainland ; form of the ground, — The land on the north of
the Galway Bay is intersected by numerous chains of lakes, bays, and
creeks ; various harbours and bays are formed by the islands and
promontories. In Sheet 113, on the west of Lettermullen, are some

O'Rrilly— On the JFaste of the Coast of Ireland, 8fc. 175

islands and rodu. Eagle rock is the largest of the group of small
isUnds lying north-west of Golam Head, its highest part being 35 feet
abore tiie main sea-level."

(p. 10.)—" To the north-west is FUh Boeh, always above water;
as is also Seal Bock, which is about half a mile to the southward.
Lettennallen ia a wild, rugged-looking island, having irregular slopes
towardi the south-west and south. Qorumna Island is the largest of
the aidiipelago studding the extensive estuary between Oreatman's
and Kilkieran Bays. Between Greatman's and Cashla Bays is the
long, naiTOw promontory which separates these harbours. At its
southern extremity are two wHd, rocky head^ between which is a
small oove, called Boleen Harbour, to which, when the wind is not
from south or south-west, small boats can resort."

(p. 12.) — *' Aran Islands. — ^From the north-east shores of the Aran
Islands the land rises in a series of clifb or huge steps, which form
continnous terraces (see section, fig. No. 1), while from the summit of
the island there is a gradual fall south-westward, it ending, however,
at the sea-board in difis that, at the present day, are being formed by
the Atlantic Ocean.

*^InisAmore is Similes long, from its north-west point to Gregory's
Sonnd, and of various widths, being only half a mile wide at Fort-
mnrry, while at Kilronan it is a Httle more than two miles across.
Viewing Inishmore from the hills west of Galway town it appears to
be three islands. This is caused by two low valleys which extend
acroas it ; one west of Killeany Bay and the other south-west of Port-
marry. The latter is so low, about 50 feet above the sea, that it has
been mistaken by Galway-bound ships for one of the channels into the
hay, for which reason it has received the name of the * Blind Sound.'
Of it O'Flaherty says: 'About the year 1640 (1639*5 sun-spot
maximum), upon an extraordinary inundation, the sea, overflowing the
hank, went across over the island to the north-west.'

** (Note.) — In Mallet's list of earthquakes for 1640, there is one
mentioned on the 4th of April, at 3.15 a.m., felt in France, Belgium,
and Holland. Ferhaps it might have been the wave resulting from
this seismic commotion which caused the inundation. On the 15th of
Augost, 1852, a large wave rolled in on the west coast of the island,
drowning fifteen persons who were fishing on the rocks."

(p. 13.) — *' At this sea cliff, on the north-west of the island, there
are two well-marked terraces, which, with four below them obser-
vable in the neighbourhood of the village of Bungowla (see fig. I),
make in all eight terraces. Such terraces are not confined to the

E.1UU PBOC., TOL. XXIY., 8B0. B.] I*

176 Proceedings of the Royal Irish Academy.

land now above the sea-leyel, as submerged terraces oconr in Galway
Bay on the north-east of the island.

** The sea-clilb on the north-east side of the island are low, and
are often replaced by strands or shingle beaches. On the south-west
they have taken a definite character, being usually perpendicular, and
often over 50 feet in height ; however, at the north-west point of the
island, under the shelter of the Brannock Islands, there is a heary
shingle beach, on which boats can land in fine weather.

«From the north-west point, south-eastward to Gregory's Somid,
the eliffs are either perpendicidar or terraced. From Mweeleenar-
ceava, a little south ot the Brannock Islands, to Doocaher, except for
a short distance, at the ' Blind Sound,' the cliffs are perpendiculsr,
although at the base of some of them, as will be hereafter mentioned,
there are sea-terraces or steps below the high- water mark of spring
tides. At Doocaher the clifEs are about 100 feet high, and from that
towards the north-west they gradually rise to 234 feet at Corker,
from which they lower by degrees to the ' Blind 8ound ' ; but north-
west of this, at Dun iBngus, there is an Ordnance height of 265 feet,
and they attain their greatest altitude (300 feet), about a mile further
north-west, a Httle south-east of Polladoo. From Doocaher towards
the south-east to the point called Illaunanaur there are sea-temoed
difis (excepting a few short breaks), which are surmounted by a
rampart formed of large blocks. This is called, in Professor King*8
MS. account of the Aran Islands, the ' Block Beach.'

'^ From Illaunanaur to Portdeha, on the west of Gregory's Sound,
the difPs are perpendicular or terraced ; but on the north of the latter
place the Sound is bounded by a strand."

Lord Dunraven, ** Notes on Irish Architecture," vol. i., p. 1
(1875-77).

^*Dun ^ngusa^ on the greater Island of Aran. — Landing on
Aranmor, the largest of the three islands, and commencing his walk
at the southern end (the visitor) should keep along the edge of the
difis, which gradually increasing in height as he advances, seem to
form a grand barrier to the ocean that beats for ever at their feet.
They are of limestone, and are marked by long parallel horiEontal
lines or fissures, so that where they break, they seem to shape
themselves into huge masses, squared as if by giant hands. Here
and there, where in bold promontories they advance into the sea, they
have become sepai-ated from the land, and rise like towers from the

O'Rbilly— On the Waste of the Coast of Ireland, Sfc. 177

wsveB. Passing upward and onward towards the liighest point, the
trUTeller will begin to perceive that this precipice is crowned with a
<:ircalar wall, ' grej, weatherbeaten and wasted/ whose broken and
serrated edge stands dark against the sky."

(p. 2.) — <*The Bolitude and grandeur of the scene are un-
speaLible."

(p. 3.) — *'Dun JSngusa or the fort of ^ngus, is named from one
of the sons of Hua Mor, a chieftain famed in the earliest period of
Irish legendary history. It occupies an angle of the cliff, and is
therefore protected by it on two sides to the north and west. It is, in
plu, irregularly concentric, composed of three areas or wards, each
within its wall. The interior of the fort proper is half an ellipse, 142
feet on the short diameter which rests on the cliff, and 1 50 feet deep,
being half the long diameter, which projects northward from the
cliflL The containing wall is 8 feet to 12 feet thick. The entrance
is on the west-east, 90 feet from the cliff."

(p. 4.) — " The inner doorway is a rude, flat-topped opening, 8 feet
4 inches wide. Only its upper 3 feet is now viaible (1870-75), the
lower part being covered up with rubbish. When Dr. 0*Donovan
saw the doorway in 1839 it was perfect. It has since shared the
inoumful fate which awaits the whole structure.

*'The annexed drawing made in the spot in the year 1857, by
Hr. Frederick William Burton, was then a faithful representation of
this doorway ; but since that time a great change has been effected,"
(p. 9.) -** 'Duhh Cathair; * The hlaohfwt; Aranmor."
(p. 10.) — ** Dr. O'Donovan observes that this iortiflcation would
appear from its colossal rudeness to be many years older than Di&n
'fngas or Dtin Conor. The guide, an old man, who accompanied me to
the place, informed me that he remembered the wall nearly perfect ; but
^ a great part of it had fallen in a storm a few years ago. Scarcely
any of the inside face of the wall now remains, and the force of the
Atlantic waves has swept away the lesser buildings which it enclosed.
One wave he described as rising in such a vast body of water above the
<^, that it overran the hollow within the wall of the fort, and flung
the stones on all directions.

^^Inishmaan is Uiree miles long from the north-east to the south-
west, and half a mile wide between Trawtagh on Gregory's Sound, to
Trawbetteragh on the Foul Sound."

(p. 14.)—^' The north part of the island is bounded by low clifb,
itrands or shingle beaches. On the south-west from Trawtagh to
Aillinera, the clifb rise in steps, at the latter place being perpendicular,

178 Proceedings of the Royal Irish Academy.

and about 200 feet bigh, from which, southwaids, the smfaoe
giadiially falls to about 40 feet at the cliff opposite Taunabroff, the
south-western extremity of the island. From the south part d
Aillinera, where the cliff is about 170 feet in altitude, southward to
Taunabruffy and then north-east to Clogharone, the cUff is cut into
sea-terraces and surmounted by a ' block beach.'

^' Inishere is less than two miles across from the shore, north of the
Tillage of Balljhees, to Faidarris Point, and 2^ miles from Trawkeera
Point to Tonefeehnej. Captain Bedford says of this island, 'Its
shores are everywhere rocky, except at its nort^-east side, where there
is a small sandy beach called the North Strand.' "

(p. 26.) — '' The subjacent rocks of the Aran Islands are limestones,
with which are interstratified some thin shales and day beds. The
shales and clay seem in a great measure to have guided the denudation
that carved out the terraces, for at the base of many of the well-
marked cliffs, shale or clay beds occur.

*' The terraces are more or less undereut, and may have been
formed by marine action, but of this there is no direct proof ; if thej
were, the force of the waves would seem to have come from the east-
north-east, while, at the present day, it is from the south-west."

(p. 32.) — '' Sea Cliffs. — Of the cliffs of the north-east and north-
west of this island, scarcely more can be said than that already
mentioned in the general description, but those on the south-west are
peculiar, as in places they are terraced by the waves of the Atlantic.
Moreover, some of them are surmounted by the previously mentioned
' block beach.' This peculiar accumulation of blocks does not occur
at all on the north-east shore, while to the north-west it was only
found at the point due east of the north point of Brannock Islands.
On Inish-Eeragh, the westernmost of the Brannock Islands, there is
also a block beach, which is thus described by Captain Bedford:
^ On all but the eastern side there is a margin of massive blocks of lime-
stone, upheaved by the violence of the sea, and which now form a
sort of barrier against its farther encroachment. The highest part
of the island is the summit of the upheaved beach at the north-west
side, which is 36 feet above the mean level of the sea.'

** The north part of the north-west coast of Inishmore, as before
mentioned, is a perpendicular cliff that either extends upwards from
the sea-level, or has at its base a few steps.

" The vertical diff seems to be caused, in a great measure, by
vertical master- joints, some of which cut through all the visible beds,
while others only reach the shale beds. In the former case the cliffs

O'Ebilly— O/i the Waste of the Coast of Ireland, Sfc. 179

are peipendicular down to the sea, wbile in the latter there are steps
or 66a>teiTBce8 at the base of the clifb. South-east of PoUadoo, there
are four sets of steps at the base of the cliff, and the note made on the
ground is as follows : — * CM over 250 feet high. Two shale beds.
The cliff rises perpendicularly from the highest.' South of Portmurvy
there are from four to six of these sea-terraces, and the cliff is less
tban 50 feet high. South of Gurtnagapple, tiie cliffs are low but
perpendicular ; hereabouts nearly east and west master-joints occur
about 2 yards apart, and as the sea undermines the diff , masses of
rock, tons in weight, that are disconnected by these joints, topple over
and fall, forming a break-water at the base of the cliff. This
breakwater extends for about i a mile."

(p. 33.) — " At Corker, there is a perpendicular diff formed by
east and west master-joints. In the face of the diff there are two
abale partings about 40 feet asunder, the lower being about 60 fee
aboye half neap tide.

*' South-east of Nalhea there are four or five sea-terraces at the base
of the diff ; and at Whirpeas the diff is about 140 feet higher than
the levd of neap tide, a shale bed occurring about 40 feet above this
lereL To the east of this, at Pollnabriskenagh, the limestones are
tnrened by east and west master-joints, and the sea yearly causes
great destruction of the rock. This diff, which is about 1 00 feet high,
is undercut at the base. At Bensheefrontee, the point a little north-
west of Doocaher, ' the block beach ' sets in, and extends to the
south-east point of the island, having only five small breaks in it :
three at the ' cooses ' or small bays in the vicinity of Doocaher, one
at the ' coose ' called Doughatna, and one about 40 yards wide at the
Glasaan Bock, in all o! which places the base of the diff is undercut,
while that part which is surmounted by the ' block beach ' is stepped.
However, although it is undercut, and forms a cave at Doughatna,
yetbdow the cave, there are six very low steps. The highest part of
the cliff on which this beach occurs is in the vicinity of Doocaher, and
about 100 feet above the sea level, while the lowest part, a quarter of
a mile west of the Glassan Bock is about 35 feet. These steps at the
hase of the cliffs are usuaUy from 4 to 7 in number, seemingly having
been cut, one by low water of spring tides, another, by low water of
neap tides, another by high water of neap tides, another by high water
of spring tides. At one place, east of Carrickurra, there is a step
above hi^ water of spring tides on which the 'block beach' rests.
At this place the diff is about 50 feet high.

" The stones forming the 'block beach' are cast up during the

180 Proceedings of the Boyal Irish Academy.

winter gales, and some of them are of considerable size. A little
Bonth of Doughatna, the following observation was made — ^ Great
quarrying seems to be going on here during the gales. Blocks, 30
by 15 by 4 feet tossed and tumbled about.' And again, half way
between Doughatna and the Glassan Eock there is this note. A block
15 by 12 by 4 feet seems to have been moved 20 yards, and left on a
step 10 feet higher than its original site. East and west of the
Glassan Eock, there are two caves which run for a considerable
distance inland, and connected with both are ' puffing holes.' The
western puffing hole is 85 yards from the sea margin, and the eastern
33 yards. On the north-eastern side of the latter there is a small
*' block beach,' the blocks in which have all the appearance of being
yearly tossed about by the waves, while more are added to it, and ws
may suppose some sucked into the abyss below. Other puffing holes
were observed further north-west, but none so large as those just
mentioned."

(p. 34.) — *' The sands are very considerable, occurring in all the
islands. They are ever changing their positions ; and in 0'Elahertie*s
'History' we find mention of various churches, tombs, and fields,
now covered or nearly covered by them. Dunng the examination of
Inishmaan, tombs were pointed out near its shore that had only a
a few months previously been discovered, as up to that time they
were covered with sand, which now has been blown away.

*' At Trawmore, on the south of Salleany Bay, proofs have been
lately discovered (1860-67), not only of the movement of the sand-
hills, but also that this part of the land, since the islands were first
inhabited, has changed its level, as human structures are found under
the strand, and extending out seaward. In the history of the islands,
by the then vicar, Eev. W. Kilbride, it is stated : ' This movement of
the sands has gradually uncovered the ruins, which consist of two
*' cloghauns " or stone-cells, with beehive-shaped roofs and structure,
in every point similar to those usually called Leahuidh Diarmaid ayu9
Grains, or ^'Dermot and Graine's Bed," and old wall or single stone
fences, dividing the ground into regular fields and gardens, evidently
under cultivation in former times. These walls extend out seaward,
and all the structures, until very lately, were completely covered over
by sand from 10 to 20 feet high. They must apparently have been
buried a long time ago, for it cannot be less than a thousand years
since Eany's church was first erected on part of this sandbank which
still remains.' This author also mentions other places to which the
sandhills have moved during the historical period, one of the most

O'Bbillt— On the Waste of the Coast of Ireland, 8fc. 181

remarkable being at St. Colman's burying-plaoe on Inisheer, when a
'hillock,' which in O'Flahertie's time (a.d. 1684) was a nice ' green
plain,' ifl now onlj a mass of sand."

(p. 35.) — ^^Inukmaan^ — Seren continuons terraces occur, but
whether they are the continuation of the terraces on Inishmore or not»
it was impossible to prore.

'* SM-^lifi, — ^On the west coast the sea-dlifPs rise in steps^ as they
followed southward from Trawtagh, until at Aillinera they reach their
maximmn height (nearly 200 feet), southward of this they gradually
bH to nearly the sea-level at Ailyhaloo, the south-west point of the
island. Immediately south of AiUinera, at a height of about 170 feet,
there is a ' block beach,' which is continuous from this point round
the west and south-east sides of the island."

(p. 36.) — *' The clifb below it are always in steps very similar to
those described below the < block beach ' on the south-west of Inish-
mtm. In one place steps were observed over high-water mark of
ipring-tides, on which the following record was made. ' At Tauna-
bmfl the limestone is thin bedded, and the winter storms have formed
seven low steps between the high- water mark of spring tides and the
'< Uock beach." '

'* On the west coast some of the blocks are remarkable for their
sise, and the distance and height to which they have been moved by
the force of the waves. The following are the most notable: —
' Aboat one hundred yards southward of PoUnashedaun, ''large blocks
hAve been ' quarried ' by the sea, the largest measured being 15 by 5
by 4 feet." South of Taunabruft a block 20 by 5 by 1 feet has been
ndaed 20 feet, and moved 31 yards from its natural site.' A little
aontli of ibis, near Ailyhaloo, a block 19 by 8 by 3 feet was raised
5 f eet» and moved 8 yards ; and another 27 by 9 by 4 feet was raised
4 feet, and carried 9 yards.

'' On the south-coast the ' block beach ' is peculiar, being formed
of small blocks ; also, in other places, the blocks seem to be re-arranged
yearly, while here and there they do not seem to have been moved for
yeaxB, aa&d the impression formed at the time was, that the tidal
current cannot hereabout now set as strong in the same direction
against tlie coast as formerly, because samphire, sea-pink, etc., now
grow freely on the two terraces below, as well as amongst the blocks
forming the beach ; moreover the blocks seem not to have been stirred,
or added to by the sea, for years." This beach was found extending
■a far towwds the east as Clogharene, the south-east extremity of this
inland.

182 Proceedings of the Boyal Irish Academy.

(p. Z^.)—^^ Islands off Erriaainhagh: Inishtreh. — This Ib a small
island on the extreme north-west of the Emsainhagh promontory ;
during low water it is joined to the mainland bj a bank of grayeUj
slungle, which also covers the greater part of the island — ^the rock is
porphyritic granite.

*' Freaghillauny Rush Island, and Inishbigger, — These islands lie in
the entrance to and west of the Moyrus Boat Harbour. The south
and south-west shores of Freaghillaun are covered with large rect-
angular blocks of an even-grained granite, with black mica, similar to
the rock about Mall Head.

'< SL Maedara^s Island^ or as it was anciently called Croach Mic
Dara, t. 0. ' Maedara^s stack or rick* — The shape of the island may
partially be due to ice erosion, aa many of the rocks have the appear-
ance of being ice-dressed. However its slopes, more specially those
northward and southward, appear in a great measure to be due to the
structure of the rock, as it is inclined to spHt off nearly everywhere
in plates a few feet thick. This remarkable weathering can be well
examined at the south-west of the island, where the waves of the
Atlantic are yearly quarrying largely, and hurling up the blocks above
high-water mark, thereby forming a beach of huge blocks, and one of
considerable size that was measured, gave 21 by 21 by 2 feet as its
dimensions.

'' Mason Island lies a little east-south-east of St. Macdara's. On
the west there are numerous angular blocks and boulders scattered
along the shore, while on the east the rock is covered with sand.

'' Wherron, Avery ^ and Ardnacross Islands, — The two former lie
north-west, and the latter east of Mason Island, from which it is
separated by a creek, which is fordable during low water, while all
are nearly covered by the tide at high water.

^^Mweenish Island^ Inishtroghen^ and Tidal Hocks. — ^Mweenish
bounds Ards Bay on the south, and is connected with the mainland
by a pass, that is fordable at half-tide.'*

(p. 38.) — " Duck Island lies about half a mile south of Mweenish,
and its subjacent rocks are similar to those on the south part of that
island.

*' Finish Island. — This island bonnds Mweenish Bay, on the south-
east, and is connected to the mainland at low water by a strand. A
mile south of this island is Inishmuskerry.

** Birmore and Birhsg Inlands and Tidal Hocks. — ^Birbeg is situated
about a mile south-south-west of Ardmore Point, and Birmore south
of the latter, the two being connected during low water."

O'Bbillt— On the Waste of the Coast of Ireland, 8fc. 188

(p. 50.) — " Inieherky Dinish, Furnace, lUauneoihin, and IHaunanar-
roor. — These are tidal, being joined dining low water."

(p. 66.) — ** Zettermore Island, with IHaunroe, and Inehagham. —
Lettermore Island, consists of two hills, one to the east and the other
to the west, with a flat bog between them. To the north, at Cashla
point, the nnweathered veins stand up 2 inches above the surface of
the Yock. A little east of Cashla Point is a north, 10^ east, dyke of
quartziferous porphyry ; and a little farther east, into which a small
bey has been cut by the action of the sea, is a course of rotten granite,
running nearly north and south. Hereabouts the unweathered portions
stand 3 inches above the mass, while inland south of this and due east
of the trigonometrical point, A 864, they are only one inch high.

'' Iniehlay and Inehmakinna. — These islands lie south-east of Letter-
more in Fearmore Bay. The former is joined by a bank to Gk>rumna
during low water, while the latter lies near the east shore.

** Annaghoaan, InMtravin, Illaunakirla, Beaghy, lUaunard, north
island, and the adjacent carrigs {rocks) and cairigeens {smaU rocks and
half tide rocks). These form a small archipelago at the junction of
Fearmore, Camus, and Kilkieran Bays."

(p. 57.) — '' Of KinneUy Islands, Mr. Cruise says — these are tidal
islands, and are situated about a mile and a quarter due west of Inish-
traven."

(p. 71.) — ^* Here it may be mentioned that a register of the amount
of weathering of some of the granites, since the ice disappeared from
this country, would seem to be recorded by the veins which traverse
these kinds of rocks ; as these veins are usually unweathered, retain-
ing their glaciated surfaces, and stand up above the mass of the rock ;
near the coast being usually from 2 to 3 inches high, while more inland
they only average 1*5 inches in height. This weathering would
seemingly also suggest that in the neighbourhood of the sea, the
atmospheric influences are different to those inland, not only in the
amount of work done, as shown by the greater height of the veins
near the sea, but also in respect to the colouring matter, in some of
the rocks, for, as previously mentioned, the purplish gray or greenish
f elstones near the sea weather red, while the same rock inland weathers
a dull yellowish white."

(p. 71.) — " On the sea-coast, about a quarter of a mile south-east
of Foal Island, there is a remarkable kitchen-midden about 50 feet in
diameter, 15 feet in height, and forming a flat-topped conical hillock.
It seems to be nearly altogether formed of the shells of the Patella
oulgata and the Littorina littorea; no excavations was made into it.

184 Proceedings of the Boyal Irish Academy.

There seemB to be added to it yearly a few more shells, by the people
who visit the site of a church, and two holy wells dedicated to St.
Columbkill, which are in its vicinity."

Memoir Geolog. Survey of Ireland (Sheets 105, and part of 114),
1869. (North of Galway Bay.)

(p. 41.) — ** In places along the shore of (Jalway Bay, peat with the
roots of trees is found below high-water mark ; this might not prove
that the land has sunk, for, at the present day, about two miles west
of Gbdway, between Blackrock and Blake's Hill, is a morass below
high-wat^ mark, in which peat and shrubs are growing. This morass
is divided from the sea by a barrier of shingle. However, against this
theory, we find in a half tide bog west of Blake's Hill, an oak stem,
12 feet long and 2 feet in diameter, immediately above the * corker *
or butt. This tree could scarcely have grown on ground below sea
level ; moreover on the Aran Ishmds at tiie mouth of the bay, there
are proofe of the islands having sunk since they were first inhabited.'^
{8ee Mmoir Sheet 113.)

Memoir Geolog. Survey of Ireland (Sheets, 114, 122, and 123)
(1868). The sheets contain the north-western extremity of the
County Glare, lying on the south side of Galway Bay, and the island
of Inishere, the smallest and most easterly of the south Isles of
Aran. They include the coast line from the southern shore of Liscan-
nor Bay to the Head of Ghdway Bay.

(p. 5.) — '^ To the north of Liscannor Bay, in the promontory of
Hag's Head, the ground rises to heights of 500 and 600 feet, especially
along the coast, which exhibits a line of magnificent precipices nearly
three miles long, and rising in one part, quite perpendicularly, to a
height of 668 feet ; these form the well-known cliffs of Moher (see

fig. 1)."

(p. 11.) — **Much of this removal has been caused by the wearing
action of the sea, when the land stood at a lower level. The escarp-
ment which runs round the foot of the limestone hills, is as much Hke
an old sea-coast on the east side as on the west where the sea is still
beating on it. The action of the sea on the high Coal Measure land
may be well observed still going on at Hag's Head and the cliffs of
Moher, the waves eating away the lower part of the cliff, and constantly
causing fragments of the upper part to fall for want of support. This

O'Rbillt— On the JFaate of the Coast of Ireland, 8fc. 185

action is conBiderably aasiflted by the great yertioal joints, which tra-
Terang the rock, diyide it into blocks, rendering the work of destruction
a far easier matter than it would otherwise be. The best instance of
this is at Ailleenasharragh, at the olifPs of Moher. A steep and wind-
ing pathway leads the explorer to the foot of this magnificent cliff,
and the most casual observer cannot fail of being struck by the
immense accumulation of Hhrie which forms a talus on the beach,
huge masses of grit, shale, and flagstone lying piled together in
wild confusion. Here the cliffs are constantly decreasing in alti-
tude, inasmuch as the ground slopes inland from the coast ; where-
ever on the other hand, the slope of the surface is seaward, the height
of the diff is increasing."

(p. 12.) — *' There is a tradition among some of the peasantry that
at one period Hag's Head was connected with the southern shores of
liscannor Bay by dry land, and that about midway stood the church of
8t. Seoithsen ; that by means of an earthquake, land and church suddenly
disappeared ; and on clear days, when the sea is calm^ it is said that
the ruins of the church are sometimes visible at the bottom. (Note — I am
indebted for this tradition to the late Professor O* Curry. It is alluded
to in the ' Annals of the Four Masters,' translated by him. — ^F. J. F.)."

(p. 19.) — *< The Coast section. — The almost continuous section
along the coast exhibits the structure of the whole district."

Memoir Geological Survey of Ireland (Sheets 115 and 116), 1865.
(No available detedls given.)

Memoir Geological Survey of Ireland (Sheets 131 and 132), 1860.
The district described forms part of the western side of the County
Clare. (No available details given.)

Memoir Geological Survey of Ireland (Sheets 140 and 141), 1860.
The district described includes the south-west part of the County
Clare, lying north of the mouth of the Shannon, and a small part of
the northern comer of the County Kerry, on the south of that river.

(p. 5.) — *' This part of the County Clare is an undulating tracts
stretdiing away westward in a long narrow promontory, the termi-
nation of which is Loop Head. The length of the promontory from
Poulnaaherry Bay to Loop Head is 16 miles in a direction about

186 ProeeedingM of the Royal Irisk Academy.

west 30^ south. Its greatest breadth (a little west of PouhiasheiTy
Baj) is fiye miles, from which pcnnt it Taries considerablj towaids
the west on aocount of the iiTeg;iilar form of its southern shore.
Thus at Garrigaholt it is nearly three miles broad, at Eilbaha, 1^,
while approaching Loop Head it rapidly contracts, so that half a mile
before reaching that point its breadUi is little more than \ mile. It is
bounded by precipitous difb, which in many places assume fantastic
forms, resulting from the action of the sea on the rocks. On the
north-west shore these cliffs attain in some places to an elevation of
200 feet perpendicularly aboye the sea ; but along the shore of the
Shannon they do not exceed 100 feet."

(p. 6.) — '' The shore of the Shannon, south-east of Eilrush, is
yery yaried in form. In some places the ground terminates abruptly in
cli^ ranging from 40 to 100 feet in height, while at others it slopes
almost imperceptibly towards the river."

" The width of ^e Shannon between the shores of Glare and Kerry
in this map varies from two to three miles."

(p. 13.) — "At the north side of the Loop, the cliffs are 200 feet
high. Dermot and Graine's Rock separated from the mainland by a
chasm 95 feet in width at the top, forms a striking object. (Fig. 4)."

Memoir Geological Survey of Ireland (Sheet 142), 1860. That
part of the Sheet which lies north of the Biver Shannon belongs to the
County Clare. At the south-west comer of the Sheet there is a small
portion of the County Kerry. At the south-west comer of the Sheet
there is a small portion of the County Kerry, in which is the little
town of Tarbert. (No available details given.)

Memoir Geological Survey of Ireland (Sheet 143), 1860. The
Biver Shannon runs with a general bearing east and west through
the northem part of the district described. (No available details
given.)

Memoir Geological Survey of Ireland (Sheets 150 and 151), 1859.
The principal features of the district described are, the promontory of
Kerry Head on the west, rising to the height of 700 feet. The
promontory of Kerry Head may be described as a regularly formed
hill, upwards of twelve miles long, from its westem extremity to the

O'Ebillt— On the Waste of the Coast of Ireland, Sfc. 187

p<ymt where it sinkB eastward into the plain. Its highest points are
Tiisk and Maolin Mountains, which are upwards of 700 feet above
the sea. From these points the ground slopes away yery gently
towards the sea on all sides except the east, terminating in rugged
cli&, which attain in some places, to a height of 200 feet, but
generally yary from 50 to 100 feet.

(p. 10.) — " The action of the water on the north coast of Kerry
Head has caused the cliffs to assume yarious curious forms in many
places, such as cares and natural arches. Near Ballingarry Island
are some good examples, as also at lUaunamuck. (Fig. 1)."

Memoir Geological Survey of Ireland (Sheets 140, 161, 171,
and part of 172), 1863. That promontory of Kerry which stretches
on the north side of Dingle Bay, and south-west of the Bay
of Tralee, happens to be divided among four of the Sheets, but can
obviously be only described as one district. The general form of the
ground is that of a broken ridge, traversed by several large valleys
and ending westward in the precipitous islets and rocks known as the
Bhiskets.

(p. 8.) — ''This north and south ridge of Mount Brandon rises
gradually from the sea near the town of Dingle, till, in the course of
2 or 3 ndles, it attains an altitude of about 2000 feet. Still further
north it rises to 2764 feet, in Brandon Peak, where it has in some
places so narrow a crest that a man may sit down astride of it. A
mile further north it reaches with a broader crest, the extreme
altitude of 3127 feet, at the point called Brandon Hill or Mount
Brandon. From this point it declines towards the north, but still
maintrftiiiH an altitude of 2500 feet to within frds of a mile of the
sea, and terminates in the grand difls of Brandon Head. The north
and south ridge of Brandon Mountains looks to the west over much
low ground, which is indented by Dingle and Yentry Harbours on the
south, and by Smerwick Harbour on the north. Between these there
runs an undulating ridge, which rises in some points to the height of
900 or 1000 feet high. This is within a mile of the sea-shore, and
the promontory terminates in Slea Head, which is 766 feet high, and
the lower Dimmore Head, which is the most western point of the
mainland. (See fig. 3, p. 10)."

(p. 9.) — " From the cliffs of Brandon Head, a line of similar but
lower cliffs is continued along the north coast of the peninsula, broken
only by the entrance to Smerwick Harbour. One summit of these

188 Proceedings of the Royal Irieh Academy,

cliffs, near the old signal tower of Sybil Point, almost hangs oyer the
sea from a height of 688 feet. Although not so peipendicnlar, yet
the increased height of those near Brandon Head, some of them
rising to over 1200 feet as steeply as their jagged and shattered
state will allow, makes them perhaps still grander objects. Many
of these cliffs consist of a mere heap of ruins, caused by great
landslips, huge crags of rock resting discordantly, one on the other,
with broken gullies and clefts between them. Standing on some of
the highest points of these cliffs, it is curious to mark what a
straight line their most striking features preserve along the coast
from near Brandon Head to Sybil Point for a distance of 12 miles,
and how these features re-appear in the same straight line 5 miles
beyond Sybil Point in the island called Inishtooskert, which rises
abruptly from the sea into a jagged peak 573 feet high. (Fig. 2.)

'* The central ridge of the promontory in like manner shows its
submarine continuation in the great Blasket Island, running off
irom Dunmore Head, and rising to height of 960 feet. There is an
almost absolutely perpendicular precipice of that height on the
north side, which keeps a height of 900 feet for a distance of about
a mile. Still farther out to sea, the Tearaght Island (see fig. 9,
p. 47) rises abruptly to 602 feet, the other Blasket Islands being
400 or 50 feet, and finally, the larger of the Foze rocks, 1 1 miles
from the mainland, juts up to 103 feet from water of twice that depth.
These islets, and the deep sounds and stormy straits between them,
give us, doubtless, a picture of what every part of the mountains of the
mainland have been in their turn in the successive stages of their last
alow elevation above the sea."

(p. 29.) — ** Brandon Head, — The wide indentation in the coast, to
the east of this point, extends to the western base of Knocknabrestee
Mountain (2500 feet high), which terminates at Brandon Head, and
here we find the Old Bed Sandstone concealed by a mass of rubbish
formed from itself, the result of an enormous landslip, which covers
the seaface of the cliffs for a width of 750 yards or nearly half a
mile."

(p. 45.)—** The Blaeket lelands.— The Great Blasket Island, which
lies at the distance of one mile to the west of Dunmore Head, is
three miles and three-quarters in length, with an average width of
half a nule."

(p. 49.)—" And here on the shore of Tralee Bay, at the distance
of about two miles east of Castlegregory, we find large roots and stems
of fir trees standing upright in the sand and shingle of the shore."

O'Rbillt— On the Waste of the Coast of Ireland, 8fc. 189

Memoir Geological Survey of Ireland (Sheet 162), 1859. The
desGription includes a part of Kerry, round the towns of Tralee and
Gastleisland. (No available details given.)

Memoir Geological Survey of Ireland (Sheet 173), 1861. The
whole of the area described lies in the County of Kerry.

(p. 7.) — ** The so-called harbour of Castlemain is choked with mud
flats and sand-banks, through which the rivers wind their way at low
water."

(p. 24.) — ''The local elevation of the land now occupied by the
Lower Lake of Killamey, and a large extent of its shores, in very
recent geological times, is not a mere supposition ; it is clearly demon-
strated by the fact, that we find some of the limestone bosses in the
pasture land of the southern part of Cahemane Demesne very much
water-worn at the base."

Memoir Geolog. Survey of Ireland (Sheets 182, 183, 190, and parts
of 172 and 191), 1861. The district described belongs entirely to the
County of Kerry, and forms the promontory of Iveragh and Dun-
kenon, between Dingle Bay and Kenmare Bay.

(p. 5.) — " The most northerly of the other ridges runs along the
south shore of Dingle Bay from Rossbehy to Valencia, rising at Dmng
Hill and Knockadober to heights of over 2000 feet. It makes Doulus
Head to the west, at which place it is interrupted by the sea forming
the entrance to Valencia Harbour, but afterwards re-appears in the
Island of Valencia, terminating at Bray Head, a cliff of 588 feet."

(p. 7.) — ^* The coast line of this district, in the neighbourhoods of
Eosbehy, Cahersiveen, and Aghadda, within the bays of BallinskeUigs
and Derrynane, and in some places along the Kenmare river, is low,
muddy, and sandy, or edged by a gravelly beach ; but in other and
mure exposed places it is high and rocky, or rises into abrupt preci-
pices, as near Hog's Head, at the seaward side of Scarriff Island,
between Bolus Head and Ballynablona, at Puffin Island, and from
thence to Portmagee, where there are some cliffs so high as 867 feet,
round the western shores of Valencia Island, from Bray Head to
Keenadrolaun Point (near the latter of which places, the Fogher cliffs
are about 700 feet high), on the west side of Beginish Island, and
from Doulus Head to beyond Kells' Bay."

190 Proceedings of the Royal Irish Academy.

(p. 13.)-—" Now if any of the islands near the shore be examined,
where faults do not exist, their bedded rocks will be found to resemble,
more or less, in dip and direction, those of the adjacent mainland ; and
even in the distant Skelligs the strike of the beds has the same general
direction as that of the rocks, of which the whole promontory is com-
posed. From this it appears that the rocks forming these islands were
once in continuation with those of the land, and are even now con-
nected with them by intermediate portions beneath the sea, some of
which have projections still above its surface, such as the Lemon
Eock, between the Skelligs and the shore, Beginish, with the adjacent
islands in Valencia Harbour and those lying between Scariff and
Deenish, at the northern entrance to Kenmare Kiver, and not far from
Lamb Head, near Derrynane. It was the gradual but ceaseless action
of the sea-breakers which cut off the islands from the mainland, and
it was a similar action of erosion, exerted upon the rocks now forming
this mountainous promontory, as they were gradually rising above the
level of the sea, which scooped out its valleys, and, taking advantage
of the numerous joints and master- joints found in all stratified rocks,
formed all the cliffs and principal features of the ground, which have
since that time been modified to some extent by atmospheric influences,
and by the glaciers which have left their marks in so many of the
glens.

'< The supposition that the wearing action of the sea is sufiiciently
powerful to have produced these results is much strengthened by con-
sidering the force with which this coast is assailed by the storm-waves
from the Atlantic Ocean. An examination of the shore-line will show
that they have produced cliffs of a bolder character, though not of so
great a height as some of those which occur inland ; while in some
instances they have undercut the hard rocks forming these cliffs, and
have removed portions of them, so that the rest overhangs the sea;
and in other places caves and long fissures have been worn beyond
the coast-line far into the land.

" (Note.) — As an illustration of the fury with which the breakers
act upon this coast it may be mentioned that during an autumnal gale
from the north-west I have seen the sea break clean over Lamb Island,
in Valencia Harbour, which has a height of 78 feet above low water,
and then runs down its eastern slope in sheets of foam and spray. It
is stated, too, that water-tanks, or butts, near the upper light-house,
on the Great SkeUig Rock, close to which, a height of 380 feet, is
marked upon the Ordnance 6-inch map, have been washed from their
places in the course of severe gales ; and that the Horse Island, at the

O'Bbilly— On the Waste of the Coast of Ireland^ 8fe. 191

west side of Ballinskelligs Bay, has not been very long separated
from land. A little to the north of this island, the shore of the bay,
there composed of ' drift,' has within the historic period been so much
worn and carried away by the sea, that the foundation has been
washed from under part of the ruined abbey of Ballinskelligs, which
was probably built at some distance from the water' s-edge, and the
skeletons of people buried in the adjoining grave-yard exposed. —
A. B. W."

'' The Ancient and Present State of the County Kerry." Charles
Smith, Dublin, 1756.

(p. 102.) — <' The sea towards the bottom of Ballinaskelligs Bay is
making great devastations and encroaching on the land every winter.
The cliffs are very high, but are unable to resist the fury of the ocean,
as they are only formed of difPerent strata of clay."

(p. 103.) — <* At Ballinaskelligs are to be seen the ruins of a very
ancient abbey or friary of the Order of Augustine Canons: it was
formerly removed hither from the island called the Great Skelligs,
where there was a monastery, consisting of several cells, dedicated to
St. Michael the Archangel, and is mentioned by Giraldus Cambrensis.
The time of its foundation is not known, but it must have been of
great antiquity, probably as early as the sixth century. The < Annals
of the Abbey of Innisf alien,' in Lough Lane, in this county, say that
Flann M'Callach, abbot of 8k$U%gy died in the year 885. At what
time the monks quitted the island is uncertain, but by the large traces
of ruined buildings, which the sea is continually demolishing, it
appears that this abbey had been formerly a very large edifice.

*' There are some traces of a town still remaining, besides a small
castle, built formerly on an isthmus to defend the harbour against
pirates, who had done considerable mischief hereabouts."

(p. 187.) — '* Between the Harbour of Smerwick and Ferriter's
Creek, the land lies low, and hath been much covered with sand
by the sea and wind of late years. This isthmus is hardly a mile
broad, growing narrower every winter, and will probably become
an island.

"The Great Blasket Island, opposite to this place, is said by tradi-
tion to have been formerly joined to the continent, and the country
people show an old ditch, which they say points to an opposite one at
Bunmore. The sound between that island and the mainland is of
great depth, and the currents of both ebb and flood set through it
with prodigious force and rapidity."

B.I.A. PROC, VOL. XXIV., SBC. B.] Q

192 Proceedings of the Royal Irish Academy.

(p. 208.) — '' Fenit Island, — Towards the north point of Fenit are
BevCTal sunken rocks, and also one above water called the Boss. This
island at low water is contiguous to the mainland ; but in all former
charts and maps is placed at a great distance from the shore.

^^BaUyhsigh Bay and Strand. — The land towards the bottom of
this bay is very flat, soft, and boggy, and hath no other defence but
the above-mentioned sandbanks from the fury of the ocean, which
almost every winter breaks through them in many places ; and,
therefore, a considerable tract of this part of the country will in a few
years (from 1756) be probably overflown. The neighbouring inhabi-
tants show rocks visible in this bay only at low water, which they
say are the remains of an island that was formerly the burial-place
of the family of Cantillon, who were the ancient proprietors of
Ballyheigh.

'' It is easy to see that if the land were depressed some 2400 feet,
the sea would then surround the mountain tops, transforming them in
time into islands like the Skelligs, and that if, as the ground arose from
the sea, the elevating action occasionally ceased, or went on very
slowly, the mountain cliffs and steeper declivities would be formed
by an action precisely similar to that which is acting on the present
eoast."

(p. 18.) — *< Purple grits and slates, with a general strike of about
€ast 25° (?) are seen on the mountain slopes on both sides of the road,
from Cahereiveen to Goomnahincha and Coonanna Harbour ; and the
continuation of the ridge from this to Doulus Head exposes along the
line of sea-clifPs to the north and round Doulus Head to Laght Point,
■a series of remarkable contortions in reddish purple girths, amongst
which are some slate beds. Some of these contortions are seen in Uie
annexed sketch of Doulus Head from the west. A north and south
fissure crosses the headland where the figures 10 to 70 are engraved
upon the map between the points marked 355 and 921. At the south
end of this, and running inward along it, is the Doulus Lane. Due
west of Glanlean (in Valencia Island) is the hill called Geokaun,
rising to the height of 888 feet above the sea, and presenting to the
north-west the bold sea-cliffs of Fogher, which are some of the
finest of the kind in the whole district, being nearly 700 feet in
height."

(p. 22). — ^* In the cliff to the south of Dromgour, which is the
highest sea-cliff in this district, being 867 feet above the level of the

O'Rbillt— 0» the WasU of the Coast of Ireland, 8fc. 193

the sea, there is a mass of rock that looks like a greenstone dyke.
Puffin's Island is chiefly composed of pui^le slates, which are well
exhihited in the cliffs all round it.'*

(p. 33.) — '< The Borlace of the country round Sneem is stated to
have been once a smooth turf bog. Subsequently to the deposition of
the drift and accumulation of the bogs, two actions, one of elevation,
and the other of depression, seem to have occurred in parts of the
district, if not over the whole of it. Owing to the latter action, the
sea at Beenagappal, in Valencia Harbour, flows at every tide over
part of the bog which so extensively covers the mainland. This
could not have been formed in the situation it now occupies beneath
the sand at low-water mark, but must have grown above high« water
mark, thereby showing a depression of at least 25 feet to have taken
place. A portion of the boggy flat at the north-west comer of the
map, as has been already stated, is laid under water by the higher
tides."

Memoir Geological Survey of Ireland (Sheets 197 and 198 and
south-east part of 191). The district described is the termination of
the promontory between Bantry and Kenmare Bays, from Kilmakiloge
Harbour, and the eastern part of Bear Island to Dursey Head.

The mountainous ridge that separates Bantry Bay from Kenmare
Bay (or river, as it is often called), has its loftiest eminences to the
east of this district; its crest, however, still retains an altitude
of over 1900 feet, south of Kilmakiloge Harbour, forming a rather
flat-topped ridge, from which proceed broken lateral spurs and deep
valleys, with sides that show many cliffs and precipices of bare rock.
About one and a half miles north of Castletown Bearhaven the crest
of the ridge sinks rather suddenly down to a level of 300 feet above
the sea, forming an open pass between the hills already spoken of and
the MirVifth and Knockoura Mountains, which rise to 1272 feet and
1610 feet respectively. The ridge is then continued to the west,
gradually sinking down to Dursey Sound, which is another pass (the
floor of which is now below the level of the sea), between the main-
land and Dursey Island, the summit of which is 825 feet above the
sea.

(p. 6.) — " It would be difficult to describe and almost impossible
to exaggerate the picturesque beauty of much of this high rocky
ground, commanding views over Bantry Bay, on the south, and the
still more lovely, Eenmare Biver, to the north, backed by the Kerry

194 Proceedings of the Royal Irish Academy.

mountains, each bay spreading out into the broad expanse of the
Atlantic towards the west.

" Although the cliffs round Dursey Island and Ballydonegan and
Coulagh Bays are often lofty, and the land above them mountainous,
they are not generally so precipitous as those which stretch eastward
from Blackball Head along the north shore of Bantry Bay, or from
Kilcatherine point, along the south shore of Kenmare Bay. There
are black jagged cliffs, often quite perpendicular for 300 or 400 feet.
They, are, however, more broken into than the former by narrow
passages, giving admission to sheltered harbours, instead of open bays.
The beautiful harbours of Ardgroom and Kilmakiloge, in Kenmare
Bay, are intances of this, and a still more striking one is Bearhaven,
lying between Bear Island, and the main."

Memoir Geolog. Survey of Ireland (Sheets 200, 203, 204, and
205, and part of 199). The district described belongs wholly to the
county Cork. Cape Clear and Mizen Head, the two southern promon-
tories of Ireland, are comprised in it.

(p. 6.) — '' As a subordinate feature of the main central ridge may
be mentioned the small ridge which runs on each side of Skull Harbour
and from thence to Toormore Bay. To the south- west of Toormore
Bay it is again met with, and continues out south-west to Mizen
Head, where it forms a bold cliff nearly perpendicular and over
300 feet in height

'* The islands that fringe the coast on the south lie in lines parallel
to the hills just mentioned, showing that they also are the summits
of ridges which are partly submerged, the islands answering to the
peaks, and the straits to the longitudinal and transverse valleys. The
most northerly of these partially submerged ridges is that forming
House Island, Castle Island, Long Island, Goat Island, and the part
of mainland that lies to the south of the beautiful land-locked harbour
of Crookhaven, and is only prevented from being an island by a sand-
flat between the head of Crookhaven and Burley Cove.

'' Another forms InisodriscoU and the Calves ; and on the line of the
most southerly is Sherkin and Clear Island and the Fastness Rock.
The Fastness Bock is remarkable, not only for being the most south-
erly portion of Ireland, but also for its aspect, as it rises, with nearly
perpendicular sides to a height of 97 feet from the water, with not
much more than room for the base of the lighthouse that stands on
it."

O'Bbilly— On the Waste of the Coast of Ireland, 8fc. 195

(p. 7.) — '' The indentations into the land along the south coast,
especially those that form harbours, often lie in the lines of the
previously mentioned transverse valleys. Four miles on the east of
Mizen Head, however, is the long narrow bay called Crookhaven,
wMoh coincides with a longitudinal valley. The bay called Skull
Harbour, which lies a few miles further east, is partly sheltered from
the swell of the Atlantic by the islands off the coast. Koariugwater Bay,
still further eastward, runs for more than two miles up into the land.

''About 3 miles on the east of Baltimore is Lough Hyne, a
picturesque salt-water lake, out of and into which the tide ebbs and
flows with tremendous force, as the entrance is very narrow
compared with the capacity of the Lough. Further north-east are
the long narrow harbours called Castlehaven, Glandore, and the Bay
of BoBscarbery. This last-named is gradually filling up with sand
and slob. (Note.) — ^When Smith wrote his * History of the County
Cork,' it was a tradition that, ' the harbour of Rosscarbery was
formerly navigable for ships.' "

(p. 18.) — "A little on the west of the north harbour (in Glare
Island) there is the ruin of a castle called Doonanore, which shows
the encroachment of the sea during the historical ages. Smith, in
his 'History of Cork,' written in the year 1750, vol. i., p. 286,
mentions that ' there is a very narrow passage about a yard broad,
and 10 yards in length, leading to the castle.' Now (I860) this
passage is nearly all gone, and only half of the castle remains, as the
other half was carried away with the rock on which it was built.''

(p. 19.) — "There are small Muoial flats along some of the
rivers. Under the slob on the east of the town of Rosscarbery,
Mr. DuNoyer has noted that they dig peat bog for fuel, and also at
the head of Tralong Bay, which Lies on the coast 2 miles south-
west of Bosscarbery. This shows that the land here must have sunk,
besides which we have a well-marked record of the encroachment
of the sea during the historical period of the old castle of Doneen,
which was built on a small island in Castlebay, a mile and a quarter
south of Bosscarbery. Half of the castle and the rock on which it

I built have been gradually carried away by the sea."

Memoir Geological Survey of Ireland (Sheets 192 and part of
Sheet 199), (1864).

(p. 5.) — " The country described comprises the ground round the
head of Bantry Bay, and a large part of the mountain ground

196 Proceedings of the Boyal Irish Academy.

between Bantry Bay and Xenmaie Bay on the north, and the rocky
promontory between Bantry Bay and Dnnmanoa Bay on the south.
The gronnd round Bantry Bay belongs wholly to Uie County Cork,
while that on the Kenmare side Uea in the County Kerry.

" The south-west comer of Ireland presents several high rugged
promontories running out to the west-south-west, terminating in
precipitous headlands with rocky islands and islets, each promontory
separated from the other by a picturesque bay, running far into the
land.

'* Dunmanus, Bantry, and Eenmare Bays are the three which are
most regular in general form, while they are at the same time the
most beautiful, on account of the loftiness and varied outlines of the
ground intervening between them. Dunmanus Bay opens between
Three Castle Head and Sheep's Head, and runs in about 13 or 14
miles, with a width never exceeding 2 nules. The opening of
Bantry Bay lies between Sheep's Head and Bear Island, from which
it runs about 20 miles into the land, with an average width of 3 miles.

*' The promontory between Dunmannus Bay and Bantry Bay varies
from 2 to 2^ miles in width, and rises in rocky lidges to a height
of about 1000 feet above the sea.

''The promontory between Bantry Bay and Eenmare Bay is
about 10 miles wide, and is much loftier and more rugged than that
to the south of Bantry Bay."

G. Smith, "History of the County Cork," vol i., p. 286 :—

" On the north side (of Cape Clear Island) stands the ruins of a castle
(buOt on a rock in the sea) called Dunanore — (' The (Golden Fort.')
There is a very naitow passage, about 1 yard broad and 10 yards in
length to this castle. This path is high and steep on both sides, the
sea on either hand being very deep, so that few but persons well used
to it will venture to walk it over."

(p. 269.}— "jBM*«iri«y. — The harbour, according to Camden, was
formerly navigable for ships, but in his time (1586) it was quite choked
up with sand ; and it is now so shallow that no vessel can come up to
the town."

(p. 274.) — " On the west of Glendore Bay, not far from the Cape,
by tiie working of the sea, a large portion of the hill fell down, on
which grew several trees ; this piece formed an island olabout twenty
yards in circumference, and the trees continued to grow, but it is now
(1750) almost quite washed away."

(p. 111.) — ''Near Ring several large horns were dug up in this

O'Rbilly— On the Waste of the Coast of Ireland, 8fc. 197

strand (which belonged to the Moose deer) by Mr. EUtyman, near
Youghal. This strand to the land is terminated by a large extended
bay by which was continued, before it was encroached upon by the
ocean, a great way beyond the lowest ebb.

*' Clay Castle, — On this strand is a yery bold sudden rising ground
or rather small promontory composed of loose, sandy clay, which had
also been encroached upon by the sea yery considerably within these
few years (1745-50). This hill stands about a mile south-west from
the town of Youghal."

(p. 256.) — " Near Dunworley to the west the coast is all bold,
high shore, abounding with stupendous cliffs, which astonish while
they please us.

'' On most of this coast are great yariety of caves worked by the
sea ; these caves are generally the habitations of wild pigeons, gulls,
and other sea fowl, who live in the upper crevices, while porpoises,
seals, and other monsters of the deep have their abode below." --^

(p. 241.) — " Old Mead, — ^Four miles south of Slinsale in the barony
of Courcy 's is a promontory running far into the sea called * Old Head,^
A mile from its extremity is an ancient castle of the Lords Kinsale,
built from one side of the isthmus to the other, which defended all the
land towards the Head. The place was formerly called Daneearma^
and was an old seat of the Irish kings. The isthmus by the working
of the sea is quite penetrated through, so that there is a stupendous
arch under which a small boat may pass from one bay to the other."

(p. 254.) — '< Caurtmaesherry. — On both sides are prodigious high
difb towards the entrance to this bay, where eagles, hawks, and
herons build their nests."

Memoir, Geological Survey of Ireland (Sheets 194, 201, 202),
(1862). The district described includes the coast line from Galley
Head to the Old Head of Kinsale.

(p. 6.) — '^ Small transverse valleys, at right angles to the general
bearing of the ridges, may be noticed in the Clonakilty estuary, in
the fiat entering Dunworley Bay, in the mouth of Kinsale Harbour,
and in the little deep blind cove, that runs for a mile into the high
land just to the west of the mouth. It is indeed but a further
extension of the same feature which produces the indentations of
Clonakilty Bay and Gourtmacsherry Bay, separating the once con-
tinuous ridge that stretches across these bays into the promontories of
Galley Head, the Seven Heads, and the Old Head of Einsale. The

198 Proceedings of the JBoyal Irish Academy.

extremity of the Old Head of Kinsale, which rises to 139 feet above
the sea, is begun to be cut ofP from the land by a subterranean sea
passage, through which the light can be seen from each of the two
indentations of the land, which are hence called ' Holeopen Bays.*

'' Oalley Head is in like manner nearly cut off from the mainland,
and formed into an island ; and this is, doubtless, one of the ways in
which the rocky islets with vertical cliffs, like those of the neigh-
bouring coasts, have been formed.

*'The Sovereign Islands, off the mouth of Oyster Haven, afford
examples of these, of one of which the following figure is a sketch
taken by Mr. DuNoyer during a gale of wind. (Fig. 1)."

(p. 27). — '' On the western side of Courtmacsherry Bay, as on so
many parts of the south coast of Ireland, a submerged bog is found at
dead low water of spring tides, and is then cut for turf by the neighbour-
ing farmers. How much further it may extend below the sea is, of
course, not known. There is a large bog at the back of Dunworley
Bay, the surface of which is but Uttle above the level of high water
mark, respecting which Mr. J. Good, of Dublin, informs me th*at a
rod was put down for a depth of more than 50 feet without reaching
the bottom. These, and other similar facts to be found round all the
coasts of Ireland, seem to point to a recent depression of the whole
island."

Memoir Geological Survey of Ireland (Sheets 187, 195, 196), (1864).
The district described includes the mouth of the Biver Lee, Cork
Harbour and the surrounding district.

(p. 5). — " The two principal longitudinal valleys may be called
here, the valley of Cork and the valley of Cloyne. The latter runs
across the district of Bally cotton Bay, to Ballinhassig, and beyond that
to Bandon and Dunmanway. The principal part of Cork Harbour
lies in it."

George Smith, " History of County Cork," vol. ii., p. 11 : —
'' In the latter end of March 830, Hugh Demdighe being monarch
of Ireland in this year, there happened such terrible shocks of thunder
and lightning that about 1000 persons were destroyed by it between
Corca-Bascoin and the sea-side. At the same time the sea broke
through its banks in a violent manner, and overflowed a considerable
tract of land. The island on the west coast of this country called
Innisfadda (a 'Long Island') was forced asunder and divided into thr^^

O'Bkilly— On the WoiU of the Cowt of Inland, 8fc. 199

parts. This island lies contigaous to two others, yiz. Hare Island
and Castle Island, which, lying in a range, and being low ground,
might have been very probably then rent by the ocean/'

Memoir, Geological Sonrey of Ireland (Sheets 188 and 189), (1861).
The district described includes some of southern ports of the counties
of Cork and Waterford, which lie on each side of Youghal.

Youghal Bay and the mouth of the Slackwater Biyer.

(p. 5.) — " The coast differs in different parts, excepting a line of
frequently inaccessible cliffs, where it runs across one of the longitu-
dinal ridges,and low sandy or marshy shores, where the valleys strike
out upon it. The cliffs rise to between 150 and 200 feet near Mine
Head, while at Ardmore Bay the coast consists of a gravelly sandy
beach, backed by low vertical banks of clay. Between Ardmore and
Ardrogma Heads it again presents cliffs, some of which are 190 feet
high ; these are broken at Whiting Bay into low rocky shores, with
sandy beaches, but appear again beyond it, and continue as fiir as the
mouth of Youghal Harbour. From Youghal Harbour a gradually
widening strand stretches to the south-west for four miles, along low
drffs d marly clay, which in one place at the end nearest to Youghal,
rise to a height of 90 feet, but further to the south-west sink down so
as to permit the tide to encroach considerably inland. Knockadoon
Headland again is edged by vertical cliffs witii heights of 130, 170,
and 200 feet, but south of Kilcredan, another strand commences where
the Cloyne Yalley comee out upon the coast."

Charles Smith's «< History of Cork " voL i. (1750) :--
(p. 109.) — <* The large extended strand of Youghal, as far as the
lowest ebbs uncover it, and probably much farther, is no other than a
conunon turf bog, covered over with sand and pebbles, from whence
not only good turf is dug every season, but also great quantities of
timber, trees, as fir, hasel, &c., are found. Some years ago a skeleton
of a m<mstrou8 animal was discovered in this strand ; I saw one of the
shoulder bones in Youghal; it is 3|^ feet long, and weighs about
1 CO lbs. The remainder of the skeleton, and (as I am informed)
another of the same kind, lie still buried in the strand. When they
were first discovered, it happened to be a very low ebb. These bones
lay in a turfy soil not far from the surface. They undoubtedly
belonged to some fish of the cetaceous fainily, which seems the more

K«IWL FBOa, VOL. XXIT., SBC. B.] ^

200 Proceedings of the Boyai Irish Academy.

piolMible from their beiiig thick, short, and pondeioiis ; and not to an
elephant or land animal, as was conjectured bj those who discoyered
thenu About eighteen years ago (1760 - 18 = 1732), this strand was
entirelj diyested of all its sand and gxayel, and being left quite bare
b J violent high winds, great quantities of roots of yarious trees then
laj exposed to yiew. The sea has greatly encroached on this part of
the coast, and is likely to gain more ground, as the land within the
strand lies low and flat. At the entrance of the harbour of Youghal
may be seen the remains of the foundation of a mill standing on a
rock, which shows that the ocean has greatly exceeded its limits on
this shore."

(p. 343.) — '' About a mile east of Doneraile is Castle Saffron, so
called from the la^e quantities of it formerly planted h€*re, being
greatly used by the Irish for dyeing their shirts, ftc"

Honoir, Geological Sunrey of Ireland (Sheets 167, 168, 178, and
179) (1865). The surface described includes a portion of the County
Wexford, ending in Hook promontory.

(p. 8.)— << The entire coast westward from Credan Head (202 feet),
past Brownstown Head (158), is quite precipitous, and it presents the
same aspect for the distance of fifteen miles westerly, from Kewtown
Head (110 feet) to Ballyyoyle Head, where it is 243 feet in elevation.
Helyick Head, at the south side of Dungarvan Bay, is 255 feet above
the sea, and its shores are also abrupt. Broken ground running
easterly from Dnnbrody (in the County Wexford) attains a height of
257 feet ; further south, the country, tiiGugh rarely level, attains here
and there to elevations of 250 feet, and on the east coast of Waterford
estuary to the south of Buncannon, the cliffs reach a height of 128
feet, and at Broomhill Bay 208 feet above the sea. On the east coast
of Hook promontory at Houseland the clifb are, some of them, over
140 feet in height, but to the south of this the promontory becomes
lower, with an average height of about 40 feet reaching to 60 feet at
the cliffs between the village of Slade and the extremity of Hook
Head."

Kemoir, (Geological Survey of Ireland (Sheets 148 and 149), (1887).
The greater portion of the district described is situated in the County
Wexford; (No available details given.)

O'Rbillt— Of» the Waste of the Coast of Ireknidj Sfc. 201

Memoir, Geological Survey of Irelaiid (Sheets 158 and 159). The
area descrihed is a tract lying east and west in the County Wexford*
(No available details given.)

Memoir, Geological Survey of Ireland (Sheets 169, 170, 180, and
181), (1879). The area described is a portion of the county Weirford,
and fonuB the south-eastern extremity of Ireland.

(p. 5.) — *' Of Bannow there is a tradition that the ancient city was
buried in the sand; this, however, cannot be correct, as the sands are
only a few feet deep, and could only obliterate the foundations of the
houses.

** Off the south coast, are the Saltee and Eeragh Islands, with
various sea rocks, while south-east of Greenore, on the east coast, is the
Tuscar Bock and lighthouse,"

From the consideration of the foregoing series of extracts taken
from the Memoirs of the Geological Survey of Ireland, it may be re-
eognixed, that in one of them alone is the waste of the coast of
Ireland specifically alluded to and dealt with, that is, in that of the
CountieB Wicklow and Wexford, by Mr. George Kinahan. He not only
oonaiderB the data presented by the coast from that point of view, but
also gives measurements and details which would allow of further
waste and change being defined and measured as to extent and
character.

In his memoir dealing with the Aran Islands he also gives specific
details as to the size of the rocks moyed by the waves, and enters
into interesting details as regards the formation of the block beaches,
and their significance. As regards all the other memoirs, the question
of the waste or wear of the coast is not specifically considered, although
in many of them, the characters of the cliffs, chasms, caves, &c.,
resulting from the action of the ocean waves on the coast line, are
incidentally considered, and more or less fully detailed, but without
any attempt at measuring the action of the waves, or of generally
defining the outline of the coast by prominent points so as to furnish
to some extent points of comparison for future investigations. From
the whole of the remarks it can be clearly inferred that considerable
sad continnous waste of coast line is going on day by day, and that

202 Proceedings oj the Boyal Irish Academy.

the extent of the area of Ireland is being alowly reduced. It ia evident,
therefore, that unless a new and special survey of the entire coast of
Ireland be undertaken with a view of accurately determining its
present outlinei and of thus leaying a basis of observation to which
future changes of outline and further losses of ground may be referred,
and thus become capable of precise measurement, this yery important
question must remaininaperfectly undetermined state, andina condition
most unsatisfactory, not only from a scientific point of view, but also
from the larger and more important one, that of the administration of
the country. With the facilities afforded by photography at the
present time, it should be possible to so represent pictorially the
present outline and state of the coast as to render easy of definite de-
termination the future encroachments of the sea, and thus allow of
continuous and accurate observations being systematically carried oat
all round the island.

It would therefore be worthy of the Jtoyal Irish Academy to pio*
mote by its action and influence the undertaking of such a survey,
and to bring to bear the services of the Ordnance Survey in conjunc-
tion with those of the (Geological Survey officers, so as to allow of the
attainment of tiiat important objecti giving at the same time the
fullest importance to all local traditions or personal observations
beanng on the subject, as also to all the details to be gathered from
the public records now being so carefully examined and published by
Government, and by many public bodies and learned societies.

[ 203 ]

IV.

ON COMPOSITE GNEISSES IN BOTLAGH, WEST DONEGAL.

By GRENVILLE A. J. COLE, F.G.S., Professor of Geology

and Mineralogy in the Royal College of Science for Ireland.

(Plates I. to V.)

[Read Mat 26, 1902.]

I. iKTBODUCnOK.

The observations on which the present paper is based are in direct
continuation of those recorded in 1900/ and have been carried on with
the assistance of a grant made by the Royal Irish Academy for the
study of metamorphic and other rocks in the north-west of Ireland.
The barony of Boylagh (Baeighellach) includes the country between
the Owentocker and t^e Gweedore River, with the long sea-inlet of
the Gweebarra in the midst of it. New roads have been cut of late
years across some of the wilder portions, and the bridge made by the
Congested Distncts Board over the Gweebarra River has greatly facili-
tated communications. '
The earlier discussions' as to the nature of the granite masses
which play so important a part in Donegal may now be regarded as set
at rest. Dr. Callaway^ showed, in 1885, that the granitoid gneiss of
northern Donegal was an intrusive rock, and his observations were again
and again verified throughout the whole county during its detailed
examination by the Geological Survey, which resulted in a series
of maps and memoirs published between 1887 and 1891. The move-
ments that have undoubtedly left their impress on the granite in many
parts of the county of Donegal, as, for example, in the Bamesbeg
area, are in all probability of Caledonian age ;^ that is to say, they

' *' On Metamorphic RockB in Eastern Tyrone and Southern Donegal.'* Trans.
Royal Irish Acad., vol. xxxi. (part xL), pp. 431-470.

' See references in above paper, pp. 449-450, and the good Bibliography in
GtoL Surrey of Ireland, Mem. to sheets 3, 4, &c., p. 23.

' « On the Oranitio and Schistose Bocks of Northern Donegal." Quart. Joum.
G«dl. Soc. Ixmd., vol. zli., p. 221.

* The recent attempt of some English writers to lise the terms '* Caledonian, "
** Hercynian, " etc., so clearly defined by Bertrand and Suess to express the trend
ol folds in general, rather than folds of a particular epoch of mountain-building
tands to deprive European geology of a rery valuable piece of nomenclature.
a.i.A. raoc., tol. xxnr., sic. b.] S

204 Proceedings of the Royal Irish Academy.

are pre-Deyonian and post-Silurian ; and the trend of the great granite
masses across the country implies that these were intruded about the
time that the folding was in progress. If, then, in the present paper,
we are led to ascribe the foliation of the granite over a wide area
to original conditions of flow and intrusion, we must fairly recognise
that such flow took place under the influence of the Caledonian
processes of mountain-building.^ At yarious points the inyading
granite had doubtless become solid and resisting before the earth-
movements came to an end ; and here deformations and new structures
were set up, and must be attributed to true dynamic metamorphism.
In Boylagh, however, it appears that foliated gneisses arose during
the ordinary course of igneous intrusion, though their structures were
emphasised to some extent by subsequent pressure. The more pro-
found changes that we observe in the contact-zone in the field were
not brought about by mineralising emanations from the granite, nor
yet by molecular re-arrangements in the heated sedimentary rocks.
They were due, rather, to that bodily intermingling and incorporation
of two dissimilar masses, which results in the formation of eompo9iie
gneiss. The gneisses of Boylagh retain, then, to a considerable degree
the structures both of an igneous and a stratified material. This, at
any rate, is the argument of the present paper.

II. — The Dome op Abdaiia.

Our first observations may be made in the area cited by Mr. J. R.
£ilroe' as clearly exhibiting the effects of pressure in produtang
foliation in the granite. Mr. Kilroe, in his memoir, and in the
beautiful maps prepared by him in conjunction with his colleagues,'
shows how the foliation of the schists runs parallel to the curving
margin of the granite between Ardara and Clooney, and how similar
foliation occurs, also parallel to the margin, at certain points within

^ Dr. S. Haughton, forty years ago, came to a very similar conduaion regarding
the granite of Donegal, when he stated that in the centre it was probably igneous,
** deriving its cleayage-phmes and gneissose character from the pressure exercised
upon it by the stratified rock, which has been lifted, to the north and south, to a
nearly vertical position. '* ( '< Experimental Researches on the Granites of Ireland,*'
part iu., Quart. Joum. Geol Soc. London, vol. xviii, 1862, p. 406). The remark
is of importance, as showing that this author was not a convinced advocate of a
metamorphic origin for the granite as a whole.

s Memoir to maps of south-west Donegal (sheets 22, 23, 30» and 31), Geol.
Surv. Ireland (1891), pp. 28-30.

* GeoL Surr. Ireland, sheets 16 and 23.

Cole — On Composite Oneisses in Boylaghy West Donegal. 205

the granite mass. Eut the general trend of this foliation is also that
of the strike of the uptilted sedimentary schists. There is no mystery
here as to the origin of the schistose series, which admittedly consists
of shales, limestones, and sandstones, invaded by basic igneous rocks,
and subsequently metamorphosed by pressure. The dynamic action
has here and there induced a foliation oblique to the bedding, while
thrust-planes have allowed some blocks to move over others ; but in
most localities the original bedding is traceable, and there has been no
general rolling out of the complex mass into mylonitic schist or
gneiss.^ In the promontory, for instance, leading to Loughros Point,
west of Ardara, obvious deformations have gone on, with consequent
production of gametiferous mica-schist; but in many places the
stratification of shales and limestones is left perfectiy clear, with a
puckered foliation running obliquely through the shales. Even where
great quartz-veins come up along the surfaces of foliation, the original
bedding may still be seen, and the character of phyllites, rather than
of mica-schists, is preserved by the crumpled shales.

The folded layers, then, of this stratified '' Dalradian " series,
although inversions and repetitions had already taken place, furnished
the important structural surfaces of the distinct at the time of the
npwelling of the granite. West of Glenties, the igneous rock has
<M>me up laccolitically along one of these surfaces, forming a low
granite dome north of Ardara. On the south side the dip of the
Dalradian series is away from the elongated dome ; but on the west
and north it is towards tiie granite, and the schists are doubUess there
prolonged to some distance beneath the igneous rock. Trusklieve, in
Banny, forming the conspicuous mass north of the Gweebarra, with
its bare tabular granite overlying schist, and sending off intrusive veins
into this sedimentary substratum, is another example of an igneous
boas behaving as a laccolite on its margin. (Compare left-hand end
4>f figure on p. 225).

Along the margin of the granite dome which stretches north from
Ardara, foliation appears in patches in the granite. According to an
older view, this indicated that the granite had been formed by the
progressive metamorphism of stratified schists. On the dynamo*
metamorpbic view it indicated, on the other hand, a crushing and

> Mr. Kiiroe describes the considerable movements and dislocations in this series
m the Glenties szea (Mem. 6. W. Donegal, pp. 19-23). 1 attribute, however,
etftain features of the metamorphism described by him to the oontact-aotion and
mftmsion of the granite, working on a regional scale.

206 Proceedings of the Royal Irish Academy,

streaking out of the consolidated igneous rock. But it cannot fail to
be observed that the clearly foliated masses are formed of dark granite,
rich in elongated flakes of biotite. Dr. J. S. Hyland^ regarded thia
excess of biotite in the gneissic granite of Donegal as a secondary
feature, ''as its frequency increases with the intensity of the meta-
morphism." Mr. Kilroe,' however, from his field-observations, holds
that shearing was facilitated where the biotite originally was abun-
dant. He elsewhere remarks,' "where the mica abounded, as in the
more basic portions, the rock has yielded most freely " to the earth-
pressures. This presumes a previous process of differentiation in the
mass, whereby more basic knots had gathered here and there. In the
field, however, inclusions of schist, drawn out along the foliation-
planes of the granite, again and again accompany the ''metamorphosed''
granite. I doubt if they are absent from any of the points where the
foliation has been thought of sufficient importance to be recorded on
the maps of the Geological Survey. Near Eilgole, immediately north
of Ardara, these inclusions are obviously connected with and emphasise
the foliation. In Garvegort Glebe, again, nearer to Glenties, a few
rugged exposures behind the school-house show a well-developed biotite-
gneiss, induding similarly elongated fragments of mica-schist, some of
which contain tourmaline. The occuirence of this mineral completes
the resemblance of the inclusions near Ardara to those caught up in
the Caledonian granite near Ballyoorus, in County Dublin.

In both the above-mentioned cases of foliated granite between
Ardara and Glenties, the undoubted Dalradian series lies only a
hundred yards or so away upon the south. At Eilgole the contact-

1 Mem. to sheets 3, 4, 15, etc. (1891), p. 135.

« i«rf., p. 78.

> Mem. S.W. Donegal (1891), p. 29. Dr. Hanghtoa noticee that the gneiasoee
gnmite of Ardara contains black mica "in large quantity'* (Quart. Joum. Geol.
8oc. LondoD, toL xriii., 1862, p. 408). Mr. £. H. Scott, discuasing the analysia
of one of theae Ardara maaaea, which oontaina only 55*20 per cent, of ailica, aaya
that it 18, << properly apeaking, not a granite at all. " (Joum. Geol. Soc. Dublin,
vol. iz., 1862, p. 287 ; see also Dr. J. 8. Hyland, Survey Mem. S. W. Donegal,
p. 64). The analysiB* which waa also published in Haughton'a paper, shows us a
Toek alUed to Boaenbuach'a Minette-Kenantite aeriea, verging on keraantite, but
with 4*63 per cent, of soda and only 3*17 per oenL of potaah. Since soda ia not
more abundant than potash in ordinary phyllites and mica-schists, it seems highly
probable that this particular darkened and gneissoee granite of Ardara became
modified by the absorption of an epidiorite or hornblende-schist, as Is the case
with the gneiss of Cara and Lough Derg in southern Donegal, and the gnoiaste
granite of Derkbeg to be deacribed later.

OoLB — On Compmie Gfneisses in Boylagh^ West Donegal. 207

effect has produced a very fine-grained biotite-gneiss along the junction.
There is in the field no question whatever of basic segregations in the
granite, such as are so often relied on to explain variations in an
intrusive mass. The phenomena are those of intrusion along the
pre-existing foliation-planes of a schist; and a good gneissose rock
results, set with inclusion-flecks, in which these foliation-planes are
still apparent. Subsequent pressure may have intensified the eye-
etructures, and may have broken up some of the residual and included
layers of schist ; but this gneissic border to the granite is none the
less of composite origin, and its main characters remain due to the
circumstances of its original fiow.^

On the north side of the Ardara dome, towards Gweebarra Bay,
the same phenomena are repeated. TJncontaminated granite, like that
of Cam in southern Donegal, is seen in the veins that come up
through the Dalradian series at Portnoo. The specific gravity of
this beautiful white rock is 2*59. It consists almost entirely of
quarts, orthoclase, and microcline, with a mere trace of greenish
biotite. Muscovite has developed, however, as an alteration-product
in the felspars. Another type at Cashelgolan, east of Clooney, is a
pale granite with well-developed primary muscovite ; here the rook
intrudes into a schistose series.

The granite at Portnoo, on the shore on the east side of Dunmore
Head, cuts in numerous veins and dykes across a tough diorite, and
similarly invades the gray crystalline limestone of the headland. In
the former case littie absorption takes place, except on the margins of
some of the dykes, and the rock remains undarkened ; in the latter
ease the granite has detached small fragments of the limestone, and
has even entered along the fissures where the successive beds in a
small anticlinal have '' sprung " apart. But no appearance of inter-
lamination on a large scale has arisen, and the granite remains devoid
of foliation. Gray garnet and wollastonite, the latter in microsoopic
prisms, have arisen in the limestone, some layers of which have
become so rich in silicates as to be practically flinty. The variation
in the type of alteration in alternate layers is, however, due to original
^fferences in the composition of the strata, and not, as might be
supposed in the field, to parallel intrusions of the granitic magma.

> Compue the obaenratioiis of Lacroix on the gnnite with elongated inoluaioDi
io the vaUey of Boutadiol, etc., ** Le granite dee Pyrte6es et Bee ph^nomdnes de
contact," 2^ m^moire (1900), Bull. Carte g6oL de la France, No. 71, pp. 21
and 25.

208 Proceedings of the Royal Irish Academy,

Here and there an argillaceous layer occiirB, eonYerted almost wholly
into mica at its contact with the granite vein.'

Opposite Swan Mount, above the Portnoo Hotel, a granite appears
amid crumpled mica-schists, and is darkened, toughened, and altogether
modified by numerous inclusions of mica-schist and aphanite. There
10 not the slightest doubt as to the nature and origin of these inclu-
sions, and the rock resembles the similar instance at Castlewellan,
Go. Down. A large specimen, with inclusions in various stages of
absorption, has the characteristic specific gravity of 2*77.

Higher up, on Cashel Hill, veins of granite, with green mica,
penetrate a mica-aphanite ; and biotite-granite appears in force near
the summit of the hill. All this serves to correlate the Portnoo granite
with that of the main mass further east ; but it must be borne in mind
that pegmatitic veins cut all the rocks of this area, including the
foliated granite of the Ardara dome, and that the Portnoo granite may
possibly belong to this later series of intrusions. Some faulting has
occurred since the intrusion of the granite veins into limestone near
the road on the north side of Narin Hill,' and these veins may belong
to the older granite ; but the typical pegmatites occur, cutting across
the foliation of dark schists, as near at hand as Clooney, and also
freely throughout Ballyiriston. From the point of view of general
principles, however, the masses of pure and modified granite at Portnoo
are, of course, available, whatever their age, as links in the argument
concerning the composite gneisses in the dome of Ardara.

When we come east of Clooney, we are dealing with exposures on
the true north flank of the dome. The junction of the Ardara granite
and the schists is well displayed in a little quarry by the main road,
just east of Cashelgolan Hill. The mica-schist is here delicately
penetrated by sheets of muscovite-granite, which have been forced
along the almost vertical planes of foliation. This foliation is parallel
to that noticeable in the granite itself on the south side of the road.
The junction shows very various features. In one place the granite,
in which muscovite is the common mica, intrudes in delicate sheets

> Compare Lacioix on the alterations of calcareous strata in the Pyrenees, ** Le
granite des Pyr^nte, etc.," I*' mte. (1898), Bull. Carte g^. de la France, No.
64, p. 15.

* This is probably the locality near Naiin where Mr. E. H. Blake noted the
•difficulty of saying *' where the slate ends and the granite commences" (<< On the
primary rocks of Donegal, " Joum. GeoL Soc. Dublin, vol. ix. (1862;, p. 296).
The apparent passage bum the one roek into the other was reoogmsed early in
Donegal, as elsewhere in Europe.

Cole — On Campimte Cfneisses in Boylaghj West Donegal. 209

along tbe foliation-planes of the mica-achiat. The latter rock containa,
in addition to the naoal pale biaxial mica, a green-brown biotite,
probably developed as a product of the igneous contact.^ A specimen
measuring 7 cm. by 8 cm. by 2 cm., from west of the fann of Ard-
lougher, and formed of clearly defined sheets of muscovite-granite
and interlaminated mica-schist, in apparently equal proportions, has a
specific gravity of 2*74. Specimens in which far more subtle inter-
mingling has gone on have much the same density, despite the more
coarse development of their crystalline constituents. The crystalline
associations probably arise at an early stage, and the growth of larger
crystals is a process of rearrangement of the groups already formed.

The contact-zone occasionally shows a yellow-brown composite rook,
in which brown mica is abundant, but in which granitic characters on
tile whole predominate. This would bo styled by French authors a
Leptynolite.' The biotite, which is almost uniaxial, and which dis-
plays grey-brown to rich yellow-brown face-colours, does not occur in
the granite itself, nor is it the same variety as that in the adjacent
mica-schist. Muscovite is also present in the leptynolite, and separated
oat a little before the biotite ; the latter certainly does not represent
in this case mere patches of residual material derived from the mica-
schist. The complete graduation of this rock into the granite, and
also into various types of interlaminated composite gneiss, makes it
dear that it also is essentially a composite rock, in which absorption of
the schist and recrystallisation have occurred. A '' leptynolite,'' or
more precisely, a fine-grained " granitite " with oligodase and two
micas, and a specific gravity of 2*70, has resulted from an interming-
ling that must have amounted in this case to interfusion (PL iv., fig. 1).
This zone of leptynolite, as observed by myself east of Cashelgolan
ffill, is not more than 10 cm. thick ; but I see no reason to doubt that
similar effects may be produced elsewhere on a far more important
scale.

Those who, with Mr. P. D. Adams,* have urged that such leptyno-
Utes arise from progressive metamorphism of the constituents of a shale,

^ SalomoD, « Oeologiache Studien am Monte Ayiolo," Zeitsohr. d. deutach.
geol. OeseU., Bd. zUi. (1890), p. 471, describet the formation of a brown biotite
at the ezpenae of chlorite, in phyllites invaded by the tonalite of Monte AdameUo.

* Compare Lacitnx, cp, eit. (1898), p. 8, and Ma pi. I, figs. l-6,>nd W. Salomon,
" Bami de nomenclatme des rochet m^tamorphiquea de contact, " Congr^ g6ol.
iAtimaL, Gomptea rendua, ^iii* aeaaion (1901), p. 843.

' ** The excursion to the Pyreneea in connection with the 8th Intematioiial
Geological Congreaa, " Joomal of Geology, vol. iz. (1901), p. 44.

210 Proceedings of the Royal Irish Academy.

without any addition from the igneous rock, have cited chemical
analyses to prove the possibility of such a change ; but they seem to
me to overlook the evidence of the field itself, and also the interming-
ling traceable mth the microscope on the margin of inclusions and of
veins that look sharp enough to the unaided eye. When a passage
from an inclusion to the surrounding rock is clearly visible, the inclu-
sion is, as previously remarked, liable to be treated as a ''basic
segregation."^ But it is impossible to assert that the contact-schists in
such a case as that of CashelgolanHill are ''basic segregations" from
the granite. The latter rock has none the less intermingled its magma
intimately with their crystallizing materials. In such a case, the
microscope merely refines and carries further the conclusions forced
upon us in the field.

On the rising ground in Ballyiriston, south-east of this junction,
there are rapid variations in the constitution of the granite. Very
pure types, free from biotite, merge into darkened types with pink
felspars and nests of dark green biotite. (PI. in., fig. 1). The mode
of aggregation of this biotite at once suggests its foreign origin ; and
this is confirmed by the frequency of lumps of schist, streaked out
parallel to the east-and-west foliation, even as far south as half a
mile from the visible junction on the road. Here, then, on the north
side of the dome, the phenomena of Kilgole and Garvegort Glebe are
repeated ; the foliated granite is clearly of composite origin.

If any evidence were required, in addition to that which is so
obvious in the field, to show that the inclusions are not the oft-dted
" basic segregations," it would be found in the fact that veins run
from the surrounding granite into these inclusions, and take advantage
of pre-existing foliation-surfaces in them. A lump of muscovite-
biotite-epidote-rock in the granite south of Bonnyglen Lough is thus
penetrated by zigzag veins and tongues of biotite-granite. These

^ W. Salomon, op. eit., Zeitschr. d. deutach. geol. Gesell., Bd. zlii. (1890), pp.
476 and 493, describes certain large masses, resembling concretions, in the tonalite
of Monte Aviolo, as shading off into the igneous rock, but being none the less
inclusions from the adjacent schists. Great crystals of biotite and plagioclase, like
those of the tonalite, occur within them ; their outer portions have been melted,
and the tonalite-magma has flowed in along cracks, taking foreign constituentB
into itself and undergoing thus a chemical modification. 8alonion*s observationB
deserve quotation beside those of Lacroix, SoUaa, and others, whose conclusions
this author to some extent anticipates ; and they have more value from the fsct
that he elsewhere denies that any extensive modification of the tonalite-magma has
occurred through absorption of schist upon a large scale (*' Ueber Alter etc. der
periadriatischen Massen," Tsch. MittheU., Bd. xvii., 1898, p. 173).

Cole— 0» Composite OneisBea in Boylagh^ West Donegal, 211

caused a farther development of greemsh biotite in a thin contact-
zone along their junctions with the schist ; in fact, this biotite seems
to have been deposited as a first product of cooling from the veins
themselves, just as the augite in the dolerite veins traversing older
dolerite at Portrush has a tendency to gather on their margins. Then
the felspar, which is mostly orthoolase, formed a zone on each side,
leaving the quartz, which has separated last, to occupy a central band.
Some large crystals of biotite occur, set irregularly in the veins. This
little block, measuring some 7 cm. by 7 cm. by 6 cm., shows us that
many lumps of schist must have been altogether cut to pieces and lost
during the invasion by the igneous mass«

The dark microdine-granite in Ballyiriston shows under the micro-
scope the clustered groups and flecks of biotite, associated with epidote,
with which one soon becomes familiar along such intermingled contact-
zones (PI. in., fig. 1). A little sphene and green hornblende have
developed, indicating the approach to more basic types. The specific
gravity of this granite is slightly raised, and is here 2*69, that of
the normal granite of Boylagh, as tested from various localities, being
dose on 2*60. The pure microcline-granite is here, then, modified
towards quartz-diorite ; at the same time, many of the inclusions in
it become so commingled with matter from the granite as to pass
at their margins quite insensibly into the igneous mass.

III. — Cabbahx, veab Qlentiss.

A concrete example of the structure of Boylagh occurs on Carbane,
a hill 450 feet in height above the sea, and half a mile east of the
little town of Glenties.* The Dalradian series in ** the Rock," as the
rough street at the south end of the town is styled, consists of vertical
well-bedded quartzites and black micaceous shales. Along their strike,
as they swing round to the north-east, they show crumplings, and the
ahale-layers pass in a quarter of a mile into foliated and wrinkled
mica-schist. The pressure under which this change was brought
about is evidenced by the folding of the metamorphosed strata round
small eyes of amphibolite (epidiorite) ; these doubtless represent the
characteristic accompanying sheets of dark igneous rock, diorites and
aphanites, which have become broken up in the more yielding shales.
** The Rock " at Glenties is probably itself an eye on a large scale, a
patch of 8trata«that has escaped crumpling and deformation.

* This is the area referred to in the Survey Memoir to S.W. Donegal, pp. 29
and 63.

212 Proceedings of the Royal Itish Academy.

Carbane rises on the south-east of this metamorphosed area, and
consists of granite, which penetrates the shale. On the southern slope
of the hill the contacts are excellently seen. The granite is a pale
pink aplitic rock, consisting almost entirely of pink orthoclase, larger
and less decomposed colourless microcline, and quartz. Its specific
gravity is 2*59. I^ear any large inclusions of schist, and near the
junction generally, it becomes darkened and gneissose ; but eyen the
uncontaminated specimens show a delicate foliation, especially on their
weathered surfaces. Specimens intermediate in structure between
the darkened granite and those which exhibit distinct yeins of aplite
penetrating schist are only intelligible when their relations are followed
out in the field. The schist, which is here not much more than a dark
micaceous shale or phyllite, occurs in situ, dipping south-east, about
midway between the road from Glenties and the summit of Carbane.
The granite not only invades it, but is so worked up in it that crystals
of quartz and microcline lie as white oval specks in a dark ground of
phyllite, which flows round them. Tongues of granite shade off into
mixed rocks of the most diverse character — here into a schist which
has become set porphyritically with constituents from the granite ; here
into a fine-grained composite gneiss, in which the former sediment is
represented by delicate waving sheets of biotite, with aplite layers,
less than 0*5 mm. thick, between them. Hence the same contact-
zone gives us granulitic biotite-gneisses, and rocks that, by themselves,
might be regarded as felspathio ash-beds. The more uniformly inter-
mingled masses show the same type of darkened granite as in the
dome of Ardara, and occasionally obvious flecks of schist indicate the
origin of the darkening. The composite rock, in a handsome specimen
specially examined, has a specific gravity of 2*73, and is seen under
the microscope to contain bent and streaky groups of biotite and yellow
epidote, caught up between the constituents of the normal granite.
Sphene, a common accompaniment of such contact-action, occurs
freely. Some subsequent pressure-effects are traceable, in the pro-
duction of mylonitic envelopes about certain felspars, and the alter-
ation of quartz crystals to granular aggregates ; but, both in the field and
in the section, the actual intermingling is seen to be due to igneous flow.
We pass by gradations from this rock to those which resemble felspa-
thio ash, and see under the microscope how an excess of sedimentary
material and a deficiency in granite has produced this extreme compo-
site type. (PI. n., fig. 2). Earth-pressures no doubt assisted the pene-
tration of the granite-magma into the schist ; as usually happens, the
igneous rock followed, but did not originate, the upheaval, and it»

CoLB — 0## Composite Ghieisses in Boylagh^ West Donegal. 213

consolidation occnired under the yery inflnences that had driven it
from its subterranean caldron.^ But the continuation of these pressures,
when consolidation had begun, broke up the tiny sheets of granite,
squeezed the yielding layers of biotite-schist between the crystals,
and gave us the interesting porphyrolde of Garbane as a parallel with
the ''crush-conglomerates" that occur so often on a larger scale.
But, here again, the essential intermixture of materials occurred
during the igneous flow. In one specimen, a lump of biotite-epidote-
gneiss occurs among the schist-layers, shifted by the dynamo-meta-
morphic movements from the position it once occupied, but showing
that gneissic rocks had arisen in Carbane by intermixture prior to
these particular movements.

All the rocks examined from this contact-zone show signs of
pressure-alteration subsequent to their having received a foliated
structure. Were not the evidence satisfactory in the field, it would be
easy to attribute the principal foliation also to dynamic action. This
combination of igneous penetration with shearing movements seems
a common feature along granite-contacts in the Pyrenees.* But
again and again, even in hand-specimens, we see that no dynamic
movements could have produced such differentiation in successive
layers of the rock. This is markedly the case in certain epidiorites
of the Dalradian series, which are found on Carbane delicately inter-
foliated with wavy and fluidal sheets of aplite. The final movements
have faulted some of these sheets, and the planes of fracture cut
across their foliation (PL ii., fig. 1) ; but the original igneous inter-
penetration, and the consequent production of a hornblende-gneiss
with strongly differentiated layers, are as clearly traceable here as in
the instance elsewhere cited by me from Cregganconroe, in the county
of Tyrone.' Sphene and epidote occur in the composite rock ; and
the latter mineral, occasionally appearing in large patches in the
fluidal granite veins, is doubtless tihere of primary origin, owing to
the conditions under which the ultimate consolidation of the rocks
took place.^

1 Compare Weinschenk on the Alps, Congr^ g^ol. intemat., Comptes rendus,
▼iii* teasion, p. 340.

' See LacToiz, *' Le granite dea Pyr^^ea et sea ph^nomdnes de Contact," l***
m/bm.j pp. 6 and 40, and 2"^ m^m., p. 18.

s "Hetam. Bocks in £. Tyrone, etc.," Trans. R. Irish Acad., toI. zzzi.^
p. 440, and pi. xxvi., fig. 1.

* See Weinschenk, **M6moire but le dynamometamorphisme, etc.," Congrda
gaol, intemat., Comptea rendni, viii* sefaion (1901), p. 340.

214 Proceedings of the Royal Irish Academy.

The microscopic evidence, in fact, sustains to the full the broader
evidence in the field. The handsome red granite of a portion of
Carhane, with flakes of mica-schist in it at intervals, passes clearly into
darkened gneissic types along its margin. The more we study the rela-
tions of the granite and the Dalradian series in Boylagh, the more
influence we may ascribe to the structure of the latter, prior to its
invasion by the granite. Even where not already foliated by the
earth-movements which reared successive mountain-chains, even where
it remained as little altered as in " the Eock " at Glenties, the sedi-
mentary series provided surfaces for the penetration of the granite,
and the previous structure of the country originated that of the
banded gneiss.

Mr. Kilroe^ has already laid stress on the metamorphosed condition
of the strata in Eoylagh prior to the intrusion of tlie granite. I
cannot help thinking, in considering the modifications of this igneous
mass, that too great importance has been attached to the dynamic
action which followed on its consolidation, and too little to the con-
tact-phenomena and interpenetration, of which Carbane serves as so
typical an example.

IV. — Thb Mabgik op the Granitb fbom EAXKr to Debkbbo Hill.

In the southern half of Sheet 15 of the Geological Survey Map,
Mr. Eilroe has represented in remarkable detail the intrusive and
serrated margin of the granite, where it invades the Dalradian series
near the mouth of the Gweebarra. Limestones, diorites, quartzites,
and schists are, as it were, dovetailed into the granite across four
miles of country. North of the Gweebarra River, their strike is
fairly parallel to the granite-margin; but on the south side their
outcrops swing round to the north-east, and are crossed at right angles
by the igneous rock. The prominent tongues of the granite, how-
ever, run out along the strike, and show liow potent an influence the
lie of the sedimentary rocks exerted on the flow of the invader.

Mr. Kilroe' has similarly noted that at Dimlewy, further north,
where the granite cuts across the foliation of the Dalradian series,
its offshoots invade that series, '^ usually along the strike." Masses
are described as having been detached from the metamorphic series,
and remaining as bands within the granite.

> Mem. S.W. Donegal, p. 30 ; also p. 27.

' Memoir to Sheets 3, 4, 5, 9, 10, 11, 16, and 16 (1891), p. 71.

OoLB — On Cotnpasite Oneiases in Boylaghj Went Donegal. 215

In our present area, commencing with the bare white hog's-back
of Trusklieve, we find that the granite has included a great eye of
schist, a quarter of a mile long, above Banny Lough, and has caused
kyanite to develop throughout it by contact-action. According to the
interesting views of Weinschenk/ this mineral implies that the
intrusion was accompanied by earth-pressures, sufficient to determine
the formation of the denser aluminium silicate rather than andalusite.
The whole rock thus comes to have a specific gravity as high as 2-86.
This schist, viewed under the microscope, is a very handsome quartz-
mica-gamet rock, rich in brown biotite, which has a small optic
axial angle, and with a little plagioclastic felspar. The longer axes
of the kyanite crystals lie in the surfaces of foliation, and tufts of
aillimaoite have developed extensively in the micaceous patches, and
spread throughout the granular quartz. The latter mineral thus
comes to resemble cordierite when seen in section ; but its granules
are, of course, uniaxial.'

The main boundary of the granite occurs at Banny Lough, and the
rock becomes darker, showing the usual gneissose bands of biotite as
it intrudes among the schists ; but another band of granite comes up
along the strike of the schists and epidorites at Eelmurry Lough, a
third of a mile to the south-west, while small veins in the towuland
of Farragan point to the continued proximity of the igneous masses
below. The two great dykes, however, at the old stone fort in Cor,
still further south against the Gweebarra, belong in all probability
to the later and unf oliated pegmatite series. They have given rise to
interesting phenomena of admixture and recrystallization in the massive
amphibolite which they traverse.

At the north end of Toome Lough, and the south-east comer of
Trusklieve, the marginal granite contains abundant inclusions of the
schist, which here dips under the main mass of Trusklieve. A bold
banded gneiss has been produced by the intrusion of parallel sheets of
pegmatite along the foliation planes of the schist (PI. i., fig. 1). Itft

1 « Memoire sur le dynamom^tamorphisme et la pi^zocristallisation," Congi^s
g«ol. intemat., Comptet rendus, viii* seaBion, Paris (1901), pp. 329 and 341.

' Salomon {op, cit., Zeitschr. d. deutsch. geol. Gesell., Bd. xlii., p. 624)
nmilariy indicates the intimate association in his Italian contact-rocks of aillimanit&
and biotite, the former mineral appearing to replace the latter. The sillimanite
also penetrates the quarts in these rocks, both in needles and dense bundles. The
connexion between biotite and sillimanite was observed by Levy, as far back as
1879 (*' Formation gneissique du Morran,*' Bull. soc. g^ol. de France, 3"« s^rie,
t. Tii., pp. 861 and 869).

216 Proceedings of the Royal Irish Academy.

composite origin is, however, almost too obvious ; and some authors
might, on this ground, exclude it from the category of gneiss.

In the east of Toome the rock-surfaces of the granite show a
feeble fluidal foliation, with some inclusions of schist lying at various
angles, and not drawn out in the mass. Blocks of schist and diorite
are common as inclusions down the east side of Toome Lough ; and
thence we may proceed to trace the junction on the south side of the
Oweebarra.

East of the iron bridge (which is not represented on existing
Ordnance maps) a small section in the cliff illustrates how the promi-
nent foliation of the granite may be due to flow, and not to shearing
movements affecting it in company with the associated schists. The
rocks invaded by the granite are limestones and shales preserving
their original bedding ; but the granite has become foliated parallel
to its junction with the sediments, and across their planes of
stratification.

On the picturesque ascent of Cleengort Hill through Kincrum, the
granite contains biotite, and is locally foliated ; lumps of schist are
freely included along this serrated junction. On Cleengort Hill, the
schist is more highly altered and more obviously micaceous.^ In part
it contains kyanite and garnet, and these are accompanied by the same
handsome brown biotite as occurs in the kyanite-schist of Trusklieve.
The kyanite-schists are distinctly felspathic, and probably owe some
material to the granite veins associated with them ; these are clearly
traceable in hand-specimens, but tend to disappear in microscopic
sections, owing to the delicate character of the intermingling.

On the crest of Cleengort Hill there is an easily traceable junction
of granite and jichist ; the former rock is foliated parallel to the junc-
tion, and is at the same time rich in obvious inclusions from the schist.
It would seem unnecessary to multiply such examples had not the
later metamorphic stresses been called in to account for the phe-
nomena across so wide a stretch of country. Where fragments of the
aphanites or epidiorites from the schist-series have been caught up as
inclusions in the granite, they prove to be rich in dark green horn-
blende, which ophitically includes some of the associated quarts. An
older saussuritised felspar is seen in microscopic sections, side bj side
with recrystallized plagioclase and granular quartz. Seeing that the
original pyroxene-dolerites of the Dalradian series were metamor-
phosed into epidiorites prior to the intrusion of the granite, it is diffi-

1 Compare J. B. Kilroe, in Mem. to Sheets 3, 4, 16, etc., p. 46.

CoLB — On Composite Gneisses in BoyUtgh^ West Donegal. 217

cult to Bay, in the ease of such inclusions, what features may be due
to the subsequent contact-metamoiphism. Sphene and biotite have
often been noted as products of contact-alteration and of marginal
absorption of amphibolites or hornblende-schists^; and thus the
epidiorites of Cleengort doubtless yield some material to the adjacent
biotite-gneiss. At other times we may expect derived hornblende
to remain, and eyen to revive in the granite magma, as Mr. Eilroe*
recognised in an example south-west of Ardara. Such instances
probably arise where the proportion of derived material is large in
comparison with that of the granite magma.

An inclusion of epidiorite from the Cleengort gneiss has a specific
graTity of 3*06; iron ores are not prominent, and the amphibole is
clearly of high density. It displays dark green and yellower green
face-colours, strong axis-pleochroism, and extinction-angles of as
much as 20^ away from the vertical axis o. The study of a mass
such 88 this has an important bearing on the hornblende-granites
and hornblende-gneisses farther to the east.

As we cross the bog of Derkmore, and climb to the geat dyke of
Cainozoic dolerite on tlie pass leading over towards Glenties, we see
the complex Dalradian series invaded by sheets of graphic granite
along the foliation-planes, and also by granite dykes cutting across
them. Dr. Hyland' has recorded sheared granite from near Meena-
largan Hill in this locality, and evidences of the later earth-movements
are doubtless traceable here, as at Garbane. But, on rounding the
pure granite masses that form the south side of Derkbeg Hill, we
come upon patches of foliated granite which have all the characters of
floidal composite gneiss. Included layers of schist are clearly seen,
striking north-east, and these are, as usual, surrounded by biotite
gneiifl, its foliation striking in the same direction.

In places, again, we have a delicately injected schist, which passes
at its margins into gneissoid granite. A sample of this granite, in
which both light and dark mica occur in strings and layers, shows a
few *' strain-shadows " tmder the microscope, but no sign of shearing
or mylomitic flow. It is a white aplite, with a few micaceous additions
from the schist. In contrast, one of the much darkened portions of the

1 See referencefl in " Metam. Bocks in £. Tyrone, etc.," Trans. B. Irish Aoad.,
ToL xxzi., p. 464, and plate zzvi., fig. 6.

* Mem. S.W. Donegal, p. 63. Also Mem. Sheets, S, 4, 15, etc., p. 78. For
mmMroiiB references to quarts-homblende-diorites, formed by hiyasion of granite
into Iwne loeki, tee paper in Trans. B.I.A., above referred to, pp. 438 and 439.

* Hflm. Sheeto, 3, 4, 16, etc., p. 184.

218 Proceedings of the Royal Itkh Academy.

granite proves to be almost a '' granitised " amphibolite. Deep green
hornblende, of the type discussed in connexion with Cleengort, occurs
in it in rough foliation-layers ; but this mineral forms so much of the
rock as to carry the specific gravity up to 2*99 — the mean of three
closely agreeing determinations. Andesine is present, and, like the
hornblende, has probably recrystallized from an epidiorite, which
renewed its youth in the granite magma ; while quartz and orthoclase,
representing the aplitic granite, are more abundant in the joint mass
than its specific gravity would suggest (PI. ni., fig. 2). A chain
of observations, and especially those made in the field, teach us to
regard this rock as composite. It is, then, one of the most extreme
modifications met with in the intrusive granite of Boylagh.

It is interesting to observe how hornblende crystals, already formed
in andesitic lavas as products of the ^' first consolidation," become
frequently reabsorbed by the magma when it ascends or is poured out
upon the surface. They then leave, as von Lasaulx long ago pointed
out, a mere skeleton formed of grains of magnetite. But, in our
instances of the absorption of amphibolites by a granite magma, the
final crystallLzation takes place under considerable pressures. The
conditions are favourable for the crystallization of biotite, or even for
the reproduction of the hornblende, and a quartz-diorite with some
orthoclase results. Were this mixed rock again melted and thrust ont
on the surface, a rhyolitic andcsite with pyroxene would probably be
formed, in which a few corroded hornblendes from the previous con-
solidation might remain.

Pegmatite veins and masses, belonging to a later granite, and
probably of Devonian age, are common throughout Derkbeg andDeny-
loaghan ; but undoubtedly the most striking features are those con-
nected with the foliation of the older granite. The constancy of strike
in this foliation becomes, if the foregoing observations are correct, a
record of the trend of the folded Dalradian series, which has here left
mere traces in the heart of the granite which attacked it.

V. — The Gneissic Stetjctuees msiK Finntowk, and thbib bea&ikg ok
THE Geological Histoey oe Gneissic Granite th&ouohovt
THE County op Donegal.

In the eastern part of Galwolie, on the road from the village of
Lettermacaward to Doocharry Bridge, a red gneiss, containing musco-
vite as its predominant mica, reminds one of the crushed and slicken-
sided granite of Bamesbeg in northern Donegal. It similarly owes

OoLB — On Composite Oneiases in Boyhghy West Donegal. 219

its present characters to dynamic metamorpliism. It is cut, however,
by a brown granite, with a specific gravity of 2*60 ; the foliation in
this does not agree with that of the older mass, but is clearly due to
igneous flow, running as it does parallel to the margins of the veins.
This granite does not show metamorphic deformation under the micro-
scope ; its foliation is due to the arrangement of plates of light and
dark mica as it flowed. Here we have, in the coarser and earlier rock
of Ghdwolie, the type of foliation that has been accepted as that of the
Donegal granite as a whole. But hitherto our observations in Boylagh
have shown us the phenomena of the fluidal veins occurring on a largo
scale, and emphasised by inclusions of schist within the granite.
It becomes of considerable interest to inquire to which type the
foliated structures belong, which are developed on a still broader scale
south of the Oweebarra, between Derryloaghan and Finntown. Our
conclusions in regard to them may justly affect our views of similar
gneissic districts throughout central Donegal.

On Sheet 15 of the Geological Survey map, the great granite mass
west of Finntown has foliation marked on it in some thirty places,
and in most cases a dip is assigned to the foliation-planes. South-
easterly dips prevail, but north-westerly ones occur near the Gwee-
barra, and predominate on the other side of the river. The general
structure, then, is that of an anticline, measuring some four miles
across.

In the strike of this foliation, streaks of epidiorite, limestone, and
a little quartzite have long ago been noticed, especially on the south-
east flank of the mass in the neighbourhood of Finntown. Mr. E. H.
Blake^ thus recorded vertical layers of limestone and mica-slate within
the granite ; and Mr. E. H. Scott' deduced from those in Glenleheen
the metamorphic origin of the granite, in the sense that it was derived
from alteration of the ancient sediments in place. Like Dr. Haughton,
he favoured this view for the greater part of the granite of Donegal,
though he recognised that even the gneissose granite occasionally
pierced the other rocks in the form of veins.' Mr. Scott, in the first
paper quoted, provides a section along Glenleheen, in which the bands
of sedimentary rock are shown going vertically down in continuous
layers, with the granite cleanly interstratdfied between them.

* " On the Primary Bocks of Donegal, " Joum. Geol. Soc. Dublin, vol. ix (1862),
p. 296.

* ** On the Granitic fiocks of South- West Donegal, " ibid,y p. 290.

* « On the Granitio Books of Donegal, " 2nd notice, ibid., vol. z (1864 ; paper
read 1862), p. 20.

MJUA* noG., VOL. XXIV., BBC. B.] T

220 Proceedings of the Royal Irish Academy.

The papers of Scott and Haughton influenced Mr. W. B*. Stacpoolo
Westropp,^ who urged that there was an essential difference between
the granites of Wicklow and those of Donegal, the former being clearly
intrusiye and the latter metamorpbic. Dr. Haughton* bad to some
extent emphasised the difference by regarding the granite with black
mica as typical of County Donegal, which may be true of the surface
as now revealed to us ; I believe, however, that this apparent minera-
logical difference is due to the position of the broad natural sectiona
in regard to the intrusive mass as a whole. The prevalence of black
mica, from this point of view, simply indicates our nearness to the
schists along the crests and flanks of great arches which are now
occupied by the granite (see figure on p. 225).

When, in 1871, the late Professor A. H. Green,' in ignorance
of the literature already published on the subject, supported the
metamorpbic view in somewhat exaggerated detail, his paper met with
a cold reception. His opponents, however, knew but little of the
petrological difficulties of the country which he had visited ; and his
paper, as finally published in the Geological Magazine, to some
extent explained the position he had adopted. He dealt with the
Dunlewy district, which may be cited as an illustration of that which
we are now discussing ; and his error, in face of the frequent inter-
lamination of the granite and the schists, seems excusable when we
consider the controversies of more than thirty years ago. He was
without microscopic assistance, and seems to have done little to
deserve the intemperate and personal onslaught made on him by Prof.
David Forbes* a few months later.

Dr. C. Callaway^ recognised that the granite of Bamesbeg had
penetrated along the foliation-surfaces of the associated schists ; but
the apparent sharpness of the included fragments, when viewed with
the naked eye, seems to have deceived him, as it has done observers
in other countries. He was thus led to deny any process of absorption^
and consequent modification of the granite, though the microscope is
not needed for its appreciation at Cashel Hill near Portnoo, and at

^ Letter on « The Origin of Granite, " Qeol. Mag., 1867, p. 522.

> Op. eit.f Quart. Joum. Geol. Soc. London, yoI. xviii. (1862), pp.410 and 417.

' Abstract of eleven lines in Quart. Joum. Geol. Soc. London, vol. zxvii. (1871),
p. 449 ; and ** Notes on the Geology of Part of Co. Donegal," Geol. Mag.,
1871, p. 653.

« « On the Geology of Donegal," Geol. Mag., 1872, p. 12. On Dunlewy, see
J. B. Kilroe, Mem. sheeU 3, 4, etc. (1891), p. 71.

fi Oj). eit,, Quart. Joum. Geol. Soc. London, vol. xli. (1885), pp. 224 and 229.

C()LB — On Composite Oneisaes in Boylayhj West Donegal. 221

yariouB other places in Boylagh, as we have already noted. Dr. Callaway
held that the main gneissic structure was due either to igneous flow
of molten material, or to flow of some kind under earth-pressures.
The latter type of flow might have occurred — (i) during consolidation,
or (ii) after consolidation. Dr. Callaway* concluded in favour of
dynamic metamorphism after the granite had forced its way as an
intrusive rock among the schists.

It was in the light of these previous opinions and observations
that the officers of the Geological Survey approached the 'interesting
area between liaas and Finntown. Mr. A. M*Henry,' in noting the
masses of limestone and schist caught up here and enveloped in the
granite, explains their parallelism with the foliation in the igneous
rock by stating that ** both the granite and enclosed masses have been
subsequently foliated at the same time by the last great shearing
forces that affected this region.''

In a previous section of the present paper, on the other hand,
certain foliated granites towards the Maas end of the district have
been explained as due to imperfect incorporation of the homblende-
and mica-schists in the igneous rock (p. 216). Mr. J. A. Cunningham,
B.A., A.R.C.SC.I., who accompanied me througbout these observations,
subsequently visited the foliated granite on the road from Gienties to
Boocharry Bridge on the south side of the Gweebarra. A photograph
taken by him shows how much work remains to be done on the non-
homogeneous " banded gneisses " which are boldly developed across
this area. At present, however, I propose only to furnish details from
personal observation on the historic roadside sections between Glen-
leghan (Glenleheen) and Finntown. I am willing, however, after
experience of similar materials in the Pettigo area, to take these
exposures as representative of a much wider district.

On approaching Glenleheen by the highly picturesque road from
Doocharry Bridge, the intrusive character of the granite is manifest
at the top of the long rise from Adder wal, and just within the town-
land of Meenmore West. The granite has penetrated the schist along
the foliation-planes, and has converted it into a ^Ucptynolite," in
which felspars are visible to the naked eye. A band of limestone,
marked upon the Survey map, is included in the granite, and has
undergone the usual type of metamorphism. Bed garnet, diopside

< Quail. Joum. G«ol. Soc. London, vol. zli. (1886), pp. 230 and 239.
3 Mem. iheeta 3, 4, 15, etc. (1891), p. 69.

222 Proceedings of the Boyal Irish Academy.

altering to actinolite/ quartz, and a lime-scapolite, hare arisen in it,
together with sphene, which Mr. Scott' always associated with altered
limestone in Donegal. Labradorite felspar occurs abundantly in parts
of this rock, side by side with quartz, but seems to vary in amount
in inverse proportion to the scapolite.

This is probably the locality where Mr. Scott' discovered scapolite
in 1861, in conjunction with sphene, pyroxene, orthoclase, and quartz.
After passing Garbal Gap, across the Glenleheen Kiver, a fine section
is seen, partly quarried, on the north-east side of the road, in the
townland of Loughnambraddan. The structural planes are vertical,
and a remarkable variety of rocks can be gathered within a few feet
of one another in traversing the strike. After a little scrutiny, two
types of rock become sorted out, the one a pink euritic aplite, with
the characteristic specific gravity of 2*59, the other a dark
homblende-biotite-schist, with a specific gravity of 2*89.

The former shows under the microscope the structure of a mildly
fluidal gneiss, without banding, but in which a few of the felspars
assumed ovoid contours, from continued movement after crystallisation
had commenced (PL v., fig. 1). The quartz settled down in angular
interlocking grains, like those of a metamorphic quartzite. Only the
merest trace of biotite occurs, and the aplitic character is complete.
Iron ores are represented by rounded grains of pyrite.

The dark rock, on the other hand, is an obvious schist,
almost slaty in places, though more distinctly crystalline than
some of the masses on Carbane (p. 212). Under the microsoope it
shows a predominance of hornblende over biotite ; these minerals are
associated with abundant epidote, sphene, and pyrite. The three
last-named constituents seem alike to have existed in the rock before
its invasion by the granite. The usual basic felspars, and a few
granules of interstitial quartz, form a second association of minerals,
interwoven, as it were, with the ferriferous ones. So far, the rock
is a typical schistose epidiorite (PL v., fig. 2).

But other bands in this striking roadside section show a coarse-
grained granite with pink orthoclase, intruding up the general vertical
planes. In hand-specimens this rock resembles some of the handsome
gneisses of the Outer Hebrides; but its most foliated portions are

1 The distinct green colour of Uub paramoiphic product indicates the presence of
iron also in the diopside.

' Op. eiUy Joum. 6eol. Soc. Dublin, vol. ix., pp. 288 and 289.
3 Op. eU., ibid., vol. x., p. 21.

Cole — On Composite Onemes in Bot/lagh, West Donegal, 223

marginal, and are due to sheets of schist entangled in it. Mr. Scott'
aptly compared this type of gneiss near Finntown with the veins at
Castle Caldwell on Longh Erne; I have little douht that in both
localities these coarser granitoid rocks belong to the later series of
granitic intrusions. They may thus be of Devonian age.

When, however, these coarser rocks associate themselves with the
earlier and more fine-grained granites, as they do also at Cam, near
Pettigo, the composite rock formed of hornblende- or mica-schist,
penetrated along its foliation-planes by coarse and fine parallel veins,
presents a remarkable imitation of many Archaean gneisses. Yet it
owes its characters to the original flow of granite under pressure up
the most easily found planes of parting.

Again and again, the shrinkage of the original uptilted sediments,
as metamorphism went forward, a feature on which Mr. Joseph
BarrelP has laid considerable stress, may have helped the intrusive
mass to spread upwards from below. The previous crumpling and
compression, however, of the Dalradian series probably drove off a
part of its volatile constituents, and so prevented any further marked
reduction in bulk under the heating action of the granite.

Short of the local bands of pseudo-Hebridean aspect, we have
every variety of intermingling between the granite and the schist.
Biotite is developed in these grey gneisses and granites, partly at
the expense of hornblende. The epidote of the partially absorbed basic
rock remains intact, in association with much biotite, a little green
hornblende, and sphene. These minerals are grouped in flakes and
patches, which give the rock its gneissic aspect. The sphene is so
prominent in some of these mixed rocks that it has probably developed
during the epoch of contact-metamorphism. The rock-section selected
for illustration (PI. iv., fig. 2) shows foliated structure on a conve-
niently small scale ; otherwise it represents the more granitoid and
leas basic type of the composite gneiss of Loughnambraddan. The
specimen from which it was cut shows a granite vein penetrating the
schist, and losing its identity in so doing, thus affording a complete
parallel with some of the specimens from Carbane.

Another of the grey composite rocks examined in detail has a
specific gravity of 2*74. It is, when considered apart from its mode
of occurrence, a member of the Tonalite series, with zoned orthoclastio
and plagioclastic felspars, quartz, biotite, hornblende, epidote, and

1 Op. eU.f Joum. G«ol. Soo. Dublin, yoL z., p. 18.

s <« Th« Physical EiEbcts of Contact Metamorphinn," Am. Joum. Bci., vol xiiLi
(1902), p. 294.

224 Proceedings of the Royal Irish Academy.

handsomely developed crystals of sphene. There are signs of defer-
mation in some of the felspars ; but the gneissic structure is clearly
due to other than dynamic causes. As in previous cases, the epidote
is in constant association with the streaks of biotite or hornblende.

It is noteworthy that the rounded granules of pyrite observed in
the pure euritic aplite of this locality occur also in the more aplitic
layers of the composite rocks.

There is good evidence, then, that the varieties of grey and red
gneiss above Pinntown are portions of granite masses locally modi-
fied by the conditions of their intrusion. The phenomena of Ballyir-
iston and Derkbeg Hill are here carried out on a still more convincing
scale. It seems highly probable that the ^'Homblende-Biotite-Granites
(Homblende-Granitites) " described from other parts of Donegal by
I'r. Hyland* have the same origin as those studied in the present
paper.

What, then, is the general conclusion that we may come to in
regard to the gneissic patches, often half a mile or so long, which
occur with so constant a foliation- strike in the granite mass south of
the Gweebarra ? Are they not, equally with the strips of epidiorite
and limestone, the relics of strata that formerly occurred, in metamor-
pliic wrinklings, in the crown of some great anticlinal arch' ? It is
not necessary to urge that the whole of the space now occupied by
granite was formerly filled with folded schists, and that the solution
of the latter provided room for the granite in its ascent ; the magma
of the granite may have at first welled up into the spaces provided for
it by the Caledonian folding, and then, under continued earth-pressures,
have been forced, with destructive effect, against its bounding roof
and walls.

Instead of representing, as Mr. Scott was tempted to do, the
foliated and sedimentary strips of rock near Finntown as vertical
beds between vertical layers of granite, may we not rather regard the
present surface of the mass between Lough Finn and Doochany
Bridge as exhibiting a cross-section of the upper zone of inter-
action? Some masses from the roof have survived, and give us a
profound impression of the material that has been altogether lost
within the granite caldron (fig. 1). Had denudation worn away the

^ Mem. to Sheets 8, 4, 16, etc., p. 136.

* Mr. Eilroe speaks of detached areas and bands of schist, etc., in the gnuiite
east of Dungloe, wliich form a cunred series when regazded as a whole, as
"obviously fragmenU of a schist series, which formeriy extended from Tor
westward*'(/^., p. 43).

Cohn— On Composite Onet'gseit in Boylagh, West Donegal. 225

N O

226 Proceedings of the Royal Irish Acadefny.

composite mass even down to the present sea-lerel, onr knowledge of
the banded gneiseeeofBojhigh might have been limited to thoee which
arose npon the side-waUs of the caldron.*^ As it is, from specimens a
few centimetres sqnare up to moorlands that are not to be traTcrsed in
a day, we may see thronghout Boylagh what attacks are made from
below npon the materials of a rising monntain-chain. The final out-
come is a consolidation of the anticlinal ridges, by the intimate pene-
tration of igneous material, which crystallizes within their cores ; and
ribs are added to the Earth's crust, like those of Donegal, which
successfuUy resist later systems of folding, and still hold their own
among the rugged highlands of the world.

Fio. 1. Ideal section to illustrate the structure of the granite mass
south of the Oweebarra. — ^A great group of sediments has become
folded into a complex anticlinal mass, with production of schistose
features in most of the rocks. Granite has intruded during the
formation of this compound arch, and especially into the antidinala,
where pressure is relieved ; it has found its way most easily along the
planes of bedding or foliation, as the case may be, in the overlying
mass. Farts of the latter mass are absorbed ; but flakes remain, pro-
ducing a composite rock, and imparting a gneissic structure to the
granite. Denudation, acting continuously daring these changes, has
now worn down the rocks to the suiface indicated by the line XZ.
Above the point A, the parallel intrusions suggest on this surface
that we are on the edge of an uptilted laccolite. At this point it
would be very difficult to determine how far the metamorphism of the
schists was previous to, and how far due to, the intrusion of the
granite. Above B, we have a granite moorland with occasional
gneissic structure. As we approach C, the origin of this structure
becomes again traceable ; and ultimately composite '' leptjmolites '*
and '* granitised schists " are seen to pass into the ordinary schists on
the right-hand side of the complex anticlinal.

VI. — Conclusion.

The references above made to the work of others show that the
explanation now put forward for the gneissic structures in Boylagh
is one that has raised a certain amount of controversy in the case of
other European areas. It is not to be expected that all gneisses

1 Compare '* Metam. Bocks ia £. Tyrone and S. Donegal/' Trans. Boy.
Irish Acad., vol. zxzi., pp. 468 and 469.

Cole— On Composite Oneissea in Boylaghy West Donegal 227

have been formed bj the same processes ; and even banded gneisses
may in certain instances represent stratified materials crystallized
and modified in place. This would, at anj rate, be the logical
deduction from the Tiews of Mr. F. D. Adams, ^ which would bring
us back to some of the oldest and half -forgotten theories respecting
metamorphic rocks. Fascinating as the dynamometamorphic theory
has been, it may be questioned if strongly marked banding can be
produced in crystalline rocks by the agencies thereby invoked.
Mylonitic destruction, rather than banding, results, as a rule, from
earth-pressure combined with movement; and the distinctions between
adjacent layers tend to become obliterated. Professor Judd,^ in 1898,
called attention to the slow processes of '' statical metamorphism,''
whereby rocks which are kept stationary underground may be modi-
fied, not only in mineral constitution, but even in chemical composition.
Crystalline layers, their individual characters dependent on those of
the successive original strata, might thus eventually arise, and would
even produce a banded gneiss. In Boylagh, however, the phenomena
of igneous injection and intimate penetration have played by far the
most important part ; and there is no particular mystery as to the
mechanical or chemical nature of the process, the stages of which
are often traceable with the naked eye.

While believing with L6vy and Lacroiz that granite does not
come into its final position without a considerable absorption of
material from the walls of its caldron,' I naturally admit, from
considerations of geological structure, that the caldron itself most
commonly originates in the arch of an anticlinal.' As Salomon^
perceived in the case of the Adamello chain, the position where the
igneous rock ultimately manifests itself is determined by the oppor-
tunities allowed it during the larger movements of the crust. But are

1 « Some recent papers on the Influence of Granitic IntrusionB upon the
Development of Crystalline Schists," Joum. of Geology, vol. v. (1897), pp.
293-302.

s « On Statical and Dynamical Metamoiphiam," Geol. Mag., 1839, p. 243, etc
Thii subject has been greatly developed by Van Hise, <' Metamorphism of Bocks
and Kock-flowage," Bull. Geol. Soc. America, vol. ix. (1898), p. 269.

'Compare Lacioiz, op, eiL, Bull, de la Carte G6ol. de la France, No. 64
(1898), pp. 1 and 62.

* Compare T. A. Jaggar, jr., «* The Laccoliths of the Black Hills," 21st Ann.
Rep. U.S. Geol. Survey, Pt. iii. (1901), p. 173.

' " IJeber Alter, Lagerungsfonn, und Enstehungsart der periadriatischen
granitiBch-komigen Massen," Tscherm. Mittheil., Bd. zvii. (1898), pp. 173-4.

K.LA. PBOC, VOL. ZXIV., SBC. B.] V

228 Proceedings of the Royal Irish Academy.

we to believe that sach upwellings, implying local relief from pressure,
are unaccompanied by incorporation and assimilation on a considerable
scale ? Eyen Salomon himself, who shows that the variations in the con-
stitution of the enormous mass of tonalite are not related to the nature of
the contact rocks, presents us with a section^ illustrating '* lit par lit "
injection above the Poglia valley, where the igneous rock assumes a
bedded structure, and includes residual and parallel strips of altered
Triassic limestone. The resemblance of such structures to those near
Finntown is apparently complete; and the author attributes the
position of the tonalite between the sedimentary bands to the solution
of certain shaly layers, as the tonalite sent off apophyses into them.
Salomon still concludes, as in an earlier paper,' that the gneissic
structure of the tonalite is due to subsequent dynamic action, despite
the occurrence of a little true fluidal structure here and there ;' but
his work is nowhere opposed to the views above stated in explanation
of the gneiss of Boylagh.

More than twenty years ago, Mr. G. W Hawes* called attention to
the production of mixed rocks on an important scale at the contact of
granite in New Hampshire ; and there is little difference between
his statements and those made so clearly by Lacroix' in 1898 con-
cerning the composite gneisses of the valley of BaxoxdUade. Con-
siderable stimulus will now be given to such enquiries by the remarks
of Mr. Teall' in his Presidential Address to the Geological Society of
London in 1902 ; and it is probable that the importance of composite
gneisses will be recognised in many areas, where the prevalent
structures have hitherto received other interpretations.

» " Ueber Alter, Lagemngsfonn, und Entstehungsart der periadriatiachen
granitisch-koniigen Massen," Tscherm. Mittheil., Bd. zyii. (1898), p. 159.

' *'Neue Beobaohtungen aua den Gebieten der Cima d'Asta und dea Monto
Adamello," Tscherm. Mittbeil., Bd. xiL (1891), p. 411.

' Op. eit., Tscherm. MittheU., Bd. zvii., p. 131.

* "The Albany Granite and its Contact Phenomena," Amer. Jour. Sci., vol.
TO. (1881), pp. 31 and 82.

* Op, eit, (1898), p. 49.

* Proc. Geol. Soc, pp. Ixxiv and Ixxviii, in Quart. Joum. Geol See. for 1902.

Cole — On Composite Oneisaes in Boylaghy West Donegal. 229

DESCEIPTION OF PLATES.

Plate L

Fi6. 1. — Composite gneiss, produced by intrusion of granite into
the scliistose series near the base of Trusklieve, on the northern shore of
Toome Lough (p. 215).

Fig. 2. — Fegmatitic granite cutting and sending veins into horn-
blende-schist, which has previously become considerably''* granitised."
North-east slope of Derkbeg WJl (p. 218, and PI. m., fig. 2).

Plate IL

Pig. 1. — Section showing margin of^athin sheet of granite in horn-
blende-schist, south-west side of Carbane, Glenties (p. 218). x 9.

This microscopic section represents in miniature^the structure of
the granitic and schistose country in Boylagh. The granite has pene-
trated the schist, after the latter had become foliated and crumpled.
A fluidal structure along the margin has converted the granite locally
into gneiss ; this is occasionally emphasised by the presence of flakes
removed from the schist. Subsequent movements, represented in the
section by the faulting, have had some effect upon the joint mass ;
but the gneissic structure is connected with original conditions of
intrusion.

Pig. 2. — Section showing junction of phyllite and gneissic granite,
south-east side of Carbane, Glenties (p. 212). x 7.

The coarse granite is seen below, with streaks of biotite, due to
inclusion of material from the phyllite. When seen over a wider
field, these give a well-marked gneissic structure to the granite. The
phyllite contaiuB numerous crystals from the granite, and resembles
a " porphyroide " or a felspathic ash. This intermingling seems due
to earth-movements acting after the crystallization of the granite, but
in oontinuation of those under which the igneous rock was intruded.

230 Proceedings of the Royal Irish Academy,

[Plates in.-v. are from photograplis taken with the microscope
by Mr. T. Crook, a.b.c.sc.i.]

Plate III.

Fig. 1. — Section of granite containing patches of ,biotite, horn-
blende, and epidote, derived from the materials of the schistose series.
South of Ardlougher, near Clooney (p. 211). x 10.

Fig. 2. — Section of composite rock (quartz-diorite), produced at
junction of granite and epidiorite. • North-east slope of Derkbeg Hill
(p. 218). X 10. The hornblende has revived and recrystallized unda
the influence of the intruding granite magma.

Plate IV.

Fig. 1. — Section of composite rock, with brown biotite ("Lepty-
nolite"), produced at junction of granite and mica-schist. Small
quarry on road from Clooney to liaas (p. 209). x 10.

Fig. 2. — Section of composite gneiss, inclining towards granite,
with delicate streaks of biotite, epidote, sphene, and some hornblende.
Quarry in Loughnambraddan, above Finntown (p. 223). x 10.

Plate V.

Fig. 1 . — Section of pure euritic aplite, with ovoid forms <rf consti-
tuents. Quarry in Loughnambraddan, above Finntown (p. 222). x 10.

Fig. 2. — Section of schistose diorite (epidiorite), into which the
aplite intrudes in the quarry in Loughnambraddan, above Finntown,
producing a variety of composite gneisses, one of which is illustrated
on Pl. IV., fig. 2 (p. 222). x 10.

Proo. R. I. A., Vol. XXIV., Sec. B.

Plate I.

Fig. 1.

Fig. 2.

"r3c RIAcad Vol XXIV Section B.

Plate IJ

r» A J C:.-. de

Proc. R. I. A., Vol. XXIV., Sec. B.

Plate III.

Fig. 1.

Proc. R. I. A., Vol. XXIV., See. B.

Plate IV.

Fig. 1.

Fig. 2,

Proc. 11. 1. A., Vol. AXIV., Sec. B.

Plate V.

Fi>r. 1.

Vitr 9

^.S.rolS'O^.SOl^V gi:p VS1903

A LIST OF IRISH ECHINODEEMS. Br A. R. NICHOLS,
M.A., BEING A REPORT FROM THE FAUNA AND FLORA
COMMITTEE.

[Read Junb 23, 1902.]

Thb first list of the Irish Eehinoderma was published by R. Templeton
in the ninth Tolume of the Magaaiine of Natural History in 1836. The
next list was contained in the report on the Invertebrate Fauna of
Ireland (drawn up by W. Thompson at the request of the British
Aflsodation), and published in the British Association Report for 1843 ;
a more detailed account of the distribution of Irish Echinoderms and a
few additional species were subsequently included in his ^'Natural
History of Ireland," yoI. iv., 1856. At the meeting of the British
Association in 1858, a Report on the Marine Fauna of the south
and west coasts of Ireland was communicated by Prof. E. P. Wright,
M JD. and Prof. J. R. Greene, and in this Report is included a tabular
list of Irish Echinoderms (excluding Holothurioidea).

Since the publication of this Report in 1859, no attempt seems to
liaTe been made to compile a complete list of the Irish species of
Eehinoderma, though a large number of species have been added to
the Irish Fauna as a result of the deep-sea explorations that have
been carried on off the western coasts of Ireland. These investi-
gations began with the first cruise of H. M. S. ** Porcupine " in 1869,
and have since been occasionally carried on by various smaller expe-
-ditions, during which the richness of the Echinoderm fauna was often
commented upon.

In conformity with the British marine area, as defined by the
Committee of the British Association in 1888, the Irish marine area
may also be regarded as consisting of two portions, viz, : a shallow-
water district with a depth ranging from 0 to 100 fms., and a deep-
water district whose depth ranges from 100 to 1000 fms. ; the boun-
daries of the shallow- water district are the 56° parallel of latitude on
the north, a lino half way between Ireland and the opposite shores of
.Scotland, Wales, and England on the east, the 49^ 30' parallel of

B.I.A. PKOC., VOL. XXIV., SEC. II.] X

: 32 Proceedings of the Royal Iri^h Acadetny,

latitude on the south and the 100 fms. line on the west. The deep-
water district is practically confined to the western coasts and
comprises the area contained between the 100 fathoms and the 1000
fathoms lines and the parallels of latitude 56^ and 49^ 30'. For the
conyenience of students of geographical distribution the known range
of the species round the coast of Ireland has been divided into the
same six provinces which I adopted in the paper on the Marine
MolluBca of Ireland (Proc. Roy. Ir. Academy (3) vol. v., pp. 477-662,
1900).

lliese provinces are —

i. North-east. From Malin Head, Co. Donegal to St. John's Point,

Co. Down,
ii. East. From St. John's Point to Camsoi*e Point, Co. Vex-

ford.
iii. South. From Camsore Point to Cape Clear, Co. Cork,
iv. South-west. From Cape Clear to Loop Head, Co. Clare.
V. West. From Loop Head to Erris Head, Co. Mayo,
vi. North-west. From Erris Head to Malin Head.

The names of the species which have not been found at a less
depth than 100 fms. off the Irish coast are placed within 9qmre
brackets and can thus be readily distinguished. Whenever a species
is found in shallow-water in some and only in deep-water in other
provinces, then the numbers denoting these latter provinces are included
in square brackets.

I have followed the classification, and with one or two exceptions,
the nomenclature adopted by Bell in the '^ Catalogue of the British
Echinoderms in the British Museum," 1892 ; but for the convenience
of reference I have given the names used by Forbes in his " History of
British Starfishes " whenever these names differ from those of the
British Museum catalogue.

The general distribution of each species is given very briefly, and has
been derived principally fix)m the ** Challenger" Ileports and the writings
of Agassiz, BeU, Cams, Hoyle, Ludwig, Lyman, Norman, Verrill, &e.

The total number of species of Echinoderms found in the seas
surrounding the British Islands is 134, and the number in this list
is 87, so that the Irish Fauna includes nearly two-thirds. If we
exclude from the Irish list the deep-water forms which belong perhaps
more fetiictly to the general Atlantic Fauna, and confine the British list
to the species which have been found at less than 100 fms. depth on.

Nichols — A List of Irish Echinodenm.

233

the coast of the British Tales, then the Irish list contains 60 out of a
total of 81 British species or nearly three-fourths ; most of the British
ahallow-water species, that are absent from the Irish list, are northern

ROCKALL
BANK.

/ "4e*'3()

Ifap ■howins the Shallow- and Deep-water District^ and the six Prorinces.

A — Malin Head. C — Carnsore Point. E — Loop Head.

B— St. John's Point. D— Cape Clear. F— Erris Head.

fonns, and several of them are recorded as British only from off the

north of Scotland.

634 Proceedings of the Royal Irish Aeeulemy,

The only species peculiar to the Irish shallow- water Fatma are two
douhtful species of Cucmnaria, riz. : C, andrewsi and C, saxicola;
the former species was named by Farran in 1852^ from specimens taken
at Clonea, Co. Waterfoid, and the latter by Brady and Robertson in
1871, from specimens taken in Birterbuy Bay and Westport Bay.

Fiye deep-water species of Ecbinoderma, viz. : Hohthuria asp^a,
Astropeeten sphenoplax^ Pentagonaster greeni, Sgmenaster gigan/eus and
Cidaris gracilis have, however, hitherto only been obtained off tlie
western coasts of Ireland ; the last two species having been described
by Sladen from specimens dredged in 750 fms. in the expedition
organized by a committee of the Koyal Irish Academy in 1888.
Pteraster personatus, Pentagonaster halteatus and P, eoneinnus wero
also described by Sladen from specimens obtained at a depth of
750 fms. in this expedition; P. halteatus and P, eoneinnus aie
however regarded by Bell as identical with P, granularis Betz., but
by Ludwig as possibly identical with P. hystricis Marenzeller, and
Pteraster personatus has since been recorded from the Bay of Biscay.

Strongyloeentrotus lividus^ the purple sea-urchin, is a very character-
istic species of the western coasts of Ireland, from Malin Head to tbc
south coast of Co. Cork ; a specimen is stated by Dr. Dickie to have
been cast up on the shore at Carrickfergus, but this is the only record
for the east of lrc4and. It is a southern species ranging from north of
Ireland and south-west of England, southwards to the Azores and
Canaries. JIblothuriaforskahli (J?, nigray auct.), the nigger or cotton-
spinner, is another soulhem species that is generally distributed on the
western coasts of Ireland, and there is no record of its oceuirence on
the eastern coasts.

The two species of Luidia, viz. : Z. ciliaris and L. sarsi, are generally
distributed and common off the western coasts of Ireland, and range
from Faroe and Norway respectively to Cape Verde but do not appear
to have been recorded from the east coast of Ireland.

BiBLIOQBAPllT.

Agassiz, a. :

'72-*74. Ecvision of the Echini. Illust. Cat. Mus. Comp. Zool.,

Harvard, "No, vii.
'81. *' Challenger" lleport. Zoology. Vol, iii. Part IX.
Echinoidea.
Alcogx, T. :

'65. Notes on Natural History Specimens lately received from
Connemara. Proc. Lit. and Phil. Soc, Manchester, iv.,
pp. 192-208.

NiCHOi^ — A List o/L-ish Echinodei^nis, 235

Bailt, W. H. :

'65. Notes on Marine Invertebrata, collected on Portmamock
Strand. Proc. Nat. Hist. Soc, Dublin, iv., pp. 251-258.
Echinodermata, pp. 254, 255. •

Beauvovt, W. I. :

1900. The Fauna and Flora of Valencia Harbour on tbe West
Coast of Ireland. Part II. YII. Report on the Results of
Dredging and Shore-coUecting. Proc. R. I. Acad. (3), v.,
pp. 754-798.

Bklpast Katuilalists' Field Club :

'74. Guide to Belfast and the adjacent Counties. Echinodermata,
pp. 125-128.

Bkll, F. J. :

'86, First Report on the Marine Fauna of the South-west of

Ireland. Holothuroidea. Proc. R. I. Acad. (2), iv., pp.

620, 621.
'89. Echinodermata in Report of a Deep-sea Trawling Cruise off

the S. W. Coast of Ireland, under the Direction of Rev.'W.

Spotswood Qreen. Ann. & Mag. Nat. Hist. (6), iv.,

pp. 432-445, plates xviii., xix.
'90. Notes on the Echinoderms collected by Mr. Bourne in Deep

Water off the South-west of Ireland in H. M. S.

''Research." Joum. Mar. Biol. Assoc, (k. s.), i.,

pp. 324-326.
'91. AateriM ruhens and the British species allied thereto. Aon.

& Mag. Nat. Hist. (6), vii., pp. 469-479.
'92. On the Echinoderms collected by the S.S. ''Fingal" in 1890,

and by the S.S. "Harlequin" in 1891, off the West

Coast of Ireland. Sc. Proc. Roy. Dublin Soc. (k.s.), vii.,

pp. 520-529, plates xziii., xziv., xxt.
'92a. a Contribution to the Classification of Ophiuroids, with

descriptions of some new and little-known forms. Proc.

Zool. Soc, London, pp. 175-183, pis. xi., xii.
'92b. Catalogue of the British Echinoderms in the British

Museum (Natural History), London.

BxRirxrr, E. T. :

'27. Notice on a peculiar Property of a Species of Echinus.
Trans. linn. Soc. zt., pp. 74-77,

236 Proceedings of the RoyaJ Irish Academy.

BouKKE, G. C. :

'90. Report of a Trawling Cruise in H. M. S. " Research'* off
the South-west of Ireland. Jour. Mar. Biol. Assoc.
(k.s.), i., pp. 306-821, 826, 827.
Bbadt, G. S., and Robertson, D. :

'69. Kotes of a Week's Dredging in the West of Ireland. Ann.

& Mag. Nat. Hist. (4), iii., pp. 353-374.
'71. Description of two new species of British Holothuroidea.
Proc. Zool. Soc, London, pp. 690-692.
Beowne, E. T. :

1900. The Flora and Fauna of Valencia Harbour on the West
Coast of Ireland. Part I. I. Notes on the Pelagic Fauna.
Proc. R. I. Acad. (3), v., pp. 669-693.

Carpenter, W. B., Jeffrets, J. G., and Thokson, Wttillb :

'70. Preliminary Report of the Scientific Exploration of the
Deep Sea in H. M. Surveying- vessel ** Porcupine" during
the Summer of 1869. Proc. Roy. Soc. London, xviii.,
pp. 397-492.

Dtceib, G.:

'58. Remarks on the Distribution and Habits of JSehintu Uvidut.

Rep. Brit. Assoc, for 1852 (pt. 2), pp. 72, 73.
'58. Report on the Marine Zoology of Strangford Lough, County

Down, and corresponding part of the Insh Channel. Rep.

Brit. Assoc, for 1857, pp. 104-112. Echinodermata,

p. 111.
Farran, C. :

'52. On the Discovery of a new Irish Pontactes. Saundert* J^ewi-
Letter, March 16, 1852.
Foot, F. J. :

'60. Kotes on some of the Marine Animals to be met with on the
Shore at Spanish Point, Miltown Malbay, County of Clare.
Nat. Hist. Rev., vii. (Proc), pp. 392-394.
Forbes, E. :

'39. On the Asteriadae of the Irish Sea. Mem. Wem. Soc, viii.,

pp. 114-128.
'41. A History of British Starfishes, and other animals of the
Class Echinodermata. London.
Gamble, F. W. :

'96. Notes on a Zoological Expedition to Valencia Island, Co.
Kerry. Irirfi Naturalist, v., pp. 129-136.

Nichols— -4 Lial of Irish Eckinodenm, 237

Orseitb, J. Bray :

'58. Additions to tlie Irish Fauna. I7at« Hist. Bey., v. (Proc),
pp. 191, 192.

EADDoir, A. G. :

'86. Preliminary Eeport on the Fauna of Dublin Bay. Proc. B.

I. Acad. (2), iv., pp. 523-531. •

'86a. Becent Contributions to the Marine Inyertebrate Fauna of

Ireland. Zoologist (3), x., pp. 1-8.
'86b. First Beport on the Marine Fauna of the l^outh-west of

Ireland. Echinodermata (part). Proc. B. I. Acad. (2),

iv., pp. 618-620.
'88. Second Beport on the Marine Fauna of the South-west of

Ireland. Narrative of Cruise. Proc. B. I. Acad. (3), i.,

pp. 31-45.

HAnrss, Db. :

'53. Note on Holothuri». Ann. & Mag. Nat. Hist, (ir.s.), xi.,
pp. 155-157.

Hassali, a. H. :

'42. A List of Invertebrata found in Dublin Bay and its
vicinity. Ann. & Mag. Nat. Hist.yix., pp. 132-134«

Hebapath, W. B. :

'65. On the Genus Synapta, with some new British Species.
Quart. Joum. Micr. Science (n.s.), v., pp. 1-7.

Hkbdieak, W. a. :

'91. The Biological Besults of the Cruise of the s.y. "Argo"
round the West Coast of Ireland in August, 1890. Proc.
Liverpool Biol. Soc, v., pp. 181-212.

Holt, E. W. L. :

'92. Survey of Fishing Grounds, West Coast of Ireland, 1890-
1891. Beport on the Besults of the Fishing Operations.
Sc. Proc. Boy. Dublin Soc. (n.s.), vii., pp. 225-387.

Horn, W. E. :

'84. Beport on the Ophiuroidea of the Faroe Channel, mainly
collected by H. M. S. "Triton" in August 1882, with
some Bemarks on the Distribution of the Order. Proc.
Boy. Soc. Edinburgh, xii., pp. 707-730.

238 Proceedings of the Roijal Irish Academy.

HoTLB, W. E. :

< '85. A Reviged List of British Ophiuroidea. Proc. Eoy. Phys,

Soc. Edinburgh, viii., pp. 136-155.
'91. A Revised List of British Echinoidea. Roc. Eoy. Phys.

Soc. Edinburgh, x., pp. 398-436.

HUUFHBEYS, J. D. :

'45. Memoranda towards a Fauna of the County of Cork. Echino-
dermata. ,

HTin>xAy, G. C. t

'58. Report of tho Proceedings of the Belfast Dredging Com-
mittee. Rep. Brit. Assoc, for 1857, pp. 220-237.
'59. Report of the Belfast Dredging Committee. Rep. Brit.
Assoc, for 1858, pp. 282-291.

KrvAHAK, J. R. :

'53. Capture of the Spiny Cross-fish (U. glaeialis) in Dalkey

Sound. Zoologist, pp. 3990, 4021.
'59. On the Distribution of the Irish Echinodermata. Nat.

BSst. Rev., vi. (Proc), pp. 368-371.
'60. Report of the Committee appointed to dredge Dublin Bay*

Rep. Brit. Assoc, for 1859, pp. 80, 81.
*61. Report of the Committee appointed to dredge Dublin Bay.

Rep. Brit. Assoc, for 1860, pp. 27-31.
'63. Kotes on the Marine Fauna of the Coast of Clare. Proc.

Nat. Hist. Soc, Dublin, iii., pp. 99-103.

Leach, W. E. :

'12. Echinus lithophagus. Tilloch's Phil. Mag., xxxix., p. 151.

Ltkan, T. :

'82. "Challenger" Report. Zoology. Vol. v. Part XIV.
Ophiuroidea*

Mackiittobh, H. W. :

'78. British Association Guide to the County of Dublin. Part 11.

Echinodermata, pp. 7, 8.
'84. Report on Irish Zoophytes. Part I. Proc. R. I. Acad. (2)^

iv., pp. 52-58.

MofiB, A. G. :

'70. Report on the Collections made in Kerry [during the sum-
mer and autumn of 1868 J. Joum. Roy. Dublin Soc, t.,
pp. 389-395.

Nichols — A List of Irish Echinoderms. 239'

NoBMAKy A. M. :

'56. Note on Comatula rosacea. Zoologist, p. 5288.

'65. On the Genera and Species of British Echinodermata* Ann..

& Mag. Nat. Hist. (3), xr., pp. 98-129.
'93. Holothuria nigra^ Gray, and its Synonymy. Ann. & Mag.

Nat. Hist. (6), xii,, pp. 409-411.
'93a. Cueumaria montagui (Fleming) and its Synonymy. Ann. &

Mag. Nat. Hist. (6), xii., pp. 469-473.

Sladsk, W. p. :

'89. •* Challenger " Report. Zoology. Vol. xxx. Asteroidea^
'91. Keport on a Collection of Echinodermata from the South-
West Coast of Ireland, dredged in 1888 by a Committee
appointed by the Eoyal Irish Academy. Proc. E. L
Acad. (3), i., pp. 687-704, plates xxv-xxix.

TncpLEiov, B. :

'36. A Catalogue of the Species of Annulose Animals, and of
Bayed Ones, found in Ireland, as selected from the
Papers of the late J. Templeton, Esq., of Cranmore, witli
Localities, Descriptions, and Illustrations. Mag. Nat.
Hist, ix., pp. 233-240;

Th6el, H. :

'82. "Challenger" Beport. Zoology. Yol. iv. Part XIII.

Holothurioidca (Part I).
'86. " ChaUenger" Beport. Zoology. Yol.xiv. Part XXXIX.
Holothurioidea (Part II).

Thompson, W. :

'40. Contributions towards a Knowledge of theMoUusca nudibran«

chia and MoUusca tunicata of Ireland, with Descriptions of

some apparently new species of Invertebrata. Ann. & Mag.

Nat. Hist., v., pp. 84-102.
'40a. Additions to the Fauna of Ireland. Ann. & Mag. Nat.

Hist., v., pp. 245-257.
'44. Beport on the Fauna of Ireland : Div. Invertebrata. Bep.

Brit. Assoc, for 1843, pp. 245-291.
'44a. Additions to the Fauna of Ireland. Ann. & Mag. Nat.

Hist., xiii., pp. 430-440.
'45. Additions to the Fauna of Ireland. Ann. & Mag. Nat. Hist.,

XY., pp. 308-322.

240 Proceedings of the Royal Irish Academy.

Thoxfsok, W. :

'46. Additions to the Fauna of Ireland, including a few species

unrecorded in that of Britain, with the Description of an

apparently new Glossiphonia. Ann. & Mag. Nat. Hist.,

xviii., pp. 883-397.
'47. Additions to the Fauna of Ireland. Ann. & Mag. Nat. Hist.,

. XX., pp. 169-176.
'53. Supplementary Beport on the Fauna of Ireland. Bep. Biit.

Assoc, for 1852, pp. 290-296.
'56. The Natural History of Ireland, vol. iv.

Thomson, Vyville :

72. On the Crinoids of the "Porcupine" Deep-sea Dredging

Expeditions. Proc. Boy. Soc. Edinburgh, vii., pp. 764—

773.
'73. The Depths of the Sea. London.
'74. On the Echinoidea of the "Porcupine" Deep-sea Dredging

Expeditions. Phil. Trans., cbdv., pp. 719-756.

"Weight, E. P., awd Obeeks, J. B. :

'59. Beport on the Marine Fauna of the South and West Coasta
of Ireland. Bep. Brit. Assoc, for 1858, pp. 176-181.

HOLOTHUBIOIDEA.

Family — Stnaptid-s:.

Synapta inhserens (0. F. Miiller).

i., ii., . iv., v., vi.

i. Shores of Belfast and Strangford Loughs, in mud banks, at, and a
little above, low-water mark, abundant (Wyv. Thomson in
Quart. Joum. Microscop. Science ii. (ir. s.) 1862) : Holywood,
Belfast Lough, in considerable numbers in the sand banks
(Belfast N. F. C. Guide '74).
. ii. Balbriggan, on the beach, a specimen found by Mrs. Hancock after
a storm (Thompson '56) : Malahide (Jacob and Dixon, f . Haddon
'86) : Sandycove, Dublin Bay (Kinahan in Nat. Hist. Bcv.
vii., 1860).

Nichols — A List of Irhh Echiuoderma. 241

iy. Boyal Iiisli Academy Exp. 1885, Berebaven, 10 fms. (Bell '86) :
Yalentia Hm-bour, and off Bray Head, 45 fms. (Beaumont
1900).
T. Birterbuy Bay, 12-14 fms., not uncommon, and Clew Bay, 4 fms.

(Brady & Kobert'son '69).
tI. Mulroy Bay, Donegal (Praeger in Irisb Naturalist, 1894).

Distributum, — Both sides of the North Atlantic. Mediterranean.

Synapta digitata (Montagu). {Chirodota digitata Forbes "British

Starfishes.")

i., • . iv., v., vi.

2. A specimen on the sand between tide-marks near Carrickfergns
Castle (Hyndman and Thompson, f. Thompson '56) : Cairiok-
fergus (Herapath '65 sub Synapta thomsanii n. sp.).
iv. Dingle Bay, 36 fms. (Beaumont 1900).
V. Boyal Dublin Society Fishing Survey 1890, Eoundstone Bay,

5 fms. (Bell '92).
Ti. R.D. S. Fish. Survey 1890, Killybegs, shallow water (Bell '9 2).
Distribution, — ^British coasts to Mediterranean. N. America.

Family — Dendsochihota.

Cucumaria hyndmani (Thompson).

i., ii., . [iv.], v.. .

i. Belfast Lough (Hyndman, f. Thompson '40).
ii. Dublin Bay (Kinahan '61).

[iv.] •' Porcupine " Exp. 1869, 251 fms. (BeU '92 B) : R. I. A. Exp.
1888, off Dursey Head, 345 fms., a single young example (Sladen
'91).
T. Boundstone (M'Calla, f. Thompson '56) : Killary Bay, numerous
(Forbes '41) : " Argo " Cruise 1890, west of Ireland (Herdman
'91).
Distribution, — Norwegian and British coasts to Spain. Medi-
terranean.

Cucumaria planoi (Brandt).

. • • iv., . .

iv. Yalentia Harbour (Beaumont 1900).

Distribution. — ^British coasts to Mediterranean.

242 Proceedings of the Royal Irish Academy.

Cncumaria pentactes (? Linne).
i., ii., iii., iv., v., vi.

i. Belfast Lough (Getty, f. Thompson '56) : Bangor (Drummond, £.

Thompson '56): Donaghadee, 10 fms. (Bmmmond, f. Thompson

'56 sub Cucumaria fusiformia) : Strangford Lough, 15 fms., very

rare (Dickie '58 sub Cfusiformis),
ii. Dublin Bay (Kinahan '61 sub C, fusi/arniis) : Malahide (Dublin

Mus.).
iii. Dungarvan Bay (A. R. N.) : R. I. A. Exp. 1885, off Glandore,

40 fms., and off Baltimore, 30 fms. (Bell '86).
ir* Bantry Bay, 15-30 fms. (M'Andrew, f. Thompson '56): R.I. A. Exp.

1885, mouth of Kenmare River, 44-47 fms. (Bell '86) :

R. D. 8. Fish. Survey 1890-91, Kenmare River, 10-21 fms.

(Holt '92) : Valentia Harbour (Beaumont 1900).
V* Lahinch, co. Clare (Kinahan '63).

vi. R. D. S. Fish. Survey 1890, Inver Bay, 6-10 fms. (Holt '92).
DistrilmUon. — Arctic seas to Mediterranean.

Cuoumaria lactea (Forbes & Goodsir). {Ocnus laeteus and
0, hrunnem Forbes B. S.)

i., . iii., iv., v., .

i. Korth-east coast (Thonapson '56) : Belfast Lough (Drummond, f.

Thompson '56 sub 0. hrunnetu) : Strangford Lough, common

(Thompson '56 and Dickie '58, sub 0. hrunneus),
iii. Glandore (Allman, f. Thompson '56).
iv. R.I.A. Exp. 1885, Berehaven, 10 fms. (Bell '86): Valentia

(Gamble '96, Beaumont 1900).
V, Lahinch, co. Clare (Thompson '56) : ** Argo " Cruise 1890, west of

Ireland (Herdman '91 sub 0. brunneus),
Distrihution, — Norway. Sweden. British coasts, Brittany.
N. America.

[Cucumaria hispida (Barrett).]

V. '< Porcupine" Exp. 1869, 422 fms. (Theel '86 sub Echinocueumis
typica).
Distribution, — Arctic seas to Bay of Biscay. West Indies.

Nichols — A Lut o/Ifkh Echinoderma. 243

Cucnmaria andrewsi (Farran).

. . iii., . ?v., •

iii. donea, Dungarvan, two specimeiis (Farran '52).
?v. A species dosely allied to, if not identical with Thyone andretoiii,
on north-west coast of co. Clare (Einahan '63).
Distribution. — Ireland.

Cucnmaria saxioola, Brady & Hobertson.

V. Birterbuy Baj, a specimen dredged in 15 fms., and six specimens
in holes of limestone boulders between tide-marks, Westport,
CO. Mayo (Brady & Eobertson '71).
Distribution. — W. Ireland.

Thyone fusus (0. F. Miiller). {Thyone papillosa Forbes B. S.)
i., ii., . iv., v., .

i. Belfast Lough (Thompson, f. Forbes '41) : off Lame, 70-90 fms.,
two specimens (Hyndman '£t9) : Strangford Longh (Thompson,
f. Forbes '41 ; Hyndman and Thompson, f. Dickie '58 ; Belfast
N. F. C. Guide '74) : among Killinchy oysters in Belfast market
(Thompson '56).

ii. Dublin Bay (Kinahaii '60).

iv. Bantry Bay, 15-30 fms. (3I*Andrew, f. Thompson '56): E. I. A.
Exp. 1885, Berehaven (Bell '86).

T. Boundstone, numerous (M'Calla, f. Forbes '41): Birterbuy Bay,
12-14 fms., not uncommon (Brady & Ilobei-tson '69) : XiUary
(Forbes '41).
Distribution. — ^Loffoden to British Isles. France. Mediterranean.

Thyone raphanus, Diib. & Kor.

. . . iv., . .

iv. Bantry Bay, 15-30 fms., a specimen (M^Andi-ew, f. Thompson '56) :
R. I. A. Exp. 1885, Berehaven, and off Great Skellig, 70 fms.
(Beir86): R. D. S, Fish. Survey 1890, Dingle Bay, 7 fms.
(Bell '92) : Dingle Bay, 30-40 fms. (Beaumont 1900).
Distribution* — Sweden. Norway. British Isles. Brittany*
Mediterranean.

244 Proceedings of the IU>yal IiHsh Academy.

Thyone porthchii, described by Forbes ('41) from a specimen found
by Capt. Portlock in Belfast Lough, was regarded by Th6el as
possibly identical with Phyllopkarus drummondi, but is beliered
by Norman ('93 a) to be a synonym of Cueumaria tnantayui
(Fleming), which may be a distinct British species (Maren-
zeller in Ann. and Mag. Nat. Hist. (6) xii., 1893 and
Norman '93 a).

Psolus phantapus (Struss.).

i. Bangor, co. Down, a single specimen (Thompson '56) : N. E. coast
(Belfast N. F. C. Guide '74).
A young specimen of P. phantapus, or F, fabricii, was dredged in
500 fms., off Black Rock, Blacksod Bay, in K. D. 8. Fish. Survey 1891
(Bell '92).

Distribution. — Arctic seas to British Isles and New England.

Phyllophoros pellucidus (Diib. & Kor.). (Cueumaria hyalina
Forbes B. 8.)
i., . . iv., . .

i. Strangford Lougli. 15-20fms. (Hyndman and Thompson, f. Dickie
'58 sub C, hyalina),
iv. K.I. A. Exp. 1888, 50 fms. (Sladen '91).

Distribution. — ^Arctic seas to British Isles and West Indies.

PhyllophoniB dnunmondi (Thompson). ( Cueumaria drummondii, &c.y

Forbes B. 8.)

i., . iii., • v., .

i. North of Ireland (Thompson '56 sub Cueumaria communis) t
Belfast Lough (Brummond, f. Thompson '40).
iii. Youghal (Ball, f. Forbes '41 sub C. communis ; Humphreys.
'45 sub C. communis).
V. Roundstonc (M*Calla, f . Thompson '56 sub C. communis).
Distribution. — Loffoden to British Channel.

Thyonidium duheni Norman, which is stated by M*Intosh in " The
Marine Invertebrates and Fishes of St. Andrews," 1875, to have
, been found by Norman on the coast of Ireland, is probably to be
referred to P. drummondi.

Nichols— -4 List of Irish Echinoderms. 245-

Family— AspiDocHiEOT^.

[Holothnria intestinalis, Asc. & Bathke.]

. . . iv., v., .

IT. R I. A. Exp. 1888, 750 fms. (Sladen '91).
T. R. D. 8. Pish. Survey 1890, off Achill Head, 220 fms., probably
(BeU '92).
Distrihutian, — ^Finmark to British Isles. Bay of Biscay. Azores.
Mediterranean. W. Indies.

Holothnria tremnla, Gunneros.
. . . iv., [v.], [vi.]

iv. B.I.A. Exp. 1886, 110-325 fms. (Haddon '88): E. I. A. Exp.
1888, 346 fms. (Sladen '91) : ** Flying Fox" Exp. 1889, 100-
fms. and 315 fms. (Bell '89) : H. M. 8. ** Research " 1889, 70-
400 fms. (BeU '90): R. D. 8. Fish. Survey 1891, off Bolus
Head, 220 fms. (Holt '92).
[v.] R.D.S. Fish. Survey 1890-91, off AchiU Head, 127-220 fms-
(Holt '92, BeU '92).

[vi.]'R^D.S Fish. Survey 1891, 45 miles off Black Rock, Blacksod
Bay, 375-500 fms. (Holt '92, BeU '92).
Diitrihution, — ^Norway to Spain.

Holothnria forskahU, DeUe Chiaje. {M. nigra BeU B. M. Gat.)

. . . iv., v., vi.

iv. Kenmare River (Bell '92 b) : Valentia (Haines '53 sub IT.
iuhuhsa; Noiman'93; Gamble '96; Beaumont 1900; Dublin
If us.) : R. D. S. Fish. Survey 1890, Dingle Bay, 4 fms. (BeU
'92).
V. MUtown Malbay (Foot '60) : West coast of Ireland (Todhunter, f.
Thompson '56):-R.D.S. Fish. Survey 1891, Casheen Bay,
7 fms., Cleggan Bay, 4-9 fms., andDavalaun Sound, 13-16 fms.
(Holt '92, BeU '92, '92 b): ** Argo" Cruise 1890, Inishbofin,
a single specimen, between tide-marks (Herdman '91).
vi. R. D. S. Fish. Survey 1891, Donegal Bay, 30 fms., and KiUybcgs
Harbour, 14-16 fms.' (Holt * '92): ?Tory Island, a specimen
(Hyndman, f . Thompson '56).
Diiirilutian. — W. Ireland and S. W, England to Mediterranean.

.*^a> f icw^a*^ of the Boyal Iruk Academy.

[^Hidotiniiut asperm. Bell.]

Vv >3^ F.-x ** Exp. 1889, 1000 fms., a smgle specimen (Bell '89).
^, V^ •*:."*•- S. V. Ireland. Deep water.

[Stiehopm natans (H. Sars}.]

^ L A- E^P- 1888y 750 fms. (Sladen '91) ; doubtful if the specimen

w^ azi e:Eample of S. natam (BeU '92 b).
r**r ' > »: wj». — Loffoden. Xorway. S. W. Ireland.

Pamilj — ThtfiftTm,^

[LflBtmogone Tiolaeeft^ Theel.]

> • ■ *■ •> • •

^^ It L A. Exp. 1888, 750 fms. (Sladen '91).

r**}tnhution. — Faroe Channel. S. W. Ireland. Off Sydney
••* Challenger " Exp.

CRE^OIDEA.
Family — ^Aittedohiiub.
Antedon hiflda (Pennant). ( Comatmla rouMa Forbes BJ3.)
i., ii., iii., ir., v., .

i^,v. Generally distributed in these provinces, and probably also in
province vi., but there does not appear to be any record of its
having been found on the X. W. coast of Ireland. Two speci-
mens dredged at the unusual depth of 250 fms. in " Flying
Fox" Exp. 1889 ;BeU '89).
BUtrihution, — Shetliind to Mediterranean.

Antedon milleri (J. Miiller).

i. Belfast (Wyv. Thomson, f. Xorman '65).
DUtributiofi, — Arran. Mouth of Mersey. Milford Haven. Belfast.
It is very doubtful if this species can be satisfactorily diagnosed
(BcU '92 b).

[Antedon phalaaginm (J. MuUer).]
• . • iv.j • 9

iv. " Flying Fox " Exp. 1889, 1:50 fms., a single specimen (Bell '89).
I/utributiofu — ^Hebrides tq Madeira and Mediterranean.

Nichols — A List of Irish Echifiodeftns. 247

ASTEROIDEA.

Family— Abchasterib^.

Pontaster tennispinTU (Dub. & Kor.).

• • • *»•> • •

iv. ** Porcupine" Exp. 1869, off Valentia, 100-150 fms. (Sladen '89
sub P. tefiuUpinm var. platynota and P. limbatus) : B. I. A.
Exp. 1886, 214 fms. and 325 fms. (Haddon '88): "Flying Fox"
Exp. 1889. ?250 fms. and 316 fms. (Bell '89): H.M.8. "Re-
seai-ch" 1889, 90-400 fms. (Bell '90) : R.L A. Exp. 1888,
345 fms. (Sladen '91 sub P. limbatm).
Distribution. — Both, sides of North Atlantic.

[Plutonatter bifrons (Wyv. Thomson).]

iv. li. 1. A. Exp. 1888, 750 fms. (Sladen '91).
Distribution, — North Atlantic, Mediterranean.

Family — Astrop£Ctinid£.

Aitropecten irregularis (Pennant). {Aster ias aurantiaea Forbes B. 8.)

i., ii., iii., iv., v., vi.

Genemlly distributed round the coast from about 10 fms. to
1000 fms.

Dint ribut ion, — Scandinayia to Liberia.

[Astropecten spbenoplax, Bell.]

. . . . v., vi.

V. K. D. S. Fish. Survey 1890. off Achill Head, 600 fms. (Holt '92).
vi. R. D. S. Fish. Survey 1891, off Bkck Rock, Bkcksod Bay, 500 fms.
(BeU '92).
Distribution. — West Ireland. Deep water.

lt.I.A. FHOC.. VOL. XXIV., SEC. P.] Y

248 Proceedhu/B of the Royal Irinh Aaule.my.

[Psilaster andromeda (M. & Tr.).]

. . . iv., . vi.

iv. H,M.S. ** Research" 1889, 400 fms. (BeiroO): R. I. A. Exp.

1888 (Sladen '91).
vi. R. D. S. Fish. Survey 1891, off Black Rock, Blacksod Bay, 500
fms. (BeU '92).
DistriOution. — Scandinavia to Bay of Biscay. Azores. Cape Yerde.
N, America.

Luidia ciliaris (Philippi). {L, fragilimima Forbes B. S. (pai*s).)
. . iii., iv., v., vi.

iii. Youghal (Ball, f . Thompson '40 a ; Humphreys '45) : Glandoie,
notunfrequent (Allman, f. Thompson '56) : south coast of Ire-
land, not unfrequent (Wright & Greene '59).
iv., V. Generally distributed.

vi. R.D. S. Fish. Survey 1891, Donegal Bay, 32 fms. (Holt '92).
Distribution. — Faroe to Cape Verde. Mediterranean.
A single specimen of Bipinnaria aslenffera Sars (Larval stage of
Luidia) was tuken in Vulentia Harbour in ^Xovember 1895
(Browne 1900).

Luidia sarsi, Dub. & Kor. {L, fragillimina Forbes B. S. (pars).)
. . . iv., v., vi.
iv., V. Generally distributed.

vi. R. D. S. Fish. Sui-vey 1890-91, Donegal Bay, 33-37 fms. ( i ell '92),
Distribulian. — Norway to Cape Yerde, Mediterranean.

Family — Pentaqokastehidjj:.

[Pentagonaster granularis (Retz.).]

. . . iv., . .

iv. R. I. A. Exp. 1888, 750 fms. (Sladen '91 sub I\ balteatusn. sp.
and P. eoncinnm n. sp.).

Distribution. — Both sides of North Atlantic.

Ludwig regards P. halteatm and P. concinnus as possibly identical
uith P, hystricis Marenzeller of the Mediterranean and Bay of
Biscay, and not as synonyms of P. granularis^

NiCHoi^ — A List of Ii'Uh Echinoderina. 249

[Pentagonaster greeni (Bell).]

. . . iv., . .

iv. "Flying Fox " Exp. 1889. 1000 fms., one specimen (Bell '89).
DUtnhution. — S.W. Ireland. Deep water.

[Hymphaster subspinosus (Ferrier).]

. . . iv., . .

iv. H. I. A. Exp. 1888 (Sladen'91 snh Nymphatterprotefiius): "Flying

Fox" Exp. 1889, 315 fms., five specimens (Bell '89 sub N.

proUntus): H.M. 8. "Kesearcb " 1889, 400 fms. (Bourne '90).

DUtrihution. — S.W. Ireland. Canaries " Challenger." West Indies

"Bbike."

Family — GniKASTEann^.

Porania pulyilluB (0. F. MUller). {Gonimier templetoni Yovbcs B. S.)

i., ii., iii., iv., v., vi.

i. Belfast Lougb (Grimshaw, f . Templeton '36 sub Asterias equedris
Sow. ?) : Strangford Lougb, a specimen (Thompson '56).

ii. East of Ireland (Wright & Greene '59).

iii. Youghal (Wright & Greene '59 ; Bell '92 b) : Nymph Bank (Ball,
f. Forbes '41) : ?ll. D. S. Fish. Survey 1890-91, off Ballycottin
(Bell '92).

iv. Off Yalentia (Bell '92 b) :. E. 1. A. Exp. 1886, 70-80 fms.
(A. 11. N.) : H. M. S. " Besearch " 1889, 70 fms. (BeU '90).

V. Coast of Clai-u (Gabbet, f. Thompson '56; Dublin Mus.) : "Por-
cupine" Exp. 1869, 106 fms. (Skden '89): R. D. I<. Fish.
Sur%-ey 1891, Davalaun, 30 fms. (Bell '92).

vi. K.D.S. Fish. Survey 1891, Killybegs Hai-bour, 14-16 fms. (Bell
'92) : Tory IsUnd (Hyndman, f. Thompson '56).
JjUtribution, — Scandinavia to British coasts. Bay of Biscay.

Family — AsTEuiNiDiB.

Asterina gibbosa (Pennant).

i., ii., iii., iv., v., .

i. Coasts of Antrim and Down (Thompson '56) : Strangford Lough
(Thompson '56; Bell '92 b) : A i-dgbss (Thompson '56).

ii. Coast of Dublin (Thompson '56) : Lambay Island (Thompson '56 ;
Bell '92 B) : Dublin Bay (Mackintosh '78) : Grcytones (Mack-
intosh '84).

250 Proceedings of the Royal Irish Academy.

iii. Southern shores tBall, f . Thompson '56) : Youghal (Miss M. Ball, f .
Thompson '56) : Cork (Humphreys M5) : Kinsale and Glandore
(Ball, f . Thompson '56) : S. W. coast of Cork (Allman, f . Thomp-
son '56).

iv. K.I. A. Exp. 1885, Bantry Harhour, 4-6 fms. (Haddon '86 b) :
Bantry Bay (Dublin Mus.) : Valentia Harbour (Beaumont
1900): R. D. S. Fish. Survey 1890-91, Smerwick Harboiu-
(Bell '92, '92 b).

V. Kilkee (Lord Chancellor Brady, f . Thompson '56) : Lahinch (Thomp-
son '56 ; Bell '92 b) : Lahinch and Ballyvaughan, common
(Kinahan '63) : Clare (Dublin Mus., coll. by Miss O'Brien) :
Roundstone (M'Calla, f. Thompson *o6): Connemara (Alcock
'65) : *' Argo" Cruise 1890.1ni8hbofin (Herdman'91): R.D.S.
Fish. Survey 1890, Killeany Bay (Holt '92).
Distribution, — European and N. African coasts. MediteiTanean.

Palmipes placenta (I'cnnant). {P. membranaeeus Forbes B. S.)

i., ii., iii., iv., . .

i. Off Larne, 70-90 fms., a very small specimen (Hvndman '59) :
Belfast Lough (M'Calla in 8th Ann. Rep. Dublin Nat. Hist.
Soc, 1845-6; Thompson '56) : Ballywalter, co. Down (Bell
'92 b) : Strangford Lough (a specimen, Templeton '36 ; Thomp-
son '56 ; rare, Dickie '58).

ii. DundrumBay, numerous (Kinahan *59) : a specimen dredged about
7 miles off the Dublin coast (Warren, f. Thompson '56).

iii. Youghal (a specimen, Ball, f. Forbes '41 ; Humphreys '45).

iv. Kenmare River (Dublin Mus., coll. by Miss Birch ; Bell '92 b) :
R. D. S. Fish. Survey 1890, Kenmare River, 10 fms. (Holt
'92) : H. M. S. ** Research" 1889, 70 fms. (Bell '00).
Distribution. — British and French coasts. Mediterranean.

Family — SxiCHASTKRiDJi.

Stichaster roseus (0. F. Miiller). {Cribella rosea Forbes B. S.)

. ii., iii., iv., [v.], .

li. East of Ireland (Wright & Greene '59): Dublin Bay (Kinahan '61).
iii. Nymph Bank (Ball, f . Thompson '40 a) : Youghal (Humphreys
'45 } Wright & Greene '59). *

NiCHiLS — A List of Irish Echinodenm. 251

iv. B. I. A. Exp. 1888, 50-54 fms. (Sladen '91) : H. M. 8. " Ee-

search " 1889, 200 fms. (BeU'96) : S. W. coast of Ireland, 55

fms. (BeU '92 b) : S. W. Ireland (Orenfell, f . BeU '92 b).

[v.] R. D. 8. Fish. 8urvey 1890, off AchiU Head, 144 fms. (Holt '92).

Distribution, — Norwegian and British coasts. Boscoff. Bay of

Biscay.

[Heomorphaster talismani (Ferrier). {N. eustiehus'^QW. £. M. Cat.)]

iv. B.I. A.Exp. 1888, 750 fms. (Sladen '91).

Distrihution, — S. W. Ireland. Bay of Biscay. Azores. Canaries.
Beep water.

[Zoroaster fnlgens, Wyy. Thomson.]

. . . iv., . .

iv. B. I. A. Exp. 1888, 750 fms. (Sladen '91).

Jhstrtbution. — ^Both sides of the Atlantic. Deep water.

Family — SoLASTEBiD-fi.

Solaster papposus (Fabr.).

i., ii., iii., iv., v., vi.

Generally distributed all round the coast and common.
Distribution. — Arctic seas to coast of France and to New EngLmd.

Solaster endeca (Linni).

i., ii., iii., . v., .

i. Coasts of Antrim and Down (Thompson, f. Forbes '41) : Belfast
Lough (M'Callu in 8th Ann. Bep. Dublin Nat. Hist. Soc,
1845-6 ; Thompson '56 ; BeU '92 b) : Strangford Lough (Dickie
'58 ; BeU '92 b).
ii. DubUn Bay (BaU, f. Forbes '41 ; HassaU '42 ; Mackintosh '78).
iii. Toughal (BaU, f. Forbes '41 ; Humphreys '45).
v. B. D. S. Fish. Survey 1890, Clew Bay, 15 fms. (Holt '92).
Distribution, — Arctic seas to British Isles and to Nova Scotia.

Family — Pt£ea8T£bid£.
[Pteraster personatas, Sladen.]

iv. B. L A. Exp. 1888, 750 fms. (Sladen '91).

Distributum.—^. W. Ireland. Bay of Biscay. Deep water.

252 ProcecdingH of the Royal Ivkli Academy,

[Hymenaster giganteus, Sladcn.]
• • • iv»> • •
iv. R. T. A. Exp. 1888. 750 fms. (Sladen '91).
Distribution, — S. W. Ireland, Deep water.

Family — Echikastkkjo2e:.

Henricia Bangninolenta (0. F. MtiUer). ( CriheUa octdata Forbes B. S.)

i., ii., iii., iv., v., .

i. North of Ireland (Thompson '56) : coast of Antrim (Ord. Surv.
Coll. in Dublin Mus.) : off Larne, 70-90 fms. (Hyndman '59) :
Strangford Lough (15-20^ fms., Hyndman and Thompson,
f . Dickie '58 ; 5 fms., Bell '92 a).

ii. Generally distributed.

iii. Youghal (Ball, f. Forbes '41 ; Humphreys '45).

iv., V. Generally distiibuted in these provinces ; and probably also on

the X. W. coajit of Ireland, but I am not aware of any record

of its occurrence. Yar. ahys$ieola, Nonnan. R. I. A. Exp. 1888^

750fm8. (Sladen'91).

Distribution. — Arctic seas to British coasts and New England.

Bay of Biscay. Azores. ? Mediterranean.

Family — Astekiid-f..

Aflterias glacialLi, Linne. ( Uraster ghcialis Forbes B. S.)

i., ii., iii., iv., v., vi.

i. N. E. Ireland, a few small specimens (Thompson '56) : Belfast

Lough (Drummond, f. Forbes '41 ; Kinahan '59 ; Bell '92 b) :

Strangford Lough, occasionally inside and outside (Dickie '58).

ii. Dublin Bay, 12-30 fms., very common (Kinahan '59): Dalkey

Sound (Kinahan '53 ; Corrigan in Nat. Hist. Rev. i., 1854).
iii., iv., v., vi. Generally distributed.

2>w<rf^tt<io».— Finmark to Cape Verde. Mediten-anean.

Asterias mbens, Linn^. ( Ur aster rubens and U. violaeeaYovhes B. B.)
i., ii , iii., iv., v., vi.

Generally distributed all round the coast and common. Dredged by
H.M.8. " Research " 1889, at the great depth of 200 fms. (Bell '90).
Distribution. — Finmark to Senegal. ? Mediterranean.

Nichols— ^i Liai of Irtish Evhinoderms. 263

Aflterias nmrrayi. Bell.
.... v., .
V. R. D. S. Fish. Survey 1891, west coaat of Ireland (BeU '92, '92 b).
2)i><rf*tt<i*o».— West coasts of Scotland and Ireland,

Aflterias hispida, Pennant. ( Ur aster htspida Forbes B. S.)
i., ii., iii., iv., v., .
i. Belfast Lough (Portlock, f. Forbes 41) : coast of Down (Thompson
'56 ; Bell '92 b) : Strangford Lough and Channel, occasionally
(Dickie '58).
ii. East of Ireland (Wright & Greene '59) : Dublin Bay (Kinahan '59).
iii. Trabulgan, co. Cork, a single young specimen (Greene '58).
iv. South-west Ireland (Wright & Greene '59).
V. North-western coast of co. Clare (Kinahan '63) : Roundstone
(M'Calla, f . Thompson '56) : west of Ireland (Wright & Greene
'59).
l)istrih$U'on,— British coasts. Sweden (Lonnberg).

Family — Bkisingid^.
[Brisinga endecacmenos, Asbj.]

• • . IV. , • •

iv. Off Valentia and ** Porcupine" Exp. 1869, 458 fms. (Thomson
'73) : R.L A., Exp. 1886, 325 fms. (Haddon '88).
DUtrihution, — Norway to Spain.

[Brisinga ooronata, G. 0. Sars.]

iv. ** Porcupine" Exp. 1869, 458 fms. (Sladen '89) : R.I.A. Exp.

1888, 345 fms. (Sladen '91) : "Flying Fox" Exp. 1889, 1000
fms., an injured specimen (Bell '89) : H. M. S. "Research"

1889, 200 fms., fragments only (Bell '90, Bourne '90).
Di$tr%buti<m, — ^Loffoden to Azores. Mediterranean.

OPHIUROIDEA.

Family — Ophiolepidid^.

Ophiora ciliaris (Linn6). ( 0. texturata Forbes B. S.)

i., ii., iii., iv., v., vi.

Generally distributed and common ; very large specimens are often

obtained off the S. W. coast of Ireland.

Di9iributian. — Eastern side of North Atlantic. Mediterranean.

864 Proceedings of the Royal Irish Academy.

OpMnra albida, Forbes.

i., ii,, iii., iv., v., vi.

Oenerallj distributed and common.
JHstribution, — North European seas. Mediterranean.

Ophiura sand, Lutken.

■ • • r 1 « *! • •

?iv. ** Porcupine/' Exp. 1869, 75 fms., a young specimen (Hoyle
'84).
Distribution, — Both sides of North Atlantic.

[Ophiura ngnata (Yerrill).]
• ■ • r 1 v.| • •

f iv. E.I. A. Exp. 1888, 345 fms. (Sladen '91, but rather doubtful of
the determination).
Distribution, — N, E. America. Faroe Channel. ? 8. W. Ireland.

Ophiura affinis, Lutken.

iv. B. I. A. Exp. 1885, Berehaven, mouth of Bantry Bay, 40 fms.,
and off Great SkelHg, 110-120 fms. (Haddon '86 b) : R. I. A.
Exp. 1886, mouth of Bantry Bay, 37^ fms. (Haddon '88).
Distribution, — ^Both sides of North Atlantic. Mediterranean.

[Ophiocten sericeum (Forbes).]

. . . iv., . .

iv. H.M. S. "Research" 1889, 400 fms. (BcU '90).

Distribution. — Eastera side of N. Atlantic. Arctic seas. Off
Massachusetts (Lyman). ? Marion I. " Challenger."

Family — AMPHnrRiDJE.

[Ophiomusium lymani, Wyv. Thomson.]

. . . iv., . .

iv. " Porcupine" Exp. 1869, 180 fms. (Thomson '73).

i^M^ri^^ilwt. —Atlantic and Pacific Oceans. Deep water.

Nichols — A List of Irish Echiiwdenns, 255

Ophiocnida braeUata (Montagu). {Ophiocoma hrachiata

Forbes B. S.)

i., ii., . iv., . .

i. Belfast Lough, a young specimen (Thompson, f. Forbes '41) :
Strangford Lough, in rock -pools (Thompson, f. Forbes *41).
ii. Off Dundnim, co. 1 Jown, two specimens (Hyndman and Thompson, f ,
Thompson '56) : several specimens from stomachs of haddock
taken off Newcastle (Thompson '56),
iv. Kenmare River (Sir P. Egei-ton, f. Bell '92 b).

Distribution, — North European seas. Mediterranean.

Amphiura chifgii, Forbes.

i., • • iv., v., .

i. Bangor, co. Down (Bell '92 b),
iv. "Porcupine" Exp. 1869,75 fms. (Hoyle '84) : 11. L A. Exp.
1885, Bcrehaven, 5-12 fms., common (Haddon '86 b).
V. Killary Sound, very abundant (Norman, f. Hoyle '85).
Distribution. — North European seas. Mediterranean.

Amphiura filiformis (0. F. Miiller). ( Ophiocoma filifonnis
Forbes B. S.)

i., ii., iii., iv., v., vi.

L This species was apparently dredged from 50 fms. off South Book,
CO. Down (Thompson '56) : off Strangford, rare (Dickie *58) :
in quantity from haddock off Killough (Thompson '56).

ii. From stomachs of haddock taken off Newcastle, co. Down
(Thompson '56) : Dublin Bay, a single specimen (Hassall'42) :
Kingstown Hai'bour (Ball, f. Thompson '56).

iii. R. I.A. Exp. 1886, off Ballycottin, 39 fms. (A. R. N.):
Courtmacsherry Harbour (Allman, f. Thompson '56) : R. I. A,
Exp. 1885, off Gkndore, 40 fms. (Haddon '86 a).

iv., V. Generally distributed.

vi. North-west Ireland (Wright & Greene '59).

Distribution, — North European seas. Mediten-anean.

256 ProceeduujH of the Royal Irish Acadeinf/,

Amphiora elegans (Leach). (Ophtocoma negJecta Forbes B. S.)

i., ii., iii., iv., v., vi.

i. Coasts of Antrim and Down (Templeton '36 sub Ophiura minuta ;
Hyndmnn and Thompson, f. Forbes '41): Belfast Lough and
Bangor (Thompson '56) : Strangford Lough, abundant
(Hyndman and Thompson, f. Forbes '41) : Strangford Channel,
25 fms., rare (Dickie '58) : common between tide-marks at
Ardglass (Thompson '56).

ii. Common between tide-marks at Annalong (Thompson *56) : Kush
(Duerden in Irish Naturalist, 1894) : Lambay (Thompson '56) :
Dublin Bay (Kinahan '59 ; Mackintosh '78) : Greystones
(Mackintosh '84).

iii. Ardmore (Kinahan '59): Trabulgan, co. Cork (Greene '58 sub
Amphiura leachii n. sp.).

iv. R. LA. Exp. 1885, Berehaven (Haddon '86 b) : Valentia
(Kinahan '59; Beaumont 1900): R.I.A. Exp. 1886, lOa
fms. (A. R. N.) : S. W. coast, 70 fms. (Bourne, f. Bell '92 b).
V. Common on the west coast (Forbes '41): Lahinch (Thompson
'56) : north-west const of co. Clare (Kinahan '63) :
Roundstone (Dublin Mus.) : Killary Bay, 3-12 fras., and Clew
Bay, 3-10 fms. (Thompson '56).

vi. Tory Island (Hyndman, f . Thompson '56).

Distribution. — North Atlantic. Mediterranean. Cape of Good

Hope and Australia ** Challenger."

Ophiaotis balli (Thompson). ( Ophioeoma hallii and 0, goodtiri

Forbes B. S.)

i., ii., iii., iv., . .

i. Giant's Causeway, adherent to Pinna (Bell '92 b).
ii. Dublin Bay (Kinahan '61) : Dalkey Sound (Ball, f, Thompson '40 ;.

Bell '92 b).
iii. Nymph Bank (Thompson '56): R.I. A. Exp. 1886, off

Ballycottin, 39 fms. (A. R. N,).
iv. R. I. A. Exp. 1885, 80 fms. (Haddon '86 b) : Dingle Bay, 30-40
fms. (Beaumont 1900).
DistrihxUion, — Scandinavian and British coasts. Bay of Bisc^iy
rKochler).

NicHoi^ — A Lint of Inah Ec/iitiodertns. 267

Ophiopholifl acnleata (LinnS). {Ophiocoma hellis Forbes B. S.)

i., ii., iii., iv., v., .

i.-v. Generally distributed, but does not seem to have been recorded
from province vi.
Distribution, — Both sides of N. Atlantic from Arctic Ocean to
British Isles and New England.

Family — Ophiocomid^.

Ophiocoma nigra (Abilg.). ( 0, granulata Forbes B. S.)

i., ii., . iv., v., vi.

i. Belfast (Kinahan '59) : const of Down (Thompson '56) :

Strangfoi-d Lough (Thompson, f. Forbes '41; Dickie '58;

Bell '92 b).

ii. Dublin Bay, common : Grey stones (Mackintosh '84).

iv. Bcrehaven (Dublin Mus.) : 11. 1. A. Exp. 1885, Berehaven, 5-12

fms., common (Haddon '86 b): R. I. A. Exp. 1886, 70-80

fms. (Haddon '88) : Valentia (Kinahan '59; Beaumont 1900).

V. R. D. S. Fish. Survey 1890-91, Kilkieran Bay, 4-8 fms., and

Birterbuy Bay, 7-13 fms. (Holt '92).
vi. R. D. S. Fish. Survey 1891, off MalinHead, 22-23 fms. (Bell '92).
Distribution, — Korth European seas.

Ophiopsila annnlosa (Sars).

. . . iv., v., .

iv. Kenmare Bay (More, f. Norman in Hoyle '85).
V. Birterbuy Bay, a single fragment of a ray (Brady & Robertson
'69).
Distribution, — W. Ireland. Mediterranean.

Family — Ophiothwcid^.

Ophiothrix fragilis (Abilg.). ( Ophiocoma rotula and 0. minuta
Forbes B. S.)

i., ii., iii., iv., v., vi.

Generally distributed all round the coast and very common, often
on oyster- and other shell-banks.

Distribution. — Finmark to Mediterranean. Azores " Hirondelle."

258 Procee(li»i(j8 of the Royal Irish Academy,

Ophiothriz Itttkeni, Wyv. Thomson.

. IV.,

IT. "Porcupine" Exp. 1869, 180 fms. (Thomson 73): R. I. A.
Exp. 1885, 80 fms., two specimens (Haddon '86b) : R. I. A.
Exp. 1886, 93-160 fms. (Haddon *88): ** Flying Fox" Exp.
1889, 200-315 fms. (Bell '89, '92b): H. M. S. "Research"
1889, 200 fms. (Bell '90 sub 0. pentaphyllum, '92 b) :
R. D. S. Fish. Survey 1891, off Bolus Head, 115 fms. and 220
fms. (Bell '92).
Distribution, — S. W. Ireland. Azores •* Challenger."

SxREPTOPHiraJE.

[Ophiobyrsa hystriois, Lyman]

. . . iv., v., .

iv. H.M. S. "Reseai'ch" 1889, 400 fms., one specimen (Bell '90) :
?R. I. A. Exp. 1888, 345 fms., fi-agments of rays only (Sladen
'91) : " Flying Fox " Exp. 1889, one sijecimen, 315 fms. (Bell
'92a).
V. R. D. S. Fish. Survey 1890, off Achill Head, 220 fms., a young
form, which may, perhaps, be referred to this species (Bell
'92).
Distribution, — Faroe Clianuel and off western coasts of Ireland.
Deep water.

ECHINOIDEA.

Family — Cidakid.e.

Cidaris papillata, Leske.

. ii., . iv., [v.], [vi.]

ii. "Porcupine " Exp., off Wexford, 30-40 fms. (Bell '92b).

iv., [v.], [vi.] Generally distiibuted in the deep-water marine ai-ea
off the western coasts of Ireland, and in some places very
abundant. Dredged in 93 fms. in R. I. A. Exp. 1886
(Haddon '88).
Distribution. — Atlantic and Pacific Oceans.

Nichols — A Lkt of Irish Echinodenns. 259

[Cidarifl gracilis (Sladen).]

. . . iv., . •

iv. R. I. A. Exp. 1888, 750 fms., a single specimen, probably imma-
ture (Sladen '91) ; perhaps when a mature specimen is found,
it will be seen to be only C, piirpurata Wyv. Thomson (Bell
'92b).
Ih'strihutian. — S. W. Ireland. Deep water.

Family — Echinothxtriidjb.

[Aflthenosoma hystrix (Wyr. Thomson).]

.... v., vi.

V. " Porcupine " Exp. 1869, off west coast of Ireland (Thomson 74).
vi. R. D. S, Fish. Survey 1891, 45 miles off Black Kock, Blacksod
Bay, 500 fms. (Bell '92, '92b).
Bistrihutton. — Hebrides to Portugal and Azores. W. Indies.

[Phormosoma placenta, Wyv. Thomson.]

. . . iv., v., vi.

iv. ** Flying Fox" Exp. 1889, 1000 fms., six specimens (Bell '89) :

R. I. A. Exp. 1888, 750 fms. (Sladen '91).
v. ** Porcupine" Exp. 1869, off west of Ireland, fragments and
spines (Thomson '74) : R. D. S. Fish. Survey 1890, 54 miles
off A chill Head, 500 fms. (Bell '92).
vi. R. D. S. Fish. Survey 1891, 45 miles off Black Rock, Blacksod
Bay, 375-500 fms. (Holt '92).
DUtrtbtUion, — Both sides of Atlantic. Deep water.

[Phormosoma uranus, Wyv. Thomson.]

. . . iv., . .

iv. R. I. A. Exp. 1888, 750 fms., a single fine example (Sladen
'91).
Distn'bution, — Both sides of Atlantic. Deep water.

•W^ P^\tcfe(HinjH of the Royal IrM Academ/f.

Family — EcHuriDiB.

Bchinus aoutns, Lamarck. {£. fUmingii Forbes B. S.)

. . iii., iv., v., .

iii. Off Youghal (in very deep water, Ball, f. Forbes '41 ; Humphreys
*45) : ** TorcupiDe " Exp., south of Ireland, 100 fms, ( Agassiz

iv ** Flying Fox " Exp. 1889, 55 fms., 1 10 fms., and 500 fms. (Bell
»89) : H. M. S. ** Research " 1889, 70 fms. (Bell '90) : Dingle
Bay, 30 fms. (Dublin Mus.}.
V R. D. S. Fish. Survey 1890, Gal way Bay, and off Achill Head,
126 fms. (Bell '92).
j)istribution. — Both sides of Atlantic. Mediterranean. Kermadec
I. «• Challenger."

Echinus norregicus, DUb. & Kor.
. . . iv., . .

iv. ** Porcupine" Exp. 1869, off Yalentia, 808 fms. (Hoyle '91):

** Porcupine" Exp. 1869, 80-110 fms. (Bell '92b): R. I. A.

Exp. 1^88, 345 fms. and 750 fms. (Sladcn'91): H. M. S.

•* Research" 1889, 400 fms. (Bell '90). Var. $ angina, Sars.

Off Yalentia, 110 fens. (Thomson '73; Xorman Coll., f. Hoyle

'91).
LUtrihutian. — Both sides of N. Atlantic. Off Japan and off Pata-
gonia •* Challenger."

Echinus microstoma, Wy v. Thomson.

. . . iv., [v.] , [vi.]

iv. ** Porcupine " Exp. 1869, off Valentia (Norman Coll., f . Hoyk' '91) :
R. 1. A. Exp. 1885, 80-120 fms., common (Haddon '86b) :
R. I. A. Exp. 1886, 93-214 fms. (Haddon '88) : R. I. A.
Exp. 1888, 345 fms. (Sladen '91): "Flying Fox" Exp.
1889, 110-500 fms. (BeU '92b).

[v.] " Porcupine" Exp. 1869, off west coast of Iieland, 150-400 fms.,
very abundant (Thomson '74).

fri.^ 11. D. S. Fish. Survey 1891, off Black Rock, Bbcksod Bay, 250
fms. and 500 fms. (Bell '92, '92b).
hidrihtUion. — West coasts of Scotland and Ireland. Bay of

Nichols — A Lust of Irish Echinoderim. 261

Echinus miliaris, Gmel.

i., ii., iii., iv., v., vi.

Generally distiibuted on all the Irish coasts.
Dutribution. — Itelund and Norway to Mediterranean.

Echinus esculentus, Linne.

i., ii., iii., iv., v., \'i.

Generally distributed round the coast in the shallow-water marine
area, and common. " Flying Fox " Exp. 1889, 110 fms. (Bell *89).
Diitrihution, — Spitzbergen and Iceland to Mtditcn-anean.

[Echinus elegans, Diib. & Kor.]
. . . iv., v., vi.

iv. '* Porcupine** Exp. 1869, off Valentia, 110 fms. (Thomson '73;

iiorman Coll., f. Hoyle '91): ** Flying Fox" Exp. 1889,

250 fms., four specimeus (Bell '89).
v. K. D. S. Fish. Survey 1890', 54 miles off Achill Head, 500 fms.,

eight specimens (BeU *92).
vi. R. D. S. Fish. Survey 1891, off Black Hock, Blacksod Bay,

250 fms. and 500 fms. (Bell '92, '92b;.
Distribution, — Both sides of Atlantic. Mediterranean. Off Admi-
ralty I. ** Challenger."

Strongylocentrotus lividns (Lamarck). {Echinus lividm Forbes B. S.)

i., . iii., iv., v., vi.

i. A specimen cast up at Carrickfergus (Dickie '53).
iii. Cork (J. Humphrey, f. Bell '92 b): Lough Hyne (Kinahan * 59 ;

Haddon '88) : south-west coast of Cork (AUman in 9th Ann.

Kep. Dublin Nat. Hist. Soc, 1849).
iv., v. Generally distributed,
vi. Bundoran, Donegal Bay (Hyndman, f. Forbes '41 ; &c.) : Tory

Island (Hyndman, f. Dickie '53) : Bay of Dunfanaghy (Rev.

Mr. Gallaher, f. Dickie '53) : Malin Head (Dickie '53).
Distribution, — North of Ireland to Canaries. MediteiTanean.
Brazil

262 Proceedings of the Royal Ivkh Academy,

Familj — Cltpeastbid^.

Echinocyamus pnsillus (0. F. MtiUer).

i., ii., iii., iv., v., .

i. l^orth of Ireland (Portlock, f. Forbes '41) : off Maidens, 70-90 fms.
(Hyndman '58, '59) : Belfast Lougb (Hyndman and Thompson,
f. Thompson '56; Hui'st in Irish Naturalist, 1896): Castle
Chichester, 6-10 fms. (Bell '92 b) : Strangfonl Lough and
outside in 15-25 fms., rather rare (Dickie '58).

ii. Fortmamock (Thompson, f. Forbes '41 ; Bell '92 b) : Dublin Bay
(Hassall '42 ; &c.) : off Bray Head, 23 fms. (Haddon '86).

iii. Cork (Humphreys '45) : west coast of Cork (Allman, f. Thompson
'56): Baltimore (A. R.N.).

iv. Bantry Bay (Thompson, f. Forbes '41): Ii. I. A. Exp. 1885, Bantry
Bay, abundant (Haddon '86 b) : Berehaven (Dublin Mus.) :
** Argo " Cruise 1890, off Dursey Head (Herdman '91) : Dingle
Bay, 54 fms. (Adams, f. More '70) : Kerry (More '70) : Valentia
Harbour (Beaumont 1900).

V. North-west coast of co. Clare (Kinahan *63) : lloundstone (M*Calla,
f. Thompson '56) : Connemara (Alcock '65) : off Achill I.
(DubUn Mus.).
Distribution, — Norway to Azores. Mediterranean. "West Indies.

Brazil '' Challenger."

Family — SpATANOiDiE.

Spatangns purpureas, 0. F. Miiller.

i., ii., iii., iv., [v.], .

i. Off Maidens, 90 fms., small (Hyndman '58) : Carrickfergus (Dublin
Mus.) : entrance of Belfast Lough, dredged alive (Hyndman, f,
Thompson '56) : Belfast Lough (20-23 fms., Thompson '56 ;
Hurst in Irish Naturalist, 1896) : Strangford Lough (Hyndman,
f. Thompson '56 ; Dickie '58) : outside Strangford Lough, rare
(Dickie '58),

ii. Newcastle, co. Down, from the stomach of a haddock (Thompson
'56) : Fortmamock Strand (Baily '65) : Dublin Bay (Temple-
ton '36 ; &c.) : off Bray (Ball, f. Thompson '56) : coast of
Wicklow (Dublin Mus.).

iii. Cork (Humphreys '45): K. I. A. Exp. 1886, off Cork, 52 fms.
(A. E. N.).

Nichols — A List of Irish JSchinodeiTna. 263

iy. E.LA. Exp. 1885, mouth of Bantry Bay, 35-40 fms. (Haddon
'86 b): R, I. A. Exp. 1888, 64 fms. (Sladen '91): «« Flying
Fox" Exp. 1889, 50-60 fma. (Bell '89): H M.S. "Research"
1889, 70-400 fms. (Bell '90): R. D. S. Fish. Survey 1891,
Kenmare River, 21 fms. (Holt '92) : off Blasquets (More '70) :
Yalentia (Dublin Mas., pres. by Mrs. E. Waller) : off Yalentia
Harbour, 20-45 fms. (Beaumont 1900).

[v.] R. D. S. Fish. Survey 1890, off Achill Head, 220 fms. (Bell '92).
Distrihutim. — Norway and Iceland to Azores. Mediterranean.

Bermuda.

Spatangos raschi, Lov^n.

. . . iv., [v.], [vi.]

iv. "Porcupine" Exp. 1869, off Yalentia, 110 fms. (Thomson '73;
Norman Coll., f. Hoyle '91) : R. I. A. Exp. 1885, 80-120 fms.
(Haddon '86 b) : R. I. A. Exp. 1886, 93-214 fms. (Haddon
'88): R. I. A. Exp. 1888, 345 fms. (Sladen '91): "Flying
Fox" Exp. 1889, 100-180 fms. (Bell '89) : R. D. S. Fish. Survey
1891, off Bolus Head, 115-220 fms. (Holt '92): off Yalentia
(Dublin Mus., pres. by Mrs. E. Waller),
[v.] R.D.S. Fish. Survey 1890-91, off AchHl Head, 127-500 fms.

(Holt '92).
[yi.] R. D. S. Fish. Survey 1891, off Black Rock, Blacksod Bay, 875-
500 fms. (Holt '92).
Di$tributxon, — Norway to Azores. Cape of Cbod Hope "Chal-
lenger."

Ecfainocardiiim oordatum (Pennant). (Amphidotus eordatus

Forbes B. 8.)

i., ii., iii., iv., v., vi.

Generally distributed, and very common on sandy shores after
storms.

BUtrtbutum. — Both sides of Atlantic, from Norway to Spain, and
Carolina to Brazil. Mediterranean.

Echinocardium pennatifldum, Norman.

iv. Off Yalentia Harbour, 20-45 fms. (Beaumont 1900).

Dutribution. — British Isles. Bay of Biscay, Mediterranean
(Eoehler). W. Indies and Florida (Agassiz).

R.I.A. PBGC. VOL. XXIV. 8X0. B, Z

264 3*roceedmff8 of the Royal Irish Academi/,

Echinocardinin flayesceiiB (0. F. Mliller). {Amphidotus roseus
Forbes B. 8.)
i., ii., iii., iv., v., .

i. Off Maidens, 70-90 fms. (Hyndman '58, '59) : near Belfast (Forbes

'41) : off Strangford Bar, 20 fms., rare (Dickie '58).
ii. Dublin (Ball, f. Forbes '41).
iii. Yougbal (Ball, f. Forbes '41 ; Humpbreys '45) : south of Ireland

(Wrigbt & Greene '59 ; " Porcupine " Exp., f. Agassiz '72-74).
iv. Bantry Bay (Wrigbt & Greene '59) : off Blasquets (More '70) :

R. I. A. Exp. 1885, 5-79 fms. (Haddon '86 b) : R. I. A. Exp.

1886, 37 fms. (A. R. K) : R. D. 8. Fish. 8urvey 1890, off the

8kemg8, 80 fms. (Holt '92).
y. West of Ireland (Thompson '44 ; Wright & Greene '59).

Distribution. — ^Norway to Cape of Good Hope. Mediterranean.
Florida.

Brissopsis lyrifera (Forbes). {£ris9us lyrif$r Forbes B. 8.)

i., . iii., iv., . .

i. Outside Strangford Bar, 25 fms., a single specimen in mud
(Dickie '58).

iii. Off Cork, 40 fms. (M'Andrev, f . Thompson '56).

iv. Off Cape Clear, and from 30 fms. in Bantry Bay (M'Andrew, f.
Thompson '56): Bantry Bay (Dublin Mus.): "Porcupine"
Exp., off Valentia (Agassiz '72-'74): R. I. A. Exp. 1885,
Berehaven, 10 fms., and oft Great Skellig, 70-79 fms. (Haddon
'86 b) : R. I. A. Exp. 1886, 70-80 fms. (Haddon '88) : R. I. A.
Exp. 1888, 5 fms. and 54 fms. (Sladen '91): "Flying Fox"
Exp. 1889, 5 fms. (BeU '89).
Distribution. — ^Both sides of N. Atlantic, from Greenland to Azores

and W. Indies. Mediteiranean. Cape of Good Hope " Challenger."

Nichols — A List of Irish Eehinoderms.

265

INDEX.

[Tk»Jlgw99 refer to (hepage$,'\

Amphidotui :

eordaiutf 263.

roseus, 264.
Amphiura:

chiajiiy 255.

elegana, 256.
' fiHformis, 255.

leaehii^ 256.
AmpliiaridiB, 254.
Antedoii:

bifida, 246.

milleri, 246.

phalangiamy 246.
Antedonidao, 246.
ArchasteridflB, 247.
Aspidochiroto, 245.
Asterias:

auraniiaca, 247.

equetirUf 249.

glacialis, 252.

hispida, 253.

munayi, 253.

rabens, 252.
Asteriids, 252.
Asterina:

gibboaa, 249.
ABterinidfls, 249.
Aateraidea, 247.
Astbenofloma :

bystrix, 259.
Astiopecten:

iiregukris, 247.

spbenopkx, 247.
AstropectmidflBy 247.

Bipinnazia :

asterigera, 248.

OQTonata, 253.
endecacmeno0, 258.

Brinngidfls, 253.
BriasopsiB:

lyrifera, 264.
Sriseuii

lyrifer^ 264.

Chirodoiai

digitaia, 241.
CidaridA, 258.
Cidaris:

gracilis, 259.

papiUata, 258.

puipurata, 259.
Glypeastrida, 262.
Comatula :

roMoeeaf 246,
CribeUa:

oeulata, 252.

rosea, 250.
Grinoidea, 246.
Cucumazia:

andrewri, 243.

communis, 244.

drummotidii, 244.

fusiformis, 242.

bispida, 242.

hyalina, 244.

byndmani, 241.

lactoa, 242.

montagui, 244.

p6iitp.cte8, 242.
• planci, 241.

saadcola, 243.

Ddmatidfe, 246«
Dendrochirotie, 241..

Ecbinaatezidtt, 252.
EcbinidiB, 260 •

add

Proceedings of the Royal Imh Aeademp.

Kohfaocardimn :

ooxdatum, 263.

iUyetoeni, 264.

p«nnatiflduiii, 263.
JSphiHccueumit :

tyfU>9, 242.
BohinooyamuB :

pusillua, 262.
JBohinoidea, 258.
Sohinothuiiidie, 269.
Xchinuf:

aoutus, 260.

eleganB, 261.

esoulentua, 261.

fUmingii, 260.

lividutt 261.

microstoma, 260.

miliaris, 261.

nonregicus, 260.

templetoni, 249.
GymnasteiiidM, 249.

Hemicia :

Bangiunolenta, 252.
Halothuria:

aspen, 246.

fonkahli, 245.

iatestinalis, 245.

niffra, 245.

tremula, 245.

tuiulo»a, 245.
flolothuiioidea, 240.
Hymenaster:

giganteus, 252.

Lntmogone :

Tiolacea, 246.
Loidia:

oiliaris, 248.

fra^iUitiima, 248.

•arsi, 248,

Neomotphaster :
ntstiehus^ 25 U
Uklismaniy 251.
Nymphaster;

rotentut, 249.
inosus, 249.

Oenui :

hrunnetttf 242.
laeUut, 242.
Ophiactis:

balU, 256.
Ophiobyrsa :

hystricis, 258»
Ophiocnida :

brachiata, 255.
Ophiocoma :

MHi, 256.

bMit, 257.

brachiata, 255.

JUiformii, 255.

goodiiri, 256.

granulata^ 257.

minuta, 257.

negleeta, 256.

nigra, 257.

roiula, 257.
Ophiocomidie, 257.
Opbiocten :

sericeum, 254.
Ophiolepididie, 253.
Ophiomusium :

l3rmani, 254.
Ophiopholis :

aculeata, 257.
Opbiopsila :

annulosa, 257.
OphiothrioidflB, 257.
Ophiothxix :

fragilis, 257.

liitkeni, 258.

pentaphyUum, 258.
Ophiura:

affinis, 254.

albida, 254.

ciliaris, 253.

mimUay 256.

sani, 254.

signata, 254.

tezturata, 253.
Opbiuroidea, 253.

Palmipes :

mefnbrattaceuif 250*.
placenta, 250.

Nichols— -4 List of Irish Echinodei^ms.

267

Pentagonaster :

bakMiut, 248.

eoneinnuif 248.

gianulariB, 248.

greeni, 249.

hjBtzicis, 248.
PentagonafteiidsB, 248.
Fhonnosoma:

placenta, 269.

iiTanus, 269. '
Fhyllopborus :

drummondi, 244.

pellucidus, 244.
Flutonaster:

bifrons, 247.
Pontafiter:

limbaiut, 247.

tenuispinuB, 247.
Porania:

pulvillus, 249.
Polaster:

andiomeda, 248.
Psolus:

fabricii, 244.

pbantapuB, 244.
PteiBsters

penonatos, 261.
PteraaleridflB, 261.

SolaBter:

endeca, 261.

pappoflusy 261.
SolasteridflB, 261.
Spatangidn, 262.

Spatangus :

puxpureus, 262.

raachi, 263.
Stiohaster:

roseua, 260.
Stichastoridie, 260.
StiohopuB :

natana, 246.
Streptophiuna, 268.
Strongybcentrotus :

lividiu, 261.
Synapta:

digitata, 241.

inlifloiens, 240.

thomsonii, 241.
SyxuiptidflB, 240.

'J'hyone :

andrnotii, 243.

fusixB, 243.

papiUota, 243.

portloekiif 2ii.

rapbanoSi 243.
Thjfonidium :

dubenif 244.

UratUri

fflaeialia, 262.
hispiddf 263.
rubent, 262.
violaeea, 262.

Zoroaster:

fulgenB, 261.

[ 268 ]

VI.

SOME KEMARfiS ON THE ATLANTIS PROBLEM.

By R. F. SCHARFF, B.Sc., Ph.D,

[Bead Notbmber 10, 1902.]

The problem of tlie former existence of a land beyond the pillars of
Hercules has occupied the mind of man since the early dawn of
history. Plato was the first to record the story of this mysterious
land, to which the name of '^ Atlantis " was given. According to his
narrative, Solon is said to have visited the city of Sais in Egypt, and
there to have heard, from priests, of the ancient Empire of Atlantis
and of its final overthrow by a convulsion of nature. From the
account given by Plato, this Atlantis was a continent lying in the
Atlantic ocean beyond the Strait of Gibraltar.

Quite a flood of literature has appeared on this subject since it
was first handed down to us by Plato. By some it was scou|:ed as a
vague and inconsistent tradition ; while others believed in the story,
and republished the account with many fanciful amplifications of
their own. Others again, in their zeal for speculation, enlarged the
Atlantis so as to make it join the New World and the Old across the
Atlantic, and argued that the early races of man must have migrated
on this land-bridge from Europe to America, and have peopled the
latter continent in this manner.

Although the original narrative has thus led to extravagant
theories, thoughtful men have, from time to time, expressed their
conviction either that it rests on some actual historic basis or that the
legend was a vestige of a widely-spread tradition. I need only mention
in this connection the names of Humboldt and Sir Daniel Wilson.

The Atlantis problem, however, was only raised to scientific
importance when modem research revealed the fact that the living, as
well as the extinct, fioras of Europe have quite a number of types in
common with North America.

The first naturalist who attempted the solution of the Atlantis
problem from a botanical point of view was Prof. linger, an Austrian
botanist. The Swiss naturalist, Prof. Heer, elaborated XJnger's^

ScHARFP — Some Beniarka on ihe Atlantic Problem. 269

theories, and argued (p. 183)* that the prominent European character
of the Atlantic Islands, as shown hy their plants and insects, proved
that they were formerly connected hy land with the continent of
Europe. But hesides these forms, he noticed that certain American
types occurred in all the islands, and that the flora of the latter, in
some respects, resemhled the tertiary flora of Europe, which, again,
was allied to that of America. These remarkable features were
explained hy Heer by the supposition that, during the tertiary era, the
continents of Europe and America were joined across the Atlantic,
and that the plants travelled on this old land-connection from the one
to the other. The plants of the Atlantic Islands, he thought, were
more European in character than American, because the islands had been
united with the Old World much longer than with the New (p. 185).

Oliver denied altogether the necessity for what he called *'the
Atlantis hypothesis," and insisted that the American element in the
flora of the Atlantic Islands played only a subordinate part, whereas
mediterranean, with a proportion of peculiar or macaronesian types,
greatly predominated (p. 1 63). He explained the relationship between
the flora of Europe and that of the New World — ^as has been done
more recently by Prof. Engler (p. 82) and Dr. v. Ihering(A. p, 43) — by
the supposition that the plants wandered across a land-bridge which
formerly joined Eastern Asia to North America. Christ, on the other
hand, atbibutes the American element in the flora of the Canary
Islands to the action of the Gulf stream (p. 515); and Trelease, in his
careful account on the Botany of the Azores, remarks that, so far as
the peculiar species were concerned, their ancestors seemed to have all
been introduced by such accidental means as drift or migratory birds
(p. 87).

Edward Eorbes maintained that, at the close of the Miocene Epoch,
a vast continent extended far into the Atlantic from the coast of
Portugal, past the Azores, and bounded on the north by Ireland
(p. 14). While adopting Eorbes's hypothesis, Murray enlarged the
area of this Atlantic continent as far as Newfoundland, Greenland,
and Spitsbergen (A, p. 37).

Imbued with the belief in the permanence of the great ocean
basins. Dr. A. R. Wallace vigorously attacked Heer, Forbes, Murray,
and others in his Presidential Address to the Entomological Society of
London, deliTcred in January, 1871. Neither Murray nor later

^ A lUt of the works and papers referred to in this Essay will be found in the
Appendix.

270 Proeeedinga of the Royal Irish Academy.

writers, he contended, had really grappled with the facts as a whole.
In 1900, he republished his address, stating, in afoot-note (A. p, 250),
that it may serve as a reply, not only to the arguments advanced by
the late Andrew Murray, but also to similar views still occasionally
put forward.

As many writers have expressed themselves against Dr. Wallace's
views since the first publication of hi:s address, I venture again to
give the salient points advanced by him, and to bring forward a few
arguments which appear to me to favour the older doctrines of Murray
and others. I have also paid special attention to the fauna of the
Atlantic Islands as a whole, with the view to reinvestigating the
•* Atlantis problem."

Dr. Wallace's address deab only with the beetles of Madeira, one of
the Atlantic islands ; but he maintains that the opinion he enunciates,
and which is founded on a study of these insects, explains the origin
of the Madeiran and of other insular faunas.

One of the most striking characters of the coleopterous fauna of
Madeira is the unexampled preponderance of wingless species on the
island. Darwin's ingenious explanation of this remarkable pheno-
menon is that the act of flying would expose the insects to being
blown out to sea, and thus destroyed. Those insects which flew least
would therefore remain behind in increasing numbers; and by a
continuous process of natural selection, a race of wingless forms would
thus, in the course of ages, become established on the island.

If Madeira, asks Dr. Wallace, were the remains of a continent,
once continuous with the south of Europe, and deriving its fauna from
such continuity, how are we to explain the absence of extensive
genera very abundant in southern Europe, and, from their being
wingless, specially adapted to the peculiarities of Madeira ? Such, he
continues, are Carahus, LampyriSy Pimelia, Akis^ and many others.
The genus Carahtu — ^a prominent member of the large family of
ninning-beetles — ^possosses, according to Dr. Wallace, 80 species in
southern Europe and northern Africa, while not a single species has
crossed to Madeira. Many other similar facts are set forth by the
same author; and these seem to him quite inconsistent with the
theory of the distribution of insects having been effected by a former
land-connection with Europe. Their transmission appears to him to
have been brought about, not by means of drift-wood or ocean
currents, but by a passage through the air when assisted by gales and
hurricanes.

Now it is evident that if Madeira had ever been connected by land

ScHARFF — Some Bemarks on the Atlantis Problem. 271

with Europe, the part of the Continent with which it wonld have
been joined must have been either southern Spain or Portugal. I
venture to think, therefore, that Dr. Wallace should have drawn his
comparisons of the Madeiran coleopterous fauna with that portion of
our Continent and not with southern Europe and northern Africa
generally. It is perfectly correct, as Dr. Wallace affirms, that not a
single species of Carahus has ever been discovered in Madeira ; whereas
a large number of species of that wingless genus occur in the Mediter-
ninean region. But when we subject the known range of the genus
Cardbui in Europe to a little further scrutiny, the unexpected circum-
stance reveals itself to us, that the number of species decreases rapidly
as we proceed westward from the east. Out of about 153 European
species, seventeen species are now known from the Spanish PeninsuLi,
and only eight have so far been detected in Portugal.^ In the extreme
south of Spain, in Qibraltar and vicinity, three species have been
found by Mr. Champion, and only two in Marocco. To judge from
the geographical distribution of Carahus, it would seem as it the genus
had originated somewhere in the east, and had only invaded western
Europe in comparatively recent times.

Let us now examine the range of the second genus mentioned by
Dr. Wallace — ^viz., LampyrU. Of nineteen European species, two
occur in the Spanish Peninsula, and only one in Portugal.

Pimelia, the next genus, is more largely represented in the Spanish
Peninsula ; for, out of forty-one European species, no less than eighteen
have been traced in Spain, though not a single one seems to be known
from Portugal.

Even a larger percentage of species of the genus AJcis occurs in
Spain ; for, out of eleven European species, seven have been taken
there, of which one has penetrated into Portugal.

These facts tend to show that, even if Madeira had formed part of
the Spanish Peninsula, not one of the species belonging to the above-
mentioned genera would probably have reached that island. The
absence of those genera of beetles from Madeira certainly cannot be
adduced as a proof that the island has never been connected with
Europe.

However, Dr. Wallace founded his views upon three other impor-
tant statements, which require to be dealt with. He pointed out, in
the first instance, that the Atlantic Islands were entirely composed of

1 These and other results have heen ohtained from the " Catalogus Coleopterorain
EuropflD (1891)" hj Heyden, Reitter, and Weise.

272 Proceedings of the Royal Irish Academy.

Tolcanic rocks ; secondly, that they were surrounded by great depths
of water; and, lastly, that they possessed no indigenous land-
mammalia.

As regards the fact that tlie Atlantic Islands are composed ot
volcanic rocks, it does not necessarily imply that they could not,
therefore, have formed part of the continent of Europe in former
times ; for even Hartung, who made a special study of the Geology of
Ifadeira, looked upon the Atlantic Islands as the summits of sub-
merged mountain -chains (p. 175); while two other geologists — ^viz.,
Quppy (p. 496) and Neumayr (p. 547) — maintain that these islands
are the remnants of a great continent which united the Old World
with the New. Lyell, on the other hand, arguing from the supposed
great depth separating the Atlantic Islands- from the Continent, does
not consider it possible that they could have been connected with
Europe (p. 411). However, as Dr. Blanford reminds us, in his
interesting address to the Geological Society of London (p. 34) : << The
occurrence of volcanic islands does not prove that the area in which
they occur is not a sunken continent." " If," he continues, ** Africa
south of the Atlas subsided 2000 fathoms, what would remain above
water ? So far as our present knowledge goes, the remaining islands
would consist of four volcanic peaks — the Camaroons, Mount Kenia,
Kilimanjaro, and Stanley's last discovery, Euwenzori, together with,
an island, or more than one, containing part of the Abyssinian table-
land, which, like the others, would be entirely composed of volcanic
rocks."

Dr. Wallace*s second statement, that the Atlantic Islands were
surrounded by great depths, is only partially correct. On the little
map he published in 1900 (A, p 253), Dr. Wallace indicates a depth of
12,000 feet between Madeira and Europe, though it is now twenty-
five years since Commander Oorringo, of the United States' Navy,
discovered the Gettysburg Bank, and demonstrated the undoubted
fact that there are shallow banks only a couple of hundred feet below
the surface of the sea in that area. He also suggested that a sub-
marine ridge probably connected the island of Madeira with the
coast of Portugal {cf. J. J. Wild, p. 877).

The Azores seem to be separated by much greater depths from the Con-
tinent ; and Dr. Wallace is so convinced of the permanence of the great
ocean basins that he will not allow that any very great changes of level
have taken place in former times. But that Dr. Wallace's views are
not generally accepted may bo gathered from the remark made by
Dr. Blanford that *' not only is there clear proof that some land-areas

ScHAiiFF — Some Eemarh on the Atlantis Problem. 273

lying witliin continental limits have at a comparatively recent date
been submerged over 1000 fathoms, whilst sea-bottoms now over

2iOO

inj.

Oontour-chart (after Wild), showing the subimiiine ridge between Madeira and
Portugal. The data were furnished from the soundings obtained by the
" Challenger," " Gazelle," and " Gettysburg."

1000 fathoms deep most have been land in part of the Tertiary Era ;
but there are a mass of facts, both geological and biological, in

274

Proceedings of the Royal Irish Academy.

fayoor of land-connexion having formerly existed in certain cases
across what are now broad and deep oceans " (p. 77). Moreover,

(^ \Yv[a\A ie corvi mAV^vni']
* VUonvgi

^ |nrule de uenturA
^ co\awti(

^ jnfuU htifTA^W

@ CAprAriA

g per to fAnfro]
[nfulA ^e \t3Y\Am\.
' << ln(aU dejerfte]

Map of the Azoi*e8 and Madeira, enlarged from a facsimile in Kordenakiolp*8
*'Periplu8/' of Gaglielmo Soleri's original map (published 1S85), preserred in
the State Archives at Florence.

when asked to give his opinion on this important question of '' per-
manence of ocean basins," Professor Suess remarked that the geological
evidence did not prove, nor even point to, a permanence of the great

ScHABFP— 5(ww Remarks on the Atlantis Problem. 275-

depths— at least, in oceans of the Atlantic type— and that he believed
some kind of a coast-line stretched across the present Atlantic ocean
daring part of Tertiary times (pp. 185-186).

It now remains for me to deal with Dr. Wallace's third statement—
that the Atlantic Islands possess no indigenous land-mammalia. * * It is-
true," he remarks, "that rabbits, weasels, rats, and mice, and a small
lizard peculiar to Madeira and Teneriffe, are now found wild in the
Azores ; but there is good reason to believe that these have all been
introduced by human agency " (A, p 248). Dr. Wallace does not inform
us what are his grounds for beHeving in the artificial introduction of
the species referred to. I can find no records of such introductions
having taken place, and the results of my endeavours to trace the
history of their origin on the islands point rather to some of them, at
any rate, having reached the latter in the normal way, which is by a
former land-connexion with Europe.

Take the rabbit, for example. We are naturaUy led to assume-
that this destructive rodent was brought over by the Portuguese
when they first colonized the islands. But when the Azorean archi-
pelago was discovered by a Flemish merchant sailing from Lisbon in.
1439, the most striking feature of the islands was the abundance' of
hawks, from which fact the name ** Azores" (meaning hawks) wa&.
given to them. Now these hawks, which still frequent the islands
are really buzzards (Buteo vulgaris) ; and these birds usually live upon
small mammals, such as mice, rats, and young rabbits. It seems
probable, therefore, that such animals already inhabited the Azores
when the Portuguese first set foot on the islands. But this supposition
receives confirmation from a still earlier record of the history of the
islands. Though uninhabited by man, the existence of the Azores had
already been known to earlier navigators, and had since fallen into
oblivion.

In a book {libro del eonoeimiento) published in 1345, by a Spanislk
mendicant friar, the Azores were already referred to, and the names of
the individual islands given. In fact, the islands even make their
appearance in an atlas issued about that time. The atlas is of unknown
authorship, but was probably drawn by a Genoese. Towards the end
of the same century in 1885, another atlas was published at Venice » —

iMr.Lyster kindly drew my attention to the splendid series of reproductions
of «icwnt maps in NordemAzold's " Periplus " contained in our National Libr^
of Ireland. Though the position and size of the Azores in tiiese old mapTZ
mcorrecUy marked, there can be no doubt that the early Italian naviirators had
discovered them, and had roughly sketched their bearings as indicated.

^76 Proceedifiga of the Royal Irish Academy.

that is to say, more than fifty yean before the discoyery of the Azores
t)y the Portaguese — ^in which Beven of the islaiids of the Azorean
archipelago are indicated by name. Some of these names are of yery
particular zoological interest, as they are evidently derived from the
names of animals which the early Genoese navigators disoovered on
.the islands. They are as follows : —

Caprarias goat island (now San Miguel).
Golumbiss pigeon island (now Pico).
Li Conigia rabbit island (now Flores).
Corvi marini= island of sea-crows (now Corvo).

Drouet has already directed attention to the peculiarities of the
•goat of Saint Michael islaud with its antelope-like horns ; while (Rod-
man states (p. 15) that the island of Pico, with its dense gro¥rth of
evergreen brushwood, is the home of the wood-pigeon {Columda palum-'
bus J L.). I do not know whether the rabbit still inhabits Flores,
though it is known to occur in the uncultivated districts of St. Michael ;
but it seems evident that the name " rabbit islatid^^ was given by the
early discoverers of the island, because these animals were abundant
there. As regards the expression " corvi marini," the term may pes*
sibly refer to the great shearwater (Pufflnus major\ which, according
to Oodman, may be seen throughout the archipelago, and which formerly
may have been particularly common on the island of Corvo.

The result of these historical inquiries seems therefore to justify
the supposition that mammals, such as the wild goat and the rabbit,
are truly indigenous species on the Azores, and that these islands
have received their land fauna from Europe by a direct land-connection
When we examine the general fauna of the Atlantic Islands more
<;losely, tho validity of such a supposition becomes more and more
strengthened; and it must be apparent that their colonization cannot
have been due to influences of atmospheric propulsion, even when
.assisted by hurricanes.

Even if the Azores possessed no indigenous Mammalia, their inver*
tebrate fauna alone seems to me to favour the view that the islands
had recently formed part of our Continent. Are not the Seychelles
destitute of mammals ? Yet even Dr. Wallace admits (B, p 243) that
they cannot be classed among the oceanic islands. New-Caledonia
has neither mammals nor amphibians; nevertheless its general fauna is
such that it must have formed part of larger land-areas within Ter-
tiary times {ef. Hedley's interesting observations on Placostylus).
After a few remarks on the geological aspects of the <' Atlantis "

SCHA.KFF — So7iie lUmarks on the Ailaniis Problem. 277

problem, I propose, in the following pages, to deal shortly with the
geographical distribution of the principal groups of vertebrates and
invertebrates, in so far as it affects the questions of the existence of a
former land-connection between the Old World and the Atlantic Islands,
and of a land-bridge across the mid- Atlantic.

Geology of ihb Atlaktic Islands.

Teneriffe, the principal island in the archipelago of the Canary
Islands, has always been a centre of attraction for the naturalist, on
account of the beauty of ils scenery, and from the fact that it possesses
a volcano rising to an elevation of 12,000 feet. Not only its Zoology
and Botany have been carefully studied, but competent geologists
have from time to time visited it, and have given us many valuable
observations on the nature of the rocks found upon the island.

Being entirely covered by immense masses of recent lava, no
fossiliferous deposits, if such do exist, have as yet been encountered,
the only indication of the presence of older crystalline rocks having
been obtained by blocks ejected from the Old Peak and other volcanic
vents.

On the islands of Palma, Fuerteventura, and Oomera, on older
mountain-chain, consisting of diabase — an eruptive rock generally found
in older geological formations — ^has been shown to exist. Fritsch and
Eeiss are of opinion that rocks of a similar nature also form the founda-
tion of the more recent volcanic deposits on Teneriffe (p. 315).

On Grand Canary and Palma, two islands belouging to the Canary
group, an upheaval of from 600-1000 feet can be demonstrated; and in
Madeira, one up to nearly 1400 feet, while it is surmised that there was
probably a similar rise in Teneriffe.

The Azores and the Canaries, as well as Madeira, are believed by
Fritsch and lleiss to be contemporaneous ; and these authors maintain
that, at the commencement of Middle Miocene times, a considerabla
part of the mountain masses of the islands must already have been
formed (p. 220).

Madeira differs from Teneriffe in so. far as volcanic activity has
ceased for a considerable time past ; so that even the most recent
lavas exhibit signs of long erosion. Marine deposits occur here at
a considerable elevation as well as on the Azores, and on the islands
of Porto Santo and Oran Canaria ; and geologists seem to agree as to
these deposits having been laid down towards the middle of the
Miocene Period.

Hartung, who specially studied the geology of Madeira, looked

278 Proceedings of the Boyal Irish Academij.

upon the whole Atlantic Islands as summits of submerged mountain-
chains; while Neumayr (p. 647) and Quppy (p. 496), as I have already
had occasion to mention, saw in them the remnants of a great con-
tinent.

Sir Charles Lyell maintained that Madeira originated in Upper
MioceDe times, and that it was separated from the Continent by a
great depth. He did not consider it possible, therefore, that the
two could have been ever connected (p. 411). However, I have
already shown that Lyell*s supposition was not borne out by later
researches, and that Suess believed in some kind of coast-line having
stretched right across the Atlantic durxng part of Tertiary times.
Prof, de Lapparent expresses himself even more decisively on this
subject. He favours the view of the existence of a coast-line, or at
least that of an island-chain, during the Miocene Period, connecting the
West Indies with southern Europe. The end of the Pliocene and the
whole of the Pleistocene Period, he believes, were distinguished by a
series of subsidences which resulted in finally opening up the northern
depression of the Atlantic Ocean (p. 1392).

Prom a careful study of the structure and distribution of the
geosynclines, Mr.Haug has recently come to the conclusion that the
convex arch formed by the Antilles, and the one found along the
western border of the Mediterranean, were connected, in the Tertiary
Era, by tangential chains of land. According to his view, a coast-line
stretched across the Atlantic from Yenezuelu to Marocco, and another
between the Lesser Antilles and Portugal, the intervening space being
covered by the sea (p. 635).

The hypothesis of a North Atlantic continent and of an Africano-
Brazilian one, to some extent meets the views expressed by Murray on
zoological grounds. As already mentioned. Dr. Gregory urged that
the intimate affinity between the Miocene marine species of the West
Indies and those of the Mediterranean Region could only be explained
by the assumption of the existence of a shallow-water connection
across the Central Atlantic in — at latest — Miocene times (p. 306).

Fauna of Atlantic Islands.

Kammals.

The historic reasons for the belief that the rabbit ( OryeUlagua
eunmUua) might be indigenous in some of the Atlantic Islands have
already been alluded to. If we supposed these islands to be the last
remnants of a former land-connection between Europe and America,
the rabbit might be regarded as a relic form which reached us

ScKAVLVv—Some Remarks cm the Atlantis Problem. 279

originally from the New World. That our European hares and the
rabbit have come to ub from America is suggested by Prof. Osbom
(p. 58), though he does not indicate the route they are likely to have
taken in their migrations.

I am not advocating a direct land-connection between southern
Europe and America by way of the Atlantic Islands. I think there
was only one land-bridge in southern latitudes, between the Old World
and the New, which joined Africa and South America. This must
have lain further south than the Atlantic Islands. But from North
Africa, there was frequent intercourse with southern Europe — with
which the Atlantic Islands, I believe, were connected; and South
American species would then have been able to reach Madeira and the
Azores indirectly.

To resume the discussion on the rabbit. The zoological position of
the LeporidaB was recently subjected to a thorough and most careful
revision by Dr. Forsyth Major, according to whom the rabbit belongs
to a different group altogether from that of the European hares. He
places it along with the South African Lepus crassieaudatua into the
genus Oryetolagus, the two species being characterized by a very dis*
continuous distribution, which we are assured by Dr. Wallace is a sign
of antiquity. The nearest relative of Oryctolagu9 is SylvHagtis, a genus
containing five species which range from south-eastern North America
southward to Paraguay. It does not seem unnatural, therefore, that
America should have been selected by Prof. Osbom as the original
home of the group. The first branch ( Capreolagus) was probably sent
o£E eastward to southern Europe and Asia. Of this, three species are
known, two of which are now extinct, while a third still siu'vives in
India. This genus may subsequently have given rise to the allied
form Nesolagua NeUeheri of Sumatra.

We have therefore evidence of migrations, not only between
Europe and America, but also between Africa and America.

It has already been urged by Mr. Lydekker that the hystrico-
morphous Kodents of the South A merican Kegion are so closely related
to those of Africa that a connection between the two areas must
have existed (p. 127) ; and another instance of a still more startling
character was quite recently referred to in Prof. Howes' admirable
address to the British Association at Belfast — viz., that a fossil mole
had just been discovered in the celebrated Argentine deposits of South
America which agreed with the golden mole ( Chrgsochloris) of South
Africa (p. 7).

We must also remember the affinity of the West Indian Selenodons to

B.I.A. PBOC., VOL. XXrr., SEC. B.] 2 A

280 Proceedings of the Royal Irish Academy.

the Tenrecs of Madagascar, in which island probably many of the older
forms are still preserved which have become extinct on the neighbouring
Continent. Thus the f aunal approximation between Madagascar and
South America is frequently more apparent than that between the
latter and the continent of Africa.

The intimate affinity which existed in early Tertiary times between
tbe mammalian fauna of Western Europe and that of the New World
is especially marked among the Carnivores. The genus JTyanodon,
for instance, which is confined to North America and Western
Europe, lived in both continents from Eocene to Miocene times.
StenoyaU, one of the Mustelidee, represented by five species in the
Eocene and Miocene deposits of France and Qermauy, turns up again
in the Miocene strata of North America. Among the Canidse,
Oaleeynus is found in the Miocene of North America and Switzerland;
while Thousj which is still living in South America at present and
has never been met with in North America, inhabited France in
Pliocene times. Some Felidte also exhibit a similar range. Thus
Eu9milu9 occurs in the Miocene of France and North America • and
the sabre-toothed tiger, Macharodus, which roamed over central and
southern Europe from Eocene until Pliocene times, spread eastward
into India, and also right across to North and South America
(Smilodon).

Some of these cases, no doubt, can be accounted for by the suppo-
sition that the mammals migrated from or to Europe across Asia
where they may have found a land-connection joining that continent
to America across Bering Strait ; but others seem to me to have used a
more direct route between our continent and the New World.

Messrs. Scluter recognize a distinct division of tho marine area of
the globe as consisting of the middle portion of the Atlantic which
they caU " Mesatlantis '* (p. 208). Two genera of mammals are
assigned as characteristic of this region— viz., Monaehtu, the Monk
Seal, and the Sirenian Mamtus, Now neither of these animals frequent
the open ocean, being bound to the proximity of land. Monachus
albiventer inhabits the Mediterranean, and the closely allied M.
tropicalu the West Indies, separated by the enormous expanse of the
Atlantic Ocean, where no Monachus is known to exist. Manatus
is still more permanently attached to the coast. One species,
Manatus inunyuis, has even forsaken the sea, and now lives only in
fi-esh water. Of the two other species, Manatus seneyaiensig inhabits
the coasts and estuaries of West Africa, M. americanus being found
along the South American coast and among the West Indian islands.

ScHARFF — Some Refiiarka on the Atlantk Problem. 281

The range of these mariQe mammals appears to Messrs. Sclater to
imply that their ancestors have spread along some coast-line which
probably united the Old World and the New at no very distant period
(p. 217).

Birds.

The birds of the Atlantic Islands, and the relationship of the
hii-ds of the Old World to those of the New, deserve a more careful
study than I was able to extend to them. A list of the birds
inhabiting the Azores is given in Godman's work already referred to ;
and he informs us that 91 per cent, are also found in Europe, Pyrrhula
murina being peculiar to the islands.

Dr. A. Konig has paid special attention to the birds of the Canary
Islands, and discovered in the island of Teneriffe an owl ( Olauddium
sifu) which had hitherto only been known from Cuba. He also
noticed that the Teneriffe wren differs considerably from both the
European species, and that it approaches more nearly the American
Eegulm satrapa (p. 8).

Doubts are thrown upon the correctness of the author's conclusions
as to the existence of an American element in the Canarian avi-fauna
by Dr. Hartert. He believes that the Canarian avi-fauna is ex-
clusively composed of European and North African species. He
also recoiHls the fact that the avi-fauna of the Eastern and Western
groups of islands are stiikingly different from one another as is the
fauna of Madeira from that of the Canaiy Islands generally (p. 114).

South American affinities with Europe or Africa are perhaps less
noticeable in birds than in any other group ; though there are a few
instances denoting that such exist.

Comparatively few fossil species of birds are known ; but there is
one which is of exceptional interest, as pointing to a former more
intimate connection between Europe and South America — viz., Trogon
gallicus, discovered in the Miocene deposits of France. The genus
Trogon, including some of the most gorgeous birds known, is now
entirely confined to Centi-al and South America, where twenty-five
species are still found living.

Beptiles.

Many of the existing species of Reptiles are very ancient. The
distribution of the class, as a whole, elucidates therefore, in a
btiiking manner, some obscure points in the ancient geographical
condition of the part of the world with which we are dealing at
present.

2 A 2

282 Proceedings of the Royal Irish Academy.

The reptilian fauna of the Atlantic Islands is almost altogether
European in its character, and shows scarcely any trace of an American
relationship.

The Lacertidee, of which family a few species occur on tlie
islands, are not found at all in America ; and the only genus of the
Scincidee which inhabits the Atlantic Islands is also unrepresented in
the New World. It is only among the GeckonidoB that we find a link
across the ocean. Two species of Tarentola inhabit the Canary
Islands ; while an allied form has been met with in the West Indies.
To judge from the general range of the genus, the migration has taken
place from North Africa to America.

Among this same family of Geckonidae, we meet with some inte-
resting instances of discontinuous distribution : thus Gymnodaetylus
matiriticanus lives in northern Africa, while the closely allied
G. i>' Orlignyi inhabits Chili. Only a single species of PhyJhdactylus
occurs in Europe, the remainder of the genus being confined to South
America, South Africa, Madagascar, Socotra, and Australia. Remi-
dactylus, another genus of the same family, seems to bo an exceedingly
ancient one, to judge from its distribution. One species inhabits the
borders of the Mediterranean, while a closely related one is found not
only in South America, but also in Madagascar and South Africa — yiz.,
jGT. mahouia. The other species of the genus are distributed oyer the
Cape Verde Islands, Socotra, St. Helena, South India, Mauritius, the
islands of the western Pacific, West Africa, Ceylon, Persia, South
America, and East Africa.

The Anguidee, which are frequently limbless or only provided
with rudimentary limbs, are almost entirely confined to Americn.
The genus Ophisaurus, however, which has a species in Mexico, and
another in eastern North America, is represented by one species in
southern Europe, and by another in the Himalayan mountains.

Of much interest, from a geographical point of view, arc the
buiTowing Amphisbaenidte, which, generally limbless, often spend
their entire existence underground in ants' nests. That such species
aic not likely to be conveyed across an ocean by accidental causes, in
the manner described by Wallace and others, is evident. Now this large
family of sixty-five species is quite confined to America, Africa, and the
MediteiTanean region. As very few species range into North America,
while not a single one has been discovered in Asia, the hypothesis of a
land-connection across the Atlantic explains the geographical distribu-
tion of this family better than any other theory. Take, for example,
such a case as the genus Anops, There are two species. A,

ScHAUFF^iSo//)^ lieniarks on the AUanlis Problem. 283

JTiVi^i being known from Southern Brazil, Uruguay, and Argentina,
while the allied A. afrieanus is only met with in West Africa.

Among the snakes, we find similar cases of distribution. Thus
the Boinse are mostly South American, but one genus Eryx inhabits
North Africa, Greece, and south-western Asia* The genus Boa is
confined to Central and South America, with the exception of the
two species B, Bunierili, and B. madagMeariemin^ which turn up far
away in Madagascar. This curious relationship between Madagascar
and South America, which occurs among many groups of animals,
has already been alluded to.

There are naturalists who attribute such cases to *^ convergence,"
in order to obviate the difficulties of land-connections across deep
oceans ; but I cannot see how the two species of Boa, agreeing in
every anatomical detail with the characters peculiar to the American
genus, coiild have arisen independently in Madagascar. The Boidse,
too, must be looked upon as a comparatively ancient family ; though
the Ophidia, as a whole, are no doubt a relatively young branch of
Beptiles, and, according to Dr. Gadow, essentially of Tei*tiary date
(p. 586).

This theory of convergence, so much discussed at present, appears
to me even less applicable to fresh- water than to land foims. Yet
among the PelomedusidaB, a family of fresh-water tortoises, a species
of Podoctiemis occurs in Madagascar ; while several allied species of
the same genus are commonly met with in South America.

Amphibians.

Whether Rana escuUnta^ the common edible frog of the Canary
Islands and the Azores, is indigenous or not will probably never
be known, as we cannot altogether rely on the reports of its sup-
posed introduction. To the latter islands, it is said to have been
brought in the beginning of the last century ; while the introduction
to the Canary Islands dates, according to Steindachner, from the
sixteenth century.

Besides the common frog, the tree frog {Uyla arhorea) inhabits
most of the islands of the Cauarian archipelago, and also Madeira and
the Salvages.

The distribution of the genus Hyla is very instructive, as it is
almost entirely confined to America and Australia. It is absent
from the Ethiopian liegion, and only represented by a few species in
Europe and Asia. It has probably come to us from the East, spreading
chiefly along the Mediterranean, where it has formed several distinct

284 Proceedings of the Royal Irish Academy.

Tarieties. In the Canary Islands, I found it commonly at Laguna
and eren higher in the island of TenerifFe ; and it has every appearance
of being indigenous there.

No relationship between the amphibian fanna of the Atlantic
Islands and America is traceable ; but, among the European Bufonidae,
Bufo calamita is closely allied to B. rariegatm of Chili.

As regards the connections between the New World and Africa,
the family DcTidrobatidae is entirely confined to Madagascar and
Tropical America ; while the tongueless toads occur only on the
continent of Africa and in Tropical America.

Fishes.

The lliver-eel {Anguilla fluviatilU) is the only species of
fish which has been observed in the fresh- waters of the Atlantic
Islands; so that no comparison of their fish -fauna with those of
America or Europe is possible. A few facts however might be stated
concerning the relationship of the American and African fish -faunas.

According to Dr. Giinther (p. 232), two of the most naturiil
families of fishes, the Chromides and Characinido;, are peculiar and
(with the exception of JEtrophes) restricted to South America and
Africa. The Dipnoi inhabiting the same two continents are also very
nearly related to one another ; and the Pimelodina, so characteristic of
tropical America, have three representatives in Africa. Yet, though
Dr. Giinther believes in the former union of Africa and South
America, he is of opinion that the separation is geologically of old
standing.

MoUosos.

The land and fresh-water Molluscs of the Atlantic Islands have
been studied by a number of careful observers, particularly by
Lowe and Wolkston. The latter published a most valuable work
entitled ** Testacea atlantica," on the distribution of the Molluscs in
the various islands, which, for a long time to come, will remain the
standard treatise on the subject.

After dwelling on the marked individuality of the island fauna,
he refers to the marvellous segregation of species in most of the
archipelagos, an overwhelming proportion of them being confined to
single islands, and not having colonized even their respective groups.
He also draws attention to the fact that the Mediterranean element is
much more traceable in the Canaries than in the other groups of
islands. Altogether, he believes that the Atlantic Islands have

ScHARFF — Some Bemarks on the Atlantis Prohleiu. 285

originated from the breaking up of a land which was once more or less
continuous, and wliich had been intercolonized along ridges and tracts
(now lost between the ocean), thus bringing into comparatively
intimate connection many of its parts ; whilst others were separated
by channels which served practically to keep them very decidedly
asunder (B, p. 564).

It is of importance to noto that a good many species of land-
moll u sea from the Tertiary deposits of western Europe have their
nearest living relations on the Atlantic Islands. Thus the operculate
genus Craspedopoma is quite peciiliar to these islands at the present
time ; but, on the Continent, it first turns up in the Lower Eocene of
the Paris basin, then again in the Oligocene of the Isle of Wight, and in
the Lower Miocene of Hochheim, and finally disappears in the French
i'liocene strata.

The peculiar Clausilise of the Atlantic Islands grouped in the
sub-genus Boeitgeria have many features in common with the Miocene
group Laminifera, which has still a solitary living continental
representative in the Pyrenees. Many other Atlantic Island MoUusca,
especially among the Helicidse, have their nearest representatives
in the European Miocene.

One of the most remarkable species of animals in these islands
is Plutonia ( Viqtiesnelia) atlanticay a subterranean slug-like Mollusc^
which, like Testacella, devours earthworms. It occurs on the islands
of Fayal and San Miguel under sphagnum and liverworts, and is
quite peculiar to the Azores. Professor Simroth, who finst made
known to us its anatomaticnl structure, is of opinion that it has
originated on these islands (B, p. 223). From a similarity of the slug
fauna of soutliem Portugal, southern Spain, North Africa, and the
Canaries, the same author concludes that there was probably a broad
hind -connection between these four countries, and that it must have
pei-sisted until comparatively recent times (B, p. 402).

The influence of Wallace's views are clearly traceable in Dr.
Kobclt's earlier writings, in which ho ridicules the idea of an
Atlantis and of a former union between the Atlantic Islands and
the Continent (A, p. 8). After having independently worked out the
same problem more recently, however, he came to precisely the
opposite conclusion. Comparing the European with the West Indian
and Central American faunas, he points out that the land- shells on
the two opposite sides of the Atlantic certainly imply an ancient
connection having subsisted between the Old World and the New,
which only became niptured towards the close of the Miocene Epoch.

286 Proceedings of the Royai Insh Academy.

And he contended that the West Indian terrestrial mollnscan f anna
had heen partly derived from the Miocene fauna of Europe (B, p. 147-
148). He is now convinced that the Atlantic Islands must have heen
joined to the mainland in Miocene times (D, p. 53) ; though he does not
comment upon the question whether the islands have had a continental
connection since. He still leaves it to he explained how the strong
Mediterranean element reached the islands. Dr. von Ihering solves
this prohlem in a peremptory manner by saying that no malacologist
nowadays could explain Uie presence of these continental Molluscs on
the islands in any other way but by their progression on land
(A, p. 51).

The molluscan fauna of the Atlantic Islands does not apparently
lend any support to the theory that they formed part of an ancient
land-bridge joining our Continent to the West Indies or North
America. But a study of the Mediterranean fauna reveals a certain
relationship with that of the West Indies in the genera Glandtna,
Tudora, and Leonia. This resemblance of the two faunas, liowever,
becomes much more marked when we compare the extinct land
Molluscs of Europe with those at present living on the other side of
the Atlantic. A migration from our Continent across Asia and Bering
Strait to America, in explanation of these facts, is inadmissible, as the
species in question have never been found fossil in either of the last
two continents. A direct land-connection between Europe and America
must therefore have existed across the Atlantic in Miocene times,
the Molluscs migrating from the former to the latter. Dr. Kobelt
places this land-bridge much further north than the Atlantic Islands
(B, p. 147). I should, on the contrary, be inclined to locate it to the
south of these islands, as it seems evident that Glandina, Tudora^ and
Leonia have only spread from the West Indies into North America
proper in comparatively recent times. The fact that these species,
which have representatives in the West Indies, are more or less
confined to the Mediterranean region in Europe, also points t^ a
southern connection rather than a northern one; and it is quite possible
that the migration took place along the northern coasts of the land,
which is supposed to have united Africa and South America, and in
favour of which Mr. Murray, Dr. Blanford, and Dr. von Ihering
especially have brought forward many important distributional
evidences.

Dr. von Ihering produces evidence of a strong resemblance
between the Brazilian and the West African invertebrate coast fauna.
He also draws attention to the fact that the fresh- water mussels of South

ScHARFF — Some Remarks on the Atlantis Problem. 287

America are very closely related to those of Africa, while they differ
enormously from the ^orth- American ones. A similar resemblance in
the fish-faima of Africa and South America has already been alluded
to ; and the author concludes that there can be no other interpretation
of these phenomena but that of a former land-connection between
Brazil and Africa. He assumes this connection to have been mainly a
Mesozoic one ; though he admits that it may have persisted until
Oligocene times (B, p. 135).

This large continental mass, which partly filled the southern
Atlantic, must have covered vast areas of the ocean, including
islands like St. Helena and Fernando de Noronha. These may
therefoi*e represent remnants of the Atlantic continent. Their fauna
and flora would naturally be of extreme interest in throwing light
upon this subject. But although the animals and plants of these
islands have been investigated by competent naturalists, who have
declared them to possess either African or South American affinities,
Wallace's decided views in favour of the accidental transmission of
fiptcies j<reatly influenced opinion regarding tlieir origin. Even
AVoUaston expressed himself against the theory that St. Helena had
ever formed part of a continent (B, p. 530). However, a large number
of species inhabiting that island are peculiar to it ; and Wallace himself
recognizes that the insect fauna is suggestive of a very great antiquity
(B, p. 300).

Mr. E. Smith — one of the more recent writers on the molluscan
fauna of St. Helena — detects a greater resemblance to the South-
American fauna than was suspected by either Forbes or Wollaston.
All the same, arguing from the isolated position of the island and the
depth of the suiTOunding ocean, he does not believe in its having been
formerly joined to South America. Still more recently, Dr. Kobelt
discusses the problem again from a study of the molluscan fauna of
the island, and comes to the conclusion that St. Helena is a last relic
of a Mesozoic continent (E, p. 202).

In the very interesting account of the fauna of ** Fernando de
Koronha'* by Mr. Kidley, the writer does not enlarge upon the
possibility of this group of little islands having formed part of a
continent, though the occunence of a subterranean amphisbaena {A,
Ridley i) and of a peculiar fresh- water snail {Planorhis norhonemis)
might have suggested to him a different explanation of their origin
than that of accidental importation. At any rate, Dr. von Ihering is
confident that Fernando de Xoronha has had an ancient land-commu-
nication with South America on tlio west, and with Africa on the east.

288 ProceediiigH of the lioyal Irish Acddemtj,

Insects.

The insects of the Atlantic Islands agiee perfectly with all the
other cLissfs of animals, in so far as they exhibit mostly South-
European or I^orth-African affinities. Taking the various orders
separately, the Hymenoptera — or at any rate thnt group of Hymenoptora
to which the ants belong, have been fairly well investigated and are
of particular interest from a distributional point of view.

A very minute ant {Monomorium minutum carhonarium) inhubits^
Madeira and also the West Indies and Central America ; but there is a
possibility of its having been introduced into the former locality by
man. The genus, as a whole, has a very wide range in the Tropics.
There is no other feature which might indicate a direct land-union
between Madeira or the Azores and America ; but Teneriffe is inhabited
by an ant {Zepiothorax gracilicornis) which is peculiar to it, and has
some allied species in America, though most members of the genus
are Paltearctic.

In the ^Fediterranean region, we find a minute ant in decaying
wood {Epitritus argiolus), No other species of the genus has hitherto
been detected in the Holarctic region ; in the West Indies, however
(Island of St. Thomas), another species has been met with (^E.
emma).

Special attention is directed by Dr. von Ihering to the family
Dorylidae, which are principally African, a few species penetrating
into the Oriental and southern Palaearctic regions. A single genus
{Ecifon) of this family is altogether confined to South America ; and
being closely allied to the African genus Anommay its distribution
contributes additional evidence, accoixling to von Ihering, in favour of
the theory that these two continents have formerly been connected
with one another (D, p. 418).

Our knowledge of the distribution of flies is still very incomplete.
Dr. Dahl, who has collected them in the Azores, maintains that they
are there thoroughly European in character (p. 333) ; and there is
apparently nothing else which calls for special notice.

The lepidopteix)us fauna of the Atlantic Islands has been carefully
studied by quite a number of competent observers. Rebel and Rogen-
hofer give us an excellent account of the species inhabiting the Canary
Islands ; and they inform us that the whole lepidopterous fauna of the
archipelago has probably undergone profound modification, owing to
the almost complete destruction of the native forests during the last
few centuries. Still there are 183 species left. Of these, sixty-one
occur also in Madeim. The Canaries possess 70 per cent, of their

ScHAKFF — Sfmte Hemnrks on the Atlauth Problem. 289^

butterflieR and moths in common with the Mediterranean region ; wliile
about 20 per cent, occur also in America. The authors, however, do
not place any signification upon this high percentage of American
forms, ns only seven species belong exclusively to the Canarian district,
all of which they look upon as having been imported in commerce.
One of the most remarkable forms is the small moth Setomof^pha
dMcipunctsUay which, though peculiar to the Canary Islands, belongs to
a genus represented in America and Africa, and is closely allied to the
South American genus ZtTidera.

There are many ca?es of distribution proving the existence of
Africano-American relationship. The following moy sei-ve as familiar
examples: — The nympholid genus Hypanartia is confined to South
America, Africa, and Madagascar. Amongst the Hesperidie, there are
three genera which are found in South America as well as in Africa —
viz., Oxynetra, Letieochitmea, and PardaleodeH. A particularly sug*
gestive case of distribution is that of the remarkable and most
beautiful type Urania, Though this genus has now been separated
into the two genera Uranidia and Chrysieidia^ their close relationship
is well seen by their general structure and brilliant colouration. The
former inhabits Brazil, Central America, and the West Indies ; the
other, Madagascar and Zanzibar.

Tlie Coleoptera, like the Lepidoptera, are a popular group of insects ;
and it need not be wondered at that the beetles of the Atlantic Islands
have received a good deal of attention. Indeed, Dr. Wallace, in his
paper on the beetles of Madeira, already quoted, hases his belief of the
* oceanic * natui*e of that island chiefly on the absence on it of certain
coleopterous continental forms. The arguments which I hrought
forward in the introduction against those advanced by Dr. Wallaco
need not he repeated here.

Andrew Murray (C) was the first to advance the theory, from a
study of the beetles, that West Africa and Brazil had been once united
by land — namely, at a period subsequent to the appearance of the
present forms of Coleoptera. The presence of the South- American
genus Ztain Old Calabar, of Oaniotrapis, HypolithuSf Galeritay and
AUndria in West Africa, seem to him to clearly indicate the existence
of a former Innd-communication between the latter country and South
America.

Since the time that Murray first announced these views, he found
many other genera of beetles exhibiting a similar relationship. And
he reiterated the same theories which hu had expressed before in a
later paper, reinforced by additional facts and arguments. At the

590 Proceedings of the Royal Irish Academy.

flame time, it may also be interesting to quote bis opinion on tbe fauna of
tbe Atlantic Islands. ** How tbe European cbaracter," be says, "of tbis
general fauna is to be accounted for, except on tbe supposition of a
former connection of tbem all witb Europe, and bow tbe presence of
tbese special forms of tbe same sub-fauna in all t)ic islands and
nowbere else, is to be accounted for except on tbe supposition tbat, after
tbey were disunited from Europe, tbey were still united among tbem-
«elves, it is for tbose wbo advocate tbe tbeory of dispersal by cbance
introductions to say" (A, p. 14).

Having made a careful study of tbe Coeloptera of tbe Atlantic
Islands, Wollaston states tbat bis own views are more in accordance
witb tbose propounded by Murray tban witb tbe tbeory of exceptional
atmospberic dissemination so ably advocated by Dr. Wallace (A, p. 209).
He aptly remarks (A, p. 210) tbat storms and burricanes, so mucb
relied upon by tbe latter for tbe aerial transmissal of European species
to tbe Atlantic Islands, are not only rare in tbese latitudes, but blow
almost invariably from tbe soutb ; and tbat tbe apterous, ratber unwieldy
forms, so largely represented on tbose islands, are tbe least suitable for
4itmospberic propulsion.

Ill tbeir European and Nortb African cbaracter, tbe Keuroptera of
tbe Atlantic Islands agree perfectly witb tbe above-mentioned orders
of insects. M *Lacblan says of tbe Madeiran species : ^ ' Tbat some of tbe
purely terrestrial forms may bave been introduced from Europe is
very possible ; on tbe otber baud, I see nu reason to doubt tbat some
of tbe European forms m<?ntioned may be considered true natives of
tbe islands also. It is wortby of romark tbat (witb one possible
exception) tbe wbole of tbe species of Trioboptera are peculiar to
tbe islands, altbougb belonging to familiar European genera, and
tbat tbey all inhabit running water in tbe larval stage."

Tbe Hcmiptera of tbe Atlantic Islands likewise follow tbe pre-
ceding orders in tbeir general relationsbip ; but a few cases of distri-
bution are deserving of notice, in so far as tbey tbrow ligbt on tbe
question of a former Atlantic continent. Tbe genus Noualhieria is
•confined to tbe Canary Islands, witb two species. Its nearest allies,
Mavniothaniay tSisammes^ and Erla^da, occur in Algeria, Guatemala,
4ind Cbili, respectively. Velia maderenm^ of Madeira, belongs to a
genus confined to Africa, soutbem Eui'ope, and Soutb America.
Eracbystcles bas two species in Madeira, two in Europe, and one in tbe
West Indies. We can trace tbe relationsbip between European or
Atlantic Island and American forms in many otber cases; buttbegenera,
as a rule, are widespread, and are found on otber continents as well.

ScHARFF — Some Remarks on the Atlantis Problem. 291

We do not expect to obtain mach information from the distribution'
of the Ortboptera, as they are such an ancient order, Progmoblattinct^
from the Carboniferous rocks of Switzerland being much like a modem
cockroach, except for the difference in the neuration of the wings.
The group containing the earwigs is now often separated from the
Orthoptera, and placed into the distinct order Dermaptera. Some
attention has been devoted to these orders in the Canary Islands by
Krauss, who found that the affinity between the fauna of those
islands and that of the Mediterranean region was a most character-
istic feature. Of the sixty-four species referred to by the author,
thirty-three are common to both districts ; and fifteen species peculiar
to the islands are allied to Mediterranean forms. He thinks that
the relationship with the American fauna is not so distinctly traceable
in these as in other insect orders. The two species of Orthoptera
belonging to the cockroach section certainly have originated in the
South- American region — perhaps also Periplaneta americana^ P. trun-
cata, and Zeueophaa surinamensis. Among those having near relations
in the Ethiopian and Neotropical regions might be mentioned two
species of Eolocampsa ; while the origin of the locust Orophila nuhigena
in the Canaiy Islands cannot, according to Krauss, be at present deter-
mined (p. 164).

The range of some of the European species of Orthoptera is of
special interest, though the Forficulida^ owing to their enormously
wide distribution, are of comparatively little use in helping us to
unravel the history of former geographical revolutions. The species of
Chelidura are, with us, mountain forms. Two are found on the Spanish
mountains ; two othera high up in the Pyrenees ; two on the Alps ; one
on the Italian mountains ; and one in Central Europe. It is therefore
remarkable, and particularly suggestive, that the two remaining species
known to science inhabit lower altitudes— one of them Madeira, and
the other Mexico. There arc three other genera whose range will be
of interest in connection with these inquiries — viz., Turpelia, hophya,
and Odontura. The former is confined to the West Indies, Mexico,
Brazil, and Madagascar ; while Isophi/a, of which a number of specics^
are known from the Mediterranean region, reappears again on the
opposite side of the Atlantic in Brazil. Finally, Odontura, a South
European genus, is represented by a single species in Patagonia.

Spiders and their allies.

The Azorean spiders, remarks Mr. Simon, are mostly of European
origin ; there are a few examples, however, which appear worthy of

1^92 Proceedings of the Royal Irhh Academy.

note. Theridionpulchellutn, common on the A zores and also on Madeira
and the Canary Islands, is likewise known from southern Europe, North
Ali'ica, St. Helena, and South America. Segestria JlorerUina occurs
in the Azores, the Canaries, the Mediterranean region, and St. Helena ;
while Dysdera crocata has a similar range, except that it has not yet
been discovered in the Canary Islands.

The distribution of the minute white Pedipalp ITansfiia mira-
bilis is most instructive. It lives uuder stones in southern Italy,
Sicily, and Tunis, and reappears again on the other side of the Atlantic,
in Chili and Texas.

There is an exceedingly peculiar Phalangid, Cryptostemnia^
evidently an ancient type, of which only a single species is known
to science — viz., C, Westermanni, which has been taken on the shores
of the Amazons in South America, and also on the banks of the Eiibi
river in "West Africa.

Such instances need no comment. But it may be mentioned that,
from a study of the distribution of the Scoi'pions and their allies, Mr.
Pocock has come to the conclusion (p. 230), which I have again and
again suppoi-ted in the preceding pages, that a Teiliary land-connection
must have existed between South America and Africa.

The Acarids of Central America und those of Central Europe appear
to be closely related, according to Professor StoU. In some instances
even the same species are represented in the two countries (p. 19).

Centipedes and their allies.

The Myriopod fauna of the Azores, described by Latzel, also very
closely resembles the European ; and there is, as in other groups, a
distinct indigenous element. Professor StoU draws attention to an
impoitant case of distribution — viz., that of the genus Folyxenus^
which includes very minute and delicate Millipedes living in secluded
localities, and being characterised by a nocturnal habitat und lack of
mobility. Three species are found in Europe and North Africa ; while
others occur in Guatemala, the West Indies, and the southern States
of America. A single species has been discovered in Ceylon (p. 25),

Crustaceanfi.

The terrestrial Isopod Crustaceans, or ** woodlice," are of much
value in zoogeographical research, as many of them pass their
exibtence entirely underground, and are therefore not liable to acci-
deutal dispersal. Some others, frequenting the crevices or bark of
trees, may occasionally be swept into the sea on floating timber, and on
rare occasions be stranded on a foreign shore. A few woodlice, no

ScHARFF — Same Itemarks on the AtlantU Problem, 293

doubt, arc introduced by man with plants and in packing-cases.
They not only form the exception among woodlice, but they rarely
spread far beyond human habitations, and are easily recognisable as
intruders.

From the researches of Dollf us and Norman, who have given us
valuable reports on the woodlice of the Azores and Madeira, we notice
that they are mostly identical with those of Europe and North Africa,
and that there is likewise an eudemic element. Some characters in
the fauna of these islands seem to support the view that they have
not long ago formed part of the continent of Europe. Philoxia
Cauehi — a species which occurs on the shores of the Mediterranean
under sea- weed, and extends along the Atlantic coasts of Eui'ope as
far north as the south of England — also inhabits the Azores and the
Canaiies. Among the rocky coasts of the Atlantic, wo find another
species, much more common than the last — viz., Li^ia oceanica, which
iB replaced in the Mediterranean by Z. italica ; but the latter form
tiinis up again on the coasts of the Canary Islands, the Azores, and
Madeira.

In the Canary Islands, a species of the peculiar blind woodlouse,
FiatyarthruSy has been discovered, which inhabits the subteiTanean
burrows of ants. One cannot conceive of any accidental means of
transport to an island of such a creature, and the occurrence of
FUtyarihrtts SchOhli — a Mediterranean form — in TenerifEe is a veiy
convincing argument in favour of a land-conneetion with North Af lica
in late Tertiary times. Nineteen species of terrestrial Isopods ai'e
known fi'om the Cauary Islands, sixteen from the Azores, and twelve
from Madeira.

There is little in the crustacean fauna of either Madeira or the A zores
which might lead us to believe that they were once connected by land
with America; but it is different with the Canary Islands, which
probably formed part of a laud stretching, as was suggested before,
from North Africa to South America.

The genus PlatyarthruSj including several small blind subterranean
species, is represented by three species in "Western Europe and North
Africa, one of which, as we have seen, reached the Canaiy Islands.
The only other species of the genus P. Simoni has been discovered in
Venezuela, in South America.

Take again, PorcelliOy an almost essentially European and North
African genus. We find one species peculiar to Venezuela. Metopo-
northus, evidently an ancient genus, is also mostly European ; but
one species has spread eastwaid to Sumatra, another is found in North

294 Proceedings of the Royal Irish Acadetny,

America, a third in Mexico, while still another has found means to
reach Madagascar. Similarly, Philoscia^ of which there are a number
of European species, has two in North America, ten species in South
America, one in Madagascar, one in Zanzibar, and another in Borneo.

There is, however, another group of Crustacea which irields such
decisive indications of the former land-counertion between Africa and
South America that scarcely an3rthingelBe is needed to put that theory
on a firm basis. The group refeiTcd to is that of the fresh- water
Decapods (cf. Ortmann, A), the species on both sides of the Atlantic
showing a most remarkable affinity.

There are, in the first place, to be mentioned two species of Atya,
A. scabra, occurring in Central America, the West Indies, and also
on the Cape Verde Islands, off the west coast of Africa ; while A.
gahonensis inhabits the Orinoco liver in South America, and the
Oabun river in West Africa. Next we have two species of fresh-water
Palajmon — viz., P. jamaicensis and P, olfersi^ the former being known
from South and Central America, southern California, and the West
Indies, and also from Liberia, the Niger river, and the Congo, in West
Aiiica. PalamoH olfersi has been recorded from the West Indies,
from Brazil, and from the island of St. Thomas on the east coast of
• Africa.

Dr. Ortmann points out that the crustacean fauna of the East-
American literal region exhibits a veiy marked relationship to that of
western Africa — such species as RemipeB cubemis, Calappa marmoratay
and Callinectes diacanthus occurring on both sides of the Atlantic.
This fact appears to me to be suggestive of a former coast-line having
existed acix)S8 the Atlantic, along which these shallow-water forms
migrated from one country to the other. Dr. Ortmann, however,
is of opinion (B, p. 84) that the larvae of these species had been able to
cross the Atlantic by means of ocean currents.

Worms.

There is one more group of invertebrates to be considered,
which is of importance in deciding questions of former geographical
revolutions — viz. , that of the earthworms. Land-planarians, no doubt,
might be most useful in aiding us to unravel problems of zoogeography ;
but their distribution is as yet very little known, and their relation-
ship has not been sufficiently studied for this purpose.

The ocean, according to Mr. Beddard, is an insuperable barrier to
most of the earthworms, and more effective than any other. The
latter are therefore, as that author remarks, exceptionally qualified

ScHABFP — Some Bemarka an the Atlaniie Problem. 295

for oaretul oondderatian in relation to the theories of past changes of
land (p. 57).

As regards the earthworms of the Atlantic Islands, most of them
are identical with European or North African forms. Whenever
peculiar species occur, — such as Helodrilue MUbii ol Madeira — they are
closely allied to continental ones. The particular species mentioned
has its nearest relation {H. Molhri) in Portugal. Several of the rarer
forms, — such as Dendrobana madeirensu of Madeira and Bimmtue
Meeni of the Azores, — occur also on the part of the Continent nearest to
them — viz., in Portugal.

As for any indications of an American relationship among the
Atlantic Island earthworms, there are some species — such as PherHima
caUfamiea of Madeira and P. barbadensts of Teneriffe — which may
be of American origin ; but they probably owe their existence on the
islands to artificial introduction.

The problem of the former land-connection between America and
Africa also receives some support frooi the distribution of earth-
worms. The &mily GeoscolicidsB is almost entirely confined to
South America and Africa ; only a few species reach the Paleearctic and
Oriental regions.

Among the genera which indicate the former union between the
two continents might be mentioned Gordiodrilue — which is confined to
the West Indies, the Gold Coast, and Zanzibar — and. Nematogenia^
which is only found at Lagos in West Africa, and at Panama in
Central America (cf . Michaelsen).

Very little is, as yet, known of the leech-fauna of the Atlantic
Islands, but our common European horse-leech {Eamopu %anguiiuga)
has been met with in the Azores ; while Dina Blaiiei^ a common
Mediterranean species, occurs on several islands of the Azorean
archipelago, and also on Madeira.

Conclusions.

The conclusions based upon this general survey of the fauna of
the Atlantic Islands are of more value than if only a single group
of animals had been taken into consideration. It will also be con-
ceded that, from the facts and examples I have collected, we are
entitled to form a very definite judgment on the subject, though my
interpretation of these may not appeal to all.

The great importance of such data of distribution, as factors in
solving problems of former geographical changes, is now generally
recognized. Dr. Wallace was the first to appreciate the bearing of a

11.I.A. PKOC., VOL. XXIY., SBC. B.] 2 B

296 Proceedings^ of the Royal LHah Academy.

study of zoogeography in the determination of past changes in the sur-
face of the glohe. ' ' It is certainly a wonderful and unexpected fact/' he
says (C, p. 14), " that an accurate knowledge of the distrihution of
birds and insects should enable us to map out lands and continents
which disappeared beneath the ocean long before the earliest traditions
of the human race."

That certain species are occasionally liable to be accidentally
carried away from their homes towards distant lands, has been noted
and referred to by Lyell, by Darwin, by "Wallace, and many others.
Yet, owing to a variety of circumstances, and especially the great
difficulty such species have in maintaining a foothold in their newly
acquired quarters, such species do not persist as a rule in large
numbers in any country. Altogether I am convinced that their
influence in the permanent colonization of a country has been
exaggerated. Actual observations of accidental introductions, more-
over, have only been made in exceedingly few instances ; while there
are numerous records to show that such occasional intruders rarely
become established. In order to find out whether animals could
traverse oceans and thus populate islands, Darwin and others have
attempted to determine experimentally how long certain snails could
stand immersion in sea-water. For one of their experiments Cydo-
stoma eleyans was taken, a snail provided with a lid or operculum
which can be closed over the mouth of the shell. This provision of
nature enabled the creature to withstand a fortnight's inmiersion in
sea- water ; and one would imagine such a species to be easily trans-
ported by sea to distant islands in that time. Cyelostoma elegans is
common on the western border of France and England ; but though
dead shells of the species have been cast upon the shores of Ireland
repeatedly, and probably living ones as well, it has never become
established on this island. If a species so particularly favoured by.
nature to resist the deleterious influences of sea-water is unable to
establish itself in a neighbouring island, what chances are there for
less suitably endowed forms to cross the ocean ?

But in studying zoogeographical problems such as the one I have
endeavoured to solve, it is not at all necessary to base a theory on
species which can be accidentally transported by hurricanes or marine
currents. We can confine ourselves to those whose habitats preclude
the possibility of occasional dispersal. As such I consider Plutonia
atlantiea — a slug-like creature living altogether underground in the
Azores ; the fresh- water crayfish, confined to South America and West
Africa ; the blind woodlouse Platyarthrus, inhabiting ants' nests in

ScHARPF — Same Memarks ofi the Atlaniia Problem. 297

Westem Europe and Yenezuela; the buirowing AmphisbaenidsB,
whose range is restricted to America, Africa, and the Mediterranean
Eegion ; and many others alluded to in the preceding pages.

From the facts quoted, I conclude that Madeira and the Azores, up
to Miocene times, were connected with Portugal ; and that from
Marocco to the Canary Islands, and from them to South America,
stretched a vast land which extended southward certainly as far as
St. Helena. This great Continent may have existed already in
Secondary times, as Dr. Ihering suggested ; and it probably began to
subside in early Tertiary times. But I think its northern portions
persisted until the Miocene Epoch, when the southern and northern
Atlantic became joined, and the Azores and Madeira became isolated
from Europe.

This, however, does not explain the whole history of the Atlantic
Islands. To account for the extraordinary predominance of the
Mediterranean element iu their fauna, they must have again united
with the Old World in more recent times. This took place, no doubt,
in precisely a similar manner as before ; and I believe they were
still connected, iu early Pleistocene times, with the continents of
Europe and Africa, at a time when man had already made his appear-
ance in western Europe, and was able to reach the islands by land.

APPENDIX.
List op Woeks Aim Papebs sefebbed to in the vbscsdisb Pages.

Beddabd, F. E. :

A text-book of Zoogeography. London, 1 895.

Blanfobd, W. T. :

Address delivered at the anniversary meeting of the G^logical
Society, London, 1890.

Chaxpiok, C. :

List of the Cicindelidae, &g., of Gibraltar. Trans. Entomol. Soc.
London, 1898-99.

Chbibt, H. :

Vegetation und Flora d. canarischon Inseln. Engler's botanische
Jahrbiicher. Vol. 6. 1885.

dJ92

298 Proceedings of the J^al Irish Aeadettiff.

Dahl, F. :

Die Landfauna d. A90ren. Plankton Expedition d. Humboldt-
Stiftung. 1892.

DoLLFUS, A. :

Isopodes terr. recueilliB aux Aqotcs. Bevne biol. du Nord de la
France. 1888-1889.

DaoujET, H. :

El&ncnts de la faune aQorSenne. Paris, 1861.
Ekgleb, a. :

Yersuch einer Entwicklungsgeschichte d. extratrop. Florengo-
biete d. nordl. Henuspheere. Leipzig, 1879.

FoBBES, £. :

On the connexion between the distribution of the existing fauna
and flora of the British Islands, &c. Geological Memoirs.
Vol.1. 1846.

Fbhsch, K. von. and Rein, W. :

Geologische Beschreibung d. Insel Teneriffe. 1868.
Gadow, H. :

Amphibia and Eeptiles. London, 1901.
GoDHAir, F. du Cane :

Natural Historj of the Azores. London, 1870.
GBBGOBr, J. "W. :

Contributions to the Palaeontology and Physical Geology of the
West Indies. Quart. Journal, Geological Society. London.
Vol. 51. 1895.

GUNTHEB, A. :

An introduction to the study of. Fishes. Edinburgh. 1880.
Gtjppt, R. J. L. :

Notes on West Indian Geology with remarks on the existence of
an Atlantis. Geolog. Magazine. Vol. 4. .1867.

Habtukg, G. :

Geologische Beschreibung d. Inseln Madeira und Porto Santo.
Leipzig. 1864.

Heolky, C. :

The Bange of Placostylus. l^roc. Linnoan Hoc, New South
Wales (S, 2). Vol. 7. 1892*

BcHARFF— iS(W)tf ttemarks on tt$ Atlantis Probkm. 299

Hnm, 0. :

On the probable origin of the organized beings now living in the
Azores, &c. Ann/and Mag. Nat Hist. (S. 2). Vol. 18
1856.

Hbtdxk, Beittsb, and Wxisk :

CatalogoB coleopteronun EuropsB. 1891.

HouG, E. :

Les geos3rnclinaux et lee aires continentales. Bnll. Soc. G60I0-
gique de France (3). Vol.28. 1900.

Howes, 0. B. :

Address to the Zool. Section, Brit. Assoc. Meeting at Belfast, 1902.

InxBiHG, H. Ton :

A. Das neotropischo Florengebict und seine Gtesohichte. Engler's

botanische Jahrbucher. Vol. 17. 1893.

Ihebiho, H. Ton :

B. Najaden yon S&o Paulo und die geograph, Verbreitung"',d.

Sussw.-faunen r. Sud.-Amerika. Archiv f. Naturgesch. 1893.

Ihxrivg, H. Ton :

C. Die Insel Fernando de Noronha. Globus. Vol. 62. 1893.
Ihbbisg, H. Ton :

D. Die Ameisen Ton Bio Grande do Sul. Berliner entomol. Zeit-

schrift. Vol.39. 1894.
KoKiG, A. :

TenerifiEa in zoologischer Beziehung. Verhandl. Nat Ver. d.
preuss. Bheinlande. Vol. .47. 1890.
KOBELT, W. :

A. Die geographische Vertheilung der MoUusken. Abhandl.

Senckenberg naturf. Gesellsch. 1874-76. & Jahrb. d. d.
malakazool. Gesellsch. 1878-80.
KOBKLT, W. :

B. Das Verhaltnissd. Europaisch. Landmolluskenfau&a zur west-

indisch.-centralamerikanischen. Nachxichtsbl. d. d. mala-
kozooL Gesell3ch. Vol. 19. 1887.

EOBELT, W. :

C. Die HoUuskenfaunad. makarones. Inseln. Jahrb. d. nassau.

Verein Naturl(. Vol. 49. 1896.

300 ProeeediffgB of the Royal Irish Academy*

EoBXLT, W. :

D. Die Fauna d. atlantiscben Inseln. Nachrichtsbl. d. d. mala-
kozool. GeBellsch. 1887.

EoBSLT, H. :

£. Die zoogeograpbiBche Stellung d. Insel St. Helena. Geo-
graphiBche Zeitecbrift. Yol. 2. 1896.

E&Auss, H. :

Systemat. YerzeichniBS d. cananBch. Dermapteren und Orthop-
teren. Zoologischer Anzeiger. Yol. 15. 1892.

LAPFAsnrr, A. de :

Traite de G6ologie. (3« 6d.). Paris, 1893.

Laizbl, B. :

Contrib. k P^tude de la faune d. Myriopodes d. A9oreB. Eevue
biologique du Nord de la France. 1889.

Ltxll, C. :

Principles of Geology, lltbed. Yol. 2. 1872.

Ltdxkkeb, R. :

A G^ograpbical History of Mammals. Cambridge, 1896.

MacLachlaf, R. :

Tbe Neuroptera of Madeira and tbe Canary Islands. Joum. Linn.
Soc. Yol. 16. 1882.

MuoB, C. J. F. :

On fossil and recent Lagomorpba. Trans. Idnn. Soc. London
(S. 2), Zoology. Yol. 7. 1898.

MiCHAELBSK, W.:

Oligocbseta (Das Tierreicb). Leipzig, 1900.
MuBBiT, A. :

A. The Geographical Distribution of Mammals. London, 1866.

MuuuT, A. :

B. On the (Geographical Belations of the chief coleopterous Faunas.

Joum. Linnean.Soc. (Zoology). Yol. 11. 1870.

MVBBAT, A. :

C. On the Geographical Belations of the Coleoptera of Old

Calabar. Trans. Linn. Soc. Yol. 28. 1862.

ScHAKFF—SoiPie Remarks on the Atlantis Problem. 301

TSmouayUj M. :

Eidgeschichte. Yol. 2. 1887.
NoBMAv, A. M. :

The Land Isopoda of Madeira. Ann. Nat. Hist. (7). Yol. 8. 1 899.

Oijtxb:

The Atlantis Hypothesis in its Botanical aspect. Natural History
Eeview. 1862.

ORncANir, A. E. :

A. Os Gamaroes da agoa doce da America do sul. Eevista do
Mnseu Panlista. Yol. 2. 1897.

Obtmahn, a. E. :

B' OmndzUge d. marinen Thiergeographie. Jena: 1896.
OsBOBK, H. F. :

Correlation between tertiary mammal horizons of Europe and
America. Ann. New York Acad. Science. Yol. 13. 1900.
PococK, R. J. :

The Geographical Distribution of the Aiachnida of the orders
Pedipalpi and Solifugse. Natural Science. 1899.

Rebel, H. & RoeENHOFKB, A. :

Zur Lepidopterenfuuna d. Ganaren. Annalen d. k.k. natnrhist.
Hof museums. Yol. 9. 1894.

RiBLKT, H. N. :

Notes on the Zoology of Fernando Noronha. Joum. Linn. Soc.
(Zoology). Yol. 20. 1886.

SCLATBB, W. L., & SCLATEB, P. L. :

The Geography of Mammals. London, 1899.

SxMOK, E. :

Mat£riaux pour serrir 4 la faune archnologique des lies de I'ocean
Atlantique. Annales d. la Soc. Entom'.* d. France (6).
Vol. 3. 1883.

SiHBOTH, H. :

A. Zur Eenntniss d. Azorenfauna. Archiy f. Naturgeschiohte.

64 Jahrg. Yol. 1. 1888.

SiMBOTH, H. :

B. Die Nacktschnecken d. portugiesisch-azorischen fauna. Nova

Acta d. d. Akad. d. Naturf. Yol. 56. 1891.

802 Proceedings of the Royal L-ish Acadetny.

Stsikdachnbii, F. :

TJeber die Eeptilien and Batracbier d. canarisch. loBeln. Annalen
d. \. k. nattirli. Hof museums. Yol. 6. 1891.

On the Land-shells of St. Helena. Froc. Zool. Soc. London,
1892.

Stoix, 0. :

Zur Zoogeographie d. landbewohnenden Wirbellosen. Berlin,
1897.

SuBss, £. :

Are great ocean depths permanent ? Natural Science. Vol. 2.
1893.

Tbblsasb, W. :

Botanical Obserrations on the Azores. 8th Annual Beport of
the Missouri Botanical Garden, 1897.

Wallack, a. R. :

A. The Coleoptera of Madeira as illustrating the origin of insular

faunas. Studies, Scientific and Social (chapter xii.). Vol. 1.
1900.

Wallace, A. R. :

B. Island Life. 2nd Ed. London, 1892.
Wallace, A. R. :

C. The Malay Archipelago. 10th Ed. London, 1894.

Wild, J. J. :

Atlantic Soundings. Ifature. Vol. 15. 1877.

WiLSOK, D. :

The Lost Atlantis and other Ethnographic Studies, l^dinburgh,
1892.

WoLLASTOir, T. V, :

A. On additions to the Atlantic Coleoptera. Trans. Entomol.

Soc. London, 1871.

WOLLASTOK, T. V. :

B. Testacea Atlantica. London, 1878.

[ 303 J

VII.

ABSTKACT OF A PHYSIOLOGICAL HYPOTHESIS TO EX-
PLAIN THE WINTER WHITENING OF MAMMALS AND
BIRDS INHABITING SNOWY COUNTRIES, AND THE
MORE STRIKING POINTS IN THE DISTRIBUTION OF
WHITE IN VERTEBRATES GENERALLY.

By captain G. E. H. BARRETT-HAMILTON, B.A.,
F.Z.S., M.R.I.A.

Read Mat 11, 1903.

ALTHOV0H 80 much attention has been attracted to the subject of
the winter whitening of mammals and birds, no theory has, so &r as
I know, been advanced to explain the physiological meaning of this
phenomenon. On the other hand, Naturalists seem to be perfectly
agreed as to the advantages in a protective sense which the animals of
omowy countries derive from their seasonal change, it being regarded
as one of the most perfect of known instances of adaptation to
environment. I venture to believe that scientifio investigations have
now brought together facts sufficient to shed a glimmering of Hght
upon the physiological or primary meaning of the white seasonal
colour changes. I therefore put forward the following hypothesis,
which is based upon an intimate connexion between fat and animal
pigmentation.

I. The connexion between Fat and Vertebrate Pigmentation.

While I was in South Africa I was greatly surprised at the
number of species of birds in which the fat is more or less deeply
coloured, rich yellow, orange, or even reddish. I further found a
correspondence, often quite remarkable, between the colour of such
fat and the pigmentation of the feathers. Thus in birds in which
yellow appears in the plumage the fat is usually of a correspondingly
yellow or orange tint. Instances of this may be found in the Gb*eat-
tailed Widow Bird, the Cape Long-Claw, the Masked Weaver Bird,
and the Red and Taha Bishop Birds.

In other birds, although more rarely— to include some other than
South African species — as the Chough, Stork, and Flamingo, the

ith Afncan species — as the Cj

H.I.A. PBOC., VOL. XXIT., SBC. B.]

304 Proceedings of the Royal Irish Academy.

fat is red, in conjunction with a corresponding external colonratioD of
the legs and bill. It has been shown that for the Flamingo intensity
of the red colour is proportional to the amonnt of fat or oil present in
the feathers.

Again, fat saturated with pigment may be found in birds whose
feathers are deeply pigmented, yet whose colours are not necessarily
red or yellow. Such are the Coot and Blackbird, the latter of which,
as Dr. Gadow informs me, may be distinguished in this very particular
from the closely allied but less darkly coloured Song-thrush. In the
Sacred Ibis, a white bird with the bill, bare head and neck, and the
legs of the deepest black, the fat is intensely red.

I further observed that the feather-tracts are amongst the parts of
the body where fat especially tends to accumulate.

But deeply coloured fat is not confined to birds. It occurs also in
mammals, as is well known in the case of domestic cattle. I found it
also in wild forms, such as the South African Hedgehog and the Aard
Wolf. It is present in abundance in reptiles, such as the African
Monitor, and in the Lizard Agama distanti. In the Monitor and many
other reptiles, in which the fat is deeply pigmented, there is a strong
accumulation of black pigment in the body cavity. Lastly, the
common Salmon is an instance of the same thing amongst fish.

In all these creatures fat is frequently deposited in the ovaries,
testes, or other glands, which, as a result, are often yellow, sometimes
black, and to a similar cause the yolks of the eggs of birds owe their
colour. The yellow pigment of the yolks of fowl's eggs, called
by Erukenberg coriosulf erin, is said to be, like zoomelanin, a coloured
fatty oil.

The yellow or red fat pigments belong to the same class, that of
the lipochromes, as the reds and yellows of external pigmentation.
They are also found in certain vegetable substances, such as maize.
If an animal be fed on maize, the colour of its fat is greatly heightened.
A third important pigment, but a little less intimately connected with
fat, is black. The constituents of the two are very similar, zoomelanin
being, according to Bogdanow, composed of carbon, hydrogen,
nitrogen, and oxygen, as found in the black feathers of Pica, Corvus,
and Ciconia,

When we regard mammals and birds, we find, with few exceptions,
great uniformity of pigment-colours. Black, red, and yellow, with
their inter-mixtures, are almost the only three pigmentary colours
with which we have to deal. In the lower vertebrates the effect of
white is due also to a pigment, guanin, a purely waste product of

B.-Hamilton— Winter Whitening of Mammals and Birds. 305

nitrogen. Almost all other vertebrate colours are stractnral. The
basis of these structural colours is, however, always a strong deposit
of pigment.

Finally, we have evidence of the direct part played by fat in
animal colouration in the marked change which may be brought about
in the plumage of birds, such as the Bullfinch, by the ingestion of a
fatty food such as hempseed.

It has been pointed out to me^ that the pigmentation of the skin
which accompanies the peripheral distribution of fat in human
pregnancy is a fact which supports my views. Again, in the morbid
symptoms of Addison's disease abnormal pigmentation is remarkably
associated with abundance of fat, especially disfcributed subcutaneously
on the abdomen. Most striking of all, perhaps, is the removal of fat,
and with it pigment, from certain organs of the breeding Salmon
(muscles, intestines, and liver) for deposition in the ovaries and the
skin. The result is that these organs become pigmented, to use Miss
Newbiggin's expression, *' as it were incidentally in the life-history
of the individual under circumstances which render the question as
to the inheritance of acquired characters absolutely unimportant.''*

It seems, then, hardly possible to avoid connecting this pigmented
fat of animals with their external colouration. The suggestion is not,
I think, a new one.

II. The connexion between Fat and Winter Whitening,

Let us now apply these facts to the Arctic mammalia and birds,
first briefly recapitulating the known phenomena of winter whitening.
The phenomena are not nearly so isolated as seems to be generally
believed. From the always pure white (or rather yellowish) Polm*
Bear and the Snowy Owl, through the seasonably white Polar and
Alpine SEares and Ptarmigans, there are many intermediate degrees of
winter whitening, until the commencement of the process is just
visible in the numerous instances where northern animals wear a
winter coat or plumage lighter than that of summer (Squirrels,
Auks, Guillemots, &c.)*

The manner of this change of colour has been deeply discussed,
and has caused much disagreement. The Polar Hare, in particular.

^ By Dr. Gadow and Mr. Andersoiiy of Cambridge, who have been so kind
as to read my paper.

* See Report of Fishery Board for Scotland for 1898.

2C2

306 Proceedings of the Royal Irish Academy.

has formed the subject of many contradictory assertions. Whatever may
be true of other species, Mr. J. Barcroft could find no trace of an autumn
moult in the case of a Scotch Variable Hare, kept at my request at
Cambridge under constant observation in the Physiological Laboratory
during the autumn of 1899, and which had turned almost completely
"white by January of 1900. In the very same winter a wild Variable
Hare which lived close to this house turned from half to almost
completely white within the space of a few days in late December.
The first half of the change had been accomplished earlier in the
season. It is, therefore, very hard to believe that the positive
statements made in regard to the change of colour of Zepus americanui
by moult alone can apply to the Variable Hares of Europe as regards
the autumn change. The spring change, on the other hand, appears
to be due in all cases which have been studied to a change of coat.

Summarised from a physiological aspect, it would appear that
there are two conditions of the animal's body, in one of which white,
in the other pigmented, hairs are produced.

That this is so cannot indeed be doubted in view of the experiments
of Professor Halliburton and Drs. Brodie and Pickering.^ These
investigators have shown that the condition of the intravascular blood
varies in animals which, like the Arctic Hare, are sometimes white,
and at other times in a pigmented state. Further, that the composition
of the blood of an animal which has undergone winter whitening is
similar to that of a permanent albino. The presence or absence of
pigment is then but the external evidence of changes occurring
internally in an animal possessing a varying metabolism. So the
white hairs of Arctic animals must be regarded as due to a cause
similar to that which brings about absence of pigment in albinos, and,
almost certainly also, in aged animals.

And, since it has been shown that Arctic animals possess a varying
metabolism, it seems most reasonable to suppose that the vitid
changes reach their lowest point at the same season as in the human
race, for which physicians accept a metabolism at its lowest in autumn.
That is to say, metabolism is lowest just at about the very time of
the change from brown to white (Dr. J. Netton Radcliffe in Quain's
Medical Dictionary, p. 114).

It is at this very season that there comes the shock to the system
of the onset of the cold of the Arctic winter, and, as is well known,

* Journal of Physiology, xvi., p. 136, 1894 ; xviii., p. 285, 1896 ; xx., pp. 310
to 315, 1896.

B.-Hamilton— JFintei' Whitening ofMamnmh and Birds. 307

heat and cold exert very serious influences on animal organisms. In
the human body, for instance, '' continued exposure to such degrees of
cold as is yet not incompatible with the maintenance of life, neverthe-
less keeps at low ebb activity of nutrition and function alike"
(Dr. A. E. Durham in Quain's Medical Dictionary, p. 270). The action
of the skin is sluggish, that of the kidneys more active. Under an
increase of temperature, on the contrary, the exhalation of carbonic
acid and of water is lessened ; the urine diminishes in quantity and
contains less urea and chlorides. But the skin acts much more freely,
its secretion being increased by about 24 per cent.

One of the concomitants of a sluggish metabolism is diminished
oxydisation and consequent storage of fat. For instance, fat is readily
accumulated by castrated animals in contrast to those which are in
full sexual activity. Many animalB accumulate fat during one season
of the year, and utilize it during the breeding season. In particular,
this has been shown in considerable detail by Dr. Noel Paton and his
colleagues of the Fishery Board for Scotland to be true of the Salmon,
while the African Mud-Fish {FfotcpUrua) is another weU-known
example.

I have already shown the connexion between fat and animal
pigmentation. It seems then not an unreasonable suggestion that the
temporary cessation of metabolism of fat and the absence of pigment
are part of the same process. In the autumnal season the metabolism
:grows more and more sluggish, particularly in the periphery of the
body, until there comes a time when a maximum of fat remains idle
internally.

There is, as it were, to use a graphic expression, at one time a
centrifugal, at another a centripetal, condition of fat. And, since the
pigment accompanies the fat, any hairs or feathers which grow
during the prevalence of the centripetal condition are white to an
extent which is evident more or less according to the intensity of
the physiological influences at work. Hence results a condition of
things wherein exists great opportunity for the play of such diverse
factors as heredity, individual temperament, and the influence of
external conditions. This exactly corresponds to the observed facts,
always so puzzling, often at first sight so contradictory.

But not only are new hairs white. Eventually there comes a time
when, with a constantly lowering metabolism, not only is peripheral
activity sluggish, but, as shown by Mr. Barcro|t, material is actually
recalled from the hairs, no doubt for the internal uses of the body.
The working of this process has been observed by Professor

308 Proceedings of the Boyal Irish Academy.

Metchnikoff,^ who in the case of old men and dogs saw the phagocytes
passing £rom the medullaiy to the cortical layers of whitening hairs.
These phagocytes ingest the pigment granules, and remove them into
the hody — a process which, as Professor Metchnikoff helieves, *' can be
classed under the general law of atrophy of solid parts of the
organism." And that the organism is at this time economical of its
resources, and unwilling to waste them peripherally, is probably
indicated by the fact that the siUdness of the winter hairs of animals
inhabiting cold countries indicates a fineness of texture, that is to say,
less material is appropriated ^for their manufacture in proportion to
their length than for the summer hairs. Since the animal organism
has power to recall pigment from its hair, it matters not whether or
no the physiological causes of winter whitening culminate at the time
of a moult. Once the required condition prevails, new hairs will
grow, and already existing hairs will rapidly become white.

As to the reverse process, the recolouration of the coat, years of
study have failed to supply me with an instance of its occurrence
without a moult : so that the conclusion seems hardly avoidable that
the hairs, once whitened, are dead to further chimges of colour.
Thus is explained the curious fact that in the mild climate of the district
where I write, such winter whitening as occurs in the Hares (and
it is sometimes considerable in extent) remains in force, no matter
what the weather may be, until the spring moult. As this moult does
not take place until l^e spring has well advanced, those Hares which
have undergone the most complete winter change are for some little
time incongruously conspicuous in the flowery meadows of the south of
Ireland, Trhile the April and May sunshine lights up their Arctic
livery. At the time of the spring moult, the physiological causes
which led to the whitening of the previous autumn having now passed
away, vital change being now at its high-water mark, and fat, and
with it pigment, available for constructive purposes throughout the
body, the new coat (or, in the case of birds, the plumage) comes up
of tiie pigmented summer tint.

But it is not fair to regard as typical of its kind the cycle of winter
whitening as observed in England or Ireland. Here at the southern
limit of the conditions which have called it into existence, the process
is complicated by numbers of contradictory factors, the resultant of
which is a considerable modification and obscuring of the typical
phenomena. In the Arctic regions these are nearly uniform.

^Froc. Roy. Soc., London, vol. box., p. 156, 1901.

B.-Hamilton— Winter Whitening ofMammah and Birds. 309

We have then in winter whitening an instance of peripheral atrophy
of the hair or feathers — an atrophy which manifests itself more or less
in all the members of the Fauna of cold oonntries, and which may be
partial or complete, seasonal or permanent.

Not the least remarkable feature of this atrophy of winter whitening
is the fact that the order of the parts affected by it is to all intents and
purposes the same in all mammals, even in those so widely separated
as the Stoat and Hare« Excluding, for purposes of this paper, the
head, the change begins from the base of the tail at the posterior
margin of the back, and on the flanks, just where the dorsal colour
meets the white of the underside. It then creeps up the back. In
many animals, as in British specimens of the Common Hare {Ltpus
europuBus Pallas), which frequently whitens to a slight extent, it rarely
climbs higher than the rump. In spring the moult, and with it the
brown colour, progresses in exactly the opposite order, creeping down
the back, and extending to the sides until it reaches the permanent
hairier of the white belly. It is, in fact, as if the internal physiological
condition represented by the white belly annually oycrpowers more
than its ordinary share of the animal in autumn, and in its ascent
reaches a height dependent upon its energy, to remain in possession
until driven out in spring by the way it came.

I look upon this fact as a conflrmation of my hypothesis that
winter whitening is connected with the fat of the body and its distri-
bution. For it seems more than a mere coincidence that the upward
march of this winter whitening and the order of the parts affected by
it is almost exactly indicative of the order in which fat is accumu-
•lated internally in an ox, sheep, or fowl — an order which is probably
applicable to other mammals and birds also. In oxen, sheep, and
fowls, as in man and most vertebrates, the favourite region for fat
storage is the belly, where, besides being deposited on the deeper
organs, such as the kidneys, it forms a layer known as the pannicuiui
adipoiw, lying near the surface, between the skin and the abdominal
walls. Next in order, as regards the accumulation of fat, comes
the rump, and thirdly, portions of the neck region and of the back
and ribs.

And since we know that the presence of fat is indicative of defi-
cient oxydation, it is not altogether surprising to find external
atrophy its accompaniment.

For the success of my theory two crucial tests have been suggested,
either of which might be performed by experiment. If my supposition
be true, there should (it has been thought) be more fat in hair in

310 Ptoceedinga of the Royal Irish Academy.

summer than in winter ; and, further, the fat of an animal should be
more deeply coloured in winter than in summer. It is obvious that
these experimental proofs of my theory cannot be performed all at
once. But I do not care to delay publication until I can myself
institute the necessary investigations, since some other worker may
possibly be in a position to do so. As regards the second point, I am
not at aU sure that it is actually necessary that the winter fat be more
deeply pigmented than that of the summer. The point is rather that
there should be more unoxygenated fat peripherally at the time of the
autumn than of the spring change ; and that this is so I can myself
answer in the affirmative. It is full of significance, also, that the
muscle of the breeding Salmon becomes pale on transference of the
fat and pigment to the genitalia or for combustion as a source of
energy.

III. The meaning of white in Domeetie Animalt and Vertebratee

generally.

If my conclusions be accepted, it seems that we may be hovering
somewhere near the explanation of the primary or physiological
meaning of many puzzles of animal colouration. The widespread
existence of white undersides in vertebrates — a fact only as yet
explained on purely secondary grounds (as by the ingenious suggestion
of Mr. Abott H. Thayer) — is now seen to have a direct connexion
with the main peripheral fat-tract of the body. The white rumps of
birds and mammals, the familiar *' recognition marks" of Mr. Alfred
Eussel Wallace, correspond to one of the next most important fat-*
tracts, light neck- and ring-marks to yet another.

I am inclined to push my theory even further, since I see in it the
explanation, often vainly sought for, of the marked extent to which
the white colour makes its appearance in domestic animals. Since
nearly all these animals derive their commercial value from their power
of accumulating fat, it is natural that, if my suppositions be true, the
pigmentation of the hair (or feathers) should be affected. I am aware
that in many breeds the appearance of the white patches is believed
to be quite irregular, and not to follow any definite order of fre-
quency as regards the regions affected. I feel sure, however, that
further investigations will show that this is not really the case. Thus
my own studies, unfortunately as yet incomplete, indicate that even
in such, at first sight, irregularly-marked animals as cattle, the
markings, although undoubtedly subject to very great latitude, tend

B.-Hamilton — Winter Whitening o/Mamtnals and Birds. 311

to arrange themBelves in. accordance with one or two definite patterns.
The latitude is no doubt due to the fact that, although following the
general order described labove, the pannietdut adiposut of domestic
animals is almost universally distributed over the body, and varies
only with each animal's individual idiosyncrasy of constitution. It is
noticeable, moreover, that the Hereford breed of cattle, in which the
arrangement of the fat differs from that of other breeds, it being
mainly distributed peripherally, is distinguished by regularity of
pattern, having the principal peripheral fat-tracts clearly mapped out
in white.

The accumulation of fat in a fattening ox is, however, marked,
not by loss of pigment, but of the hair, the skin becoming bare, par-
ticularly on the rump and neck, as the animal ripens. This, then, is
only another aspect of the atrophy which may accompany deposits of
fat under the skin. Here again I find more than a mere coincidence in
the fact that the bare buttocks of monkeys correspond to the light
rump-patches of many other vertebrates ; further, that the accumu-
lation of subcutaneous fat in marine mammals is correlated with
deficiency of hair, in a graduating series, from the amphibian warmly
furred Fur-seals to the completely aquatic hairless Cetacea and Sirenia.

A great difficulty for some time lay in the way of my theory,
namely, the occasional reversal of the ordinary arrangement of
vertebrate colouration, whereby the ventral is usually the lighter, the
dorsal the darker sur&ce. For instance, in the Skunks, Polecats,
and the Eider Duck, the upper surface is conspicuously lighter than
the under. These facts were not at all explained by Mr. Thayer's
hypothesis, and each case is usually argued on its own merits, the
Skunk's white back being regarded as a warning of its bearer's
malodorous nature, the Eider Duck's as protective to the sitting-bird,
and so on ; they certainly proved a stumbling-block to me. I hardly
felt bold enough to predict that the unusual arrangement would be
found to correspond with a like internal disposition of the pannietdua
adipo8UB\ and no other supposition, unless, as it were, some d&UB $x
machinay in the shape of an ingenious secondary explanation, seemed
likely to be able to pull me round the difficulty.

A second difficulty lay in the fact that the heads of vertebrates are
very frequentiy the centre of conspicuous light marks or bars, which,
while not apparently related to any internal fat-tracts, are yet so
similar in many widely-distinct forms (such as mammals and birds] as
to be without doubt due to some similar cause in all.

Most fortunately both these difficulties were simultaneously and,

312 Proceedings of the Royal Irish Academy.

as I think, most remarkably, removed by the accidental trapping of a
Badger. In this animal, as is well known, although the case is not
nearly so conspicuous as are many others, tiie deepest tints occur on
the under side. In the particular specimen (a male) the upper surface
of the body, with the whole tail, and the inguinal region, was con-
trasted by its light colouration with the remaining portions of the
under side. On examining the carcase, I found that the lightest ex-
ternal parts, viz., the rump, tail, and inguinal region, lay over the
thickest accumulation of fat. A thinner layer of fat extended over
the whole back, whereas the upper belly and breast were almost free
from fat. Thus external colouration was here directly correlated with
the distribution of the panniculus adiposus. In view of this fact, it
seems probable that certain of the colour-differences which help to
distinguish some of the foreign species of Badgers from our own are
due to further developments of the panniculus adiposus^ for they
frequently follow the line which would be taken by winter whitening
on a Hare or that iu which fat is deposited on an ox. That is to
say, when a Badger differs from our own species in regard to the
lightness of its upper surface, we may almost predict that the lightest
part will be the rump, while in another species the white will have
undergone further extension up the back. Thus a series of skins of
tlie various species of Badgers may be almost made to match (in
regard to the whiteness of the upper side), one consisting of the skins
of Hares in process of whitening.

The second difficulty was upset in a most unexpected manner when
I came to examine the animal's head. There I found that the three
white bands lay over three regions of the skull where no flesh intervenes
between the bone and the skin : the three white external bands were,
in fact, clearly marked out by three similar cranial bands of ligament
and bone. Here, then, was the most unexpected fact that not only
may deficiency of pigment be associated with the presence of under-
lying subcutaneous fat, but also with that of bone or ligament. So
that we may almost lay down the law that there is a tendency to
pigmentary aberrations at those parts of the body where nutrition of
the skin is interfered with by contact with underlying fat, bone, or
ligament, or better, that adjoining regions of uneven nutrition tend to
originate unevenness of external pigmentation — a result which Mr.
Alfred Tylor missed by very little in 1886.*

It is obvious that this conclusion, if further borne out, may exercise

* *' Colouiation in AnimalB and Plants ; " London, 1886.

B.-Hamilton — Winter Whitening ofMamniah and Birds. 313

a profound influence upon cuirent views of animal colouration. For
instance, it at once dawned upon me that therein lies the explanation
of the white '' blaze " of so many domestic animals, and in particular
of horses. This is usually situated over the frontal or nasal bones
where they lie directly under the skin. Again, the fact that in man
baldness occurs first in corresponding regions is almost certainly but
another instance of the working of the same law.

Although thus pushing my theory to lengths which have, I believe,
been imtouched by any other view, I must be the first to point out its
own restrictions. I have at present, at all events (although I confess
I begin to see light here .also), no desire to connect it with such
complicated colour schemes as the spots of the Leopard or the stripes
of l^e Tiger and Zebra. It is further evident that puzzles like the
curious arrangement of the white areas on the tails of birds, or the
restriction of pigment to the upper side of a flat-fish, are phenomena
which, although probably connected in their origin with subcutaneous
fat, seem to require some further factor for their full explanation.

In birds, for instance, without entering into detailed descriptions of
what is perfectly well known to naturalists, the light patterns on the
rectrices are frequently the sum of a series of complicated markings,
different as to each individual feather, but fitting into their place like
the pieces of a mosaic. Now, although the deficiency of pigment in
this case is, on my showing, certainly connected in a general manner
with the fat-tract of the region whence these feathers spring, it is
hard to see how all the complicated details of the pattern can be thereby
explained. But in view of Dr. Finsen's discoveries, it does not seem
too great a stretch of imagination to suppose that the exact distri-
bution of the pigment may be not unaffected by the varying amount
of light to which the different parts of the feathers are subjected,
or again that the pigmentary differences between the two surfaces
of a flat-flsh may have in a like manner been due, although exactly
how we do not understand, to unequal stimulations of the light
which they receive.

Thus, then, I have no wish to extend my arguments universally
to white colouration in nature, since there may undoubtedly be causes
other than atrophy which result in absence of pigment. It is obvious
also that many animals are not subject to the hair-atrophy which in
others follows the peripheral accumulation of fat. On the other
hand, it may well behove Zoologists to consider not only the external
advantages accruing from but the deep-seated physiological processes
involved in seasonal colour changes. Even those connected with sex

314 Proceedings of the Royal Irish Academy.

may but represent the external Bymptoms of a Yarying metabolism.
They may be, as I have elsewhere suggested, primarily but the
symptoms of a pathological or quasi-pathological condition, the
importance of which to the organism must quite overshadow any
external applications for ornament or protection.

In conclusion, I submit that my hypothesis, although it may not
explain the minutiffi of each individual case, throws a distinct light on
the phenomena of winter whitening, and through it of animal coloura-
tion as a whole. It also illustrates the possibility in nature that
characters having a definite physiological or primary meaning may be
found useful for some quite secondary external purpose.

[ 316 ]

vin.

AN ADDITION TO THE LIST OF BKITISH BOREAL^

MAMMALS,

Bt captain G. E. H. BARRETT-HAMILTON, B.A.,
F.Z.S., M.R.LA.

Bead Mat 11, 1903.

Ik the present Paper I wish to describe a remarkable Bank Yole
or Red-backed Mouse (Motomys) inhabiting the small island of
Skomer, off the coast of Pembrokeshire, Wales.

Skomer Island' is said to owe its Danish name, which, according
to some writers, signifies '* the rocky," to its rough character. It
is the haunt of immense numbers of Puffins, JFVatereuia aretica^ and
of Manx Shearwaters, JPuffinua anglomm. It has an area of about 700
acres, and, forming the southern horn of the crescent of St. Bride's
Bay, is parted from the mainland by a narrow sound some two miles
wide. There is but one house upon the island ; in connexion with
this there are about 250 acres of cultivation. The island is without
bush or tree, and is said to be very wind-swept.

I first heard rumours of the existence of a peculiar Yole on Skomer
Island in or about the year 1898. In October of that year Mr. H. W.
Marsden, of Clifton, was so kind as to send me a pair. They had
been caught by Dr. Y. H. Mills, of Haverford-West. Dr. Mills has
since obtained for me several excellent specimens, so that I now
possess a dozen in all.

I believe, however, that Mr. R. Drane, of Cardiff, deserves the
credit of having been the first to collect and recognise the interesting
character of the Skomer Yoles. Mr. Drane sent specimens for exhibi-
tion to the Linnean Society of London, but they were regarded by the
members present as '' the Common Bank Yole, M%erotu» glareoluiJ^^
Mr. Drane's own opinion, however, as expressed to me in a letter, is
both different and decided. He wrote : " They are, I contend, a local

' These details are taken mainly from the Bev. Murray A. Mathew's ** The
Birds of Pembrokeshire and its Islands," 1894, pp. xxx to tttj.

*Proc. Linn. Soo. Lond., Jane, 1899, p. 63.

' I use the term throughout in the sense given to it in the works of American
writers on geographical distribution.

316 Proceedings of the Royal Irish Academy.

variety of this Vole (i.e. Mototnys ylareolus). They certainly are not
either of the other two British Yoles; they are the common Bank
Yole, a local variety of it or a new spedes to this country."

Mr. Drane '' always took these Yoles ahout farm buildings
or within them, where one would not expect to find Yoles." When
traps were set a few hundred yards away, he never took the Yoles but
only Wood Mice, Jftu syhatiew.

The predilection of the Skomer Yoles for the neighbourhood of
houses is corroborated by Dr. Mills, who wrote that he usually caught
them in the heaps of turnips stored up for winter, and that the turnips
are their food.

The following is a description of the ** Skomer Yole," which I
propose to name

Evotomys skomerenslB, sp. nov.^:—

General Characters, — Size large ; skull of adults, about 25 mm.
in greatest length ; total length, averaging about 165 mm. ;
hindfoot, averaging 18 mm.; ratio of tail vertebrae to total length,
33 ; skulls strong, (for Evotomys) angular and ridged for muscular
attachment ; the zygomata rather heavy. Colour deep and moderately
bright. Skull of tiie same type as that of M nageri (I have no skulls
of either E. norveyieus or K vasconiae for comparison) with which it
agrees in size, angular appearance, and general massiveness, but is, on
the average, slightly smaller.

The skull of an adult male presents the following dimensions (in
mm.) : — Greatest length, 25 ; basilar length, 22*5 ; palatal length, 12 ;
length of palatal foramina, 4*5 ; zygomatic breadth, 14 ; breadth of
brain-case above zygomata, 11*75; length of molar series (both
upper and lower), 6 ; length of nasals along middle line, 8.

Coloftr. — Above between bright '^ cinnamon-rufous" and ^'madder-
brown,"* the general appearance being due to the subterminal bands
of the hairs, about 2 mm. in breadth. The hidden (and major)
portions of these hairs are '^ slate-black " and the tips black. Face,
sides of head, and flanks becoming gradually deficient in rufous, and
running through light '' hazel " or '' vinaceous-cinnamon " to a dull
greyish-bufl. Rump and upper side of the sharply bi-coloured tail,
'* mummy-brown.'' Under side of body and tail, with the legs and
feet, white (the hidden portion of the hairs again near '' slate-black "),

^ I thus aocoid this form full specific rank in order to secure uniformity with
Mr. Miller's treatment. (See footnote No. 2, p. 818.)

' Names of colours in inverted commas are taken firom Mr. Robert Bidgeway*s
<' Nomenclature of Colours," 1886.

B.-Hamilton — Addition to List of British Boreal Animab. 317

usually with a very perceptible yellowish wash. The line of demar-
cation between the colours of the upper and under surfaces moderately
defined. Ears nearly naked externally ; internally covered with light
'* dnnamon-mfous *' hairs.

The rump of the only winter specimen before me shows a much
larger area of brown than is present in any of the other specimens
(all taken in April and May). This may be an indication of a sea-
sonal change.

Type of Spegiss. — ^A male, registered No. 3. 7. 4. 3. of British
Museum Collection, presented by Dr. Y. H. Mills, Skomer Island,
April 7th, 1900.

DmBNSIONS W MlLUMBTBBB.

Head
and
body.

TaU.

Hindfoot.

Ear.

Maximum of 7 males and 6 fermdes,
Mean „ „ about
Minimum, .....

114
110
106

66-6

19
18
17

13-6

No accurate naturalist could possibly confound the Skomer Vole
with the ordinary Bank Yole of Ghreat Britain. The greyish sides,
brown rump, far larger size, general proportions, and cranial characters
of JE, skofMrensis are such as to mark it as belonging to quite a distinct
division of the genus from that containing E. glareolus hritannieuSj
Miller, the ordinary British Bank Yole. The following dimensions
of a number of the latter form will illustrate my meaning : —

Head
and
body.

Tail.

Hindfoot.

Ear.

Greatest
lenffth
of skull.

Greatest

breadth
of skull

at
xygomaU.

/ No. of specimens,

^ < Maximum,

mm.
104

mm.
66

mm.
18

mm.
12

mm.
26

mm.
14

16-26

23-6

12-75

^ / No. of specimens,
< < Maximum,

mm.
. 117

61*

mm.
18

mm.
13

mm.
24-76

mm.
13.6

91*6

16-6

11-6

23-26

12-26

318 Proceedings of the Boyal Irish Academy.

The Skomer Voles thus exceed those of Great Britain generally
by an average of 18 mm. on the body length, so that they are nearly a
quarter as large again, and these proportions are borne oat in the
average dimensions of the tail, hindfoot, ear, and sknU. There is,
besides, a difference in the proportionate lengths of the tail Tertebrae
to the total length of the two forms, that of the ordinary British
being about as 31*50, that of the Skomer Yoles as 33 per cent, of the
total length.

I was much surprised to find that the Skomer Yole is clearly allied
to the forms inhabiting Boreal Europe. It is quite closely related to
Evotamys norveyiau Miller, of Norway, JS, nageri (Schinz) of the Alps,
and E. vasemia Miller of the Pyrenees.

In a recent paper Mr. Gemt S. Miller, junior,^ has, excluding the
Arctic JSL rtUilus (Pallas) and the very distinct' E. rufoeanui (Sundevall)
of Northern Europe, divided the Bank Yoles of Continental Europe
into two sets. One of these, consisting of the three species mentioned
in the last paragraph, is characteristic of the mountains, and cor-
responds in its distribution with the Yariable Hares. The other
includes a number of smaller lowland forms, amongst which is the
British, the E. hereynicus* hritannieus of Miller.

The Skomer Yole constitutes a fourth member of the Boreal group,
which, like its allies, is totally distinct from the Yoles of the surround-
ing country. I regret that I have very few specimens of the other
Boreal forms wherewith to compare it. It is, however, less grey
than E. nay&ri, of which Mr. Oldfield Thomas has shown me several
specimens obtained by himself near Locarno in Italian Switzerland,
while it appears to have a shorter tail than E. norv^yicut. Further,
on comparison with the dimensions given in Mr. Miller's tables, it
seems to be the smallest Boreal form yet described.

I do not propose to attempt an explanation of the occurrence of
this colony of Yoles almost indistinguishable from those of Boreal
Europe, nor why they appear to be confined to a small, wind-swept
island. It is, for the present, sufficient to place the facts on record,
noting, however, that the Skomer Yole is in no sense of the word a
stunted representative of its genus such as might reasonably be

1 *< Preliminary BeviBion of the Euzopean Bed-becked Mice." Piooeedings of
the Washington Academy of Science, vol. zi., pp. 83 to 109, July 26, 1900.

' A form which Mr. Miller has, I think, somewhat unnecessarily raised to sub-
generic rank.

* I do not, howeTer, accept the validity of Mr. Miller's arguments for the
abolitioD of the well-known term glareolu* and the substitution for it of /ureynieut.

B.-Hamilton — Addition to List of British Boreal Animab. 319

considered to have definite relations to the peculiarly cramped local
conditions of a small island. On the contrary, the Skomer Yole is
remarkably robust, and apparently only slightly less so than the
correspondbig types of Boreal Europe. In its robustness it a£fords a
parallel to the long-taQed Field Mice of St. Kilda and of Lewis {Ifus
sylatieus hirtmtis mihi and M, hebridenm^ de Winton), and the House
Mouse {M, murahs mihi) of the former island.

It cannot, however, be without meaning or importance that we
hare here on this small, treeless, wind-swept islet, almost facing the
home of Zepus timidus hihemious, an animal which belongs to the same
type of Fauna as that Boreal mammal. It may be that we may yet
find amongst the Welsh mountains further colonies of these Boreal
Voles, and the possibility should at least be a stimulus to British
Field Naturalists in their collecting expeditions. Meanwhile we may
note the parallel between the occurrence of a Boreal Yole at sea-level
on Skomer Island and the similar downward extension of the range of
the Variable Hare in Ireland, accompanied as it is in the West by
the frequent descent to the plains of certain Alpine plants.

H.I.A. PXOC, TOL. XZIT., SEC. B.] 2 i>

[ 320 ]

IX.

ON THE RELATIONSHIPS BETWEEN THE CLASSES OF

THE ARTHROPODA.

By GEORGE H. CARPENTER, B.Sc. Lond., M.R.I.A.,

of the Science and Art Museum, Dublin.

[PlATE VI.]

Bead Mat 11, 1903.
Introduction,
Few zoological problems have given rise to wider differences of
opinion than that of the relationships that may exist between the
yarious classes of animals included under the term '' Arthropoda." For
many years, the existence of some rather close affinity between luBects,
Centipedes, Millipedes, Arachnids, and Crustaceans was undisputed.
LinnS, in 1758, included all these groups in his '' Class Insecta" ; and
the name '^ Arthropoda," bestowed upon the assemblage by Yon Siebold
in 1848, was intended to mark them off as a grand primary division of
the Animal Kingdom. When the evolutionary doctrine spread, and
naturalists began to go ancestor-hunting, there was no hesitation in
deriving all the Linnean '^Insecta" from a common stock. The
development of so many and diverse Crustacea from a Nauplius larva
was believed by Miiller ('69) to indicate the descent of the whole
Crustacean class from Nauplins-like ancestors; and through some
primitive Phyllopod, the Arachnida were traced back to the same
parent-stem. The six-legged larva of certain Millipedes led to the
conclusion that both Insects and ** Myriapods " had originated from a
Thysanuriform stock, which had been derived, according to Haeckel
('76) and others, from a primitive zoaea-like Crustacean.

During recent years these phylogenetic speculations have been
discredited by many zoologists. If too much weight was formerly
allowed to larval stages in the discussion of ancestral forms, the
tendency at present is to regard such stages as of hardly any im-
portance at all. Then comparisons have constantly been made
between Arthropodean and Annelidan organs — between appendagea
and parapodia, coxal glands and nephridia, tracheal tubes and dermal
glands, so that many zoologists think it more instructive to compare
various Arthropods with Annelids than with each other. And the

Garfentbr — jRelationsAips between Classes of Arthrqpoda. 321

demonstration of the Arthropodous affinities of the PeripatidsB has led
many stndents to look upon those worm-like creatures as indicating
the prohable ancestors of Millipedes, Centipedes, and Insects, and to
believe in the derivation of those classes from an Annelidan stock
quite independently of the Crustacea. Thus the opinion seems to have
been slowly gaining ground that the Arthropoda can no longer be con-
sidered as a natural group of the Animal Kingdom (Hutton, &o., '97).

The most extreme view of the multiple origin of the Arthropoda
is that put forward by Bernard ('96), who would derive each of the
great olasses independently from an Annelid ancestry. Most recent
writers, however, consider that the present-day Arthropods have
developed along two main lines of descent. Eingsley ('94), for
example, recognising, with Lankester ('81), the Arachnidan affinities
of limulus, refers the Crustacea and Arachnida to one great group, the
Insects, Centipedes, and Millipedes to another. But other zoologists
consider the manner of breathing to be the all-important character in
deciding the affinities of the Arthropod classes. Lang ('91), for
example, divides the Arthropoda into a Branchiate and a Tracheate
series, regarding Limulus and the Eurypterida as closely allied to the
Crustacea, and believing that the Arachnida were derived from the
Insectan (Tracheate) stock by the fusion of the head with the thorax
and the disappearance of the feelers.

Supporters of either of these two views agree in supposing a wide
divergence between Crustaceans and Insects ; they differ as to whether
the Arachnida should be associated with the former or with the latter
group. The special question of the affinities of the Arachnida will be
discussed later. The conflicting views of the various authors men-
tioned have been briefly sketched as an introduction to the argument
of this essay, which wUl endeavour to show that the various classes
of the Arthropoda are indeed truly related to each other, and that
ancestors with distinctly arthropodan characters must be predicated
for all of them. As has been recently pointed out by Lankester ('97),
the structural features in which aU Arthopods agree — even if the hard,
segmented exoskeleton and the jointed limbs be left out of account —
are striking and remarkable. The heart with paired openings ; the
'^ perioardium " and the secondarily-formed body-cavity made up of
greatly enlarged blood-channels ; the reduced coelom ; the variable
number of pairs of mesodermal excretory tubes ; the uniformly striated
muscle-fibres,* and the complete absence of ciliated epithelium* — all

> Except unong the Malacopoda.

22>2

322 Proceedings of the Boyal Irish Academy.

these form an assemblage of characters quite uniqae in the Animal
Kingdom. And any attempt to explain their appearance in the yaiious
classes of Arthropods as the result of convergent evolution must raise
far more difficulties than it can solve.

But the principal view maintained in this essay is one which was
suggested nearly sixty years ago by Huxley ('68), and which has been
already published in outline by the present writer ('99) — ^that Crus-
taceans, Arachnids, and Insects agree closely in the primitive number
of their segments. It has been generally believed that the fixed and
definite number of segments found in the Malaoostraca has been derived
by reduction from the numerous segments of Branchiopodan ancestors ;
that the definite segmentation of Insects has arisen by condensation
from some primitive richly-segmented Myriapod. But, as Huxley
wrote in 1858, '' I venture to think it a matter of no small moment if
it can be proved that a Lobster, a Cockroach, and a Scorpion are
composed of the same primitive number of somites." If this be the
fact, we have well-nigh demonstrative proof that the classes to which
these three animals belong are truly aldn to each other, and that their
allies with very many segments represent abnormal developments. It
is ahnost impossible that a reduction to exactly the same number of
segments in three classes of similarly-formed animals could have been
independently produced.

It is proposed, therefore, to compare the orders of the various
classes of Arthropods so far as it may be necessary to arrive at a con-
ception of the most primitive members of each class, with especial
reference to their segmentation. Then the various classes as a whole
can be profitably compared and their affinities discussed. The writer
would express his special indebtedness to a most suggestive but
atrangely neglected paper by Hansen ('93). If students of the
Arthropoda would follow his example, and compare diligently Arthro-
pods with other Arthropods, before comparing them with specialized
Ajinelids, our phylogenetic studies might advance with greater
assurance and less controversy than at present.

Nature <^ the meet primitive Insects.

Any lengthened discussion of the relationship between the various
Orders of Insects is needless in view of the almost universal agree-
ment among entomologists in regarding the Thysanura as the most
primitive of living groups. Bernard (Hutton, &c., '97) has, indeed,
recently revived l^e suggestion that the caterpillars of Lepidoptera
and Sawflies are to be considered as representing the ancestral stock

Oarfentbr — Belatianshipi between Cbuses of Arthropoda. 323

of Insects ; and this speculation is tempting to those who seek to derive
Insects, independently of other Arthropods, from Annelidan ancestors.
But the Tiew will not hear examination. Brauer ('69), Lubbock ('74),
and, more recently, Miall ('95), and Packard ('98), have shown con-
clusiTely that the active, campodeiform, armoured larva characteristic
of the lower orders of Insects must have preceded, in the evolu-
tion of insect-metamorphosis, the worm-like cruciform larva charac-
teristic of the more specialised orders. Not only is this evident
from a study of the various orders, but from comparison between the
families of any one order. Among the Lepidoptera, for example, we
find that the caterpillar of a low-type moth, like Hepialus, has, in
addition to a chitinous tergite on the first thoracic segment, paired
tergal plates on the second and third segments, and in some species on
the abdominal segments also, while the legs are strong and relatively
long, recalling those of a beetle-larva. But the caterpillar of a high-
type moth — ^a Sphinx, for example— has no distinct tergal plates on
any body-segment, excepting the first thoracic, those of the other
segments being reduced to tubercles, while its legs are relatively
shorter and weaker than those of the Hepialus caterpillar. Thus we
see that the worm-like characters of the larva are most markedly
shown by the higher moths. Among the beeties a complete transition
from the campodeiform to the cruciform type of larva can be traced ;
while the fact that, in the life-history of certain genera, the former
type precedes the latter in the development of the individual, shows
conclusively that the active armoured grub preceded the worm-like
caterpillar or maggot, which is undoubtedly a specialized secondary
larval form. We may, therefore, safely accept the conclusion that
the primitive insects were thysanuriform.

But in connexion with the object of this essay it is of the greatest
importance to arrive at a correct view as to the segmentation and
appendages of the primitive insects. To this we are guided partiyby
morphological and partiy by embryological evidence. Taking, first of
all, the head, at least six limb-bearing segments, all primitively postoral
in position, can now with certainty be recognised. Foremost of these
is the antennal segment bearing the feelers, innervated by the deuto-
cerebral ganglia. Then comes the tritocerebral segment with evanes-
cent appendages clearly detected by Wheeler ('93) in the embryo of
Anurida, and by TJzel ('97) in that of Campodea. In the latter insect,
indeed, these appendages persist as paired tubercles in the adult. The
next postoral segment of the head is that which bears the mandibles.

The next segment has only recentiy been clearly demonstrated ;

324 Proceedings of Vie Royal Irish Academy.

and its existence is still ignored by most writers upon Arthropod
morphology. Its discovery is due to Hansen ('93), who points out
that the paired structures associated with the tongue in the Thysanura
and Gollembola, and vaguely called *' paraglosss " by most students
of those insects, are in reality a pair of jaws situate between the
mandibles and the first maxillsd. He gives to them the appropriate
name of '' maxillulsd/' It is strange that so important an observation
should have been received with marked neglect for many years, but
Howes ('02) has now accepted Hansen's interpretation. A careful
examination of these maxillulaa in the Thysanuran, Maehilis marttima^
must convince anyone that there is no escape from Hansen's conclusion.
If the mouth-parts are viewed in their natural positions, the tips of

la.

Fig.l. ^ Fig. 2.

Fzo. 1. Right Mazillula of JUaehili maritima, x90. ga., galea; la., lacinia ;
pip., palp.
2. Left MaziUula of laotoma palustrit. x 260. ga., galea ; la., laoiiiia.

the maxillulse are seen to lie just behind the mandibles, and in front
of the maxillae ; while, when dissected out, these organs show all the
appearance of a reduced pair of jaws (fig. 1 ). On the outer edge of each
maxillula is a short vestigial palp ; while the tip of the appendage has
two very distinct lobes, corresponding with the galea and lacinia of
a typical maxilla. Both lobes show a beautiful arrangement of
spicules, ridges, and pits ; and the lacinia, at least, is jointed with the
basal sclerite. The maxillulsB, at their bases, articulate with the central
tongue. In Japyx these organs are still more highly developed, with
three-segmented palps, according to Hansen ('93).

Oarfenter — Relationships between Classes of Arthropods. 325

A full description of these organs in the CoUemholan Orchesella
was three years ago published by Folsom ('99) ; and, although he
did not at that time recognise them as a pair of jaws, his figures
show that they correspond closely with the mazillulee of Machilis,
though less highly developed than the latter. In Isotoma — ^perhaps
the least specialized genus of the CoUembola — the maxillulse are
more strongly developed than in Orchesella. As observed in
Isotoma pahutris (fig. 2), the lacinia is distinctly toothed at the tip ;
and the series of spines along the inner edge of the basal region are
stronger than in Orchesella. The association of the maxillulsB with
the tongue is closer in the CoUembola than in the Thysanura, as
might be expected from the greater specialization of the former
group.

A final proof that the maziUulsB (or '' superlingusd") are indeed
a distinct pair of appendages has been afforded by Folsom ('00), who
has studied their development in the CoUemholan Anurtda maritima.
He finds that they arise from paired rudiments like those of the
mandibles and masdUse, and that their association with the central
rudiment of the tongue is secondary— exactly as would be anticipated
from a comparative study of the adult insects. Although the
maxiUular rudiments arise between rather than behind the rudiments
•of the mandibles, a special post-mandibular gangHon and a pair of
coelomic spaces are associated with them. It is evident from the
figures given by Eaton ('88) and Heymons ('96) that these appendages
iire present in the Ephemerid larva, though in a reduced state.
According to Hansen, their vestiges can be clearly made out in the
Earwigs and Hemimerus, and in a stiU more reduced condition in the
Cockroach and other Orthoptera.

The two pairs of toutiHip (^'maxiUae" and ''labium" of
entomologists) are the appendages of the two hinder postoral somites
of the head. A point of considerable interest to be noted is the fact
that, in the more generalised Insects at least, the labial segment is
incompletely fused with the head-capsule, part of its skeleton forming
the cervical sderites or so-caUed '' microthorax," very evident in the
Cockroach. This interpretation of the cervical sclerites, suggested
by Huxley ('78), has been established by Comstock and Kochi ('02).
We conclude, therefore, that in the Insectan head are six limb-
bearing segments, whereof the hindermost, at least in the more
genendised orders, is incompletely fused with the rest. It is likely,
as wiU be seen later, that an extra, primitively limb-bearing, ocular
segment in front of the feelers must also be reckoned.

Behind the head, the segmentation is comparatively simple. The

326 Proceedings of the Royal Irish Academy.

three thoracic segments, each with its pair of legs, are sacceeded by
ten abdominal segments. Of these latter, the second to the ninth
bear short nnjointed appendages in Machilis ; while the tenth, in many
of the more generalized insects, carries a pair of jointed cercopods.
Then comes a small terminal anal segment. But the researches of
Heymons ('95) have shown that the segment on which the ceroopoda
arise in the embryo, is in reality the eleventh abdominal, which, aa
growth proceeds, becomes fused with the tenth. It has long been
known that rudiments of limbs appear on the abdominal segments of
many insect-embryos. This fact, in conjunction with the abdominal
appendages of Machilis and other Thysanura, leads us naturally to
conclude that the ancestors of insects had limbs on all the segments of
the body, except the anal segment. With confidence, therefore, we
can believe that the most primitive insects possessed a head with five
post-oral limb-bearing segments, completely fused, a '^ neck " segment,
undergoing fusion with the head, three thoracic segments with well-
developed legs, and an abdomen of twelve segments, whereof the first
ten carried poorly-developed limbs, the eleventh a well-developed pair
of cercopods, while the twelfth or anal segment had no appendages.
As no insect is hatched in the winged stage, and as the young of so
many insects are Thysanuriform, there need be no hesitation in
concluding that the ancestral insects were wingless. And it is
reasonable to conclude that the pedigree might be traced farther back
still to animals with a head with paired eyes and five limb-bearing
segments, and a trunk with sixteen undifferentiated segments, whereof
all but the last carried paired appendages. (See Table, pp. 354-5.)

JSelationships between Imeets, Centipedes^ and Millipedes,

But it may readily be objected that Centipedes and Millipedes are
less highly organized than Insects — to which class nevertheless they
are related — and that they possess a larger number of limb-bearing
segments than the Insects have. Therefore, it may be argued, Insects
must have been derived from ancestors with numerous segments.
This objection, however, is by no means serious, and rests largely on
the assumption that ''rich segmentation*' must, of necessity, be a
primitive character among Arthropods. The absence of wings in
Centipedes, and the similarity of most of the body-segments and their
appendages, are doubtless primitive characters. But it is quite as
likely that, compared with the ancestral stock, the number of segments
should have increased as that they should have suffered reduction.

Carfbnter — Sehtionships between Classes of Arthropoda. 327

And an examination of the relationships of these classes and their
orders shows that the former altematiye has much eyidence in its
lavonr.

The morphological studies of Kingsley ('88) and Focock ('9dA), and
the emhryological researches of Ke3rmons ('01), have estal)lished beyond
any reasonable doubt that the ^' Class Myriapoda " must be abandoned,
the Centipedes (Class Chilopoda) being more nearly related to the
Insects than to the Millipedes (Class Diplopoda). The Centipedes
agree with the Insects in l^e simple segmentation of the body, in the
lateral position of the spiracles, in the anastomosing tracheal tubes,
and in the posterior position of the genital openings; while the
Millipedes exhibit for the most part a fusion of the segments in
couples, so that each apparent segment carries two pairs of legs, the
spiracles are ventral in position, the air-tubes are unbranched and do
not anastomose, and the genital openings are far forward on the third
body segment.

If, then, it is believed that Insects and Centipedes on the one hand,
and Millipedes on the other, have diverged from some common
ancestral stock, it is natural to inquire whether any living form can
suggest approximately what that stock may have been like. The
only animals that combine some of the divergent characters of Insects,
Centipedes, and Millipedes, are the Scolopendrellid®, now usually
regarded as a distinct class, called, on account of their annectant
characters, the Symphyla (Eyder, '80). These small, frail, somewhat
degenerate creatures, eJiow the series of similar, simple limb-bearing
segments characteristic of Centipedes, the forwardly-situated genital
aperture as found in Millipedes, and a number of body-segments
identical with that occurring in Insects. Their chief point of
specialization is the curious inequality and displacement of the
tergites. No surprise need be felt that some students of their structure,
like Packard ('98), regard them as representing the ancestral stock of
Insects; others, as Grassi ('85), that of Centipedes and Millipedes. But
if we are willing to accept the view, admitted as possible by Lang
('91), that most living Centipedes and Millipedes have become what
they are by an increase from the number of primitive segments, there
is no reason why we should not, with Haase ('86) and Pocock ('93a),
regard the Symphyla as approximate to the common ancestor of the
Inseota and the Chilopoda on the one hand, and of the Diplopoda
(including the Pauropoda) on the other. Haase particularly suggests
that the common ancestors of the three great Tracheate classes had as
many segments as Scolopendrella.

328

Proceedings of the Royal Irish Academy.

But Schmidt ('95), to whom we owe the most recent account of
this interesting animal, denies that its segmentation is primitiye, and
suggests that the pointed processes on the cozsb of its legs must be
regarded as the vestiges of pairs of limbs belonging to segments which
have become closely fused with the present evident segments. He
considers Scolopendrella, therefore, to be a very highly specialized
Diplopod, the fusion of the segments in couples being so intimate, that
the ad j acent limbs have coalesced. But this view is surely far-fetched,
when we consider in how many points of structure Scolopendrella
inx
*u la

xnxl.

Fig. 3. a. Eight Maxilla (inx.)> tongue (li.)> and mazilliila (mzl.) of SeolopendrelU
immaetdata. The tip of the right maxillula is seen in ntu ; the left
maTJllnla is exposed hy removal of the maxilla and part of the tongue.
x390.
B. Second maxillss (lahium) of Scolopendrella. x 90.

approaches the Thysanura. The antennae resemble closely those of
Campodea, and differ in the most marked way from those of any
Millipede. The head-skeleton, with its angular epicranial suture,
is quite Insectan. Latzel's figures ('84) show clearly three distinct
pairs of jaws, the mandibles, maYillae, and labium corresponding rather

Carpenter — Relat%omhip% between Classes of Arthropoda. 329

closely with the similar structures in the Gollemhola. And hy
dissection of the head, I have succeeded in demonstrating the presence
of a pair of minute maxillulsB^ associated with the tongue, and lying
between mandibles and maxillae, just as they do in the Springtails
(fig. 8, mxl.). All that is now wanting to bring the segmentation of
ScolopendreUa into perfect agreement wi^ that of the primitive Insects,
is embryological proof of the presence of the vestigial tritocerebral
appendages of the head. With confidence, therefore, we may postulate
a Scolopendrelloid ancestry for Insects, Centipedes, and Millipedes.

Among the Centipedes we find very great variation in the number
of the body-segments, Lithobius and Scutigera having only fifteen pairs
of walking-legs, Scolopendra and its allies twenty-one or twenty-three
pairs, and the (^eophiloids often more than a hundred pairs. Now,
according to the view of Haase, adopted by BoUman ('93), the fifteen-
legged groups must be regarded as the more primitive on account of
the comparative simplicity of the tracheal system in Lithobius, the
spiracles having no closing apparatus ; and it is especially noteworthy
that a correspondingly simple stage is passed through in the develop-
ment of the Scolopendridffi. Although the tracheal system of Scutigera
is highly specialised and the spiracles dorsal in position, the head and
mouth-parts of that animal retain many primitive characters.

But the important discovery recently announced by Focock ('02)
of a Tasmanian genus of Centipedes (Craterostigmus) with fifteen
pairs of legs like a Lithobioid, and twenty-one tergites like a Scolo-
pendra, is believed by him to indicate the descent of the Lithobioids
from Scolopendroid ancestors through the loss of six segments — ^the
3rd, 6th, 9th, 11th, 14th, and 17th. Ke suggests that, in Cratero-
stigmus, the tergites of these segments are still retained. Two objections
may be made to this view. It is hard to imagine a reduction in the
number of segments by the loss of a scattered series such as this.
And the derivation of the Lithobioids from the Scolopendroids, through
Craterostigmus, would destroy the remarkably close correspondence
between the position of the spiracles on the 1st, 3rd, 5th, 8th, 10th,
12th, and 14th body-segments of Lithobius and the corresponding
segments (except the Ist) of Scolopendra. As Craterostigmus exhibits
several Geophiloid characters, it is more likely that its six *' minor"
tergites should be compared to the smaller sections of the incom-
pletely-divided dorsal plates of the GeophilidsB.

> Wliile thlB paper is passing through the press, I find that the mazillulis of
Scolopendrella have been seen and figured by Hansen ('03).

330 Proceedings of the Boyal Irish Academy,

There is good reafion, therefore, for considermg that the richly-
segmented Centipedes are abnormal deyelopments from forms with a
moderate number of segments. The palsontological OTidence of the
subject is very meagre. In Carboniferous times, we know from the
researches of Scudder ('90) that Latzelia, a form resembling Scatigera»
but without the specialised dorsal tracheal system, existed. The
fossils referred by Scudder to the EoscolopendridsB are too imperfect
for any certain conclusions to be drawn from them. If the bristle-
bearing animal Falffiocampa, referred by him to a special order, the
Frotosyngnatha, were indeed a Centipede, its body-segments were but
few in number. Embryological researches on the Centipedes, the
latest of which is Heymons' exhaustiTe treatise on the development of
Scolopendra ('01 ), show the close correspondence between the Chilopodan
and the Insectan head. The presence of a tritocerebral rudiment in
the Centipedes has been established, so that the feelers, mandibles,
maxillsB, maxillulae, and labium of the insect correspond respectiyely
with the feelers, mandibles, two pairs of maxillae, and poison-jaws of
the Centipede. The freedom from the head of the segment bearing
the last-named limbs in the Centipedes shows that their ancestors must
have diverged from the primitive stock at a very early period. In
this respect, the head of Scolopendrella is specialised as compared with
the Centipede-head ; and in tie Symphylan ancestral stock of Centi-
pedes and Insects, the pair of limbs that now forms the plate-like
labium in Scolopendrella and the Collembola, must have been free and
leg-like.

One of Keymons' most startling discoveries is the presence of a
pair of pre-antennal rudimentary appendages on the head of the
Scolopendroid embryo. The segment bearing these he regards as post-
oral ; and he ranges it with the optic segment of the insect head. Its
existence strongly suggests that the eyes of the far-off ancestors of
Centipedes were stalked and appendicular. As the development of
the lateral simple eyes in Scolopendra does not support the theory that
they are degraded compound eyes, it is to be presumed that the ances-
tral compound eyes have been lost, except in the Scutigeridae.

Turning next to consider the Millipedes, we find that they, like
Centipedes, exhibit a wide divergence in their segmentation. It is
impossible to lay any stress on the hexapod condition of their larvse,
as indicating relationship to the Insects, as the segments on which the
three pairs of legs occur are not successive, and vary in different
groups. But the strongest evidence for the derivation of the Diplopoda
from the same stock as the Chilopoda and the Insecta is afforded by

Cabfbntbr — RelatiomhipB between Classes of Arthropoda. 331

the fact that in Scolopendrella the genital opening is far forward, as
in the Millipedes, while the curious group of the Fauropoda, which show
many points of correspondence with the Symphyla, have heen proved
to exhibit (Kenyon, '95) marked Diplopod affinities. Kenyon, indeed,
places the Fauropoda, together with the Fselaphognatha (FolyxenidsB),
in a group which he calls the Frotodiplopoda. Pauropus has only
nine pairs of legs ; and its segments are imperfectly fused. There can
be little doubt that this form has undergone secondary shortening; but
Polyxenus has thirteen leg-bearing segments followed by two limbless
segments, so that its segmentation agrees exactly with that of
Scolopendrella and the primitive Insects. The mandibles of Pauro-
pus and Polyxenus resemble those of the Collembola. In other
Millipedes, these jaws are complex, being composed of several sderites
— a condition which, like that of the mandibles in certain Scarabseid
beetles, must perhaps be regarded as a secondary adaptation. The
" lower lip " or ** gnathocilarium " of Millipedes seems to be certainly
formed by the union of two pairs of appendages which probably
represent the maxillulse and maxillae of Insects ; while the labial seg-
ment of Insects is represented by an embryonic limbless segment
(Heymons, '97). (See Table, pp. 364-6.)

What palsBontological evidence we possess of the history of Milli-
pedes shows that richly- segmented forms, in which the segments were
already beginning to fuse together in couples, were living in Devonian
times. But as winged Insects have been traced back to the Silurian,
we have nothing but comparative studies in living forms to guide us
as to the nature of the common ancestor of Insects and Millipedes.
Morphological evidence shows clearly that Millipedes might well have
arisen, through some form combining the primitive characters of
Pauropus and Polyxenus, from a Scolopendrelloid stock. The fusion
of segments in couples would not be likely to take place until the
number of segments had become very great. It is suggestive to notice
in this connexion that, in the Pauropoda and Pselaphognatha, the
fusion of segments has hardly begun. The earliest truly ''diplo-
podous " forms would have been elongate Juloid Millipedes. Thence,
by a reduction in the number of segments, the Glomeroid forms may
have sprung.

The difficulty that arises in bringing together two groups which, like
the Millipedes on the one hand, and the Insects'and Centipedes on the
other, exhibit a great difference in the position of the genital aperture,
will be discussed later in connexion with the relationship between
Insects and Crustacea. For the present it is enough to repeat the

332 Proceedings of the Royal Irish Academy.

fact that in Scolopendrella, despite its marked Insectan affinities, the
genital opening is far forward, as in Millipedes. Therefore the differ-
ence in ^e position of the genital opening cannot by itself indicate
a very radical divergence. Although Centipedes are more nearly
related to Insects than to Millipedes, it is likely that the kinship of
Insects to the two classes of **myriapods" is equally close. The
Insects represent the main stem, the Centipedes and Millipedes two
divergent branches.

Relatumship hetween the Orders of Crustacea.

Turning next to consider the probable nature of the most primitive
Crustaceans, we find the prevailing opinion among modem zoologists
to be that the Phyllopoda, as exemplified by the many-segmented
ApuB and Branchipus, represent, more nearly than any other living
oi^er, the ancestral stock of the class. According to this view, the
evolution of the Crustacea has been effected by a reduction in the
number of body-segments until the definite and limited number
characteristic of the higher orders (Malacostraca) has been reached.
But Packard ('82), Sars ('87), and Hartog ('88) have argued that the
Copepoda are more primitive than the Phyllopoda.

Now if we consider the lower orders (Entomostraca) as a whole,
we are struck by the quite exceptional presence of a rich segmenta-
tion. In the Phyllopoda the number of pairs of legs may vary from
four to over sixty ; whilst, in the other recent orders of the Entomo-
straca, the limb-bearing segments are always few. We now know that
Phyllopods, closely related to Apus (Protocaris), and Ostracods, had
already been differentiated in the Cambrian period. Therefore, what-
ever may have been the segmentation of the primitive Crustaceans^
there had been great modification before the dawn of the earliest life-
epoch known to us by fossil evidence. No doubt can be entertained
that such poorly-developed segmentation as is shown by the Ostracods
must be due to reduction. But has such reduction been the constant
rule in Crustacean development ? It may be of interest to consider in
this connexion that most ancient of Crustacean orders known to us —
the Trilobita.

Nearly all Trilobites are composed of a number of segments greater
than that characterising the Malacostraca. After the recent researches
of Beecher ('00) and others, there can be no reasonable doubt that
these animals were true Crustacea, and that they combined to some
extent the characters of the Branchiopoda and the lower Malacostraca.

Carpenter — Eelatianships between Classes of Arthropoda. 333

What is the history of their eyolution as regards segmentation?
Olenellus is the oldest known genns; and, according to Peach ('94), 0,
Kjervlfi is its most primitive species. This Trilobite had sixteen body-
segments, in addition to the five-segmented head — only one more than
the typical Malacostracan and Insectan number. And if we study the
segmentation of Trilobites generally, we find a slow but steady increase
in the number of segments from the Cambrian on to the dying-out of
the order in the Carboniferous. Taking from Zittel ('87) the genera
whose segmentation is clearly known, it is found that the average
number of body-segments present in the Trilobites of each great period
of the Primary Epoch work out as follows : —

Period.
Cambrian

No. of Oenera.
12

Average number of
body-segments.

17-66

Ordovician

18-68

Silurian

19-34

Devonian

20-70

Carboniferous

20-76

It is also noteworthy that in the Ordovician period, when the
Trilobites. seem to have reached their culminating point, there lived
two genera with the largest number of trunk-segments actually known
(thirty-two).

The history of the Trilobites suggests, therefore, that a steady
increase in the number of trunk-segments characterised the evolution
of that order; and, as we have seen, the most primitive Trilobite
known to us possessed only sixteen trunk-segments. It may be fairly
inferred from this that t^e ancestral Trilobites were by no means
*' richly segmented " animals. It may also be inferred tiliat the rich
segmentation of such a Phyllopod as Apus is at least as likely to be a
secondary as a primitive character, and that the most generalised
Crustacean we can imagine might have had no greater number of
segments than a modem lobster.

What further light can be thrown on the nature of the earliest
Crustacean type ? Grobben ('92) has suggested that the Entomo-
straca cannot be regarded as a primitive group to be set over against
the Malacostraca ; but that, from a Phyllopod-Hke ancestry, the various
groups of the Entomostraca that have undergone either reduction in
their segmentation or degeneration, as well as the higher Malaco-
stracan orders, must be derived. But almost every student of the
Crustacea has seen in the Leptostraca something of a connecting-link

334 Proceedings of the Eoyal Lieh Academy.

between the Malacostraca and the Fhyllopoda. Some writers, like
Clans ('72), have laid stress on their aflSnity with the former group,
others, like Sars ('87), on their relationship with the latter. Most
zoologists, regarding the Fhyllopods with their yery extended seg-
mentation as the most primitive of all Crustacea, see in the Lepto-
straca, transition-forms between the Fhyllopods and the Malacostraca.
But if this isolated and very ancient group—with representatives like
Hymenocaris and Ceratiocaris going back into the Cambrian Feriod —
combines, in many respects, the characters of the Fhyllopoda and the
Malacostraca, is it not more natural to regard it as a direct offshoot —
modified, of course, in some particulars — of the common Crustacean
stock whence both Fhyllopoda and Malacostraca sprang?

In Nebalia and its few allied genera, which alone represent the
Leptostraca at the present day, we find a head bearing stalked eyes
and the usual five pairs of appendages, a thorax with eight pairs of
simple limbs, with lamelliform exopodites and jointed endopodites,
and an abdomen with eight segments, whereof the first six bear paired
pleopods, while the eighth (anal segment) is provided with two f ureal
processes. Now a comparison between the head-appendages of Nebalia
and those of Apus or Branchipus shows that the former retains de-
cidedly the more primitive characters. In Nebalia the feelers of both
pairs are elongate and normal ; in the Branohipoda they are greatly
reduced. The mandible in Nebalia, with its long endopodite (palp),
is among the least specialised of all Crustacean mandibles. From it
the Malacostracan mandible with its reduced palp, or the Branchio-
podan with no palp, could be both readily derived, while it most cer-
tainly could never have arisen from the last-named type. So also
with the maxillae : in Nebalia those of both pairs have jointed endo-
podites, while in Apus those of the first pair are small limbless
masticating plates, those of the second vestigial with hardly recognis-
able parts. Fassing to the thoracic limbs, we find in those of Nebalia,
the protopodite with the three segments, which Hansen shows to be
the primitive number, a narrow segmented endopodite and a broad
branchial exopodite ; in Faranebalia tiie expedite is slender and fringed.
From such a limb as this can be derived on the one hand, through the
Schizopoda, the legs of the specialised Malacostraca, on the other the
complex lamellate appendages of Apus ; while it would be hard to
imagine how the limb of Nebalia could have arisen from the Bran-
chiopodan limb. The Leptostraca are specialised in the great develop-
ment of the carapace ; this probably accounts for the smaU size and
crowded arrangement of the thoracic segments, which, nevertheless,

Carfbntbr — Belationshipa between Classes of Arthropoda. 335

lemain free from each other and from the head. The abdomen of
l(ebalia shows a remarkable likeness both to the abdomen of the
higher Malacostraoa and to that of the Fhyllopoda. It agrees with
the latter and differs from the former in the reduction of the limbs
on the hinder segments and in bearing a terminal fnrca; while it
approaches the Malacostracan abdomen in the limited number of its
segments, eight being present, the last two of which are limbless. In
the Malacostraca there are seven abdominal segments, the sixth bear-
ing a strong pair of appendages. (See Table, pp. 354-5.)

Having seen that the structure of the cephalic and thoracic limbs
in Nebalia leads us to regard it as more primitive than Apus, we are
prepared to compare the abdominal region in the two animals, and
to admit that the numerous abdominal segments in the latter may well
have arisen by the multiplication of a primitively moderate number.
On the other hand, if the Malacostraca have developed, as is almost
universally believed, from Leptostracan ancestors, it is easy to con-
ceive that one abdominal segment has been lost, in connexion probably
with the strong development of the Malacostracan uropods. If these
limbs belong to the true sixth abdominal segment, then there may be
two fused segments in the telson ; or, as is perhaps more probable,
when the formation of the Insect abdomen is recalled, the uropods
may be in reality the limbs of the seventh abdominal segment which
has become united with the sixth. Future researches on the embry-
ology of the Malacostraca will doubtless dear up this point.

If we carry our investigation still further back, and speculate as
to the nature of the ancestors of the Leptostraca, we naturally com-
pare them with that other ancient group— the Trilobita. Beecher's
restoration ('00) of the appendages of the Trilobites, as suggested by
the study of Triarthus, is now weU known. The head bears a pair
of simple feelers and four pairs of biramous limbs, not differing from
the succeeding limbs of the trunk. The hinder trunk-limbs are
specially modified as swimmerets by the flattening of the endopodital
segments, and not, as in Apus, by the foliation of the protopodite.
There can be little doubt that the feelers of the Trilobite aro Crustacean
antennules ; and that the other four head-appendages represent the
antennflB, mandibles, and two pairs of maxillae. Now the distinct and
conspicuous palps of the mandibles and maxillsB in Leptostraca carry
ns some way towards the very primitive condition of the appendages
of the Trilobites, as do also the comparatively simple biramous
thoracic limbs of Paranebalia (Sars, '87). There need be no hesitation,
therefore, in deriving the Leptostraca from an ancestral form in which

R.I. A. PUUC, VOL. XXIT.y SSO. B. 2 jB

336 Proceedings of the Royal Iruh Academy.

•
all the tnmk-flegments, then not oorered by a canpace, were similar —
no differentiation between thoiax and abdomen existing — and the
head- and tronk-appendagea alike. Such ancestorB must have liTed in
pre-Cambrian times, and the segmentation of the Trilobites suggests,
as we haye seen, that similar pre-Cambrian ancestors — ^with five head-
and fifteen trunk-segments bearing limbs — ^may be most reasonably
imagined for them. As Bemard ('94) has pointed out, the head of
the Cambrian Microdiscas had apparently only four segments, suggest-
ing that the hinder head-segments were successiTely absorbed from
the trunk. It is evident that those most generalised Crustacea must
have combined the primitive characters of the Leptostraca and the
Trilobites, and that they had the typical Arthropodan number of
body-segments. A furca or a spinose telson was undoubtedly attached
to the anal segment. The eyes were stalked, the Leptostraca in thia
respect being more primitive than the Trilobites. The probability
that the stalked eyes represent an additional post-oral pair of limbs,
anterior to the antennules, will be discussed later.

Relations between the Orders of Arachnide.

Some reference has already been made to the controversy regarding
the affinities of the Arachnida as a whole. As an introduction to the
closer study of this question, some discussion of the relationship
existing between the various orders that are undoubtedly referable to
the Arachnid class is necessary. On this subject very diverse opinions
have been expressed by zoologists. The question depends largely on
how we regard the Scorpions.

That Scorpions are specialized Arachnids has been argued from two
points of view. The complete fusion of their cephalo-thoracic seg-
ments is compared with the presence of free or partially free segments
in such forms as the Palpigradi and the Solifugida. And the writers
who, like Lang and Bemard, believe that the lung-books of Scorpions
have been derived from tufted trachese, naturally regard the order as,
in that respect at least, more specialized than those Arachnids that
breathe chiefly or wholly by means of air-tubes.

It may help to clear the ground if, leaving the Solifugida and
Palpigradi aside for awhile, we compare the Scorpions with the other
prominent orders of living Arachnids, — with the Pedipalpi, the
Araneida, the Phalangida, and the Acarinida. And if that be done,
there seems no escape from the conclusions of Lankester (*81) and
Pocock ('93b), that the sequence in which these orders have just been

Garpbntbr — Meiatiomhips between Classes of Arthropoda. 337

mentioned is, on the whole, an ascending sequence. The reduction
and fusion of segments in the hind-body may be regarded as evidence
of specialization, as strong as similar fusion in the head-region. The
Scorpions have a well-developed abdomen with twelve free segments.
In the Fedipalpi the twelve segments are still recognizable, but the
hindmost are reduced and crowded. In the Spiders only a single
genus (Liphistius) retains any certain trace of abdominal segmentation ;
all other spiders have the abdominal segments fused into a compact
hind-body constricted off from the cephalothorax by a narrow waist
(the pre-genital segment). In the Phalangida the segmentation of
the abdomen may be more or less apparent, though the anterior
segments tend to disappear, or to become fused with the cephalothorax.
And lastly, in the Mites we find cephalothorax and abdomen fused into
a single ovoid mass, all trace of segmentation having vanished.

Now, it is surely a very striking fact that we find this condensa-
tion and fusion of the hind-body region correlated with a replacement
of lung-books by tracheal tubes as breathing-organs. In the Scorpions,
four pairs of lung-books are present ; in tiie Pedipalpi and the Avicu-
larian Spiders, two pairs, which belong, however, to the genital a|id
post-genital segments, and do not therefore correspond vrith any of the
Scorpions' lungs; in the vast majority of Spiders one pair only, the
hinder pair being replaced by tracheal tubes. And in the Fhalangids
and Mites, no lung-books whatever are present, the breathing being
entirely tracheal. Considering more particularly the Spiders, it might
seem needless to insist that the Dipneumonous families are higher than
the Tetrapneumonous, were it not that the true relationship of the
Arachnid orders depends so much on the appreciation of this point.
For if the relationship between the two great divisions of the Spiders
be as just stated, it is certain that among the Arachnids, lung-books
are more primitive organs of respiration than are tracheal tubes.
Compare the two pairs of respiratory slits in Avicularia, all leading to
lung-books (fig. 4a), with the similarly situated two pairs in one of the
lower Dipneumonous Spiders— Dysdera (fig. 4b), for example — in which
the hinder pair lead to tracheal tubes ; then with a higher type such
as AnyphsBna (fig. 4c), in which the hinder pair of openings have
coalesced to form a median slit, half-way back along the ventral
surface of the abdomen ; and, lastly, with a highly organised Spider
like Epeira (fig. 4i>), where the median slit is far back just anterior to
the spinners. Is it possible to regard such a series without being
forced to the conclusion that the arrangement in Avicularia is the
most primitive, in Epeira the most specialized ? And a corresponding

2E2

338

Proceedings of the Royal Irish Academy,

specialization in the secondary sexnal organs may be seen in the four
types just mentioned ; the copulatory apparatus on the male's palp is
far more complex in the two-lunged than in the four-lunged spiders,
in Anyphaena than in Dysdera, in Epeira than in Anyphaena
(fig. 4, a, b, c, d). Again, therefore, we are forced to the conclusion
that, among the Arachnida, tracheal tubes are deriTable from lung-
books — and not lung-books from tracheal-tubes. And palseontological
evidence confirms, so far as it goes, the teaching of morphology on
this question. For remains of Scorpions occur in the Silurian rocks,
of Fedipalpi and Avicularif orm Spiders in Carboniferous ; while the
Dipneumonous Spiders, Fhalangids, and Mites are not certainly known
until the Eocene. Therefore, without supposing that Spiders are
actually derivable from Fedipalpi, or these from Scorpions — each

Fig. 4.

Fio. 4. Ventral view of the abdomen in four Bpiden. A, Avioulaira (nat. sixe) ;
B, Dyadera x 2 ; G, Anyphaona, x 3 ; D, Epeira, x 2 ; S\ S*, air-open-
ings. Also terminal portion of palps of the male in the same four
genera, a, Avicularia, x 2 ; d, Dysdera, xZ; e, Anyphsna, x 8 ; <i,
Epeira, x 8 ; to show increase in complexity of the copulatory organ.

order having of course undergone specialization along its own lines
(the Scorpion's post-abdomen, for example)— we are fully justified in
placing the origin of the Scorpions lower on the Arachnid stem than
we place the origin of the Fedipalpi, and that lower than the origin
of the Spiders.

Embryological researches also support strongly the view just set
forth. The development of Spiders* lungs has been recentiy studied
by Simmons ('94) and Furcell ('95), and there can be no doubt from
their observations that the respiratory plates arise as outgrowths from

Carpenteb — Belatioushipa between Classes of Arlhropoda. 339

the hinder aspect of an abdominal appendage comparable to the gill of
Idmnlns. Purcell, in particular, has shown that the plates begin to be
developed while the appendage still stands out distinctly from the
Tentral surface of the body; as growth proceeds, the appendages
simply sink in, ** without any inversion or other complications," as
Simmons remarks, and the lung-book is thus formed. And the
tracheal tubes in spiders arise as in«puahings behind the appendage of
the third abdominal segment, the appendage bearing evanescent fold-
ings resembling those that give rise to the lung-book on the appendage
of the second abdominal segment. Could stronger evidence be desired
that lung-books are more primitive among the Arachnids than tracheal
tubes, and that they were preceded by lamellate appendicular gills?
Moreover, Purcell diows that, in most Dipneumonous Spiders, part of
the tracheal system, and in the AttidsB nearly the whole of it, arises
from the ectodermal in-pushings that form the entapophyses, so that
the tracheae have not, in all cases, a similar origin. Clearly it was
these entapophysial imaginations that gave rise to Jaworowski's
statement ('94) that the tracheal tubes precede the lung-books in the
development of Trochosa. It follows from these researches on the
development of Spiders' lungs and trachesB, that the latter are the
later development among the Arachnida, that they are not constant
in their mode of origin, and that they must have arisen altogether
independentiy of their origin among the Insects.

These considerations show that the tracheal respiration of the
Solifugida cannot be invoked as an argument that the Arachnida as a
whole are *' Tracheata," still less, as suggested by Thorell, that the
Solifugida are Insects ! For granting that, in their segmented fore-
body, the Solifugida have retained a primitive character lost by the
Scorpions, their abdominal segmentation is reduced and condensed as
compared with that of the latter, and their chelicersB are the most
powerful and specialized to be found in the whole class. Hie
existence of these very powerful limbs, and the extreme rapidity of
locomotion attained by these animals is sufficient to explain the
exceptional development of spiracles among them on the fourth limb-
bearing segment of the cephalothorax. Even if, as Bernard ('96)
claims, these spiracles suggest the presence of breathing-organs on
nearly all the segments of the primitive Arachnid, there is no im-
possibility in such a conception. But the fact that cephalothoradc
breathing-organs are found only among the Mites and the Solifugida
— ^the former in all respects, and the latter in some respects, highly
specialized forms — suggests rather that breathing-organs among

340 Proceedings of the Boyal Irish Academy.

primitiTe Aiadmids were confined to flie hind-body, as has heen ahly
maintained by Wagner ('95).

Bat whatever may be the tmth as regaids this point, theie can be
no reasonable doubt that in their free thoracic segments, the Solif agida
and the Falpigradi (Hansen and Sorensen, '97) retain a primitive
character. This then must be taken into acconnt, together with the
indications furnished by the abdominal segmentation of the Scorpions
and Pedipalpi. When we come to speculate on the nature of the
primeval Arachnids, we are clearly led, in this way, to imagine an
Arthropod with a head carrying three pairs of limbs, whereof the fore-
most were three-segmented dielicene, a thorax with three free segments,
each with a pair of limbs, and an abdomen of thirteen segments — ^the
foremost the pre-genital or waist-segment, detected in the embryo
scorpion by Brauer ('95), and in the embryo of Spiders by many
observers, represented by the metastoma of Eurypterids, and perhaps
by the sternum of the Scorpions. Each abdominal segment from the
second to the seventh, inclusive, had a pair of appendages carrying
gill-plates. Of these the two foremost pairs are represented in the
Spiders, the four hindmost in the Scorpions. There was, of course, no
specialized '' post-abdomen," as in Scorpions ; but the segments tapered
gradually towards the tail-end, the hindmost bearing some kind of
telson. This conception of the ancestral Arachnid agrees closely with
that figured by Focock ('93b), except that he supposes a completely
fused cephalothorax, thereby allowing no weight to the evidence of
the free thoracic segments in the Solifugida and the Falpigradi.

Having arrived at this result, we are now in a position to inquire
whether the ' ' Gigantostraca " — ^the Idmuloids and Eurypterids — should
be considered as belonging to the Arachnid class. That Limulus is
nearly related to the Merostomata, and can probably be derived from
the same ancestry as that order, through such forms as the Silurian
Bunodes and Hemiaspis, and the Carboniferous Belinurus and
Frestwichia, can hardly admit of doubt. And it is further evident
that Eurypterus and the Limuloids could, like a Scorpion or a
Galeodes, be derived from the ancestral Arachnid that we have just
imagined. The correspondence in the segmentation of Eurypterus and
the Scorpion is so close and remarkable, that we are forced to admit
an affinity. This point has been sufficiently ai^ed by Lankester and
others ; but to claim that both the Herostomata and Xiphosura ought
to be classed as Arachnida does not involve the belief that Scorpions
are descended from Eurypterids, still less from Eing-crabs ! In view
of the specialized fusion of both cephalo-thoracic and abdominal

Carpenter — Belationshipa between Classes of Arthropoda. 341

flegments in Limulus, anything like an ancestral standing for that
animal is unthinkable ; while the Enrypterids, although primitive in
their abdominal segmentation, had the cephalothorax completely fused,
and the sixth pair of limbs specialized as paddles. Moreover, they
were contemporary with the earliest Scorpions. Howes, in a recent
criticism ('02) on the upholders of the Arachnidan affinities of Limulus,
has pointed out that the limuloid type is simplified by the Eurypterids,
as the modem Scorpionoid type is by Falaeophonus. This is just
criticism : but it should not prevent us, led by the close correspondence
between the segmentation in the primitive Scorpion and in the
Eurypterids, from carrying the simplification still farther back, and
deriving the Oigantostraca on the one hand, and all the modem air-
breathing Arachnids on the other from aquatic Arachnid ancestors,
with three free thoracic segments. In the detailed account of the
Silurian Scorpion (Falaeophonus) lately given by Focock ('01), it is
especially noteworthy that the post-abdomen is relatively thick, and
that the appearance of the ventral surface suggests aquatic respiration
by lamellate gills.

Bemard ('96) has contrasted the ventral position of the mouth in
the Oigantostraca with its terminal position in the undoubted
Arachnida, and has founded on this contrast a plea against the union
of the former with the latter group. But this argument surely lays
too great stress on an adaptive character. It might forbid us to
derive Scorpions and other predaceous Arachnids directly from
Eurypterids, but not to trace both back to a common ancestry. And
further, if the forward position of the mouth, as in Galeodes,
characterised the most primitive Arachnids, then the cbelicerte must
be the foremost appendages, and there can be no limbs missing from
the Arachnidan head. This view has indeed been recently put
forward by Heymons ('01), and adopted by Bomer ('02); but it is
in direct conflict with the embryological observations of Jaworowski
('91 and '92), who described a pair of evanescent antennae in the
embryo of Trochosa, and of Brauer ('95), who describes and figures
two ganglionic rudiments in front of the cheliceral ganglia in the
embryo Scorpion. That the appendages of one or two segments have
been lost from the Arachnid head seems therefore certain ; if so, the
mouth of the precursors of the class must have been ventral.

The composition of the head in the Arachnida is of great
importance with regard to the subject of the present essay. Doubt
has been thrown by later observers on Jaworowski's observations;
but the appendicular nature of the vestigial antennae that he figured

342 Proceedings of the Royal Irish Academy.

('91) seemB likely, although the fact that they siiik into pits and
occupy the position of the cheliceral ganglia, makes his interpretation
uncertain. The pre-antennal segments that he descrihed later ('92)
do not seem to rest on yery dear eyidenoe ; while the two pairs of
yestigial appendages lately descrihed and figured hy Fokrowsky ('99),
in the emhryo of Pholcns, are not conyincing. But the presence of a
pre-cheliceral segment is clearly indicated hy the segmentation of the
Scorpion's hrain as interpreted hy Brauer ('95). Anterior appendage*
are perhaps to he found in the paired rudiments of the rostrum, as
suggested hy Groneherg ('80). These are conspicuous in the emhryo of
Trochosa rurieola (fig. 5), which I haye examined in the yain hope
of seeing Jaworowski's antenns. They are also clearly shown in
Brauer's figures of the emhryo Scorpion, ultimately hecoming fused

pljK.

vr 'w.

Fig. 5.

Fio. 5. Front portion of geitn-band of Troehota rurieola ; to., ch., pip., 1., rudi-
ments of roetrum, cbelicers, pedipalps, and first pair of legs ; ch. g.,
cheliceral ganglion: I.e., lateral eyes, x 28.

togethei. Lendl (fide Eorschelt ('99)) has descrihed yestigial appen-
dages hetween the chelicerse and the pedipalps of Spiders. The
former existence of a limh-hearing segment there is indicated hy the
structure of the Fycnogonida, which seem to he an aherrant order
of Arachnida. Not only do the emhryological history, as descrihed
by Morgan ('91), the possession of cheUcersB as the foremost pair of
limbs, and the four pairs of walking-legs suggest Arachnid affinities,
but also the fact that the segments bearing the three hinder pairs of
legs are, in the generalized families of Fycnogonida, as in the
Solifugida, not fused with the cephalic segment on which the four
front pairs of limbs are borne. This indicates that the pycnogonid
appendages, not present in Arachnida generally, are the palps.
Therefore, it is safe to conclude that the Arachnidan head with three

Cabfemtbb — Relatwiuhips bettceen Classes of Arthvopoda. 343

pairs of limbs, was preceded by a head with four pairs, and that
the hi'Oit ancestors of the Arachnida had a head with five pairs of
limbs, the foremost of which were feelers. (See Table, pp. 854-5.)

Relationship between Insects and Crustaceans,

Belief in the multiple origin of the Arthropoda rests chiefly on a
supposed radical divergence between Insects and Crustaceans. It has
been shown in the prerious chapters of this essay that Insects, as well
as Centipedes and Millipedes, can be traced back to ancestral Arthro-
pods with five limb-bearing head-segments, and fifteen Hmb-bearing
trunk-segments ; and that Crustaceans and Arachnids can be traced back
to an ancestral stock showing precisely similar segmentation. It has
been further pointed out that the identity of segmentation in three
distinct classes cannot be reasonably explained as the result of con-
vergence. The strongest presumption is raised for a real kinship
between Insects and Crustaceans. It is desirable, therefore, to compare
the two classes in some points of detail with the view of more firmly
establishing their relationship.

General agreement now exists among zoologists that the feelers,
tritocerebral vestiges, and mandibles of Insects represent, respectively
the antennules, antennae, and mandibles of Crustaceans. Both the
general form of the appendages and the ganglia from which they
are innervated afford evidence that this view is correct ; and Hansen
('98) has brought forward facts that tell most strongly in support
of the homology of the Crustacean with the Insectan mandibles.
He points out that there is a very close likeness, both in form and
musculature, between those jaws in the Thysanura and Collembola
on the one hand, and the lower Malacostraca, especially the
Cumacea, on the other. Indeed there is a much closer likeness between
the mandibles of Diastylis and those of MachiHs than between the latter
and the mandibles of Blatta or any winged Insect.

Most writers on Arthropod morphology have not hesitated to range
the two pairs of Crustacean maxillie with the similarly-named appen-
dages in Insects. But, as we have already seen, the researches of
Hansen ('93) and Folsom ('00), supported by the facts brought forward
early in this essay (pp. 324-5), compel us to recognise the maxillulae
of the Thysanura as a pair of jaws anterior to the two pairs of
maxiUsB. Hansen has further called attention to the points of like-
ness in detail between them and the first maxillsB of Crustaceans, and
between the first maxiHaB of Insects and the second maxiUse of Cms-

344 Proceedings of the Royal Irish Academy.

taceans. "We need not hesitate, therefore, to consider the latter corre-
spondence as accurate, and to compare the second maxillae (labium)
of Insects with the foremost thoracic legs of Crustaceans. That this
correspondence is correct is shown by the fact that the labial segment
in Centipedes (whereof the poison-feet are the appendages) is
not fused with the head-capsule, while in the more generalised
Insects it is only partially fused. Among the Crustacea, we find
that the foremost trunk-segment is added to the head in the Mala-
costraca — notably in the Amphipoda and the tsopoda ; while the same
thing seems to have happened among some of the more specialised
Trilobites— ^.^., Ogygia. Behind the head we find, in a typical
modem Insect, three thoracic and ten abdominal segments, all of
which must have carried appendages primitively. And embryological
researches have shown that eleven limb-bearing abdominal segments
and a limbless twelfth anal segment must have been present in the
ancestors of Insects. Comparing the segmentation of an Isopod with
that of an Insect, we find exactly the same number of body-segments
in each. And the hindmost limb-beaiing segment of the primitive
insect, revealed only by embryological research, is paralleled by the
extra segment in tiie abdomen of the Leptostraca. There is no
improbability in the assumption that this segment originally carried
limbs. There is every reason, then, for tracing back the Insecta and
the Malacostraca to common Arthropod ancestors with twenty-one
limb-bearing segments, and for considering that the abnormally
numerous segments found in Centipedes, Millipedes, and certain
Inlobites and Phyllopods have arisen by increase from that originally
limited number. (See Table, pp. 364-5.)

In comparisons between Insects and Crustaceans, allowance must be
made for the possibility that the stalked eyes in the former class repre-
sent yet another pair of head-appendages. The fact that they may be
abnormally developed as jointed limbs is hard to explain on any
other view. They may be matched in the Insectan branch of the
Arthropoda by the rudimentary appendages of the pre-antennal
segment recently described by Heymons ('01) in the embryo Scolo-
pendra. This segment, as previously mentioned, is compared by
Heymons with the hinder part of the protocerebron in Insects, to
which the compound eyes belong. "We see in this correspondence
suggestion for a close comparison between the Insectan and the
Crustacean compound eye.

If true kinship between Insects and Crustaceans be. thus estab-
lished, it remains that the nature of such kinship be discussed. Have

Carpenter — Relatiamhips between Classes of Arthropoda. 345

the Insects and their allies actually branched off from the Crustaceans ?
Or is it more reasonable to regard Insects and Crustaceans as divergent
stems from a common far-off Arthropod ancestry ?

With regard to this question, the conclusions of Hansen incline
towards the former view. Although (Hutton, etc., '97) he *' dislikes
ancestor-hunting and pedigrees," he expresses the opinion that *' the
lower Malacostracous Crustacea and the Thysanura are more closely
related to each other than hitherto recognised." And Lankester ('81)
admitted the possibility of deriving the Insecta from the Isopoda,
although more recently ('97) he has suggested a wide divergence
between all Insects and all Crustaceans. As we have already seen,
there is a dose agreement in the segmentation of both head and body
in the two groups. When we find further that there is a striking
similarity between the Thysanuran and Crustacean mandibles, that
Insects agree with Amphopods and Isopods in possessing sessile com-
pound eyes, and that the minute structure of these eyes in certain
Thysanura, as shown by Oudemans ('88), agrees with that of the
Crustacean eye in the presence of a hypodermal layer, wanting in the
higher Insects, between the corneal facets and the crystalline cones, it
must be at least admitted that the origin of Insects from the lower
Malacostraca is worthy of discussion. To imagine a close connexion
between Insects and Isopods is easy on account of the adaptation to a
terrestrial life shown by the OniscidsB. But in Isopods, the hinder pair
of feelers are strongly developed ; while in Insects these appendages
are represented only by embryonic rudiments ; and the Thysanuran
mandibles resemble those of the Cumacea much more closely than
those of the Arthrostraca. We seem forced, therefore, to the conclusion
that the base of the Insectan stem cannot be sought above the base of
the whole Malacostracan series, and that such characters as the sessile
condition of the compound eyes common to the Insecta and Arthro-
straca have been independently gained. Nevertheless, convergence
of BO striking a nature could only be possible in two groups somewhat
nearly akin to one another.

But if it is not possible to believe in the origin of insects from any
of the orders of Malacostraca as at present developed, there is some-
thing in favour of the view that they branched off from the immediate
ancestors of the Malacostraca. In connexion with this point, tlie
remarkable Tasmanian " Arthrostracous Schizopod " Anaspides,
lately described by Thomson ('94) and Caiman ('96), is suggestive. In
this animal, the mandible is of the Thysanuran-Cumacean type, but it
possesses a palp ; while the first maxiUa shows much likeness to the

346 Proceedings of the Royal Irish Academy.

maxillula of the Collembola. The exopodite is greatly reduced in
some of the thoracic legs, and there is less distinction between the
thoracic and abdominal regions than in most Malacostraca. For the
ancestors of Insects and Myriapods, we mnst suppose animals without
a carapace and with all the limb-bearing trunk-segments similar.
Such forms might conceiyably have existed among the earliest Mala-
costraca; and Anaspides and its FalsBozoic allies (Faleeocaris, etc.)
come nearer to the ideal than any other Malacostraca known to
zoologists. Neyertheless there is a great difficulty in supposing that
an Arthropod with similar body-segments could be deyeloped from
one in which differentiation between thorax and abdomen had set in.
On the whole, therefore, it is most reasonable to believe that the
ancestors of Insects, Millipedes, and Centipedes were an offshoot from
the progenitors of the primitive Leptostraca. The head and its
appendages show such close correspondence in the Insects and Crus-
taceans, that the Tracheate branch must have arisen above the branch
that had given origin to the Trilobites, in which there was no differen-
tiation between the head- and body-limbs. The result of our inquiry,
therefore, is to trace back the Insecta to ancestors that are essentially
Crustacea, although Crustacea of a very generalised type.

It will now be convenient to discuss the meaning of the position
of the genital aperture or apertures. In this there is a marked
difference between the Insects and the Centipedes, in which the ducts
open near tiie hinder end of the body, and the Millipedes, Symphyla,
and Crustaceans, in which they open more or less anteriorly. This
divergence has often been considered very radical, only explicable by
imagining common ancestors with a large series of paired, segmental,
coelomic ducts which served to carry off the germ-cells. Now it is
very hard to believe that the primeval Crustaceans, which seem to have
been ancestral to all these classes, could have possessed such a very
generalised reproductive system as that. It is impossible to suppose,
on the other hand, that any sudden shifting of the genital aperture from
one region of the body to another could have taken place.

Two sets of facts, when taken together, seem to give the solution
of this problem. There is sometimes not absolute constancy in the
position of the apertures within the same group. And if the Insects,
Crustaceans, and Millipedes be compared together, it is found that
the apertures of the Leptostraca and Malacostraca lie on segments of
the body situated between the genital segments of the Millipedes on
the one hand, and of the Insects on the other. It is found, taking the
antennular segment as the first, that, in the Diplopoda, the genital

Carpenter — Belationships between Clasaen of Arthropoda. 347

ducts open between the eighth and ninth or on tiie ninth segment ;
in the Symphyla between the ninth and tenth segments ; in the Mala-
costraca generally on the eleventh (female), and thirteenth (male) ; in
most Insects between the sixteenth and seventeenth (female), and on
the eighteenth (male). (See Table, pp. 354-5.)

From this snrvey a strong presumption arises that the median
position which still characterises the Crustacea is primitive, and that
a very slight shifting forwards, or a more extended but gradual shift-
ing backwards, has led to the position of the apertures in the other
classes. According to this view, the common ancestors of Insects,
Centipedes, and Millipedes had the genital ducts opening about the
eleventh segment. In the Symphyla and Diplopoda their position
has been shifted forwards, in the Insecta and Chilopoda backwards.
The structure of the ovaries in the Thysanura suggests that the genital
ducts of Insects are not the representatives of some special pair of
segmental organs, but longitudinal mesodermal vessels, analogous to
the arohinephric ducts of Vertebrates, into which the coelomic seg-
mental ducts open. There is no unlikelihood, therefore, in the gradual
shifting far backwards of the reproductive openings among the
Insects.

Relationthip between Araehnide and Cruetaceane,

The discusion in a former section of this essay on the relationship
between the various orders of Arachnida led to the conclusion that the
primitive Arachnids were aquatic animals, breathing by means of
appendicular gills. Naturally, therefore, we compare the Arachnids
with the Crustacea rather than with the Insecta. The immediate
progenitors of the Arachnida appear to have possessed a head with
four pairs of limbs, a thorax with three segments, and an abdomen
with thirteen segments and a telson, only six of which can be clearly
shown by comparative morphology to have carried appendicular gills.
But embryological evidence enables us to postulate with confidence
still more remote ancestors in which the head carried well-developed
compound eyes and five pairs of appendages, while it may be supposed
that aU the abdominal segments, except the anal, bore limbs. In these
very ancient Arthropods, all the limbs, except the feelers,hadambulatory
and branchial branches ; and one important feature in the evolution of
the Arachnida must have been the division of labour between the
anterior and posterior limbs, the former becoming specialised for loco-
motion, the latter for breathing. Another was the loss of the feelers

348 Proceedings of the Royal Irish Academy,

and the degeneratioii of the compound eyes. Thus we are led to trace
the Arachnida (including the Merostomata and Xiphosura) back to
ancestors which cannot be regarded as Arachnids, but which were
identical with the primitiye Trilobites, and near the ancestral stock
of the whole Crustacean class.

According to this conception of the primeval Arachnids, they early
lost the Crustacean antennules; and tiie primitively compound eyes
underwent more or less degradation. The foremost of the ordinary
appendages, that, in the Crustacea, became the antennsB, took on
in the Arachnida the function of chelicerae ; this homology is sup-
ported by a comparison between the antennal glands of the former
class and the poison or spinning-glands found on the chelicerse in
several orders of the latter. The Crustacean mandibles seem to be
represented among the Arachnida only by the palps of the Fycno-
gonida, while the pedipalps of typical Arachnids (ovigerous legs in
the Fycnogonida) and the first pair of walking-limbs correspond
respectively with the two pairs of Crustacean maxillee. Thus we come
to the hind margin of the ancient Arthropod head. (See pp. 354-5.)

Jaworowski's interpretation ('91) of the embryonic Hmbs of Spiders
as consisting of protopodite, endopodite, and exopodite (which
develops into the actual leg), would support the Crustacean re-
lationship of the Arachnids, at the expense of the homology of the
walking-legs in the two classes. In Crustacea, the endopodite is the
ambulatory branch, and doubtless this is also the case in the
Arachnida. The ** endopodite " of Jaworowski seems to be the basal
masticatory region of the appendage.

It is strange what differences of opinion have prevailed regarding
the possibility of near relationship between the Tiilobites and the
Limuloids. Eor example, Bernard ('94) considers such a relationship
unquestionable, deriving the Limuloids directly from the Trilobites,
and referring both orders unhesitatingly to the Crustacea. Lankester
('97) also claims a relationship between the two orders ; but he refers
the Trilobites, despite their antennae and biramous Hmbs, as well as
the Limuloids, to the Arachnida, Eingsley ('94) considers the
Limuloids akin to the Arachnida, and the IMlobites to the Crustacea,
denying any close relationship between the two orders. According
to liie view here adopted, the Xiphosura and Merostomata are to be
referred to the Arachnida, while the antennae and biramous limbs of
Trilobites oblige us to class them with the Crustacea. A direct
descent of Limuloids from specialized Trilobites cannot be maintained,
since the Merostomata, which are certainly more primitive than the

Carpenter — EelationsAips between Classes of Arthropoda. 349

lamuloids, show less than they a superficial resemblance to Trilobites.
But though the cephalothoracic carapace of Limulus does not
correspond with the head shield of a Trilobite, the former may, in all
probability, have arisen by the fusion of three trunk-segments with
the primitive head-shield of the Proto-Trilobita. There is no difficulty
in tracing back the Merostomata, the Xiphosura, and the Trilobita to
a common ancestry; and thus the Arachnida as a class, like the
Insecta, haye been evolved from Crustaceans. Except among the
Fycnogonida, where the reduction of the abdomen has necessitated a
shifting forwards, the genital openings in the Arachnida arc only one
segment in front of the position of the female genital openings in the
typical Crustacea.

The sessile eye of the Trilobites, the median telson in their more
primitive genera (contrasting with the tail-furca of the Leptostraca
and Fhyllopoda), and the trilobite-larva of Limulus, all suggest the
probability that the Arachnida arose from the base of the Trilobitan
branch, rather than from the main Crustacean stem. And the loss of
the antennules, together with other specialized characters, shows that
the Arachnida have diverged much more widely from the Crustacea
than the lowest Insecta have.

The Ancestry of the Arthropoda as a whole.

We have seen that the Insecta, the Chilopoda, the Diplopoda, the
Crustacea, and the Arachnida can all be traced back to common
Arthropodan ancestors, with a definite number of segments. The
origin of these primitive Arthropods now demands consideration.
They were distinctly Crustacean in character, so the question of their
history may practically be reduced to that of the very remote
ancestry of the Crustacea. Two theories on this subject need to be
discussed — the older view that the Nauplius larva represents the
primitive Crustacean ; and the newer, according to which the Crustacea
must be direcUy derived from Annelidan ancestors, the nauplius being
regarded merely as a modified trochophore with certain adult
Crustacean characters precociously developed.

As an introduction to the examination of the Annelidan theory, we
must try to ascertain the relationships of the Malacopoda (Feripatidae),
since they show more Annelidan characters than any other group that
can be considered as belonging to the Arthropoda. Although the
Malacopoda have not jointed limbs, there need be no hesitation in

350 Proceedings of the Royal Irish Academy,

grouping them with the Arthropoda, since they possess the more
fundamental characters of the Phylum — ^the ostiate heart, the peri-
cardium and secondarily formed body-cayity consisting of swollen
blood-spaces, and the greatly reduced coelom (Sedgwick, '87). In
their soft skin and simple segmentation, these animals strongly recall
the Annelids. Only three pairs of limbs (or two, if the feelers be, as
belieyed by some, primitively pre-oral) are carried on the head, instead
of the five usual in Arthropoda. But perhaps the most striking
feature in which the Malacopoda differ from other Arthropods, is the
presence of paired coelomic excretory ducts in aU the body-segments.
These have been constantly compared with the nephridia of segmented
worms.

The number of body-segments varies greatly in the different genera
of the Malacopoda ; and it is hard to determine whether the ancestral
stock of the class had few or many segments. Bouvier ('00) regards
the Tropical American genus Feripatus, with from twenty-two to
over forty pairs of legs, as the most primitive, pointing out that in this
genus the genital opening is situated between the penultimate legs,
while in most of the other genera it is between the hindmost existing
pair, those corresponding with the hindmost in Feripatus having been
lost. But in their method of development, the Australian Feripatoidinae,
with their large, yolked eggs, in some cases hatched outside the body
of the mother, are certainly the most primitive members of the class,
the views of Sedgwick ('88), adopted by Korschelt ('99), being much
more reasonable than the suggestion of Willey ('98) that the acquisi-
tion of yolk in this class has been recent. It is very interesting
to note, then, that the segmentation of these animals agrees closely
with that of the typical Arthropods. They have from fourteen to
sixteen pairs of legs, so that, allowing for the head appendages
and the lost pair of legs on the post-genital segment, we arrive at
from eighteen to twenty limb-bearing segments. As regards their
segmentation, therefore, the Malacopoda might have been derived
from the typical Arthropodan stock, although the number of segments
in the class is too variable to justify any definite theory on this
subject. And such very primitive characters as the set of paired
segmental organs and tlie simple nature of the eyes, obliges us to
consider the Malacopoda as an offshoot from the far-off ancestral stock
of the other Arthropodan classes. As pointed out by Sedgwick ('95)
and Lankester ('97), they stand far below the rest of the Arthropoda.
Any attempt to derive the Insecta directly from them, through the
Ohilopoda, is vain in view of the numerous correspondences between

Carpbntbr — Relationships between Classes of Arthrqpoda. 351

Insects and Crostaceans — unless we are willing to explain any like-
ness whatsoever as the result of '' conyergence."

The concealed situation in which the Feripatids liye might incline
us to the yiew that they have lost on originally firm exoskeleton.
Insect-larv® that liye in wood are soft-skinned, while their allies in
the outer world are well armoured. Yet there is an undegraded
aspect about a Peripatid that makes such a yiew hard to accept; and it
is more reasonable to regard the type as a yery ancient one that
has come down, like certain Brachiopods, from a remote period, with
yery little modification. But granting this, do the Peripatids really
help to bridge the gap between other Arthropods and Annelids ? The
soft skin, the simple eyes, and the segmental organs are really the only
distinctly Annelidan characters of the Malacopoda, and the force of the
last-named and most important of these is greatly weakened if Good-
rich's yiew ('97), accepted by Lankester ('00), be established, that the
ccslomic ducts of Arthropods (to which category the segmental organs
of Peripatus must certainly be referred) haye nothing to do with true
Annelidan nephridia. Certainly the legs of Peripatus resemble
Annelidan parapodia as little as they resemble Crustacean appendages.
On the whole, then, the Malacopoda are low-type Arthropoda, of
uncertain segmentation, but with the fundamental characters of the
phylum, and showing only a superficially Annelidan appearance.

The question of the Annelidan ancestry of the Arthropods must
remain, then, a matter for speculation. To the present writer it
seems unsound morphology to compare closely the most highly-
deyeloped class of worms (CHisBtopoda) with the most highly-organised
of aU Inyertebrates (Arthropoda). The presumption must always be,
in such cases, that each group has become specialised along its own
lines, and it is most unlikely that the one can haye deyeloped
directly from the other. Both may haye diverged from a common
ancestry; but the closed blood-vascular system and ccslomic body-
cavity of the ChsBtopoda point to the period of such an ancestry as
immensely remote.

As contrasted with the vast difficulties involved in the transforma-
tion of Polychaete worms into Phyllopods, the derivation of the
primeval Arthropods from Naupliif orm ancestors by a gradual increase
in the number of segments is perfectly simple ; and, before many years
have passed, zoologists are likely to revert to Miiller's theory of
Crustacean origins. The occurrence of the Nauplius larva, or its
representative, in all the great groups of Crustacea back to the Trilo-
bites raises the strongest presumption of some phylogenetic meaning ;

B. I. A. PBOO., VOL. XXIV., BBO. B.J 2 F

352 Proceedings of the Royal Irish Academy.

and just as the origin of a Crustacean from a Kauplius is reasonable
and natural, so the origin of the Kauplius itself from some soft-skinned,
poorly-segmented trochophore-like form, is quite possible. Perhaps
belief in the Rotif era as the connecting links between the Annelids
and Arthropods, will yet be justified, for the divergence between the
great main types of animal structure must bave begun when all the
life-forms were aquatic and all microscopic.

This view of the relationship between Arthropods and Annelids
throws light on the origin of the Feripatids. Before they became
adapted to life on land they must have been marine forms with free-
swimming larvsD. The three pairs of Kaupliar appendages correspond
with tbe three pairs of appendages in the bead of Peripatus. There-
fore it is probable that the microscopic ancestors of the Malacopoda
had acquired three pairs of appendages, but that these had not yet
become jointed or chitinised. Thus the far-off ancestors of the Arthro-
poda gave origin to a lower soft-skinned branch, whence sprang" the
Malacapoda, subsequently adapted to a terrestrial life ; and to a higher
firm-skinned, and truly <* arthropodous" branch, with a definite number
of segments, from whicb have descended the Crustacea, the Insecta
and their allies, and the Arachnida.

Conehuion,

It may be conyenient briefly to sum up the principal conclusions
set fortb in this essay : —

1. The Arthropoda are a natural, monophyletic assemblage of

animals.

2. There is exact numerical correspondence between the segmenta-

tion of typical Insects, Crustaceans, and Arachnids.

3. Such correspondence in three distinct classes cannot reasonably

be explained as the result of couTergent evolution from
ancestors with very numerous segments, which indepen-
dently became diminished to exactly the same extent.

4. The ancestral Arthropods must therefore have possessed a fixed

and definite segmentation ; and the "various forms with
very numerous segments (Phyllopods, Millipedes, &c.)
have undergone abnormal elongation.

OAVt^rEVtHtL—lteiationships between Classes of Arthropoda. 353

5. The Inaectay Cbilopoda, and Diplopoda may be derived from

common Symphylan ancestors, which branched ofl from
the primitive Crustacea (proto-Leptostraca).

6. Among the Crustacea, the Leptostraca and the Trilobita show

the most primitive characters. The proto-Trilobita had
the typical Arthropodan number of segments.

7. The Arachnida, including the Merostomata, Xiphosura, and

Pycnogonida, arose from the proto-Trilobita.

8. The Malaoopoda must be regarded as Arthropoda of low type.

They have no close relationship to Chilopoda or Insecta,
and their Annelidan affinities are doubtful.

9. The Arthropoda, as a whole, probably sprang from Kaupliform

ancestors, and not from well-developed Annelid-worms.

The genealogical *' tree " (Plate YI.) may serve to show these
conclusions in a graphic form.

[T,

TABLB SHOWINQ THB NUMERICAL GOBBESPOKDBNOB IK

ASACHKCDA.

Cbubtacsa.

Scorpionida.

Merostomata*

Pycnogonida.

Trilobita
(Olenellus}.

Leptostraca.

Ocular Segment

—

Ocular Segment

Stalked Eyes

Segment
CheHcerae

—

Antennules

2
3

Ohelicerae

Chelicerae
Palps.

1st Biramous

limbs of Head
2nd „ „

Antennse
Mandibles

Pedipalps

1st Legs

Ovigerous limbs

Sid „ .»

IstMazills

IstLegs

2nd „

1st Legs %

4th „ „

2nd „

3id „

2nd „ ?

iBt Trunk limbs

1st Thoracic Le|

3rd „

4th „

3fd „ s %

2nd „ „

4th „

Paddles

4th „ cT ?

3rd „ „

3rd M „

9
10

Pre-genital Seg-
ment
Operculum i ?

Metastoma
Operculum s ?

[Abdomen greatly
reduced & con-
densed]

4th „ „
6th „

4th „ „
5th „ „

Pectines

Ist Gill-plates

6th „ „

let Lung-books

2nd „

7th „ „

7th „ .,

2nd „

3rd „

8th „ „

3ni „

4th „

9th „ „

1st Pleopods

4th .,

6th „

10th „ „

2nd „

11th „ „

3«i „

Ist Tail-Segment

12th „ „

4th „

13th „ „

6th „

14th „ „

6th „

20
21

Anal Segment

I6th „ „
Anal Segment

Limbless Seg-
ment
Anal Segment

Telson

Furca

The line after Segment 6 indicates the hind-margin of the piimitiye Arthropodan Head.

SEQMENTATION BETWEEN THE CLASBEB OF THE ABTBBOPOBA.

Insbota

(Machilis).

Stmphtla
(Scolopendrella).

DiPLOPODA

(Polyxenus).

Ohilopoda
(Lithobius).

Malaooetraca

(Afltaous).

Stalked Eyes

Ocular Segment

—

Ocular Segment

Pre-antennal rudiments
fSoolopendra)

Antennules

FeeleiB

Feelen

Feelers

Antenna

Tritocerebral

—

Tritocerebral Segment

Mandibles

Segment
Man£bles

Mandibles

IstMaziUtt

MaxilliiLB

Mii^'niii«

IstMaxilhB

2nd „

IstMaziUs

Ist Maxille

( Gnathochilarium

2nd „

IstMaxiUipedB
2nd „

2nd Mazills
(Labium)
IstLegs

2nd Maxillo)
IstLegs

[Vestigial Seg-
ment]
1st Legs

Poison-feet
IstLegs

3rd „

2nd „

Chelie

3id „

3rd „ ''

3id „

l8t Legs
2nd „ ?
3rd „

1st Abdominal

Sesment
Ist.Abdominal
Limbs
2nd „ „

4th „ ^^
6th „
6th „

4th „
1 6th „
( 6th „

4th ,
6th „
6th „

4th „ i

3id „ „

7th „

l»t Pleopods

4th „ „

8th „

2nd „

6th „ „

9th „

1 9th „

9th „

3rd „

6th „ „

10th „

1 10th „

10th .,

4th .,

7th „ „'

11th „

nith „

nth „

6th „

8th „ „ s

12th „

Il2th „

12th „

Uropods

Cercopods

Reduced Limbs
Cercopods

13th „

Limbless Seg-

ment
Anal Segment

13th „
14th „

Anal Segment

16th „

Telaon

Genital limbs s %
Anal Segment

The signs 6 ? indicate the positions of the male and female genital openings, respeotirely.

356 Proceedings of the Royal Irish Academy.

BEEEBENOES.

'00. Beecheb, C. £.— Trilobita, in ZittePs '' Text-book of Palaeon-
tology." (EngliBh edition.) London, 1900.

'94. Bebvasd, H. M. — ^The Systematic Position of the Trilobites.
Quart. Jour. Gfoh 8oe,, toI. 50, 1894.

'96. BxENABD, H. M. — The Comparative Morphology of the Gkdeodidse.
IVant. Linn. 8oo. (2), ZooL, vol. 6, 1896.

'93. BoLLMAir, C. H.— The Myriapoda of North America, Bull. 46,
IT. S. Nat. Museum, 1893.

'02. BoBK£B, C. — ^Arachnologisohe Stndien. Zool, AnM.^ vol. 2t5,
1902.

*00. BoTTviEB, E. L. — Observations sur les Peripatus de la Collection
du Mus6e Brittannique. Quart. Jour. Mier. Sei.f vol. 43,
1900.

'69. B&ikUEB, F. — ^Betrachtong iiber die Yerwandlung der Insekten.
VerhanM. ZooUBot. Gmll Wim.Yol. 19, 1869.

'95. Bbauxb, a.— Beitrage zur Kenntnis der Entwicklungsgeschichte
des Skorpions. Z^t. WtM. JZ6ol.,Yol. 59, 1895.

'96. Calkak, W. T. — On the Genus Anaspides, and its affinities with
certain fossQ Crustacea. lY'ans. Eoyai. 8oo. JSdinb.y vol.
38, 1896.

'99. Cabpbnteb, G. H. — Insects, their Structure and Life. (Chi^. 6.)
London, 1899.

'72. Claub, C. — ^IJeber den Bau und die systematische Stellung von
Nebalia. Zeit. Wiss. Zool, vol. 22, 1872.

'02. CoHSTOCK, J. H., and Koghi, C— The Skeleton of the Head of
Lisects. Am&r. Nat., vol. 36, 1902.

'80. CnoifEBERo, A. — ^Ueber die Mundtheile der Arachniden. ArcK
filr Naturgmh., vol. 46, 1880.

'83. Eatok, a. E. — A Bevisional Monograph of the existing
Ephemerids or Mayflies. Dram, Linn. Soc. (2), vol. 3,
1883-85.

'99. FoLsoM, J. W.— The Anatomy and Physiology of the Mouth-parts
of the Collembolan Orchesella cincta. BuU. Mus. Comp^
^no^., Harvard, vol. 35, 1899.

CATiP'RVT'Rn—Relf(itans/up9 betteeen Clasiea of Arthropoda. 357

'00. FoLsoir, J. W. — The Deyelopment of the Mouth-partB of
Anurida maritizna. Id,y vol. 86, 1900.

'97. Gk>ODBicH, E. S. — ^On the Nephridia of the Folycheeta. Q^art,
Jour. Micr. Set., vob. 40, 41, 43. 1897-1900.

'85. Gbassi, B. — ^Morphologia delle Scolopendrella. Atli, R. Aeead.
Torino^ vol. 21, 1885.

'92. Obobben, C. — ^A contribution to the Knowledge of the Genealogy

and Classification of the Crustacea. Ann, Mag. Nat.

JSTm^. (6), vol. 11, 1893 (from Sitzb. K.Akad. Wm., Wien,

1892.)
'86. Haahe, E. — Beitrag zur Fhylogenie und Ontogenie der Chilopoden.

SehUi. ZeiUch. Entom. 1881.
'89. Haask, E. — ^Die Ahdominalanhange der Insekten mit Bordcksich-

tigung der Myriapoden. Marph i/aAr3., vol. 15, 1889.
'66. Haeokbl, E. — The History of Creation. (English Edition).

London, 1876.
'93. Hansen, H. J. — A Contribution to the Morphology of the Limbs

and Mouth-parts of Crustaceans and Tnsects. Ann. Mag.

Nat. Hut. (6), vol. 12, 1894 (from Zoohg. An%., 1893).

'97. Hansen, H. J. and Sobensen, W. — The Order Falpigradi and its
Eelationship to the Arachnida. JEntom. Tidik, 1897.

'03. Hansen, H. J. — The Genera and Species of the Order Symphyla,
Quart, Joum. Mter. 8ci., vol. 47, 1903.

'88. Habtoo, M. M. — The Morphology of Cyclops and the Eolations
of the Copepoda. Urant. Linn. 8oo. (2), vol. 5, 1888.

'95. Hbtxonb, R. — Die Segmentiemng des Insectenkorpers. Grund-
ziige der Entwickelung und des Korperbaues von Odonaten
und Ephemeriden. Fhgs, Abhandl. K. Akad. Wtssen.^
Berlin, 1895-6.

'96. Hetkons, B. — ^Zur Morphologic der Ahdominalanhange bei den
Insekten. Morph. Jahrh., vol. 24, 1896.

'97. Hetmons, E.. — Mittheilungen iiber die Segmentiemng und den.

Eorperbau der Myriopoden, Sitiuh. K. Fteu8s. Akad.

Wmensch, 1897.
'01. Hetmonb, E. — Die Entwicklungsgeschichte der Scolopender

Zoologiea^ vol. 13, 1901.

'02: Howes, G. B.— The Morphological Method and Frogress. Rep.
Brit. AttQC.f 1902.

358 Proceedings of the Royal Irish Academy.

'97. HuTTow, F. W., AJSTD Othvbs. — ^Are the Arthropoda a natural
Group ? Mt. 8ei., vol. 10, 1897.

'58. Huxley, T. H. — On the Agamic Beproduction and Morphology
of Aphis. IVans. Linn. Soe.^ yoI. 22, 1859.

'78. Huxley, T. H. — Manual of the Anatomy of Invertebrated
Animals. London, 1878.

'91. Jawobowski, a. v. — XJeber die Extremitaten bei den Embryonen
der Arachniden und Insekten. Zool. AnM.^ vol. 14, 1891-2.

'92. Jawobowski, A. v. — Uber die Extremitatem, deren Driisen und
Kopsegmentierung bei Trochosa eignoriensis. Zool. Awi.y
vol. 15, 1892.

*94. Jawobowski, A. v.— Die Entwicklung der sogenannten Lungen
bei der Arachniden. Zeitsch, JTiss, ZooL, vol. 58, 1894.

'96. BIenyon, F. C. — The Morphology and Classification of the
Pauropoda. Tuft8 Coll. Stud., No. 4, 1895.

'88. KnrosLEY, J. S. — The Classification of the Myriapods. Amer.
iVb^., vol. 22, 1888.

'94. KivosLEY, J. S. — ^The Classification of the Arthropoda. Id.f
vol. 28, 1894.

'99. KoBSCHELT, E., and Heideb, K. — ^Text-book of the Embryology
of Invertebrates (English Edition, by Bernard and
Woodward), vol. 3. London, 1899.

'91. Lavg, a. — Text-book of Comparative Anatomy. (Trans. H. and
M. Bernard.) Fart L London, 1891.

'81. Lankesteb, E. E. — Limulus an Arachnid. Quart Jowm. Micr.
Set., vol. 21, 1881.

'97. Lakkesteb, E. E. — ^Are the Arthropoda a natural Ghroup P Ifat.
Sei.f vol, 10, 1897.

'00. Lankesteb, E. E. — A Treatise on Zoology. Fart II. London,
1900.

'84. Latzel, E. — Die Myriopoden der Osterreichisch-Ungarischen
Monarchic. Wien, 1880-4.

'74. Lubbock, J. — The Origin and Metamorphosis of Insects.
London, 1874.

'95. MiALL, L. C. — The Transformation of Insects. Nature^ vol. 53,
1895.

CAnTJ^VTER-^Eelatianshipa between Classes of Arthrapoda. 859

'69. HuLLXB, V. — ^Facts and Figures in aid of Darwin. (Transl. by
Dallas.) London, 1869.

'88. OunsHANS, J. T. — ^Beitrage zur Kenntniss der Thysannra nnd
Collembola. Bijid. tot Dierk, Amsterdam, 1888.

'82. Faokasi), a. S. — On the Homologies of the Crustacean Limb.
Amer. Nat., vol. 16, 1882.

'98. Packahd, a. S.— a Text-book of Entomology. New York, 1898.

'94. Peach, B. N. — Addition to the Fauna of the Olenellus-zone of
the N.-W. Highlands. Q^art, Joum, QeoL 8oc,y vol. 60,
1894.

'93a. Pocock, R. I. — On the Classification of the Tracheate Arthro-
poda. Zool. An%,f vol. 16, 1893.

'93b. Pocock, E. I.— On some points in the Morphology of the
Arachnida, with Notes on the Classification of the Group.
Ann, Mag, Nat, JSist, (6), vol. 11, 1898.

'01. Pocock, E. I. — The Scottish Silurian Scorpion. Quart, Jtmrn.
Mier, 8ci., vol. 44, 1901.

'02. Pocock, E. I. — A New and Annectant Type of Chilopod. /rf.,
vol. 46, 1902.

'99. PoKBOwsKT, S. — Noch ein Paar Kopfhocker hei den Spinnenem-
bryonen. Zool, Anz,, vol. 22, 1899.

'96. Pu&CBLL, F. — Note on the Development of the Lungs, Entato-
physes, TrachesB, and Genital Dncts in Spiders. Zool,
Ann., vol. 18, 1895.

'80. Etdbr, J. A. — Scolopendrella as the Type of a New Order of
Articnlates. Amer, Nat, vol. 14, 1880,

'87. Saba, G. 0.— Eeport on the Phyllocarida collected by H.M.S.
"Challenger." London, 1887.

'90. ScuDDKE, 8. H. — New Carboniferous Myriapoda from Illinois.
Mem. Boston Soc. Nat. Hist., vol. 4, 1890.

96. ScHxiDT, P. — Beitrage zur Kenntnis der niederen Ifyriapoden.

Zeitsch, Wissen. Zoolog,, vol. 69, 1896.
'87. Sbdowick, a.— Peripatus, in the Cambridge Nat. History, vol.

6, 1896.
'87. Skdowick, a.— The Development of the Cape Species of

Peripatus. Qaart Joum. Mier. Sd., vols. 26-8, 1886-8.

B.I.A. PBOC, vol.. XZIV., SBC. B.] 2 O

360 Proceedings of the Royal Irish Academy.

'94. SiMKONS, 0. L. — DeYelopment of the Lungs of Spiders. Amer.
Joum. Sci,, vol. 48, 1894.

'94. Thokson, G. H.— On a Freshwater Schizopod from Tasmania.
Trans. Linn, Soc. (2), vol. 6, 1894.

'97. IJzELy H. — Beitrage zur Entwicklungsgeschichte von Campodea
staphylinus. Zool, An%,^ vol. 20, 1897.

'95. Wagner, J. — Contributions to the Phylogeny of the Arachuida.
Ann, Mag, Nat. Hist. (6), vol. 15, 1895. (From Jena.
Zeitsch. Naturwiss.y vol. 29, 1894.)

'93. Wheeleb, "W. M. — A Contribution to Insect Embryology. Jaum.
Morph., vol. 8, 1893.

'98. WiLLET, A. — The Anatomy and Development of Feripatus novae-
Brittaniee. Cambridge, 1898.

'87. ZiTTEL, K. A. — Trait6 de Paleontologie (traduit par Barrois),
vol. 2, part 1. Paris, 1887.

t 361 ]

THE INTKU8IVE GNEISS OF TIRERRILL AND
DRUMAHAIR.

By GRENVILLE A. J. COLE, M.R.I.A., F.G.S., Professor of
Geology in the Royal College of Science for Ireland.

Read Jttnb 8, 1903.

Thb beautifully contrasted scenery between Ballysadare and Manor-
hamilton, in the counties of Sligo and Leitrim respectively, is due to
the ridge of ancient gneiss, with its irregular and rounded summits,
which here apperrs through Lower Carboniferous strata. On the
south-east rises a broad upland, that culminates in the coal-field of
Lough Allen; on the north-west, the Carboniferous Limestone
weathers out in huge scars and terraces, from the cUfis of Glenade
and Benbulben, to the massive outlier of Knocknarea.' The con-
spicuous gneissic axis, running north-east and south-west, in continua-
tion of the line of the Ox Mountains, has been the subject of various
investigations.

By its general character and trend it is to bo classed with the
Caledonian folds of M. Bertrand, as a mass which was brought into
its present position by earth-movements in earliest Devonian times. In
this it agrees with the main axes of folding throughout the county
of Donegal ; but it is well recognised that the rocks thus brought
into prominence and re-arranged may be much older than the Cale-
donian epoch of earth-movement. Prof. Hull^ included the gneiss
now under consideration ''provisionally" in his Laurentian group;
and I cannot bring forward any conclusive proofs that it is of later
age than the close of the Archaean era. The Hercynian movements
sent earth- waves against it, which uptilted the Carboniferous strata on
its flanks, while preserving its north-east and south-west trend. As

^Comparo A. B. Wynne, '* On the Geology of Parts of Sligo, &c." Journ.
Oeol 8oe,y Dublin, vol. x. (1863), p. 34.

> « On the Laurentian Books of Donegal and of other parts of Ireland." Trans.
&. Dublin Soc., voL i. (1882), p. 262.

362 Proceedings of the Royal Irish Academy.

the Geological Survey Maps, sheets 43 and 55, so excellently show,
faults were at the same time produced, which enabled the ancient
crystalline mass to assert itself aboye the denuded Carboniferous
Lhnestone as a '* horst."

Those who have hitherto examined the gneiss of the ridge do not
seem to have greatly concerned themselves with its mode of origin.
Mr. G. H. Einahan^ classes the gneiss of the Ox Mountains, with other
western gneisses, as a highly altered sedimentary series. In dealing
with such rocks in Galway, he speaks' of schists that graduate into
" metamorphic granite and granitoid gneiss " ; even when he states
that '^ rocks of the older groups are absorbed into the granite and
gneiss/' it appears that we must not read into these words the
modem view that the granite is intrusive and is responsible for much
of the metamorphism. The words ** changed into gneiss or granite "
occur later, and indicate the prevalent attitude of the Irish surveyors
twenty years ago. Mr. E. T. Hardman's^ paper, in the same volume,
is a solid contribution to the geology of the Ox Mountains, and deals
specially with the north-east portion of the range. The gneiss is
clearly regarded as of sedimentary origin, and attention is called for
the first time (p. 358) to '^a curious band of conglomerate," near
Ballydawley Lake, consisting of ^^ a coarse granitoid gneiss, containing
lenticular blocks and rounded pebbles of diorite or homblendic rock
weathering out on the surface." The importance of these inclusions as
indicating some earlier mass of homblendic rocks is duly noticed.

The mineral notes in Mr. Hardman's paper are somewhat incom-
plete, and are subordinate to a very detailed description, by the
author and Prof. Hull, of the dyke of serpentine in the valley of
Correagh. I venture to question if olivine is disseminated in the
gneiss at any point, as is implied on p. 361 ; the granules observed
were probably a green pyroxene, like that derived from eclogite in
Glennagoolagh. Yery scant justice, on the other hand, is done to
garnet, which simply abounds throughout the range.

The Memoir to Sheet 55 of the Geological Survey of Ireland was
written by Mr. J. K. Kilroe, and was published in 1885. Simul-
taneously Mr. A. B. Wynne's Memoir to Sheets 42 and 43 appeared,
which includes the gneissic areas of the Bosses and Manorhamilton.

1 <» Palseozoic rocks of Galway and elsewhere in Ireland, said to be Laurentians,'*
Sci. Proc. B. Dublin Soc, vol. iii. (1882), p. 348.

^Jbid,,^. 353.

3 <* On the Metamorphio Bocks of Counties Sligo and Leitrim, and the enclosed
xnineialB," ibid., p. 357.

OoLB — The Intrusive Oneiss of TirerriU and Drumahair. 363

Both authors speak of the *' bedding " of the gneiss ; and the general
variations in its structare are well described. Mr. Eilroe^ notes its
tendency to pass into qnartzite, by the disappearance of felspar and
mica ; and these qnartzose areas are shown upon the map. The same
author lays proper stress on the homblendic inclusions observed by Mr.
Hardman, and gives excellent figures of them. He declines, however,
to regard the rock as a conglomerate, and makes the important obser-
vation that '' thin streaks of homblendic schist and gneiss also occur
in the same place which bifurcate, and thus become lost in the con-
taining rock." Sir A. Geikie is quoted in a foot-note as considering
the basic masses as '^ geodes — segregations of hornblende rock in the
gneiss." At that date this was the common way out of all such diffi-
culties, and Professor Sollas' was probably one of the first British
geologists to enter a protest against the assumption of local segregation
as opposed to igneous absorption and inclusion.

IVom experience gained in southern and central Donegal,* I was
led to conclude that these interesting rounded masses of amphibolite,
and the conspicuous banding of the gneiss throughout the ridge, were
phenomena of igneous intrusion, t.^., that a granite magma had
penetrated an earlier series of rocks along the axis of the Ox Moun-
tains. Nothing could be better, from this point of view, than Mr.
Kilroe's descriptions and figures of the phenomena near Ballydawley
Lough ; and it is noteworthy that this author refrained from drawing
any conclusion on his own account. I am fortunately able to add
details of similar features from other portions of the baronies of TirerriU
and Drumahair, which will, I think, materially assist in a correct
appreciation of the ground.

Taken as a whole, the gneiss of the area may be regarded as a
granitoid rock, consisting of quartz and potash-felspar, fairly free
from mica, but occasiondly containing biotite. The micaceous por-
tions are arranged in strings and bands, and sometimes impart a
superbly gnarled and striped character to the mass. In the town-
lands of Dromore, Crossboy, and Killery, east of the Correagh or
Slishwood valley, white quartz-veins have penetrated the rock along

1 Memoir to sheet 65, p. 15.

' '* Relation of granite to gabbro of BamavaTe," Trans. B. I. Acad., vol.
ux. (1894), p. 502.

' ** On metomorphic rocks in Eastern Tyrone and Southern Donegal,*' Trans.
B. I. Acad., vol. zui. (1900), pp. 453 and 464.

*< On composite gneisses in Boylagh," Proo. B. I. Acad., vol. xziv., section B
(1902), p. 203.

364 Proceedings of the ttoyal trUh Acadeiinj.

the foliation-lajers, and still further emphasibe its handsome struc-
ture. In Castleore, on the other hand, just ahove the Correagh hamlet,
the quartzose type prevails, and may easily be taken for a true quartzite
in the field. In sections under the microscope, however, the rock is
seen to be still f elspathic, and a good type of that fluidal aplite,
affected locally by pressure, which forms the basis of the gneiss of
north-west Ireland.

In Castleore, the rocky bosses of brown gneiss show a delicate
banding, which is mainly due to abundant strings of garnets carried
out along the general lines of flow (fig. 1). Where blocks of amphi-

rig.1.

Microscopic section of fine-grained gneiss (fluidal aplite) with
abundant garnet. Castleore. x 18. The garnet is derived
from the included eclogites and amphibolites.

bolite occur in the gneiss, the banding becomes emphasised, and the
fiow-surfaces fold round them. At its junction with one large mass,
the gneiss sends off dykes into the amphibolite, and cuts into it along
a zig-zag surface, the fiow-lines following the serrated margin
(fig. 2). The appearance of sharp folds thus produced in the gneiss
is due to its having worked its way into the amphibolite along
joints or planes of weakness. In one place a dyke arose ; in another
the amphibolite became deeply notched ; and the crest of the " infold "
of gneiss occupying the notch sometimes runs on as a thin sheet into

GoLB — The Intrtmve Onms of Tirerrill and Ih^mahair. 365

the crack which determined its position. The case is condnsiTe
against the production of the foliation in the gneiss by subsequent
pressure. The banding is due to primary flow ; the metamorphio
effect of the hot magma on the amphibolite can be clearly trac^ed ; and
the accumulation of garnets locally in the gneiss, at the expense of the
amphibolite, is easily observable in the field.

The absence of marked alteration in the colour of the aplitic gneiss
indicates, however, that little absorption of basic amphibolite has here
gone on. I have elsewhere* given reasons for regarding streaks and
layers of garnet in this type of gneiss in north-west Ireland as
distinctly derivative ; but we may conceive that the foreign material
absorbed in Castieore was a series of quartzites, schists, and lime-
stones, containing only a few basic igneous rocks. In this and similar
cases, we may picture the garnets as arising during the early stages

Fig. 2.

Dykes and trough-like intrusions of gneiss (fluidal aplite) in
amphibolite, Castieore. From a photograph by the author.
Width of the block shown, 80 cm.

of metamorphism of the invaded masses, and then being carried off,
and f requentiy dissolved, in the dominant intrusive rock, which in the
first instance promoted their growth along the contact-zone. In
other cases they may have formed a constituent of an already
metamorphosed and schistose series, into which a granitic magma pene-
trated, inducing the formation of sillimanite, altering amphibole to
biotite, but without effecting much else in the way of crystallisation.
In confirmation of the above observations, it should be stated
that in a section in the Geological Survey collection, cut from a granulitic
gneiss near Slishwood, the garnets are associated with patches of biotite,

1 Op, eit.y Trans. B. I* Acad., voU xzxi., p. 457 ; also p. 456.

366 Proceedings of the Royal Irish Academy.

which is probably a relic of amphibole, absorbed into the aplitic intruder.
Again, on the rocky crest about one mile from Drumahair, beside the
mountain-road to Lurganboy, the yarious stages of absorption can be
traced with the unaided eye. Lumps of amphibolite seem to swim in
the gneiss, and to fade off into it, as if melting before our eyes. The
gneiss becomes enriched with streaks of basic matter, in which biotite
begins to predominate over araphibole. Over a wide area it passes
into a banded biotite gneiss, in which the lens easily reveals the pale
brown garnets, derived from the amphibolite, still surrounded in most
cases by a retinue of biotite-flakes. A mile westward, in Stonepark,
down against the road to Lough Gill, the evidence of the intrusive
charact^ of the Leitrim gneiss is still more marked. The blocks of
dark included rock here consist of eclogite, containing a deep green
pyroxene, abundant garnet, magnetite, quartz, granular triclinic
felspar, and a variable but subordinate amount of hornblende and
brown mica. On the margins, in contact with the gneiss, biotite has
freely developed, so that some lumps, before they are broken across,
resemble mica-schist. Biotite similarly appears along the margins of
the aplitic veins sent off into the larger masses of eclogite from the
gneiss.

The gneiss of Stonepark is in consequence beautifully flecked with
dark absorption-products, grouped along the lines and surfaces of flow.
Under the microscope these black flecks prove to consist largely of
biotite and garnet (fig. 8), as in the slide prepared by the Geological
Survey from the Slishwood mass. Muscovite, however, is also present,
and here and there a prism of pale pyroxene remains. Isolated
garnets lie in the gneiss, which is also speckled by a number of
crystals of a spinelloid. This black mineral, by its red alteration-
products, seems to be ordinary magnetite, which is an abundant con-
stituent of the amphibolites.

I have similarly no hesitation in assigning a composite origin to a
rock styled '* homblende-omphacite-gneiss," No. 1966 of the Survey
collection, from the east end of the metamorphic area of the Bosses.
Another slide in the same collection, from the south of Lough Cooney,
and about one and a half miles south-west of Ballysadare, shows clearly
the derivation of gametiferous material from the amphibolite. The
label, "amphibolite penetrated by granite," indicates that a revision
of the area by the officers of the Survey would probably have led to
the conclusions expressed in the present paper. No suggestion, how-
ever, as to the relations of the granites to the amphibolites is given in
the " Guide to the Collections of Bocks, and Fossils," published in

Cole — The Intrusive Oneiss of Ihrerrill and Drumahair. 367

1895 (p. 52), though important remarks on the stmctnre of the Ox
HountainB appear on p. 42 of that valuable work. Here it is stated,
howeyer, that the amphibolites penetrate the gneiss in the region to
the west. This is contrary to my experience elsewhere. It is
of interest also to observe that the geological map of Sir E. Oiiffith,
edition of 1855, shows a patch of ^* gneiss passing into granite " on
Benbo, near Manorhamilton. This gives us no clue, however, as to
whether the granite was regarded as intrusive ; we may almost safely
presume that the metamorphic view was then adopted.

rig.3.

MicroBcopic seotion of gneiss, with inclusionis of light and dark mica,
garnet, and pyroxene, derived from the adjacent eclogites.
Stonepark, near Drumahair. x 18.

A variation on the prevalent type of composite gneiss is seen in
the strongly banded masses east of Castleore. A granulite with pale
pyroxene and biotite has here arisen, with obvious residual inclusion-
flecks containing both these minerals. I have not been able to trace
the original pyroxenic rock in this instance ; but colourless pyroxene
occurs in many of the amphibolites and eclogites of southern Donegal.
Such basic crystalline rocks arise as products of metamorphism from
very different materials, when these become invaded by and immersed
in a granite magma ; and the variety of mineral ccmstitution in the

B.I.A. PROC, VOL. XXjy.f SBC. B.] 2 S

368

Proceedings of the Royal Irish Academy,

Leitrim gneiss points to a corresponding variety in the rocks forming
the more ancient series traversed by it.

The characteristic blocks of homblendic rock in the gneiss are
well seen again at the north-east termination of the chain, in the
townland of PoUboy, H miles west of Manorhamilton (fig 4). The
basic inclusions, which do not seem to have attracted attention in this
area, are as striking as those of Olennagoolagh, near Ballydawley
Lough. They similarly weather away more rapidly than the
surrounding gneiss, leaving in places mere lenticular cavities. In
section they resemble dull and altered diorites, rich in hornblende.

Tig. 4.

Glaciated surface of banded gneiss, showing included blocks of
amphibolite (aphanite and diorite), which are often drawn out
parallel to the general flow. A dyke of later granite cuts the
whole. Pollboy, near Manorhamilton. From a photograph by
the author.

The gn^ss has here become much darkened by biotite, and is
traversed by later veins of coarse white granite, corresponding with
those so frequently seen in Donegal.

Returning now to the instances of amphibolite originally observed
by Mr, Hardman, we have no difficulty in recognising them as
inclusions in the gneiss.

The typical gneiss in Olennagoolagh is banded and rich in biotite ;
and garnet and green pyroxene occur in the micaceous bands. The
rock is obviously darkened in the neighbourhood of the inclusions,

Cole — The Intrmive Oneias of Tirervill and Drumahair. 369

which oonsiBt of amphibolite, rather poor in garnet. Neither in the
Survey collection nor my own have I found an eclogite from this
area ; yet the handsome occurrences of eclogite in Stonepark make it
probahle that some such rock has furnished the green pyroxene to the
gneiss of Olennagoolagh. The marked handing of the gneiss is again
clearly associated with an abundance of inclusions ; and we have now
sufficient evidence from various parts of TirerriU and Drumahair to
show that this is a normal characteristic.

The gneiss, then, of the ridge on the south side of Lough Oill
Tepeats the features of the granite floor of Donegal, and was probably
formed during the same epoch of intrusion. It has certainly absorbed
a Dalradian series on its margins ; and one is tempted to regard it as
of the same age as the Caledonian earth-movements. Yet we must
remember that the Gbtlandian (Upper Silurian) conglomerates of
Lough Nafooey, in County Galway, contain pebbles of granite,
associated witii quartzite, and prove that an earlier intrusion of
granite had taken place in these western highlands. It is always
possible that the composite rocks formed in Archaean times may have
been brought to the surface at a far later epoch, and that they
then underwent a certain amount of mechanical deformation. While
I do not think that such deformation is a prominent feature in
Tirerrill and Drumahair, it has been sufficient in other cases to lead to
a misapprehension as to the origin of the banding and flow-structure
throughout the gneissic mass.^

In conclusion, now that the composite origin of banded gneiss is
becoming a matter of general acceptance by geologists, it is well to
refer back to the views of M. Michel L6vy, summarised by him in
1887.' Sixteen years ago M« L6vy emphasised the similarity between
more recent ribboned gneisses, formed by parallel intrusions of granite
into metamorphosed sediments, and the ancient yet complex masses,
which were commonly regarded as the primitive crust. Sederholm
in Finland, working on the earlier masses, and Duparc and Mrazec,
dealing with far more modem intrusions on Mont Blanc, may be
cited among those who have verified the master's generalisations.
Similar views have even found their way into the text-books;
and now that Mr. A. Harker^ has given us a convincing study of a

1 (Compare op. eit,, Proo. B.I. Acad., vol. zziv, sect. B., pp. 220 and 221.

* <' 8ur Torigine des temixiB oristallinB primitifs,*' Bull. Soc. gdol. de France,
3m6. 86r., t. xvi. (1887-8), pp. 102-113.

3 « The Overthrurt Tonidonian Books of the lale of Bum, and the Associated
Gneisses,*' Quart. Joum. Geol. Soc. London, vol. lix. (1903), pp. 207-215.

2ir2

370 Proceedings of the Royal Irish Academy.

Cainozoic example in the Isle of Rami we may be sure that the
theory of the formation of banded gneisses by admixture [will receiye
adequate recognition throughout the British Isles.

As will have been seen, I am much indebted to the officers of the
Geological Survey for permission to examine the specimens and rock-
slices in their collection. Mr. A. M'Henry has been especially generous
in discussing material gathered by himself from the Ox Mountains,
which will form the basis of a report to be presented by him shortly
to the Academy ; and I am glad to think that our views are likely
to be in complete harmony as to the intrusive nature of the gneiss.

[ 371 ]

XI,

REPOKT ON THE OX MOUNTAIN ROCKS AND THEIR
PROBABLE CONTINUATION FROM GALWAY AND
MAYO INTO DONEGAL, TYRONE, AND LONDON-
DERRY.

By ALEX. M'HBNRY, M.R.I.A.

Read Junb 22, 1903.

Ths range of hills of which Slieve Gamph and the Ox Mountains torm
the main portion, begins a little to the north-west of Castlebar, and
continues in a somewhat sinuous north-east course past Fozford,
Coolaney, and Lough Gtill to Manorhamilton, where the older rocks
forming it sink beneath the Carboniferous strata of the plain, which
formation bounds the range on both sides along its whole course of
sixty-fiye miles.

The several great divisions of the metamorphosed sedimentary
rocks of the range are similar in every respect to those lying to the
west in Gal way and Mayo, and are, in fact, a continuation of them.
These rocks were originally considered by the Geological Survey to be
mainly metamorphosed Lower Silurian strata;^ and this opinion I
believe to be correct ; with the exception that in places, such as south
of Clew Bay; in Croagh Patrick Mountain, rocks of Upper Silurian age
must be included in the metamorphosed group, as was proved by my
colleague, Mr. Kilroe.'

The main divisions of this metamorphosed sedimentary series that
are so persistentiy and well recognised in other districts in Ireland
are well marked at several places in the south-west end of the range,
to the north of CasUebar, and to the east and north-east of Fozford,
in the vicinity of Loagh Talt.

> Explanfttoxy Memoin of the Geological Survey of Ireland, Sheets 78, 74,
£3, and 84 (1876), and Sheets 93 and 94 (1878).

' Annoal Bepott of the Geological Survey for the year 1896, p. 60.

372

Proceedings of the Royal Irish Academy.

These divisions consist of quartzite at the top of the series, then a
pebbly or conglomerate zone ('* Boulder Bed "), which passes down
into a limestone that is sometimes also pebbly. Associated with th&
limestone, and mostly below it, comes a series of black shales and
grits, the shales being sometimes graphitic, and the lowest of all a
great series of pebbly grits, with j>ccasional thin black shale or schist
beds, which in the sonth-westem extension of the series north of
Castlebar, around Westport, and along the south shore of Clew Bay,
have been locally designated by myself and my colleague, Mr. Kilroe,
as the ** Westport grits."

Fig.I

15 FEET

Sheared pebbly grit (quartzite) cut by early basic (B.) and later acid granulite
rocks (C), the whole series haying been subjected subsequently to still
later shearing.

West side of Denyclare Lake, Gonnemara.

This metamorphosed sedimentary series has been eztensiyely
inyaded by a complex of basic and acid igneous materials, and shows
great alteration both by contact with the igneous intrusions, and the
later earth stresses that probably accompanied, and evidently succeeded,
those intrusions. In many places the limestone, where it has been
invaded and enveloped by tiie acid igneous masses, is converted into a
whitish saecharoid marble with garnets developed in it, exactly like

M'Henry — Report on the Ox Mountain Rocks. 378

the similarly circumstanced limestones of Donegal and Oalway, or into
a serpentinoas rock when the igneous masses of contact are basie.

Examples of the former class of metamorphism occurs in the
vicinity of Lough Talt, which lies in a north-west cross glen dividing
the Slieve Gamph from the Ox Mountain portion of the range, also at
one or two places to the north-west of Castlebar, while along the
north lower slope of Croagh Patrick Hill the serpei^tinous variety is
developed, as well as to the north of Castlebar, and elsewhere at
many points in Connemara. To the north of Castlebar, and between

Fig;2

6 FEET

Induaions of earlj Bheored basic rook (B.) in later sheared add grantdite (C.)«
N. of BaUydawley Lake, 4 miles S. of Sligo.

it and Westport, the lowest pebbly grits are in strong evidence, and
show the passage stages and conditions from an original conglomeratic
grit or sandstone into a quartzite ; the metamorphism may be due,
however, to dynamic metamorphism, rather than to actual contact
with igneous masses. The quartz pebbles in this quartzose grit or
conglomerate can be seen to have been crushed and drawn out in a
remarkable manner, sometimes into riband-like forms, and even the
drawn out pebbles, puckered and contorted by subsequent movement,
and folding in the rock mass.

374

Proceedings of the Royal Irish Academy.

Interbedded with those pebbly grits, there are black shales, which
but for the amount of deformation, through moToment in the rock
mass that they have undergone, are closely similar to Ordoyician
shales, from which fossils of Llandeilo types have been obtained.

Of the two varieties of igneous intrusions, the basic is the earlier,
abundant and conclusive evidence existing at many points to prove
this fact (fig. 1). After the basic came an acid intrusion which
affected both the basic igneous and the sedimentary rocks to a
considerable extent, large masses of both these (basic rocks and
sediments) in a highly mineralized condition being included in the

Fig. 3.

(C.) Acid granulite enveloping sediments (A.) and early basic igneous rocks (B.),
with later imcrushed granite (O.)*

Near Maam Cross, Connemanu

granulitic or second igneous intrusion (fig. 2). Subsequent to the
period of this acid intrusion, intense earth stresses took place, which
resulted in a milling out, shearing, and banding of all three varieties
of rock, thus forming those banded gneissose rocks and sheared or
bedded-like rock-masses as they now exist, and which were considered
by some authorities to be metamorphosed early Palaeozoic sediments,
and by others to belong to a more ancient or Archaean period.

M^Hbnry — Report on the Ox Mountain Rocks,

375

Subsequent to the intrusion of the basic rock, and prior to the
&r8t add one, a certain amount of moyement in the rock-masses took
place, as is shown by the sheared and banded condition of the
inclusions of the earlier rock in the acid granulites (fig. 3). A
still later intrusion of granitic material took place, the masses and
Teins of which only occasionally show signs of deformation from
«arth-stress6s (fig. 4).

Regarding the probable age of the two earlier intrusions of basic
and acid materials, they are at least pre-Old Red Sandstone, as the

Fig. 4f.

20 FEET

Early basic igneouB rook (B.) and later acid graniilites (C.) invading the sedi-
mentary limestone zone, all three groups having been subsequently sheared
together, and later uncrushed granite vein (G.).

Five miles E. of Clifden, Connemara.

massive conglomerates of this latter formation are largely composed of
the detritus of all the yarieties of igneous rocks and the associated
sediments into which they intrude. Besides, they are seen to rest
truly unconformably on them at many places. Therefore, if it be
acknowledged that the altered sedimentary rocks are of Ordoyician
and eatly Silurian age, it is most likely that the period of this igneous
complex belongs to early Devonian time. In my opinion they are the

376

Proceedings of the Royal Irish Academy,

equivaleiLts of, and belong to the same period as, the granites and
associated earlier basic igneous rocks of Leinster, which are admittedly
of Devonian age.

The relations of the Old Red Sandstone and metamorphic series
can be seen in the neighbourhood of Windy Gap, to the north of
Castlebar. Here also can be noticed the effects of the post-Old Bed
Sandstone movement, or overthrusting of the later Old Red Sandstone
on to the older metamorphosed rocks, and the formation between the
two series, at this place, of a pseudo-conglomerate or zone of fault
and crush breccia, 40 feet and more in thickness, the breccia being
made up of the broken-up materials of both series, re-cemented into a
compact rock mass (fig. 5).

Fig. 5.

OverthruBtof Old Bed Sandstone on to metamorphosed grit or quartzite (Q.)*
Windj Gap, N. of Castlebar, Co. Mayo.

Proceeding north-eastward along the Ox Mountain range, the
included sediments become scarcer till they almost entirely disappear
and give place to the igneous rocks, as in the region north of Coolaney,
south of Lough Gill, and west and north of Manorhamilton. A&
previously mentioned, the simulation of bedding seen in those rocks to
the south of Sligo and elsewhere is not due to sedimentation or
deposition, as originally supposed, but, on the contrary, is entirely due
to movement and shearing of the rock masses (fig. 6).

It may be mentioned that a small ridge of similar deformed rocks
crops out from beneath the Carboniferous strata a few miles to the
north-west of Sligo, in the Rosses peninsula.

The general characters of the Ox Mountain rocks are again re-
peated in south Donegal, in the region to the north of Pettigo, and

M'Hbnrt — Meport on the Ox Mountain Rocks.

377

about Lough Derg. The intnisiye rocks of this area have been
described as Archaean on the late editions of the Geological Survey
Maps of that district. At one time I believed them to belong to this
group; but I now consider them to be contemporaneous with the
rocks of the Ox Mountain chain, %,$. early Devonian.

Continuing north and north-east into Donegal and Tyrone, and on
to Londonderry, we have repetitions of the Ox Mountain series, both
in tbe character and the conditions of the rocks, t. e, quartzite,
'< Boulder Bed," limestone zone, black schist or slate, and the
lowest pebbly grit, and with the basic and acid intrusions in the same
order of sequence.

-20 FEET

Contorted and crumpled shear rtracture in complex of basic and acid igneous

rocks, simulating bedding.

Four nulee W. of Coolaney, Co. Sligo.

In Fanad area, Donegal, to the north of Xnockalla Mountain, we
find the metamorphosed sediments (quartzite) overthrust on to the
Old Bed Sandstone, with a zone of crust or overthrust breccia along
the line of moyement, the direction of the overthrust being to the
north-west, here again proving the occurrence of great earth-stress
in post-Old Bed Sandstone times.

As in the case north of Castlebar, the intrusions of igneous rocks
were prior to the folding and shearing, and both sediments and

378 Proceedings of the Royal Irish Academy.

igneous rocks were folded and sheared out together before the disposi-
tion of the Old Bed Sandstone, as is shown by the contained stones of
the former in the latter.

Throughout Donegal, Tyrone, and Londonderry similar conditions
and varieties of rocks are found to exist, namely, a lower pebbly grit
series, with their black shale or schist beds, probably of Llandeilo age,
then a passage upwards into black shales or slates greatly deformed
by movement, and sometimes pyritous, above them the limestone
2one, probably of Bala age. On top of the limestone that remarkable
and important deposit, the '' Boulder Bed," is always found to exist.
This '' Boulder Bed " has been identified all over Donegal, Mayo, and
€kdway in varying thickness, and from its general appearance and
characteristics it is probably an older Palaeozoic Glacial Boulder Clay
Deposit which marks a possible break between the Ordovician and
Silurian series. The boulders in it are almost entirely of an
unf oliated granite, unlike any granitic rock at present seen at the sur-
face in Ireland. They are angular and sub-angular, and occasionally
rounded in form, and but thinly distributed in a fine-grained matrix ;
occasionally blocks measuring 3^ feet across can be seen, as in the
Fanad area, Donegal.

The limestone, where seen, always underlies the '' Boulder Bed,''
and is occasionally quite pebbly, as at Camdonagh and other places in
Donegal, as well as in Counties Oalway and Mayo. The known
fossiliferous Bala Limestone rocks in the west of Lreland, at Eossroe
and Tourmakeady, at Caherconree, in Kerry, and at Portrane, county
Dublin, and elsewhere in Ireland, are occasionally found to be pebbly.
It appears to me, therefore, to be a point in favour of the contempo-
raneity of those metamorphosed limestones of the west and north-west,
and the Bala Limestone of the west, south, and east, and one well
worthy of further consideration.

[ 379 ]

XII.

THE SYNTHESIS OF GLYCOSIDES: SOME DEKIYATIVE&

OF ARABINOSE.

By HUGH RYAN, M.A., D.Sc, F.R.TJ.I., and
GEORGE EBRILL, B.A., Catholic University School of Medicine,.

Dublin.

Bead Junb 22, 1903.

The first experiments recorded in the chemical journals for the syn-
thetical preparation of glycosides were made by Schutzenberger.^
From triacetylglucose and the sodium or lead derivatiyeB of saligenin
he obtained an amorphous compound, which could be hydrolysed by
dilute sulphuric acid into glucose and saliretin, but which was not
identical with salicin. In a similar manner a substance, which very
closely resembled rhamnegin, was obtained from rhamnetin.

By the interaction of the acetochloroglucose discovered by Colley,^
and the alkali salts of the phenols, Michael' obtained helicin, methyl-
arbutin and the glucosides of phenol, eugenol, and guaiacol. Using
MichaeFs method, the glucosides of thymol and a-naphthol were
obtained by Drouin,* and by a slight modification one of us obtained
the glucosides of the three cresols, i^-naphthol, and carvacrol, with
the galactoside of a-naphthol, and still later the tetracetyl derivatives
of the three cresyl glucosides and of )3-j9-naphthyl glucoside.

By the condensation of helicin in weak alkaline solution with
acetic aldehyde, Tiemann and Kees^ succeeded in obtaining o-cumaric-
aldehyde glucoside. Compounds of the aldehydes and ketones with
grape-sugar were prepared by Hugo Schifi " by the interaction of their
components in acetic acid solution.

* Annalen der Chemie und Pbysik, dz., p. 96.
3 Ann. Chim. Phjs., 1870, iv., 21, p. 363.

^ Comptes Rendus, Ixxziz., P- 365 ; Amer. Chem. Joiizn., vi., p. 366.
« Bull. Soc. Chim., czi., 13, p. 6.

* Ryan, Joum. Chem. Soo., 1899, p. 1064 ; Rjan and Hills, Joum Chem.
Soc., 1901, p. 704.

« Proc. Royal Dublin Soc, 1901, Vol. ix. (n.s.), iv., p. 608.
' B«richte, xyiii. (1886), pp. 1966, 3481.

* Ann. ccxxiv., p. 19*

^80 Proceedings of the Royal Irish Academy*

The substances obtained in this way were very hygroscopic and
•decomposed even by solution in water.

The most convenient and fruitful method for the synthesis of
glycosides was discovered by Emil Fischer.^ In the first instance, he
condensed the sugar with an alcohol or mercaptan, in presence of a
large excess of cold, strong hydrochloric acid. The later and more
convenient method consisted in heating the sugar and alcohol with a
small quantity of hydrochloric acid.

In this way Fischer and his pupils obtained the glucosides of
methyl, ethyl, propyl, and benzyl alcohols, of glycol and of glycerine,
of dimethyl acetal, of ethyl, amyl, benzyl, ethylene, and trimethylene
mercaptans ; the galactosides of methyl and ethyl alcohols, of ethyl,
benzyl, and ethylene mercaptans ; the mannosides of methyl alcohol,
of ethyl, and ethylene mercaptans ; the fructoside of methyl alcohol ;
the glucoheptosides of methyl alcohol and ethyl mercaptan; the
rhamnosides of methyl and ethyl alcohols, of ethyl, benzyl and
ethylene mercaptans; the arabinosides of methyl, ethyl, and benzyl
alcohols, of ethyl, benzyl, and trimethylene mercaptans ; the zylosides
of methyl alcohol and the sorboside of methyl alcohol.

Although the glycosides of the monatomic phenols cannot be
obtained by Fischer's method,' those derived from the polyatomic
phenols resorcine, pyrocatechine, pyrogallol, and phloroglucine with
arabinose have been so obtained. In a similar manner the glucosides
of resorcine, ordne, and phloroglucine with the galactoside, fructoside,
and mannoside of phloroglucine were synthesised.'

Another method for synthesising glycosides was discovered by
Hill,^ who found that zymohydrolysis is a reversible operation. He
obtained maltose from glucose by the action of maltase; but later
experiments by Emmerling* seem to show that the disaccharide
obtained by Hill's method was isomaltose. Similarly by the action
of the kephyr lactase on a mixture of galactose and glucose, Emil
Fischer and Armstrong' obtained galactosido-glucose (isolactose).

The discovery of a method of obtaining a crystalline mother sub*

1 Berichte, zxvi. (1893), pp. 2400, 2928 ; xzvii. (1894), pp. 674, 2483, 2985 ;
ixviii. (1896), p. 1146.

3 Emil Fischer and Jennings, Berichte, xxvii., 1894, p. 1358.

3 Berichte, xxviii., 1895, p. 24.

^ Joum. Cbem. See., Izziii, 1898, p. 634.

^ Berichte, zxziy., 1901, p. 600.

« Sitz. der £. Akad. der Wissenscb., Berlin, 1901, xH., p. 123.

Btan and Ebrill — The Syntheais of Oli/cosides. 381

stance (acetobromoglucose) by Koenigs and Enorr^ was an important
advance in the method of Bynthesising glycosides. The pure, well-
crystallized acetobromoglucose was converted into )3-pentaoetyl
glucose, )5-methyl, /3-ethyl, jS-phenyl, )3-)3-naphthyl, and jS-carva-
cryl-glucosides.

A still further advance was made by the discovery of Fischer and
Armstrong,' that anhydrous liquid halogen acids react with a- and fi^
pentacetyl glucoses to form well-crystallized a- and ^-acetochloro and
acetobromoglucoses. In this way the acetochlorogalactose, obtained
as a syrup by Colley's method," was isolated in the pure condition aa
a well-crystallized compound, and converted into )3-phenyl galacto-
side. From the a-acetochloroglucose they obtained a-dkyl-glucoddeB,
and from j9-acetochloroglucose the corresponding )3-glucosides. The
failure of Fischer and Armstrong to convert acetohalogen pentoses into
phenol derivatives is probably due to the ease with which the a-com-
pound changes into the j9-derivative in the presence of dilute alkalL*

By the action of phosphorous pentachloride and aluminium
chloride on the chloroform solution of a-pentacetyl-glucose and a-pen-
tacetylagalactose, crystallized a-acetochlorohezoses were obtained by
Skraup and Kremann.' Acetochlorolactose was obtained by Bodart,*
by the action of hydrochloric acid gas on dry lactose, suspended in
cold acetic anhydride, which, with its isomeride, was also obtained
by Fischer and Armstrong. The latter chemists also converted the
analogous acetochloromaltose into ^-methyl-maltoside.

The ^-phenyl«maltoside obtained by flscher and Armstrong'' from
)3-acetobromomalto8e was hydrolysed by emulsine to maltose and
phenol. Its behaviour towards the enzyme is different from that of
amygdalin, which is decomposed, on hydrolysis by emulsine into
glucose, benzaldehyde, and hydrocyanic acid.

Although halogen derivatives have been most largely employed
for the synthesis of glycosides, it is interesting to note that nitro-
derivatives have also been successfully used by Xoenigs and Knoir.*

1 SiU. Bayr. Akad. der WinseiiBch., 1900, p. 103.
' Sitz. der K. Akad. der WiBseiisch., Berlin, 1901, ziii., p. 316.
3 Byan, Joum. Chem. Soc., 1899, p. 1057 ; Proc. Boy. DuU. Soc., toL iz.
(k.8.)» p. 606.

* Bericbte, zzziv., 1901, p. 2886.

^ Monatsch. f. Chem., xziL, p. 376.

• r. loe. eit.

^ Berichte, zxxr., 1902, p. 3168.
" Berichte, zzziv., 1901, p. 967.

382 Proceedin(/8 of the Royal Irish Academy.

Acetonitroglucose and acetonitrogalactose have been converted into
alkyl hexosides. From acetonitromaltose heptacetyl-jS-methyl malto-
side was similarly obtained.

It has been shown by Eyan and Mills^ that, by the direct action
of acetyl chloride on arabinose, a well-crystallized acetochloroarabinose
can be obtained. Chayanne, by the same method, afterwards' re-dis-
covered the substance, obtained the corresponding acetobromoarabi-
nose, and converted it into a crystallized tetracetylarabinose.

From acetochloroarabinose, as mother-substance, we have obtained
the arabinosides of carvacrol, ortho cresol, j9-naphthol, and methyl
alcohol. The new glycosides resemble the corresponding phenolic
hexosides in their appearance and behaviour.

Preparation of Acetochloroarabinose.

The method of obtaining this compound has been briefly described
in a previous paper.' In further preparations the method which was
found most convenient was to allow acetyl chloride (4 mols.) to act
on dry, powdered arabinose (1 mol.) in a small flask (fitted with a
calcium chloride tube to prevent the entrance of moisture), until the
mixture had solidified to a crystalline magma. Dry chlorofonn was
then added, and the action allowed to go on until complete solution
was effected.

The chlorofonn solution was shaken in a funnel, washed first with
water, then with sodium carbonate, separated, passed through a dry
filter, dried with anhydrous sodium sulphate, and the chloroform dis-
tilled off in vacuo.

The yield from five grams of arabinose was generally seven and
a half grams of crystalline acetochloroarabinose, which was sufficiently
pure for conversion into phenolic glycosides. The properties of the
substance given by Chavanne {loc. dt.) are almost identical with those
previously given by Eyan and Mills {loc, cit.).

The acetochloroarabinose, which is a well-crystallized compound,
and comparatively stable in the air, is a more convenient mother-
substance for the preparation of glycosides than the impure syrupy
acetochloroglucose and acetochlorogalactose previously employed.

^ Joum. Chem. Soc., 1901, p. 706.

3 Comptea Rendos, czzziv., 1902, p. 661.

' Byan and Mills, Joum. Chem Soc, 1901, p. 706.

Hyan and Ebrill — The Synthesis of Olycosides. '68S

Owing to its stability, it is even more convenient than tlie crystallized
acetobromoglucose of Eoenigs and Knorr ; and the method of prepa-
ration is simpler than that of Fischer and Armstrong for the aceto-
halogenhezoses.

Action of Methyl Alcohol on Aestoehhroarabinosc.

Acetochloroarabinose (1*6 gram) was dissolved in warm methyl
alcohol (50 c.c.)} and the mixture was allowed to remain at the tem-
perature of the laboratory for four days. Silver carbonate (1*6 gram)
was added, and the precipitate filtered. After evaporating the filtrate
on the water-bath| the residue was dissolved in a little methyl alcohol,
refiltered, and let evaporate spontaneously in a vacuum desiccator
over calcium chloride. Leaf-like aggregates of crystals separated.
They were free from chlorine, and did not reduce Fehling's solution
before hydrolysis. After repeated recrystallizations from hot methyl
alcohol, the crystalsy when dried at 105^0., became soft at 159^ C, and
melted at 16&-168''G.

Methyl arabinoside obtained by Fischer's method becomes soft at
165^ C, and melts at 169-176^ C. It crystallizes in needles or leaf-
like aggregates, and does not reduce Fehling's solution.^

Action of an Alkaline Solution of Carvacrol on Acetochloroarabinose,

Slightly more than equivalent quantities of potash and carvacrol,
dissolved in absolute alcohol, were added to a solution of acetochloro-
arabinose (7 grams) in absolute alcohol. The mixture was allowed to
remain at the temperature of the laboratory for a few days. A white
solid separated, and the solution smelt of acetic ester. The filtrate
from the potassium chloride was evaporated on the water-bath, and
the residue dissolved in water. The aqueous solution was evaporated
a few times with addition of water, until the odour of carvacrol had
disappeared.

The residue crystallized on cooling. The reaction had proceeded
thus: —

I 0 ,

CHa OAc CH (CHOAo). CH + G6H3CH3C3H7 OE + 3CsH60H

= CHjOH (5H(CH0H)a'CH + KCl + SCHs COOCt H»

O-CeHsCHsGsH?

^ Berichte, xxyi., 1893, p. 2400.

&.X.A. FBOC, VOL. XZIT., SEC. B.] 2 /

884 Proceedings of the Royal Irish Acadewy.

The carvacryl-arabinoside was recrystallized from boiling water,
and dried at 100° C. for analysis : —

0-1229 g. sbst, : 0-2869 g. CO, : 0-0897 g. H,0.
C 63-66, H 8-19
Ci6 HtjOfi requires C 63-78, H 7-87.

Carvaeryl arahinoside (CcHjOi • 0 • CeHs • CHs • CjHt) crystal-
lizes from water in long needles, melting, wben dry, at 119-120^0.
It dissolves in alcohol, ether, acetone, and chloroform. It is insoluble
in carbon bisulphide and toluene, and sparingly soluble in cold water »
but readily in hot water. The pure substance does not reduce
Fehling's solution. When heated for a short time with dilute sul-
phuric acid it is hydrolysed to carvacrol and arabinose. Carvaeryl
arabinoside difPers from carvaeryl glucoside in its not being more
soluble in potash than in water.

Conversion of Aeetochloroarabinose into p-naphthyl-arahinoside,

Acetochloroarabinose (3 grams), dissolved in absolute alcohol, was
slowly added to a solution of 0-6 gram potassium hydroxide and
1-5 gram j9-naphthol, also dissolved in absolute alcohol. The mix-
ture after a few minutes smelt of acetic ester, and quickly became
turbid from the separation of a white solid (potassium chloride).
After remaining at the temperature of the laboratory for one day it
was heated on the water-bath for a short time, and again allowed to
remain for three days at the ordinary temperature. The yellow
filtrate from the potassium chloride was heated on the water-bath,
under the reflux condenser, for half an hour, and the alcohol then
distilled off. The cold residue became solid on the addition of a little
water. The product was dried on clay, washed with chloroform till
colourless, recrystallized from boiling absolute alcohol, and dried at
105° C. for analysis : —

0-1072 g. sbst. : 0-2562 g. COj and 0-0562 g. H,0.

C 65-18, H5-82.
CifiHieO, requires C 65-21, H 5-8.

fi-naphthyl arabinoside crystallizes from absolute alcohol in long-
branching, grouped needles, which are visible and multicoloured
between crossed nicols. It jdissolves in cold alcohol and acetic ester.
The crystals are scarcely soluble in benzene, chloroform, ether, water,

Byan and Ebrill— j^i^ Synthesis of Glycosides. 385

or petroleum ether, but yery readily soluble in hot alcohol. They
melt, when dry, at 176-177^ G. The arabinoside does not reduce
Fehling's solution before hydrolysis, but does so readily after hydro-
lysis by boiling with dilute sulphuric acid for a short time.

Conversion of Aeeioehioroarabinose into Orthooreeyl Arabinoeide,

Equimolecular quantities of orthocresol, potassium hydroxide, and
acetochloroarabinose were mixed together in alcoholic solution. The
copious precipitate which first formed was redissolved on boiling the
mixture. After remaining at the ordinary temperature for a few
days, the filtrate from the precipitated potassium chloride was allowed
to eyaporate spontaneously, and the residual oil was dissolved in
boiling water. On concentrating to a small bulk, and allowing it to
stand for seyeral days, beautiful rosettes, consisting of needle-shaped
crystals, were obtained, which were dried on clay and recrystallized
from water. When air-dried at 100° C. it melted at 124° C, and gave
an analysis : —

0-1548 g. sbst. : 0-3364 g. C0>, 0-0978 H,0,

C 59-44, H 702,
Cu Hie Oft requires C 59-95, H 6-7.

Orihoereiyl araHnosids is soluble in cold water, and very readily
soluble in hot water. It is insoluble in ether and carbon disulphide,
scarcely soluble in chloroform or benzene, and easily soluble in alcohol
or acetone, from which it separates in beautiful branching needles.
The arabinoside does not reduce Fehling's solution before, but readily
after, hydrolysis by hot, dilute sulphuric acid. The hydrolysed solu-
tion smelt of cresol.

Nomenelaiurs employed.

It has been customary, up to the present, to call a substance which
can be hydrolysed by an enzyme or a dilute acid to two or more
bodies, one of which is a reducing sugar — a glueoeide. The oldest and
best-known members of the series are deriyatiyes of glucose, and, in
these cases, the term is a correct one. When, howeyer, similar
derivatives were obtained from another hexose, such as galactose,
they should, strictiy speaking, have been termed yalaetosidea.

In most instances, this system has been adopted ; but a difficulty is
still felt in finding a suitable name for the whole series.

386 Proceedings of the Royal Irish Academy.

The term hexoside has been applied by me to all such derivatiyeB
of the hezoses (glucose, galactose, fractose, mannose, &c.). Analogous
deriyatiyes of the pentoses, tetroses, &c., may be called prntosides,
tetrosideSj fto* In general, any such deriyatiye of a polyose may be
called tLpolyoside, Although the latter tenn is probably the best for
the whole series, I haye, in the present paper, used the word glycoside^
suggested for the same purpose by yan Rijn.

Proc. R lAcad. Vol.XXIY, Saction B.

Aranel<U.

PhalBagidit.
ActfinidJi.

Xiphosnra.

Pycnogonida.

Suggested Kelalionships between the Arthropod Classc
their principal Orders.

Plaia VI.

Diptera. HymfnopUra,.

geroldea,.
CHILOPODA.
oTriddcT S. Geophiloitou

Scolopendroulea.

-eioBDeioidea.

^s and

aW«A&San% }itii.

[ 387 ],

XIII.

A LIST OF IRISH HEPATIC-H.
Bt DAVID McARDLB.

(kEPOBT FBOIC THE FAXTITA AKJ) FLOSA. COIOIITTXB.)

. Bead Juxb 22, 1903. Publiohed Januabt 28, 1904.
iNTBOBUCnON.

This paper is an attempt to give a full and reliable list of the
HepaticsB of Ireland, as they are known at the present time. It is
intended to form Part II. of ^* Cybele Hibemica," and is based on
exactly the same lines. Since the late Dr. D.Hoore's death in 1879, 1
have continued to study the subject which he first taught me, knowing
that his valuable Report on Irish HepaticsB, which he read before the
Royal Irish Academy in 1876, was preliminary to a more exhaustive
work. With financial help from the Fauna and Plora Committee of
the Academy, I have been enabled to make research in many
counties. The results I have from time to time laid before the
Academy, and for their help I offer my best thanks.

The Irish Hepaticse have been studied with great success by the
earlier botanists, notably by Dr. Taylor, of Eenmare, in Kerry, who
published the result of his researches in Port II. of Mackay's ** Flora
Hiberoica," in which seventy-five species are enumerated under the
genus Jungermania^ besides Marehantiaeea and Anihocerotaeea^ which
include eight species, making eighty-three in all.

Miss Hutchins, of Bantry, about the same period was collecting
and studying Hepaticsd in Co« Cork, with rare discriminating
power. Most of the plants she gathered were sent to Sir William
Hooker ; and one has only to turn over the pages of Hooker's grand
work on the British fyngermanud to find her name more or less
connected with the discovery of every rare Irish plant.

Dr. Thomas Power's '* Contributions towards the Fauna and Flora
of Cork," published in 1844, includes fifty species.

The late. Mr* Isaac Carroll contributed largely to our knowledge
of theoe plants in Co. Coik and elsewhere. In 1863, the late Dr.
Cdrrington, of Manchester, published his '< Gleanings among the

B.I.A. PSOO., yOL. XXIT., SBC. B.] 2 IT

388 Proceedings of the Royal Irish Academy.

Irish Ciyptogainfi " ; 110 species are enumerated, and roany varieties
collected by him when on a visit to Ireland of eleven weeks' duration,
which he spent in Kerry and Cork. He also includes in his list some
stations for rare HepaticfiB, discovered by Dr. D. Hoore ; and some
species growing in the neighbourhood of Cork by Mr. I. Carroll,
and Hr. W. Wilson, of Warrington ; about the same time Hr. Mitten
made interesting discoveries in Co. Kerry, notably on Brandon.

In 1878, at the invitation of Dr. D. Moore, the late Professor
Lindberg, of the University of Helsingfors, paid a visit to this
country, and spent the months of June and July collecting Liver-
worts, in con^pany with Dr. Moore, in Co. Kerry. Brandon, and
a large part of the Dingle peninsula, and Killamey, got a close
examination. They also collected in many parts of Co. Wicklow
and Co. Dublin. The result of their trip was a collection of eighty-
seven ppecies of Hepatice, an account of which Professor Lindberg
published in the ''Acta Societatis Scientiarum Fennica)," vol. x.,
under the heading, '' HepaticflB in Hibemia,men8e Julii, 1873, lecte "
This was an important contribution, and included several new species.

In the northern counties, the subject has been by no means
neglected, but has occupied the attention of keen observers since the
days of John Templeton, A.L.S., of Belfast, one of the most acute
naturalists of his time. Mr. Samuel Alexander Stewart, of the same
city, has studied the subject with his characteristic care. The
results of his work are published in the " Flora of the N.-E. of
Ireland," and in various Reports. In 1885, in company with Mr.
Holt, of Manchester, he visited Killamey, in search of Mosses and
Hepaticfld. The result of their trip was the discovery of several
species of Hepatic« new to science. He has also been ably assisted
by the. Kev. C. H. Waddell, of Saintfteld, Co. Down, and the Rev.
Canon Lett, of Loughbrickland, Co. Down. The Report of the latter,
which was read before the R. I. Academy in 1889, included the
Mosses, Hepaticae, and Lichens of the Moume Mountain district.
Sixty-four species of Hepatic® are enumerated, an important list of
plants collected on a wide area, which includes roughly 660 square
miles. The Rev. C. H. Waddell published a valuable paper in the
*' Journal jof Botany," 1893, on the distribution of Zefeunsa in
Ireland.

Dr. D. Moore's Report on Irish Hepatic® in 1876 included all
previous papers and work by collectors in Ireland up to that date.
As to his own investigations he writes : ** The Irish habitats may be
relied upon, as I have eoUeoted nearly eveiy one of the plants with

MgAbdlb— ^ List of Irish ffepaticm. 389

mj own hands at some time or other during the last forty yearB,
having for this purpose travelled over a very large portion of Ireland,
from east to west, and from north to south, and from sea-level to the
tops of the highest mountains. The chief merits of this Beport may
indeed be considered to consist in its giving as full an account as I
am able to render of the Irish Hepatice, and of their geographical
distribution in Ireland; 137 species of them are enumerated." It
wil] be seen from the following list that I have endeavoured to
follow closely in the footsteps of this great bryologist, and have
availed myself of every advantage offered to further the object. I
enumerate 172 species and sixty- three varieties; some of the latter
have been raised to the rank of species by authors, and they are all
more or less of botanical value. To Mr. W. H. Pearson, of Manchester,
and Mr. M. B. Slater, of Malton, Yorkshire, 1 offer my best thanks for
their help in matters of doubt when investigating critical species.

Physical Fjeatubes.

The physical features of Ireland are favourable for the growth of
HepaticsB. A large area is occupied by peat bogs both lowland and
mountain ; and large lakes lie in the central plain, with smaller and
more numerous ones towards the west — as in Conn emara. West Mayo,
and Kerry. In the north-east. Lough Neagh covers an area of 153
square miles, and is the largest fresh-water surface in the British
Islands. The Shannon is the largest river; it flows for 214 miles,
■and creates in its course Lough Bee and Lough Derg. The eastern
part uf the central plain is drained by the Bivers Boyne and Liffey,
the south-eastern part by the Bivers Suir, Barrow, and Nore ; while
the waters of the north-eastern part are collected into Lough Neagh,
chiefly by the Blackwater, and from thence discharged into the sea
by the Lower Bann. The rivers outside the central plain are short ;
the principal ones are the Erne, flowing north-west ; the Foyle and
Bann to the north ; the Slaney to the south-east ; and the JBandon,
Lee, and Blackwater flowing through Co. Cork. The bays and
marine loughs are numerous and deep, penetrating inland for a con-
siderable distance, as Lough Swilly on the north coast, Bantry Bay
in the south-west, &c.

The principal mountain ranges are near the coast. The highest
Irish mountain is Carrantuohill, 3414 feet, which is part of Magilli-
ouddy's Beeks in Kerry; while westward across the Iveragh and
Dingle Peninsulas lies Brandon, which rises to 3127 feet, and is the

2X2

390 Proceedings of the Royal Irish Academy.

bigkefit in the Dingle range; on the east coast ranges we have
Lugnaqnilla in Co. Wicklow, which reaches to 3039 feet; and the
Qaltees in Co. Tipperary rise to 3015 feet. In the connties of HayO)
Waterf ord, and Wexford, some mountains are over 2600 feet. North-
wards, the extensive Ben Bnlben range in Sligo rises to 2100 feet;
Errigal, 2466 feet, and Muckish, 2197 feet in Co. Donegal; there
are extensive ranges in Antrim and Deny; and Slieve Donard in
Co. Down rises to 2796 feet.

Cldcits.

The moist, mild atmosphere of the south-west and south is now

accounted for by the broad area covered by the south-west winds over

. the Atlantic Ocean (which drive the vapour-laden clouds which are

condensed by the Kerry Mountains), and also by the influence of the

Gulf Stream.

The mean annual temperature is about 50^ Fahrenheit. The rain-
fall is remarkable, as may be seen from the following table for ten
years. It will be observed that the increase from east to west i»
striking.

McArdlb — A List of Irish ffepatiea.'

391

<« «o

r-l

to 00

»? 1 «

i;^

CO .

»po

»-• r*

■^

04 00

•-•

»o '^

•*

•^iS

sss

S2g

•o

»b

1?-

4tir* •

«*

ii<

<o»

<o c«

00-

CO I-l

•p 1 •

•0

•^

io»o

^ 00

«>

q:s

e!.

«0 lO

COOOO

•-4

e4oe«

r»

■^

0)04

rJ»

e«o "ji

«o

i-«

opoft

(Ncom

»o

•<m

04 0«

<o F^«o

■^

•*

«oco

•^

»H

a> lo -^

»«

«-•

o»

04 CO

^^y

c«

r«

fr»

r-co

m ^00

•H

»b

coo

jjjj

>or-^

•«•

fc*

•^o*

Sfr

Oft

oo <^co

■i«

ao»o

■^ t«»o

»*

<«

-^eo

O O '^

•0

I-l

COO

cpioco

•?"?•

<»OTO

o»

(l^

•4«

OIO

U30CD

"*00

!.!

00 eo

»^

<« •«

«o

■^

04 04

«D O 04

•^

■^

K^eo

OeoS

»H

Qp04

09 ^Hr-

»H

e«

1?-

l^«Ck

kO^iO

•^

■^

"*04

:::

•
•

•

5?:

5 •• «?

•

a
1

Kerry.

Valentia Telegraph S
Mangerton Mountain
Killamey, Woodlawi

.1
II

Dublin.

61en-na-8moel Wati
Glasnevin, Boyal Bo

5°

392" Proceedings of the Royal Irish Acadkmy.

PSOULIASITIBB 07 THE IrIBH HkPATIC PlOBAS.

It is in the moist, warm, sheltered glens of the high mountain
nmges near the coast that some of the rarest species flourish. A few
of them are alpine, as Seapania nimhota and 8. omithopodioidei,
Cesias, &c. The cnrious Clasmatoeolea euneifolia often descends, as da
other alpines or sub-alpines, to low elevations, washed down by
mountain torrents, so that in few instances can we recall species that
are ezclusiyely alpine in habitat. About the Killamcy basin ther
luxuriance and beauty of some of the tropical species, such as Dumor^
tiera irriguMy are remarkable. This is accounted for by its sheltered
poBition, and the continual moist, genial atmosphere the plants enjoy
most months of the year. Similar luxuriance has been observed in
many of the glens in the Dingle peninsula, where 129 of the total 172
^ecies of the Irish Hepatic flora are known to grow.

It is remarkable that several of these plants have never
been found in fruit, and still continue to grow and increase,
as in the case of Phtgioehila tridentieulata^ the female plant of
which has not been seen. The same may be said of Clasmatoeolea
euneifolia ; neither male nor female fruit has been found, and yet it
flourishes. Of Forella pinnata, sterile plants only are found in Ireland,
and, I believe, in Europe. Adelanthus decipiens furnishes another
example ; the male plant was once found many years ago at Killamey
by the late Dr. Canington; but the female has never been seen on
Irish or British soil. The rare and beautiful Seapania omithopo*
dioideSf of which fertile specimens are unknown, luxuriates on
Brandon, in Kerry, and has been known to grow there for over one
hundred years.

Later research proves that Hepatics can propagate themselves not
only by spores, but by adventitious budding, gemmsB being produced
an the leaf-margins, or almost on any part of the plant, stem and
perianth included. • Dr. Spruoe records an instance of Jungermania
jkiniperina {-Herherta adunea) with branchlets growing out of the
leaves, whicH would in time become 'independent plants (see '^ Fhyto-
Ibgist," vol. ii., 1845, p. 85). My own investigations for a number of
jears, on this subject of their asexual mode of propagation and dis-
persal, prove that ^hey readily propagate by budding, and with more
Certainty of growth, as the gemms are often furnished with root-haira
before they become detached from the parent plant. (See McArdle,
*| On Adventitious Branching in Liverworts," ** Irish Naturalist,'^
iJoL iv., p. 81, plate 3, 1895).

McArdlb— ^ List of Ivinik Hepatka.

ALPims OK Svb-Alpivb Espaticjs.

Anthiliajulaeiay Mn^tigophwra Wbodntj Scapania omithcpodioidet^
S. nimbosa^ S. tdiginosa^ MyUa Tayhriy Phgiochila Bptnulotay F. pufuh
tata^ P. trtdentieuiata^ Jungtrmania eordi/olia, J. aipettrUf J, lyeope*
diaides Yai.Hoerlii, J. minuia^ Nardia eomprtua^ Marsupatlaiphaeelaiaf
M. Funehii^ C&sta earalloides, C, ohtusaj FoMombronia pfuiUa^ F. crutatay
F, easpitiformis.

Tbopical Typss.

JuMa Hutchifuia (Pacific I8land8)| Lejeunea haimatifolia (Gold-
bearing districts, Kynsna, Bouth Africa), Pleurwia eochleariformit
(E. Indies, Sandwich Islands) fferhwia adunea (W. Indies, Africa,
Java), Ma%t%gophora JToodaii (Himalayas), Lep%do%ia eupressina (W.
Indies), Cephalozia eonnwens (S. Africa), C. curvifolia (Mexico, 8.
Africa), C. dtvarieata (Asia), Frumohhu Turneri (California and
Africa), Adehnihus decipims (Cuba), Seapania omtthopodioides (Sand-
wich Islands, E. Indies), 8. netnoroM (Java), LophoeoUa hidentata (W.
Indies), Jung^rmania minuia (Africa, Mexico), Blasia pwilia (N. Asia),
Aneura pinguia (Cuba), Met%geria pubeseen* (Simla, Himalayas), M.
/ureata (Africa), M, canfugata (Africa), M. hamata (Asia, N. Zealand),
Marchantia polymorpha (Japan, Java), Canoeephalus eonicus (Asia,
Japan), Behoulia hemisphariea (Asia, Java, N. Zealand), Freistia com-
mututa (Asia, Japan), Lunularia crueiata (Africa,'QueeDsland), Dumar-
tiera irrigua (W. Indies), Spharoearpus terreitris (K. Africa).

Tbopical South Ambbicak Ttpbs. .

Zefeunea JlavOf Z, hamattfolia, Serhmia adunea^ Juhtda StUchtfuia^
Zeptda%ia eupressina^ Cephalo%ia ( Odantoschistna) denudata, Adelanthus
d$e%p%ensy Seapania nemorosa, Clasmatoeolea cuneifolia^ Nardia hytUina^
Blyttia Lyellii, Aneura palmatay M$t%geria fureata, if. haniata^ Tar-
gumia hypophylla.

North Amebicait Ttpbs.

We have ninety -three species, among them L^eunea serpylUfolia^
L. ealearea^ Forella Thuja, F. platyphylla, F.. pinnata, Triehoeolea
iomeniella, Lepido/gia reptans, Baaaania trierenaia, B. triangularis, B.
trilohata, Mylia Taylari, Fossamhronia pusilla, F. anguksa, F, eris-
fata, Jungertnania graeiUs, J. eardifelia, Barpanthus seutatus,
MaraupeUa sphacelata, FeUia ealyeina, Metzgeria puheseensy Rieeia
glauea, Anthoeeros lavis, A. punetatus.

394

Pxoceedmgs of the Royal Irish Academy.

British ^Fttss.

Lejiunea Maekaii, £. microseopiea, FVuUania germana, RaduHa voluia,
R. aquiUgia, Lepido%ia Pearsoniy Cephalonia pMida, Seapanta nimhota
(ScotlaDcL and Ireland), LophoeoUa 9picata^ Acroholhus Wilsoni,- Sealia
jRookeriy PdUa/oieinia hihemieay Aneura itnuata, Riocia ghuceicent.

Ibish Ttpbb.

L&jeunea Soltiiy Z. diveniloha^ RadtUa JRoltii, Bataanta Pearsoni^
CephaUmia hihemtca, Plagioohila amhagtosa, P. exigua.

DISTRIBUTION IN THE DISTRICTS.

Varieties are printed in italiei.

?rullania Tamarisoi,
atropir0tUf
eomubieay
robutta^ .

miorophylla, .

fragilifolia,

germana,

dilatata, .

JlabellatOf •

Jubula HutohinauB,
inUffri/olia,

Lejeunea Mackaii, .
aeipyllifolia, .
pUiniuteuUy
eavifoliaf .
heterophjfUa,
jtroUfiraf .
patens, .
0recta,

flaya,
Holtii, .
ovata, • *
hasLatifolia.

8 8 4 6 6 7 8 9 10 11 IS

2 8 4 6 6 7 8 9 10 11 12

2 3

__--8--ll-
4__»---lI12
4 . 6 - 8 > 10 11 12
4 6 6 7 8 9 10 11 12

--466-8-- -12
2 3 4 6 6 7 8 9 10 11 12

-3466-89

11 12
11 12

4 - -

4 6 6

8 9
8 -

- 12
11 12

- - - 6
3 4 6 6

8 -
8 9

II 12
11 12

FAOB

404
404
404
404
405
405
406
406
406
407
407
407
408
408
408
409
409
409
409
410
410
410
411
412
412

MgArdlk — A List of Irish HepaliecB.

395

Lejeunea caloarea,
Bosaettiaiia,
minutiaaima,
microsoopica,
diveniloba,
ulicinay •
calyptiifolia,

Sadula voluta,
Holtii, .
aqoilegia,
Caningtonii,
compUnata,

PoreDa lasyigata)
{nUgra,

platyphylla,

Thuja, .

rivulariB,

pinnata, •
Pleurozia cooUeaiiformiB,
Anthelia julaoea,

Herberta adunca,
Mastigophora Woodaii,
Blepharozia ciliariBy
Trioliocolea tomentella,
Blepharostoma trichophylli
Lepidozia oupressina,

reptaxu, .

Peanoni,

tertularioitUif
trichoclados, •
Bascania trilobata, .
triangularis,
itinovantf .
devgxum, .
tricrenata,
Peanoni,
Kantia Trichomanis,

arguta, .
Cephalozia catenulata,
pallida, .

um,

S 8

4 b

- 6

6 7 8 9 10 11 IS

« - - - - - 12

10 . - 12
-, - 12

2346-789 -11 12

6-. -8- - - 12

-«_---8-.10 - -

2 -
2 3

4 6

4 -

4 6
4 -

e. - 8 -.
- - 8 -

4 6 6 7 8 9 10 11 12
----^-10- -
4 _^-io.- -

9 10 - 12
- .- - 12
-. 10 - 12

11 12
11 12

^ 9 - 11 12
8 - - - -

- _ 4 - -. 7 - - - 11 12

- - 4 6 8 -. - - 12

466-8-10 11 -

2 3 4 6 6 7 8 9. 10 .11 12

2 8 4 6 6 7 8 - - 11 12

___»--8-- - -

--456-89-.- -
-346--8--U12

------89 10- -

2 3 4 6 6 7 8 9 10 U 12

2 3 4 6 6 7 8 - - 11 12

2 3 4 6 6 7 8 9 10 11 12

- - 4 ft - - 8 -. - 11 -

PAGE
413

413
414
414
416
416
416
417
417
417
418
419
419
419
419
420
420
421
421
421
422
422
423
423
424
424
424
426
426
426
426
427
427
427
428
428
429
429
429
429
429
430
431

396

Proceedings of the Royal Irish Academy.

OepfaaloBia lunulsefolia,
bicuspidata,

riffidula, .

Mtulota, .

tenuirama,

minute redptani^
Lammeniana,
hibernica,
coimiyensy
cunrifolia,
Francisci,
fluitans, .
Sphagni, .
denudata,
divaricata,

Stmrkii,
stellulifera,
elachista,
leucantha,
Prionolobua Tiuneii,
Hygrobiella laxifolia,
AdelanthuH decipiens,
Soapania compacta,
Bubalpina,

undulifoliay
SBqniloba,

aspera, .
resupinata,
nemorosa,

purpurea, .
nimboaa, .
ornithopodioides,
iindulata,

purpuraecentf

epeeioMf .

itoloba,

mqior,

laxifoliay ,

dentata,
intermedia,
irrigua, .

8 4

3 4

5 6

6 -
6 6

7 8

- 8

9 10

11
11
11

IS PAOB

- 432
12 432

- 432

8 7

- 10 -

- 4
3 4

6 -
6 6.

7 8

-. 10
-. 10

- 4

6 6
6 -

7 8

- 8

- 10
9 -

433
433

- - 433

- - 433
11 12 433

- - 434
11 12 434
11 12 435

3 4
- 4

6 -
6 6.

6 6
6 6

4 6-

7 8

- 8
7 8

- 8

- 10
9 10

U
11
11

3 4

- 4

- 4
3 4

- 8

9 -
- 10

6 -
6 6

- ~ 9 ^ 11 -

3 -

3 4

- 4

6 6
6 -
6 6.

- 8

-. 10
- 10
-. 10

11
11
11
11

3 4

- 4

6 6
6 -

- 8
7 8

9 10

11
11

12 436

- 437
12 437
12 438
12 438
~ 439
> 439

- 439

- 440
12 440
12 441
12 441

- 442
442

- 442

- 442
12 443
12 444
12 445

- 445

- 445
12 446

- 446

- 445

- 446

- 446
12 447

11 447

12 447

8 9 - - 12 447

MoAmdlb— ^ IM of Irish Hepatiae.

397

1 S

PAOK

Soapania uliginoaa, .

1 -

448

CUltfty • • • «

I -

—

448

iimbroM) . • .

. r -

-■

449

IMplophyllum alHcass, .

. 1 2

449

oUumfolium, .

1 -

450

Dioksoni,

1 -

—

460

Lophocolea bidentata, .

1 2

460

Eook^rUma,

. - -

451

cuspidata.

1 2

—

461

. I ^

•11

461

spieatay . ' .

. 1 -

—

452

damiHtocolea ouneifalia.

. 1 -

•8

462^

ChiloflOTphiiB poljanthoa.

. 1 -

468

paiUieeng,

1 2

458

riviilarU, .

1 -

—

-.•

468

HarpanthuB soutatuB, . -

1 -

464

Mylia Taylori, . •

. 1 -

—

-•

464

anomala, . . . .

1 -

465

Pedinophyllum mteiruptum,

. 1 -

455

Plagiochila asplenioidea, .

.12

456

ffitn#y*y •

1 -

—

-•

456

d^vtxOf

1 -

-•

456

humiliSf .

, 1 -

466

ambagiosa,

1 -

466

spinulofla.

1 2

12,

457

Jlagellifara, . ,

1 -

467

tfMI'MWy •

1 -

467

punctata^ •

1 -

457

tridenticulata, .

1 -

-•

468

ezigua, . . . .

1 -

—

458

^ungemumia ooidif olia, .

1 -

469^

piimila, .

1 -

459-

nparia, .

. 1 -

■ 11

460

sphflBrocazpa, •

1 -

460

lurida, .

1 -

—

-•

461

crenulata.

1 2

-•

461

graeiUima^

1 2

462

inflata, .

. 1 2

468

compaeta, .

. - -

464

laxa^

. 1 -

464

hetsroiiipaf

- -

—

464

turbinate,

1 -

-■

464

bantriensis,

. 1 -

466

398

Proceedings of the Royal IrUhAmdemy.

Jimgennania bantrienaia — camt,
MmeUeH, .
aeuia,

. ffomtehuehiinmi
capitata, .
• lucreiiatof •
Tentricoea, •

porphjfroUticat
alpeatria, .

ezflecta, .
exsectsBfonnia,
Ljoniy
gracilia, .
barbttta, .
lyoopodimdea, .
minuta, .
orcadensiay
Nardia hyalina,
oboTats, •
compreBsa,

acalaris, .
eompreMio,
dittfuuy
rivularii^ .
robutta, .
ManupeUa emarginata,
minor, •
picM,
mqjor,
sphacelata,
Funckii, .
Ceda coialloides, .
obtuaa, .
crenulata,
AcrobolbuB Wilaoni,
Saccogyna vitiouloaat
Scalia Hookeii,
Fossombronia punUa,
oehroaporUf .
cristata, .
angi^psa,

1 2 8 4 6 6 7 6 9 10 11 18 PAOB

466

- - - 466

- 11 12 466

- - 12 467
la 11 12 467
10 11 12 468
10 ^ 12 468
10 11 12 469

- . - 12 469
- - - 470

- - 12 470

- _ 4 - -. - r- -

2--.- -789
2-46-- 8-
2-45 6. -89
--4--8--

- 3 4 6 - - - 9

- 3 4 6 6 7 .8 9

- - 4 - 6. - 8 9. - -
--4-6. ---- -
--4------ -

2--6- 11

--46--8-10 11
------8- - 11 -

2-46--8-- - -

------8-- - -

- 8 4 6 - - 8 - 10 11 12
2 - 4 - 6. - 8 - 10 11 12

- - 4 6 - - .8 -. - 11 12

- - 4 --.-.- -
2846 6. 789

2 3 4 6
- - 4 -

2 - 4 -

2-46

6 7 8 9
- - 8 -

- 8

471
471
472
472
478
473
474
476
475
476

- - - 476

- - - 476
10 11 12 476
10 11 12 477

- . - - 477

- - - 477

- - 12 478

10 11 12

2 3 4 5 6 7 8 9.
-5

10 - 12 478

- - - 478

- - 12 479

- - - 480
10 .11 12 480

- - - 481

- 11 12 481

- - - 482

- - 12 482

- - - 482

MoArolb — A Li»t of Irish Hepatiem.

399

Foflsombronia puBill»— «0fi<.

Dumortieri,

cflBspitiformiSi .
Petalophyllum Ral£ui,
Pallayicinia Lyellii,

hibernica,

Blaaia puailla,
Pellia epipbjllAy

calycma, .

Neesiana,
Aneura palmata,

multifida,

ambrosioideSy

latifrons, .

flinuata, .

pingnis, .

dmticuUUa,
Metzgeria pubeacena,

furcata, .

aruffinota,
prolifera, .

conjugata,

proliftra, .

hamata, .
Marcbantia polymorpba, .
Conocepbalus conicus,
Beboulia bemispberioa,
Preisna commutata,
LuDulaiia cruciata, .
Dumortiera irrigua,
Targioxiia bjpopbjUa,
SpbsBrocarpus terrettris,
Bioeia glauca,

(OTBtallina,

Borocarpa,

glaacescens,
Ricoiella fluitans .
Ricciocarpus natans,
AntbooeroB hayiBf .

punctatoB,

8 8 4 5 6 7 8 9 10 11 18 PiioE

--4---8-
23466789

10 - -
10 11 12

2 3

7 8 9 10 11

- 8 -

11 12
12

11 12
11 12

- 7

- 10 11 12

- - 4 6 - - -

- 12

- 12
11 12

- 12

- - 6 6 -

- 11 12
10 - 12

488
483
488
488
484
484
484

---6----- -

-_4 6----- -

_-4-_^_^ _ -

---6----- -

2 - 4 6 - - 8 - 10 11 12

2 8 4 6 6 7 8 9 10 11 12 485

- 8 4 6 6 7 - 9 - 11 12 486
---_--_«. . « 4gff

486
487

--------- - - 487

-346678-- - - 487

2 - 4 ----- 10 - 12 488

2 8 4 6 6 7 8 - 10 11 12 489

-.-__--«__ _ _ 489

-_-______. 12 490

2 3 4 6 6 7 8 9 10 11 12 490

- 8 4 6 - - - 9 10 11 12
_-_6----- - -

2 8 4 6 6 7 8 - 10 11 12 491

----6---- - - 491

8 -

490
490

12-46

- 8

- 11 12

492
492
493
493
494
494
496
496
496
497
497
497
498
498
498
499
499

400 Proeeedmgii of the Moyal Iriai Academy.

Alphabetical List, ukdeb Attthohs, of the fbinoipal Books, Papebs,
AND Herbaria BELATore to the .Hepatic Flora of Ibklaitd,

WITH abbreviated REFERENCES USED IN THE PRESENT PaPER.

I. PuhlieatioHi.

Oarringtos, Benjamin, M.D. :

Gleanings among the Irish Cryptogams. Trans. Bet. 8o9. SdM., toI. tiu.

1863. (Cairington 1863.)
British Hepatice, Descriptions and Figuxes. [Only {our parts imaedJ] 1874-6.

(Carrington 1874.) ....

^tter, G. M. : ...

The Flora of the County Cork, [in F. M. Cusack's « History of die City mad

County of Cork." 1876. Enumerates twenty-nine Hepaticss.]
Hart, Henry Chichester, B.A. :

Irish HepaticsB. /o«ni. o/J?(rf., Tol. xxiv., p. 360. 1886. (Hart 1886.)
Hooker, Sir William Jackson :

British Jungermanie, being a History, with Description and Figures, of each

Species of the Grenus, and microscopical Analyses of the Parts. London.

1816. (Hooker 1816.)
Lett, Rev, Canon Henry William, M.A., M Jt.I«A. J . . .

Report on the Mosses, Hepatios, and Lichens of the Moume Mountain District.

Froc. R, I. Academy, 3rd ser., vol. i.. No. 3. 1890. (Lett 1890.)
A List, with Descriptive Notes, of all the Species of Hepatics hitherto found in

the British Islands. Eastbourne, 1902.
Lindberg, Prof, Sextus Otto :

HepaticsQ in Hibemia mense Julii, 1873, lecta. Aeta Soe. SeienL Fenniea^

vol.x. 1876. (Lindberg 1876.)
MoArdle, David :

Notes on some new or rare Irish HepatiosB. 8ci. Froe. R. Dublin Society,

voL iii., plates 5, 6. 1880. (McA. 1880.) '
HepaticsB of County Wicklow. Joum. of Bot,^ vol. zxvii., p. 267. 1889.

(McA. 1889.)
HepaticsB of Lough Bray, County Wicklow.- Jomm, mf Bot., vol. zxviii.

1890. (McA. 1890.)
Hepatic® of King's and Queen's Counties. Irith Nat., voL i., p. 69. 189S.

(McA. 1892 a.)
The Plants of Dalkey Island. IrUh Nat,, vol. i., p. 133. ; 1892. (McA. 1892 b.)
On the Hepaticss of the Hill of Howth. Froe. R.I. Academy, 3rd ser., vol. iiL,

No. 1, pktes 3, 4. 1893. (McA. 1893 a.) '
Bare Irish Hepaticsd at Leixlip, Coimty Kildare. Irieh Nat., vol. iL 1893.

(McA. 1893 b,)
A Visit to Castletown Berehaven, County Cork, Irieh Nat., voL iii. 1894.

(McA. 1894.) ...

Mosses and Liverworts [of Galway Field Club Conference.] Irith Nat., vol. iv.,

p. 244. 1896. (McA. 1896 a.)

McArdlb — A List of Irish Hepatiem. 401

JCeArdle, David— Mm^iftuMf.

Adventitious branching in Liyerworts. Iruh Nat.^ vol. ir., plate 3. 1896.

(McA. 1895 h.)
Hepatioas ooQeeted in County. Gadow. IrUh Nai.^ vol. y^ 1896. (MoA.

1896 a.)
Mosaea and HepatifiiB fX Clonbrock, County Oalway. Iruk Nat,, vol. t.

1896. (McA. 1896 b.)

Additions to the Hepatioe of the Hill of Howth, with a Table showing their
geographical distribution. Froc. B, I. Aead$m^, 3rd ser., vol. iv., No. 1.

1897. (McA. 1897.)

Report on the Mosses and Hepatioe of County Cavan. iVoe. JB. i. Aeadtm^,

3rd ser., vol. iv., plates 21, 22. 1898. (MoA. 1898.)
The HepaticsB of Ross Island, Killamey. IrUh Nat., vol. ix., plate 1. 1900.

(McA. 1900.)
Report on the Hepatiea of the Dingle Peninsula, Barony of Corkaguiny,

County Kerry. Proe. J2. J. Aeademff, 3rd ser., vol. vi.. No. 3, plates 16, 17.

1901. (McA. 1901.)
HepaticsB from County Wexford. IrUh Nat., vol. xiL, p. 132. May, 1903.

(McA. 1903.)
JCcArdle, David, and Lett, H. W. :

Report on Hepaticss collected at Toro Waterfall, Killamey. Proc. J2. /.

Academy, 3rd ser., vol. v., No. 2, plates 8, 9. 1899. (McA. & Lett 1899.)
Voore, David, Ph.D. :

Contributions to the British and Irish Musci and Hepatioss. Froe, Duhl. Univ.

Zool. and Bot. Atsoe., vol. ii., p. 80. 1863. (Moore 1863.)
[Adds Scapania undulata var. major and some new habitats.] Dublin Nat. Sitt.

Soe. Froe., vol. v., p. 89. 1866.
Report on Irish Hepaticss. Froe. B. I. Academy, ser. 2, vol. ii., plates 43, 44,

46. 1876. (Moore 1876.)
list of Hepaticss which an found in the Counties of Dublin and Wicklow,

with their principal Localities. Set. Froe. B. D. Society, vol. i. 1878.

(Moore 1878.)
Paanon, W. H. :

On Badula Carringtonii. Joum. of Bot,, vol. xz., p. 140. 1882.
Frullania mierophylla. Joum. of Bot,, vol. xxzii., p. 328. 1894.
A new British Hepatic [Lejeunea Boteettiana]. Joum. of Bot., vol. xxxvii.,

p. 353. 1899. (Pearson 1899.)
A new Hepatic [^Cephalosia hibemiea']. Irish Nat., vol. iii., p. 245, plate 6. 1894.
The HepaticaD of the British Isles, with Figures and Descriptions of all known

British Species. Vols, i., ii. London, 1902. (Pearson 1902.)
Power, Thomas, M.D. : '

Contributions towards the Fauna and Flora of Cork. Part ii. Botany. 1844.

[Fifty Species of Hepatic«.]
Scully, Reginald William :

Kerry Hepaticn. /otirn. o/^o^, vol. xxviii., p. 200. 1890. (Scully 1890.)
flpnioe, Richard, Ph.D. :

Musd Prssteriti. II. Joum* qfBot,^ vol. x., pp. 11-33. 1881. (Spruce 1881.)

402 Proceedings of the Boyal Irish Academy.

Spruee, Richard, Ph.D. — continued.

On Opkahzia, its Sub-Geoera, and tame allied Genera. Malton, 1882.

(Spruce 1882.)
Hepatics Amaaonicn et Andine. JVmns. Bat, Soe. BdM,, yol. zy., with

twenty-two plates. 1885. (Spruce 1886.)
L^eunea Holtii, A new Hepatic from Killamey. Jowm. of Bot., toL xxt.,

p. 33, plate 272. 1887. (Spruce 1887 a.)
On a new Irish Hepatic^[i?4ufM/a Soltit]. Joum. of Bot., toL zzy., p. 209.

1887. (Spruce 1887 b,)
On Lejeunea BoiHttiana, Joum. of Bot,, voh itrvii. 1889. (Spruce 1889.)
Stewart, Samuel Alexander, F.B.S.E. :

Beport on the Botany of the Idand of Bathlin, County Antrim. IVoe. B. L

Jeademy, 2nd ser., Yol. iy., No. 2. 1884.
Beport on the Botany of Lough Allen and the Sliereanierin Mountains. Broe>

B. I. Academy, 2nd ser., vol. iv., No. 2. 1886. (Stewart 1885.)
Beport on the Botany of South Clare and the Shannon. Proe, B, L Academy,
3rd ser., toI. 1., No. 3. 1890. (Stewart 1890.) •

Stewart, S. A., and Corry, T. H., M.A., F.L.S. :

Flora of the North-East of Ireland. Belfast, 1888. (Stewart 1888.)
Stewart, S. A., and Praeger, E. Lloyd, B.A., B.E., M.B.LA.:

Supplement to the Flora of the North-East of Ireland. Froc. Belfast Nat.
Field Club, 1894-5, Appendix. (Stewart 1896.)
Taylor, Thomas, M. D. :

Hepaticie (Part ii. of Mackay's Flora Stbemiea). [Eighty-two Species.]

1836. (Taylor 1836.)
Descriptions of Jungermania ulieina and /. Lyoni. Trane. Bot. Soc. Edinb.,

yol. i., p. 115. 1841. (Taylor 1841.)
On two new Species of Jungermania and another new to Britain. Trane,

Bot. Soc. Edinb., yol. i., p. 179. 1843-4.
On four new Species of British Jungermania. Trane. Bot. Soc. Edinb., yol. ii.,

p. 43. 1846. (Taylor 1846.)
Contributions to British Jungermania. Trane. Bot. Soc. Edinb., yol. ii., p. 115.
1846. (Taylor 1846.)
Waddell, Bev. C. Herbert, B.D. :

Mosses and Hepaticss of Ben Bulben, County Sligo. Irieh Nat., rol. i.

p. 194. 1892. (Waddell 1892.)
DiBtribution of Lejeunea in Ireland. Joum. of Bot., yol. zxxi. 1893.
(Waddell 1893.)
Wade. Walter, M.D. :

Plants Bariores in Hibemis inyentte. Dublin, 1804. (Wade Bar. 1804.)

II. Mffrbaria.

Belfast Museum, College-square North, Belfast.

Moore, Dayid, Ph.D. : Ordnance Suryey Collections, Counties of Derry and

Antrim. 1834-8.
National Museum, Kildare-street, Dublin.
Botanical Department, Trinity -College, Dublin.

McArdle — A List of Irish Hepaticw.

403

AbBBSVUTIOKS T78ED FOE ArTHOKITIES OTHEB THAN THOSE OITED IN THE

FOEEooixo List.

Brenan, . . Rev. 8. A. Brenan, B.A., Cusbendun.

Camngton, . . Dr. Benjamin Carrington, EccleR, Mancheater.

CaiToU, . . Isaac Carroll, Cork.

F. W. M., . . Frederick William Moore, A.L.S., Glaanevin Botanic Gardens.

Greene, . . Br. G. E. J. Greene, F.L.S., ftc, Ferns, Co. Wexford.

Holt, . . G. A. Holt, Manchester.

Hooker, . . Sir William Jackson Hooker, London.

Hunt, . . G. Hunt, Manchester.

Hunter, . . J. Hunter, Holywood, near Belfast.

Hutchins, . . Miss Hutchins, Bantry, Co. Cork.

Lett, . . Rev. Canon H. W. Lett, M.A., Loughbrickland, Co. Down.

Lindberg, . . Professor Sextus Otto Lindberg, Helsingfors Uniyersity.

Mo A., Dayid McArdle, Glasnevin Botanic Gardens.

Moore, . . Dr. David Moore, Glasnevin Botanic Gardens.

Pearson, . . W. H. Pearson, Manchester.

Praeger, . . Robert Lloyd Praeger, National Library, Dublin.

Russell, . . Rev. Canon Charles Russell, D.D., Geashill, King's Co.

Scully, . . Dr. Reginald Scully, Dublin.

Stewart, . . Samuel Alexander Stewart, F.B.S.E., Belfast

Templeton, .. John •Templeton, A.L.S., Cranmore, Belfast.

Taylor, . . Dr. Thomas Taylor, Kenmare.

WaddeU, . . Rev. C. H. Waddell, B.D., Saintfield, Co. Down.

Wade, . . Dr. Walter Wade, Dublin.

The Twelve Botakigal Disteicts of "Ctbele Hibeeitica."

I. South Atlantic. — Kerry and South Cork ; 3143 square miles.
II. Blackwatbr.— North Cork, Waterford, South Tipperary ; 3 18 1 square
miles.
III. Bakrow. — Kilkenny, Carlow, Queen's County ; 1805 square miles.
IV. Lbinstbr CoAST.—Wexford and Wicklow ; 1677 square miles.
y. LiFFBY AND BoTNB.— Kildare, Dublin, Louth, Meath ; 2230 square miles.
VI. LowB& Shannon. — Limerick, Clare, East Galway ; 3989 square miles.
VII. Upper Shannon. — North Tipperary, King's Co., Westmeath, Longford ;

2700 square miles.
VIII. North Atlantic. — ^West Galway, West Mayo; 2146 square miles.
IX. North Connauoht.— East Mayo, Sligo, Leitrim, and Roscommon; 3086

square miles.
X. Erns. — Cavan, Armagh, Fermanagh, Monaghan, T3rrone ; 3733 square

miles.
XI. DoNBOAL. — Donegal, and Derry west of the Foyle ; 1890 square miles.
XII. XJlotkr Coast. — Down, Antrim, and Derry ; 2862 square miles.

R.I. A. PROC, VOL. XXIV., SRC. B.] 2 L

404 Proceedings of the Royal Irish Academy.

Class HEPATICiE.

Order I. JXJNGERMANIACKS.

Tribe I. TOBJJLEJR.

QtemiB I. Frnllania Eaddi.

1. Frnllania Tamarisoi Linn., Dmnort.

Jungermania Tamarisei Linn., Sp. PL, 1 ed., yol. ii., p. 1 134. Hook.,
Brit. Jung., tab. 6. Dumort., Reoueil Jung., p. 13. Moore, Iriflh
Hepat., p. 610. Pearson, Hepat. Brit. Isles, p. 24, plate I. .

Districts I. II. m. IV. V, VI. VII. VIII. IX. X. XI. XII.

Sab. — On the trunks of trees in large spreading patches, on rocks
and wall-tops, from sea level to the tops of the highest mountains.

var. atrovirem Carrington.

Hah, — On rocks which are frequently inundated.

Eagle's Nest and Gromaglown, Xillamey : Carrington 1863.
Glena, Killamey, 1873: Lindberg 1875. Toro Waterfall, Sept. 1897:
McA. & Lett. Boss I., 1899 (McA. & Lett) : McA. 1900. Loughanscaul
near Dingle, rare, Sept. 1898 (Lett & McA.) : McA. 1901.

var. eomuhiea Carrington.
Sab. — On stones.
Fairhead, Co. Antrim (Lett) : Stewart 1888.

yar. robwta Lindberg.

Sah, — On rocks and on the bark of trees.

Glena and Cromaglown at KiUamey, Connor hill near Dingle,
and Lough Bray, Co. Wicklow, 1873: Lindberg 1875; McA. 1890.

2. Frullania microphyUa Oottsche, Pearson.

Ihdlania Tamarisei Linn., var. microphyUa Gottsche ex Carrington
in Trans. Bot. Soc. Edinb., vol. yii., p. 457, 1863. FnUlania micro-
phylla Gott., Pears, in Joum. of Bot., 1894. Ezsicc. Gottsche &
Eabenh., Hepat. Eur., nos. 209 & 636. Carr. & Pears., Hepat. Brit.,
fasc. 2, no. 137. Pearson, Hepat. Brit. Isles, p. 26, plate 2.

McArdlb — A List of Irish Hepaticm. 405

Districts I. II. VIII. XI. — .

Sah, — On smooth rocks and on the bark of trees, in shallow
patches, closely attached, mostly near the coast.

L Old Weir Bridge, Eillamey: Carrington 1863; Moore 1876.
Ross I., KiUamey, 1899 (Lett & McA.) : McA. 1900. Tore Water-
fall, Eillamey, 1898: McA. & Lett 1899. Anascaul and Connor
Hill : McA. 1894. On the west side of Brandon near the summit, and
Maghanabo glen near Castlegregory, April 1897 : F. W. M. & McA.
On smooth rocks on the shores of Lough Duff near Connor Hill, May
1899 (Lett & McA.) : McA. 1901. Glengariff, 1861 (Carrington and
G. E. Hunt) : Pearson 1902.

II. On rocks, Bay Lough, Knockmeildown Mountains, Co. Tipperary,
June 1902: McA.

Yin. Woods at Pontoon on Lough Conn and on Nephin, May
1901: Lett & McA. Achill and Bangore, Sept. 1901: Lett On
Alder near Ballinlough, Co. Mayo, Sept. 1901 : McA.

XI. Bathmullan Wood, July 1902: Hunter. Gartan Lake and
Cratleagh Wood, Sept 1902 : McA. On rocks. River Trillick, Bun-
crana, March 1903 : Hunter. On ffypnum eupressiforme, Errigal,
June 1903: McA.

3. Frullaiiia fragilifolia Taylor.

FruUania fragilifolia Tayl. in Ann. and Mag. of Nat Hist., p. 172,
1843, and Trans. Bot Soc. Edinb., vol. ii., p. 45, 1846. Moore,
Irish Hepat., p. 609. Pearson, Hepat. Brit. Isles, p. 29, plate 3.

Districts I. IV. XI. XII.

Sab, — On shaded rocks and on the bark of moss-covered trunks
of trees.

I. Eillamey: Moore 1876. Muckross demesne and near Dean
Bridge : Carrington. On the bark of Bettda, Eillamey, 1873 : lind-
berg 1875. Tore Waterfall, on rocks and the bark of trees, 1897:
McA. & Lett 1899. Boss I., 1899 (Lett & McA.) : McA. 1900. On
mural rocks, Dunkerron, 1829 (Taylor): Carrington 1863. Bumham
Wood between Dingle and Yentry : McA. 1901. On boulders, Bantry
Bay and Glengariff : Carrington.

IV. On the tronks of Alder with Plagiothecium JBorrerianum,
Lough Bray, Co. Wicklow, 1887 : McA. 1890.

XI. Cratleagh wood near Milford, rare, Sept. 1902 : McA.

XII. On granite rocks. Cove Mtn., Co. Down : Lett 1890. " The
Craigs," Rasharkin, Co. Antrim (Lett & Waddell) ; Glenarifl (Lett) :
Stewart 1895.

2X2

406 Proceedings of the Royal Imh Academy.

4. Frnllania germana Taylor.

Jungermania germana Tayl. in Trans. Bot. Soc. Edinb., toI. ii.
p. 45, 1846. Fndlania Tamariseiy var. germana Carr., Trans. Bot. Soc.
Edinb., vol. viii, 1863. Fndlania germana T&jl., Moore, Irish Hepat.,
p. 610. Pearson, Hepat. Brit. Isles, p. 31, plate 4.

Districts I. IV. — VI. — VIII. — X. XI. XII.

Skb. — On rocks and on the trunks of trees.

I. Dunkerron, 1832 : Taylor. Killamey : Aloore 1876. Glena and
O'Sullivan's Cascade, 1873 : Lindberg 1875 ; and (Holt & Stewart) :
Pearson 1902. On rocks at Tore Waterfall, rare, Sept. 1897 :
McA. & Lett 1899. Boss I., 1899 (Lett & McA.): McA. 1900.
Connor Hill near Dingle, 1873: Lindberg 1875. Bumham Wood
near Ventry, May 1894 : McA. 1901. On the west side of Brandon,
Sept. 1897 ; Anascaul and Mt. Eagle, 1898 ; rocks between Emalough
and Inch, May 1899: Lett & McA. Lough Duff in the Brandon
Valley, 1899 (Lett & McA.): McA. 1901. Old walls about Castle-
town Berehaven : McA. 1894.

IV. Lough Bray, Co. Wicklow : Moore 1876.

VI. Carn Seefin, Co. Clare : McA. 1895^.

VIII. On the slopes of the DeviPs Mother, and on Slievemore,
AchiU, Sept. 1901 : Lett.

X. Slieve Glah, Ballyhaise woods, and Famham woods, Co. Cavan,
1893; McA. 1898.

XI. Qlenalla hill, Rathmelton and Rathmelton Wood, July 1902 :
Hunter. Cratleagh Wood, on rocks by Columbkil Lake, Bunlin
Waterfall on trees, Sept. 1902 ; Lough Eask woods, June 1903 : McA.

XII. Eathlin Island, Co. Antrim : Stewart 1888.

5. Frnllania dilatata Linn., Dumort.

Jungermania dilatata Linn., Sp. PL, p. 1600. Hook., Brit. Jung.,
tab. 3. Frullania dilatata^ Dum., Recueil Jung., p. 13. Moore^
Irish Hepat., p. 609. Pearson, Hepat. Brit. Isles, p. 33, plate 5.

Districts I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.

Hah, — On the trunks of trees and on rocks.

Yai. Jlabellata Spruce.

ffah, — On the trunks of trees in shallow tufts, closely attached,
spreading in neat strata. Hickson's Wood near Anascaul, Co. Kerry,
very rare, 1898 (Lett & McA.): McA. 1901.

McAkdlb — A List of Irish Hepaticce. 407

var. prolifera McArdle.

HaL-^On moist rocks, side of a stream near the Baily Lighthouse,
Howth, Co. Dublin, 1896: McA. 1897.

Note. — This form shows adventitious budding, the leaf -margins
and the stems being covered with leafy shoots, which become inde-
pendent plants.

Genus II. Jnbula Dumort.

Jnbula HutchiiLsiflB Hook., Dumort.

Jungermania Hutehinsia Hook., Brit. Jung., tab. 1. IVullania
Sutehifuia^ Nees, Europ. Leberm., iii., p. 240. Moore, Irish Hepat.,
p. 608. Juhtda MtUehinsia Dum., Comm. Bot., p. 212. Pt-arson,
Hepat. Brit. Isles, p. 36, plate 6.

Districts I. IV. VIII. XII.

JSiab, — On shaded wet rocks near streams, often found within
reach of the spray of waterfalls.

I. Glengariff (Miss Hutchins) : Hooker 1816; and Moore 1876.
Ballinhassig Glen, and near Kinsale (I. Carroll) : Moore 1876.
Cromaglown (Moore), Tore Cascade, Killamey, 1873: Lindberg 1875.
In the same station on the fronds of Dumoriiera irrigua, Sept. 1898 :
Lett & McA. Caves, Dingle Bay (Moore) : Carrington 1863. Magha-
nabo Glen near Castlegregory, 1875: McA. Loughanscaul, west side
of Brandon, and shores of Lough Doon in the Dingle Peninsula,
1897-8 (Lett & McA.) : McA. 1901.

IV. Altadore Glen, Co. Wicklow : McA. 1889.

VIII. Rocks by the lake at Letterfrack, 1874 : Moore. On the
slopes of the Devil's Mother, Co. Mayo, Sept. 1901 : Lett.

XII. Rocks on the coast south of Newcastle, Co. Down (Miss
Thompson) : Waddell, Irish Nat., vol. iv., 1895, p. 190. Tollymore
Park, and by the Spinkwee River: Waddell 1892. Waterworks on
Rostrevor Mtn. (Waddell) : Stewart 1888.

var. inUgrifolia^ Nees ab Essenbeck, Syn. Hepat., p. 426 : var. jS.

N. ab E., Hepat. Java, I.e. . Moore, Irish Hepat., p. 609 (under
FruUania), with excellent figure, plate 45.

Bah. — On wet rocks, and on the larger Hepatics. Connor Hill
1873 (Lindberg & Moore) : Lindberg 1875 ; Moore 1876. Mountain
stream in the Maghanabo Glen near Castlegregory, on the fronds
of JDufnofiiera irrigua, fertile, 1^5 (McA.): Moore 1876. Tore

408 Proceedings of the Royal Irish Academy,

Waterfall, Killamey, 1889: Scully 1890. On rocks, Loughanscaul
near Dingle, Sept. 1898: Lett'.

Genus III. Lejeunea Libert.
1. Lejeunea Hackaii Hook., Sprengel.

Jungermania Mackaii Hook., Brit. Jung., tab. 53. Pkragmieomi
Maehaii Dumort. , Comm. Bot. , p. 1 1 2. Lejeunea Machaiiy Spreng., Syst.
Yeg., iy., p. 233. Moore, Irish Hepat., p. 616. Pearson, Hepat
Brit. Isles, p. 40, plate 7.

Districts I. IV. V. VI. — VIII. XII.

Hah, — Mostly on limestone rocks, often in large shallow patches,
and on decayed wood.

I. Ballylicky near Bantry, 1812 (Miss Hutchins): Hooker 1816.
Near Cork, frequent : Moore 1876. Muckross, Killamey : Camngton
1863 ; and (Moore) : landberg 1875. Ross I., 1893 : McA. ; plentiful
1899 (Lett & McA.): McA. 1900. Tore Waterfall, Sept. 1899:
Lett and McA. Loughanscaul, rare, Sept. 1898 (Lett & McA.):
McA. 1901.

IV. Dargle, Co. Wicklow, 1812 (Mackay) : Hooker 1816. On
rocks in the same station (Scully & McA.) : McA. 1889.

V. Woodlands near Dublin: Moore. Omeath Waterfall, Co.
Louth (Waddell): Lett 1890.

VI. At Kilmuryy on the Aran Islands, rare : McA. 1895 a.
Clonbrock, Co. Galway : McA. 1895 5.

VIII. On rocks by a small lake near Letterfrack, Co. Galway, 1 874 :
Moore.

XII. On old yew trees, ToUymore Park, Co. Down ( WaddeU) :
Lett 1890. Gobbins Clifk, Co. Antrim: Waddell 1893. Limestone
rocks, Redhall Glen, 1809 (Templeton); Glenariff (Lett) : Stewart 1895.

2. Lqeunea serpyllifolia Dicks., Libert.
Jungermania serpyllifolia Dicks., PI. Crypt. Brit., fasc. 4. Zefeuma
serpyllifolia Libert, in Ann. Gen. Sch. Phys., vi., p. 374. Moore, Irish
Hepat., p. 614. Pearson, Hepat. Brit. Isles, p. 45, plate 10.

Districts I. 11. III. IV. V. VI. VII. VIII. IX. X. XI. XIL
J3ah, — In damp shaded places, on the trunks of trees, on rocks and
stones in riyulets, and on damp banks.

var. planiuseula Lindberg.
JSiab, — On rocks among Mosses. 0' Sullivan's Cascade, Killamey,
1873 : Lindberg 1875. In same station : Scully 1890. Connor Hill
near Dingle, 1873 : Lindberg 1875.

McArdlb — A List of Irish HepaticcB. 409

var. eavifolia Ehrhart, Lindberg.

Sdh. — On rocks and damp banks.

Killamey: Garrington 1863; Moore 1876. Qlena and Tore
Cascade, among ITypnum eugyrium^ 1873: Lindberg 1875. On a turfy
bank among rocks, between Enudough and Inch, Go. Kerry, May 1899
(Lett & McA.) : McA. 1901.

var. heUrophyUa Garrington.

Hob. — On wet rocks.

O'SuUivan's Gascade and Tore Waterfall : Garrington 1863 ; Moore
1876. Boss I., 1899 (Lett & McA.): McA. 1900. Frequent in the
Dingle Peninsula : McA. 1901. Altadore Glen, Go. Wicklow: McA.
1889. Glenariff, Go. Antrim; Spinkwee River, Go. Down: Waddell
1892. Lough Eask woods and Bamesmore Qap, June 1903 : McA.

var. prolifira McArdle.
Mah. — On decayed bark.

Hickson's Wood near Anascaul, Go. Kerry, May 1894 : McA.
1901, and McA. 1895^.

3. Lejennea patens Lindberg.

Zefeunea patens Lindberg in Acta Soc. Fenn., vol. x., p. 482, 1875.
Moore, Irish Hepat., p. 615, plate 49, 1876. Pearson, Hepat. Brit.
Isles, p. 47, plate 11.

Districts I. — III. IV. V. VI. — VIII. IX. — XI. Xn.

Hob, — On damp rocks, on the bark of moss-covered trees, and on
the larger Mosses and Hepatices.

I. Killamey : W. Wilson. Olena and Tore Gascade, 1861 :
Garrington; and Sept. 1897: McA. & Lett 1899. O'Sullivan's
Gascade, among Thamnium alopecurum, and on Gonnor Hill, 1873:
Lindberg 1875. Between Dingle & Ventry, 1873: Lindberg &
Moore. On the west side of Brandon, and on Mt. Eagle, 1881 :
F. W. M. & McA. Goumanare Lakes, Sept. 1898 ; Derrymore Glen
near Tralee, May 1899 : Lett & McA. Frequent in the Dingle
Peninsula: McA. 1901. Dunboy Wood, Gastletown Berehaven:
McA. 1894.

III. Gappard, Queen's Go., 1891 (Russell) : MoA. 1892 ii. Wood
near Goresbridge, Go. Garlow: McA. 1896 a,

IV. Altadore Glen, Go. Wicklow: McA. 1889.

V. Carlingford Mtn., Go. Louth, very rare (Waddell) : Lett 1890.

410 Proceedings of the Royal Irish Academy.

YI. Cam Seefin, Co. Clare : McA. 1895a.

Vin. GentianHill near Galway : McA. 1895 a. Co. Galway, 1891 :
Dr. E. J. McWeeney. Nephin, May 1901 : Lett & McA. Slopes of
the Devil's Mother, Bangore, Slieyemore on Achill, Sept. 1901 : Lett.
Pontoon near Foxford, May 1901 : Lett & McA.

IX. Ben Bulben: Moore; and July 1880: McA. Glenade, Co.
Leitrim, 1875: Moore.

XI. Cratleagh Wood near Milford, very rare, Sept. 1902 : McA.
Dunree River near Buncrana, June 1903 : Hunter.

XII. Glenariff: Waddell 1893. Sallagh Braes (Lett) : Stewart
1895. ToUymore Park: Waddell 1893. Black Stairs on SHeve
Donard : ^Waddell.

var. ireeta, McArdle, I. Kat., vol. iii., p. 139, 1894.

-Sfli. — On damp peat among rocks.

Boss. I., Killamey, 1893 : McA. Connor Hill near Dingle, rare,
June 1894 : McA.

var. cochleata, Spruce (species).

Spruce, Hepat. Amaz. et And., Trans. Bot. Soc. Edinb., vol. xv.,
p. 273, 1885.

JETab, — On wet rocks, decayed wood, and on the larger Mosses
and Hepatics.

Ben Bulben, Co. Sligo, and Glenfarm demesne, Co. Leitrim,
1871: Moore. Kylemore, Co. Galway, 1874: Moore; McA. 1880;
and 1891 : McWeeney. Woodenbridge, Co. Wicklow, fertile, 1895:
McA. Glenariff, Co. Antrim : Waddell. Black Stairs on Slieve
Donald, Co. Down, and Slish Wood, Co. SUgo : Waddell 1892.
O'Sullivan's Cascade, Killamey, 1893 : McA.

Note. — The specimens collected in these localities compare
favourably with plants collected by Dr. Spruce on Mount Tanguragua,
S. America, and probably the var. eookleata should be the type, and
Z. patens the variety.

4. Lejeunea flava Swartz, Nees.

Zefeunea Jiava Swz., Prodr. Fl. Ind. Occ, p. 144, 1788. Nees,
Nat. Eur. Leberm., iii., p. 277, 1839. Zefeunea serpyUifoUa var.
thymifoUa Carrington, Irish Crypt. 1863. Lejeunea Jfoorei Lindberg,
Hepat. Hib., p. 487, 1875. Moore, Irish Hepat., p. 615. Pearson,
Hepat. Brit. Isles, p. 49, plate 12.

Districts L — m. IV. V. VL — VIII. XI. XH.

McArdle — A List of Irish Hepaticce. 411

Sah, — On damp moss-covered rocks, on decayed wood, and on the
larger Hepatics.

1. Cromaglown and Glena, 1873 : Lindberg 1875. At the Hunting
Tower, Cromaglown, 1862 : Moore; and 1875 : McA. O'Sullivan's Cas-
cade (Moore ; G. A. Holt ; M. B. Slater) : Pearson 1902. On decayed
wood and on the fronds of Mstzgiria^ Tore Waterfall, very scarce,
1897: McA. & Lett 1899. Ross L, 1893: McA.; and 1899, rare
(Lett & MoA.) : McA. 1900. Brandon, April, 1897 : F. W. M.
& McA. Among Hypnum^ Lough Nalachan on Brandon, 1899 ; and
Mt. Eagle Lake, Sept., 1898 : Lett & McA. Hickson's Wood near
Anascaul, 1894 : McA. In the Dingle Peninsula: McA. 1901. Bun-
boy Wood, Co. Cork, typical : McA. 1894.

lU. Goresbridge, Co. Carlow : McA. 1896 a.

lY. Altadore Glen and Luggielaw : McA. 1889. Dargle, Co.
Wicklow: McA. & Scully. Wood by the Slaney River near
Enniscorthy, and Killoughrim Oak Forest, Co. Wexford, typical, rare,
1899: McA.

V. Howth demesne, Co. Dublin, April, 1895 : McA. 1897.

VL Cam Seefin, Co. Clare: McA. 18'95a.

VIII. Near Letterfrack, Co. Galway, 1891 : McWeeney. Pontoon,
Co. Mayo, May, 1901 : Lett & McA.

XI. Very fine in wood by the seashore, Rathmullan, and at
Macamish Point, July 1902 : Hunter. Cratleigh Wood and wood
at Mulroy Bay, Sept. 1902 : McA. Bamesmore Gap, very scarce,
June 1903: McA.

XII. Glenariff, Co. Antrim, and ToUymore Park, Co. Down :
Waddell 1893.

5. Lejeimea Holtii Spruce.

Lejunea IToUii Spruce, Joum. Bot., vol. xxv., p. 33, plate 272.
1887. McA. & Lett, Hepat. of Tore Waterfall, KiUamey, Proc. R.
I. Acad., 3rd ser., vol. v., no. 2, 1899. McArdle, Report on the
Hepat. of the Dingle Peninsula, Proc. R. I. Acad., 3rd ser., vol. vi.,
no. 3, 1901. Pearson, Hepat. Brit. Isles, p. 51, plate 13, 1902.

Districts I. .

ITab. — On wet rocks in shaded places, among Mosses and on the
larger Hepatics, on the bare wet rocks in neat strata.

I. Tore Waterfall, KiUamey, June 1885 (Holt): Spruce 1887a.
On TriehoeoUa and Mettgeria^ also mixed with Lejeunea Mackaii, on
bare rocks within the spray of Tore Waterfall, Sept. 1897 : McA.
& Lett 1899. In the crevices of rocks among Fistid&ns taxifoliu*,
Loughanscaul near Dingle, May 1894: McA. Rocks near the lake

412 Proceedings of the Royal Irish Academy.

on Mt. Eagle, and shores of Bamanaghea Lougli near Dingle, Sept.
1898 : Lett & McA. Lough Nalachan on Brandon, May 1899 (Lett &
McA.) : McA. 1901.

6. Lejennea ovata Taylor.

Jungermania serpyllifolia var. cvata Hook., Brit. Jong., tab. 42.
Lefiunea ovata Taylor MS., G. L. N., Syn. Hepat., p. 376. Moore,
Irish Hepat., p. 612. Pearson, Hepat. Brit. Isles, p. 42, plate 8.

Districts I. VI. — Vni. — — XI. XII.

]3ab, — On moss-covered trunks of trees, and on the larger Hepatic»
and Mosses.

I. Xaiamey: Carrington. Tore Waterfall: Taylor; and Sept. 1897:
McA. & Lett. Cromaglown abundant, and through the Xillamey
district : Moore. Among Rhacomitrium on Connor Hill near Dingle,
1873 (Lindberg and Moore) : Lindberg 1876 ; and 1881, F. W. M. &
McA. On the N.E. side of Brandon, 1875 : McA. Frequent in tlie
Dingle Peninsula : McA .1901. Bantry : Taylor. Dunboy Wood and
near Pulleen Cove, Co. Cork : McA. 1894.

VI. Cam Seefin, Co. Clare, plentiful: McA. 1896 «.

VIII. Nephin, May 1901 : Lett & McA. Bangore, slopes of
Devil's Mother, and on SHevemore, Achill, Sept. 1901 : Lett.

XI. On Frullania Tamarisci on Goat Island, Lough Eask, plentiful,
June 1903: McA.

XII. Near Belfast (Dickie) : Moore 1876. Slieve Donard: Lett
1890. On the stem of Holly at the Black Stairs on Slieve Donard,
Co. Down (Waddell) : Lett 1890. Glenariff, Co. Antrim: Waddell
1893.

7. Lejennea hamatifolia Hook., Dumort.

Jungermania hamatifolia Hook., Brit. Jung., tab. 64. Lejeunea
hamatifolia Dnm., Comm., p. Ill, 1822. Moore, Irish Hepat., p. 611.
Pearson, Hepat. Brit. Isles, p. 43, plate 9.

Districts I. — in. IV. V. VI. — VIII. IX. — XI. XII.

Sab. — On the trunks of trees and on bare moist rooks, and on the
larger Hepatics, such as Frtdlania.

I. Xillamey Woods, plentiful: Moore. On bare rocks, Connor
Hill near Dingle, with Lejeunea ealyptrifolia^ July 1873 (Lindberg and
Moore) : Lindberg 1875. In same locality, very fine, with perianths,
Sept. 1898 : Lett & McA. Mt. Eagle, July 1881 : F. W. M. & McA.
Loughanscaul, on Radula Carringtonii, Sept. 1898 : Lett & McA.
Brandon, June 1 900 : Lett & McA. Frequent in the Dingle Peninsula :
McA. 1901. Bocks near Bally bunnion : Stewart 1890.

McArdlb — A List of Irish Hepaticce. 413^

III. Brittas demesne, Queen's Co., 1891 (Bnssell): McA. 1892 a.

IV. Fowerscourt, Co. Wicklow : AToore.

V. Woodlands, Co. Dublin (Taylor) : Moore 1876. Close to the
waterfall in Omeath Glen, Co. Louth, rare: Lett 1890.

VI. Gleninagh, Co. Clare: McA. 1895 «. Tycooley Wood, Clon-
brock, Co. Oalway : McA. 1896 h,

VIII. Pontoon near Foxford, May 1901 : Lett & McA. Bangore,
slopes of Deyil's Mother, and Doolough, Co. Mayo, Sept. 1901 : Lett.
Kylemore, 1874: Moore.

IX. Olenfann demesne, Co. Leitrim, 1875 : Moore.

XI. Woods by River Trillick, Buncrana, March 1903 : Hunter.
Ooat Island, Lough Eask, June 1903 : McA.

XII. Olendun, 1836: Moore, and Glenarm and Colin Glen, Co.
Antrim, 1837: Moore. Glenoriff, Co. Antrim, 1889 (Waddell) :
Stewart 1895. Near the waterfall at the Black Stairs on Slieve
Donard: Lett 1890. ToUymore Park, Co. Down: Waddell 1893.

8. Lqeunea oaloarea Libert.

Lejeunsa ealcarea Libert, in Bory de St. Vine, Ann. des Sc. Nat.^
yol. Ti., p. 873, no. 1, tab. 96, fig. 1, 1820. Jungermania hamatifoUa
yar. eehinata Hook., Brit. Jung., 1816. Zefeunea echinata Taylor MS.,
G. L. K, Syn. Hepat, p. 345, 1844. Moore, Irish Hepat., p. 612.
Pearson, Hepat. Brit. Isles, p. 58, plate 16.

Districts I. V. XII.

Hob, — On limestone rocks and on Mosses.

I. Muckross demesne, on Thatnnium akpeeurum, 1863 (Carrington) :
Moore 1876. In the same place, and on the same moss, 1873 : Lind-
berg 1875. Boss I. : Scully 1890. Limestone rocks near Tralee, 1875 :
Moore. On rocks, Mt. Eagle, July 1881 : F. W. M. & McA.
Connor Hill near Dingle, 1897 : Lett & McA.

V. Limestone rocks at Woodlands, Co. Dublin (Taylor) : Hooker
1816. Omeath, Co. Louth : Lett 1890 ; Waddell 1893.

XII. Wall at the base of bridge over the Shimna River, ToUymore
Park, Co. Down, very rare (Waddell): Stewart 1881*. Glenariff,
Co. Antrim, 1893 (Lett and Waddell) : Stewart 1895.

9. Lejeunea Sossettiana Massalongo.

Lejeunea Bossettiana Massal., Nuovo Giom. Bot. Ital., vol. zxi.,
p. 487, 1889. Pearson in Joum. Bot., vol. xxvii., p. 352, tab. 292,
1889. Pearson, Hepat. Brit. Isles, p. 60, plate 17. McA., HepaticsB of

414 Proceedings of the Royal Irish Academy.

KoBs Island, Killamev, Irish Naturalist, vol. ix., p. 23, plate 1, figs.
1-6, February, 1900.

Districts I. V. .

Siab. — On limestone rocks and among Mosses, often mixed with
Z. ealearea.

I. Muckross demesne, May 1861 : Carrington. Boss I., KiUamey,
1889 : Scully 1890. On decayed stems of Mrtca^ and on damp peat
in the same place, yery rare. May 1899 : Lett & McA.

V. At Woodlands near Dublin, 1830 : Taylor.

10. Lejennes minutlMims Smith.

Lejunea minutissitna Smitb, Eng. Bot., tab. 1633. Hook., Brit.
Jung., tab. 52, excepting fig. 5, which is probably X. ulieina. Junyer-
mania ineanspieua Baddi, in Atti Soc. Mod., 1818. Moore, Irish
Hepat., p. 613 (under Lefeunea). Lejeunea minutissitna Sm., Pearson,
Hepat.'Brit. Isles, p. 61, plate 18.

Districts I. — III. VIII. — X. — XII.

Hah, — On the trunks of trees and on decayed wood, and on the
larger Mosses and Hepatics.

I. Xenmare : Taylor. On Ash trees, Muckross demesne : Carring-
ton 1863. Near Muckross Hotel on Beech, among ZyyodofUium,
and on rocks at Glena on Lefeunea Mackaii, 1878 : Lindberg 1875.
Boss I., 1893: McA. 1900. Tore Waterfall, on Met%geria^ and on
bark of trees, Sept. 1897 : Lett & McA. Brandon : Moore. Connor
Hill, 1873 : Lindberg 1875. Bumham Wood near Ventry, May 1894 :
McA. 1901. Hickson's Wood near Anascaul, Sept. 1898: Lett &
McA. Co. Cork (I. Carroll) : Carrington 1863.

III. Brittas demesne, Queen^s Co., 1891 : Eussell & McA.

VIII. Slievemore, Achill, Co. Mayo, Sept. 1901 : Lett.

X. On trees among Metzyeria, Famham demesne, and very fine
on Fndlania ; and at Eillakeen, Co. Cavan, on Hypnum eupressijfbrme :
McA. 1898.

XII. GillhaU, Co. Down (Waddell), CoUn Glen near Belfast
(Moore): Stewart 1888.

11. Lejennea microBCopics Taylor.

Jungermania mioroscopiea Taylor, in Mackay's Fl. Hib., part ii.,
p. 59. Taylor, in Hooker's Journal of Botany, vol. iv., p. 97, with
excellent figure, tab. 29. Z^'eunea mieraseopica Moore, Irish Hepat.,
p. 613. Pearson, Hepat. Brit. Isles, p. 63, plate 19.

Districts I. YI. — Vni. XII.

McArdle — A List of Irish Hepaticm. 415

Eah. — On the bark of trees, decayed wood, Filmy Ferns, Mosses,
and Hepatics.

I. Wood at Gortagaree, near Eillarney, on Hypnum hreutn, 183&
(Taylor) : Moore 1876. Cromaglown, 1849: Taylor and W. WQson.
Tore Cascade with Lophocoha hidentata : Carrington 1868 ; and Sept.
1897: McA. & Lett 1899. Killamey, common up to 1500 feet
on Slieve Mish: Scully 1890. O'Sullivan's Cascade 1873 : Lindberg
1875. Ross I., on Hymenophyllum, Nov. 1893 : McA. 1900. Glen
on Brandon : Moore. Very fine on the K.E. side of Brandon on
Diplophyllum alhteans^ June 1900 : Lett & McA. On Frullania, Connor
Hill, 1873: Lindberg 1875; and Sept. 1877: McA. Loughanscaul,
Sept. 1898 (Lett & McA.) : McA. 1901.

YI. Cam Seefin, Co. Clare, on Flagioehila spinulosa : McA. 1895 a.

VIII. On the bark of Alder and on Frullania Tamarisei, fertile,
Pontoon near Foxford, and on Nephin, May 1901 : Lett & McA.
On Deyil's Mother, and on Slievemore, Achill, Sept. 1901 : Lett.

XII. Glenariff, Co. Antrim, July 1889 (Waddell) : Stewart 1895.

12. Lejennes diverBiloba Spruce.

L^eunea diversiloha Spruce, Joum. of Bot., vol. xxv., p. 38, 1887.
LfjeuMa tninutissima var. mq/or Carrington, Trans. Bot. Soc. Edinb.,
vol. viii., p. 468, tab. 17, fig. 1. Lejeunea diversiloha Pearson, Hepat.
Brit. Isles, p. 56, plate 15.

Districts I. .

Siah, — On moist rocks and on the trunks of trees among Mettgeria^
&c., and on Mosses.

I. Tore Waterfall, 1842 : Spruce; and 1885 (Holt) : Spruce 1887a.
Very rare here, Sept. 1897: McA. & Lett. Tore Waterfall, Glena
and Eagle's Nest, Eillamey : Carrington 1863. On damp rocks among
Mettgeria conjugata, Connor Hill near Dingle, July 1881 : F. W. M.
& McA. ; and Sept. 1898 : Lett & McA. Loughanscaul near Dingle,
Coumanare Lakes and Bamanaghea Lough near Anascaul, Sept. 1898 ;
and at Lough I^alachan on Brandon, rare, 1899 : Lett & McA.

13. Lejeunea ulicina Taylor.

Jung&rmania tdieina Taylor, in Trans. Bot. Soc. Edinb., vol. i.^
p. 115, 1841. Lejiunea uUeina Taylor, in G. L. N., Syn. Hepat^
p. 387. Pearson, Hepat. Brit. Isles, p. 54, plate 14.

Districts I. II. IIL IV. V. — VII. VIII. IX. — XI. XII.

Hob. — On the bark of trees and on mosses.

I. Abundant in the Eillamey woods : Moore 1876. Loughanscaul

416 Proceedings of the Royal Irish Academy.

near Dingle among Hypnum on damp rocks, Sept. 1898 : Lett & McA.
Eare in the Dingle Peninsula: McA. 1901. About Cork, frequent
(Carroll): Carrington 1863.

II. In woods, Scarriff, Galtees, cum. per., plentiful ; and Glengarra
Wood, Galtees, Co. Tipperary, June 1902 : McA.

III. Cappard, Queen's Co., 1891 : Russell.

IV. Luggielaw and Powerscourt, Co. Wicklow : Moore. Killough-
rim Oak Forest, Co. Wexford, May 1899 : McA.

V. Woodlands near Dublin : Moore.

Vn. Lake at Brittas, King's Co., 1892 : McA.
YIII. Eylemore Castle demesne, 1874 : Moore. On the bark of
Alder at Pontoon, and on Nephin, May 1901 : Lett & McA.
IX. Glenfarm demesne, Co. Leitrim: Moore 1876.

XI. Lough Eask Woods, on Fndlania Tamariseiy June 1903 : McA.

XII. GiUhall and Castlewellan, Co. Down : Waddell 1893. Colin
Olen near Belfast : Moore.

14. Lejeonea calyptrifolia Hook., Dumort.

Jungermania calyptrifolia Hook., Brit. Jung., tab. 43. £ng.
Bot., tab. 2538. Zefeunea ealyptrifolia Dam., Gomm., p. 111.
Colura ealyptrifolia Dum., Recueil, p. 12. L^'eunea ealyptrifolia Moore,
Irish Hepat., p. 611. Pearson, Hepat. Brit. Isles, p. 64, plate 20.

Districts I. V. VIII. XII.

Hah, — In minute yellowish -green tufts on the bark of trees, on bare
moist rocks, on the stems of UUx near the ground, and on Fhdlania.

I. GlengarifP, Co. Cork (Miss Hutchins) : Hooker 1816. On rocks.
Upper Lake, Killamey, 1873 (Lindberg& Moore) : Moore 1876. Near
Dunkerron, 1836: Taylor. Tore Mountain on the stems of PtntM :
Wilson & Carrington. O'Sulliyan's Cascade, 1893 : McA. Near the
Hunting Tower, Killamey : Scully 1 890. Connor Hill, on bare moist
rocks by the " Doctor's Well," 1873 : Lindberg & Moore ; also July
1881 : E. W. M. & McA. ; and Sept. 1897-8 (Lett & McA.): McA.
1901. In Hickson's Wood near Anascaul, on the stems of AhieM
and PinfM, fertile: McA. 1894, and I. Nat, vol. iv., p. 73, 1895.
Loughanscaul, on Frtdlania, Sept. 1897 : Lett & McA. Bamanaghea
near Lispoll, on Diplophyllum albicans and Z. ovata, June 1899 (Lett
&McA.): McA. 1901.

V. Luttrellstown, Co. Dublin (Templeton MSS.) : Waddell 1893.

VIII. Co. Mayo, Sept. 1901 : Lett.

XII. Slieve Donard, Co. Down, very rare : Waddell in Brit. Assoc
Guide to Belfast, 1902.

McArdlb — A List of Irish Hepaticce. 417

Tribe 2. JTTVOEEMAiriKfi.

Sub-tribe 1. EADXTLRfi.

Genus lY. Badula Dumort.

1. Badnla voluta Taylor.

Radula voluta Taylor, G. L. K, Syn. Hepat., p. 253. Radula

xalapensis N. M., Lindberg, Hepat. Hib., 1875. Moore, Irish Hepat.,

p. 616. Radula voluta Tayl., Pearson, Hepat. Brit. Isles, p. 67,

plate 21.

Districts I VIII. — X. .

Hah. — On moist rocks and on the trunks of trees.
I. Dunkerron: Taylor. On boulders below Tore Waterfall
(Spruce) : Garrington 1863. Yery fine there and plentiful, forming
large yellowish patches, Sept. 1897 : McA. & Lett 1899. Rocks
bdow the Eagle's Nest, Cromaglown (G. E. Hunt), near Derry-
cunighy Cascade and Gortagree (Moore), O'Sullivan's Cascade and
Glena, 1873 : Lindberg 1875. Glena (Stewart & Holt) : Pearson 1902.
Derrynane and Mangerton, and to 2500 ft. on the Reeks : Scully 1890.
Bumbam Wood near Yentry, 1894 : McA. Mount Eagle Lake on
rocks, rare, Sept. 1897 : Lett and McA.

YIII. Pontoon near Foxford, May 1901 : Lett & McA.
X. Shores of Lough Cultra, Co. Cavan, 1893 : McA. 1898.

2. Badula Holtii Spruce.

Radula Holtii Spruce, in Joum. of Bot., vol. xxv., p. 209, 1887.
McAidle & Lett, HepaticsB of Tore Waterfall, Proc. R. I. Acad., 3rd
ser., vol. vii., no. 2, 1899, p. 328, plate 9. Pearson, Hepat. Brit. Isles,
p. 72, plate 24.

Districts I. YIII. .

Hob, — On moist shaded rocks and epiphytic on Dumorti&ra irrigua^
Jubula Sutehtnsiaf and Lejeunea Maekaii,

I. Tore Waterfall, KiUamey, June 1885 (Holt): Spruce 1887 «.
Within the spray of .same waterfall, associated with the same plants
and Radula Carringtonii^ very rare, Sept. 1897 : McA. & Lett.

YIII. Bengorm north of Killery Bay, Sept. 1901 : Lett.

3. Badiila aquil^^ Taylor.

Jungermania aquiUgia Tayl., in Trans. Bot. Soc. Edinb., vol. ii.,
p. 117, 1846. Radula aquiUgia Taylor, G. L. N., Syn. Hepat., p. 260.
Jungermanta eamplanata var. minor Hook., Brit. Jung., t. 81, iig. 17.

418 Proceedings of the Royal Irish Academy.

Radula aquiUgia Tayl., Moore, Irish Hepat., p. 617. Pearson, Hepat.
Brit. Isles, p. 74, plate 25.

Districts I. II. VIII. XII.

Wah. — On damp rocks and on the trunks of trees which are
often suhmerged, and on the larger Hepatics such as FndUmia
Tamarisei.

I. Not rare in the Killarney district: Moore 1876; and Carrington,
cum. per. Glena, male plants, 1873 : Lindberg 1875. Knockavohill
(Taylor): Pearson 1902. Near Waterville, Mangerton, Slieve Mish to
1500ft.: Scully 1890. Coomashana Lake, Dingle: Carrington 1863.
Connor Hill, 1877 : McA. On the west side of Brandon, April 1897 :
F. W. M. & McA. Mount Eagle 1898 : Lett & McA. Anascaul, on
Frullania Tamarisci: McA. 1894 ; and 1898 : Lett & McA. Bumham
Wood near Ventry, and Lough Nalachan, 1899: Lett & McA.
Derrymore Glen near Tralee, May 1899 : Lett & McA. Near Bantry
(Miss Hutchins): Hooker 1816.

II. On rocks, Galtees, Co. Tipperary, July 1902: Lett.

VIII. Bangore, slopes of the Devil's Mother, and on Slievemore,
Achill, Sept. 1901 : Lett.

XII. On wet rushes in the chasm below the waterfall at the Black
Stairs on Slieve Donard (Waddell) : Lett 1890.

4. Sadula Carringtonii Jack.

Radula Carringtonii Jack., in ** Flora," vol. Ixiv., p. 385, 1881.
Radula aquiUgia Taylor var. major Carrington, in Trans. Bot. Soc.
Edinb., vol. vii., p. 455, 1863. Lindberg, Hepat. Hib., p. 491. Badtda
Carringtonii Jack., Pearso^, Hepat. Brit. Isles, p. 76, plate 26.
McArdle & Lett, Hepat. Tore Waterfall, Proc. R. I. Acad., vol. v.,
no. 2, plate 8, 1899.

District I. VEIL .

Hah. — On damp shaded rocks, on the trunk of trees, and on the
larger Hepatics.

I. Tore Woods, Cromaglown, Glena, Tomes Woods, Killarney,
June 1861 : Carrington. Rocks near Tore Cascade, male plant, April,.
1892 : G. E. Hunt. On damp rocks with Metzgeria eonjugata
and Zefunea Mackaii, 1899: McA. & Lett 1899; and June 1885,
male plants: Stewart & Holt. O'Sullivan's Cascade, Glena, and
Cromaglown, Killarney, 1873 : Lindberg 1875 (under E.aquilegiarar.
major). Anascaul near Dingle : McA. 1894 ; and May 1898, fertile :
Lett & McA. Mt. Eagle, May 1898 : Lett and McA. Brandon near
the summit, on north-east side, 1900: Lett & McA. Lough.

McArdlb — A List of Irish SepaiietB. 419

Nalacban on Brandon, and on the rocky shores of Lough Duff in the
Brandon Valley, 1899 (Lett & McA.) : McA. 1901.

Ylll. Bangoroand Beyil's Mother, Co. Mayo, Sept. 1901 : Lett

5. Badula eomplansta Linn., Dumort.

Jun^ermania compUmata Linn., Sp. PL, 1699, 1753. Hook.,
Brit. Jung., tab. 81. Radula compUmata Dum., Comm., p. 112. Moore,
Iriah Hepat., p. 617. Pearson, Hepat. Brit. Ides, p. 78, plate 27.

Districte I. n. in. IV. V. VI. vn. viil ix. x. xi. xii.

Hah. — On the trunks of trees and on rocks.

var. minor McArdle (non Hook.).

A very minute and fragile plant of a light yellow colour, leaves
more convex, found fertile among Mettgeria on Boss I., Eillamey,
1899 : Lett & McA. On Slieve Glah, and in Eillakeen Woods, Co.
Caran, Oct. 1893: McA. 1898.

Sub-tribe 2. POSELLSJE.

Genus V. PoreUa Dillenius.

1. PoreUa IflBvigata Schrad., lindb.

Jungermania lavigaia Schrader, Sammlung, ii., no. 104, p. 6^
1797. Hook.,lBrit. Jung., tab. 35. Madotheca Uevigata Dumort.,
Comm., p. Ill, 1822. PoreUa Umgata lindberg, Muse. Scand., p. 3,
1879. Moore, Irish Hepat., p. 617. Pearson, Hepat. Brit. Isles»
p. 80, plate 28.

Districts I. IV. X. .

Hob. — On the trunks of trees near the ground, on stones and rocks
which aro often submerged.

I. Mountains near Bantry (Miss Hutchihs) : Hooker 1816. Near
Cork (I. Carroll): Moore 1876. O'Sullivan's and Tore Cascades,
Killamey, 1873: Lindberg 1875. Upper Lake and Dean Bridge:
Carrington 1863. Near Waterville : Scully 1890.

IV. Damp rocks, Lough Bray, Co. Wicklow (Moore) : McA. 1890.

X. Rocks below Benburb Castle, Co. Tyrone, 1880 : Lett.

Tar. inUgra Dill., Lindberg.

Glena, Killamey, on inundated stones by the margin of the Lower
Lake, 1873 (Lindberg) : Moore 1876.

m. I. A. PBOC., TOL. XXIT., SBC. B.J 2 Jf

420 Proceedings of the Royal Irish Academy,

2. Porella platyphylla Linn., Lindberg.

Junyermania platyphylla Jadh,, Sp. PL, ed. 1, p. 1134. Hook.,
Brit. Jung., tab. 40. Madotheca platyphylla Dumort., Comm., p. 111.
Porella platyphylla Lindberg, Muse. Scand., p. 3, 1879. Moore, Iruh
Hepat., p. 618. Pearson, Hepat. Brit. Isles, p. 84, plate 30.

Districts I. IV. V. VI. — IX. X. — XII.

Mob. — On rocks and stones, on the trunks of trees, and on moss-
covered banks.

I. Killamey, frequent: Carrington 1863. Tore Waterfall, 1897:
McA. & Lett 1899. Brandon, 1881 : F. W. M. & McA. Mount
Eagle Lake, 1898: Lett & MoA. Anascaul, 1894, and other places
in the Dingle Peninsula : McA. 1901.

IV. On rocks and trees, Co. Wicklow : Moore 1 876. On stones
at the summit of Dermott McMurrough's Castle, near Ferns, Co.
Wexford, 1897 : Canon Gibson & McA.

V. On rocks, Beauparc, Co. Meath, Sept. 1893 : McA.

VI. Common in South Clare: Stewart 1890. Kilronan, Aran
Islands: McA. 1895a.

IX. Ben Bulben, Co. Sligo: McA. 1880. Lough Allen, Co.
Leitrim, frequent : Stewart 1885.

X. Killakeen Woods near Cavan, 1893 : McA. 1898.

XII. On dry exposed rocks, Co. Antrim, 1834 : Moore. On a
bank at the railway embankment north of Dromore, Co. Down :
Lett 1890.

3. Porella Thiy's Dicks., Moore.

Junyermania Thuja Dicks., PL Crypt. Brit., fasc. 4, p. 19. Made-
thfica Thuja Dumort., Comm., p. 111. Moore, Irish Hepat., p. 618.
Pearson, Hepat. Brit. Isles, p. 82, plate 29.

Districts I. IV. XII.

Hckb. — On rocks and stones ; on the trunks of trees.

I. I^ear Lough Mangei*ton, 1885 (Holt & Stewart) : Pearson 1902.
Lough Fennehy, Dunkerron (Taylor) : Moore 1876. Brandon, 1864 :
Moore; and 1899 : Lett & McA. Anascaul, May 1894 : McA. Meant
£agle Lake neai* Dingle, 1898 : Lett & McA. Side of Gougane-Barra
Lake near Bantry (Miss Hutchins) : Hooker 1816.

IV. Lough Bray, Co. Wicklow : Moore ; and McA. 1890.

XII. On moist rocks by the side of a waterfall near Carrickf ergus,
Co. Antrim (Templeton) : Hooker 1816. Woodbum and Colin Glens,
Co. Antrim : Waddell in Brit. Assoc. Guide to Belfast, 1902.

McArdle — A List of Irish Hepaticce, 421

4. Porella rivalaris Nees, Lindberg.

Madothsea rivtdaris Nees, Nat. Eur. Leb., iii., p. 196. PoreUa rtvu-
hris Nees, Lindberg, Muse. Scand., p. 3. Pearson, Hepat. Brit. Islee,
p. 87, plate 31.

Districts X. — XII.

Hic^. — On wet rocks and stones and on trees near water.

X. Castlesbane, Co. Monaghan : Waddell.

XII. On trees by the River Lagan at Drumcro, Co. Down (Waddell) :
Stewart 1888. Saintfield, Co. Down (Waddell) : Stewart 1895.

5. PoreUa pinnata Dill., Lindberg.

Jungermania Porella Dicks., Trans. Linn. Soc, vol. iii., p. 230.
Jungermania Cordeana Hiiben., Hepat. Germ., p. 291. Madoth$ea
Porella Nees, Nat. Eur. Leb., p. 201. Porella pinnata Dill., Lindb.,
Hepat. Hib., p. 493. Pearson, Hepat. Brit. Isles, p. 89, plate 32.

Districts I. .

Hah, — On damp shaded rocks, on stones, and on aquatic mosses.

I. Near Fermoy, Co. Cork (T. Chandlee) : Moore 1876. South of
Ireland (Taylor & Moore) : Pearson 1902. Lough Mangerton (Stewart &
Holt): Pearson 1902. Connor Hill, 1873: Lindberg 1875. On wet
rocks, Anascaul near Dingle, 1894 : McA. Lough Adoon near Clohane
on Cinelidotuif Sept. 1897 : Lett & McA. Lough Nalachan on
Brandon, rare, Sept. 1899 (Lett & McA.) : McA. 1901.

Genus 6. Plenrozia Dumortier.
1. Pleurozis cochleariformis Weiss, Dumort.

Jungermania cochleariformis Weiss, PL Crypt., p. 123. Hook.,
Brit. Jung., tab. 68. Jungermania purpurea Scop., Fl. Carm. ii.,
p. 847. Phyeiotium eochleariforme Nees, Nat. Eur. Leb., iii., p. 79.
Pleurwda cochleariformis Dumort., Syll. Jung. Eur., p. 38. Moore,
Irish Hepat., p. 620. Pearson, Hepat. Brit. Isles, p. 91, plate 33.

Districts I. ly. V. VI. — VIII. XL XII.

Sab. — On damp bogs among heaths, and on the boggy mountain
sslopes in quantity.

I. Mountain near Lough Guitane, and Glena: Carrington 1863.
Connor Hill, 1873 : Lindberg 1875. Mangerton, 1873 : Lindberg 1875;
and the male plant (Stewart & Holt) : Pearson 1902. Common in
the Dingle Peninsula : McA. 1901.

IV. Lough Bray, Co. Wicklow, 1887: McA.

V. Dublin mountains, frequent: Moore 1876.

2M2

422

Proceedings of the Royal Irish Academy.

VI. Very fine on bogs near Clonbrock, Co. Galway: McA. 1896 i.

VIII. Ifaam Turc, Connemaxa, fertile in September : Wade Rar.
1804; and 1874: Moore. Croagbpatrick and Murrisk, Sept 1901:
McA. Pontoon near Foxford, and on Nepbin, May 1901 : Lett k
McA. Slopes of Devil's Motber, Doolougb, and on SHeyemore, Achill,
Sept. 1901 : Lett. Gentian HiU near Galway : McA. 1895«.

XI. Errigal and Bamesmore Gap, and plentiful near the ¥Mte
Cliffs, Lougb Belsbade, June 1903 : McA. Bog at Drumfriee, foot of
Slieve Snacbt. and Bulben Mountain, Marcb 1903 : Hunter.

XII. Abundant on moist boggy moors tbrougb Co. Antrim, 1836:
Moore. Rasbarkin bog, Co. Antrim (Lett & Waddell) : Stewart 1895.
Evisb Mountain (Lett): Stewart 1895. Rocky Mountain and Hen
Mountain, Co. Down, very rare : Lett 1890 (under Fleuroziapurpurts].
Abundant in Co. Deny (Moore) : Stewart 1895.

r?pi

Sub-tribe III. Ptilideee.

Genus VII. Anihelia Dumortier.

1. Anfhelia jolacea Linn., Dumort.

Jungermaniajulacea Linn., Sp. PL, ii., p. 1 60 1 . Hook., Brit. Jung.»
tab. 2. Anthelia julacea Dumort., Becueil, p. 18. Moore, Irish Hepat.
p. 636. Pearson, Hepat. British Isles, p. 94, plate 34.

Districts I. IV. VIII. XL XII.

Hah, — In the crevices of moist rocks and in large patches on damp
peat in alpine and sub-alpine situations.

'I. Magillicuddy's Reeks: Moore. Connor hill, 1873: Lindbei^

1875. Horse's Glen, Mangerton: Scully 1890. Common in the
mountainous parts of the Dingle Peninsula: McA. 1901.

IV. Lugnaquilla: Hart 1886.

VIII. Maam Turc, Co. Galway: Wade Rar. 1804; and Hoore

1876. Kylemore, Co. Galway, 1874 : Moore.

XL On rocky bank. River Trillick, Buncrana, March 1^02 :
Hunter. Bamesmore, Lougb Belsbade, Errigal, Goat Island nt^i
Lougb Eask, June 1903: McA.

XII. Slieve Donard and Bloody Bridge River, Co. Down(¥addtill ; :
Stewart 1888. Wbite River Glen, Slievenabrock, Slievenamaddy,
Bloody Bum glen. Hare's Gap, Deer's Meadow, Shansliere, Ccu
Down : Lett 1890.

var. gracilis Hook.

Hook., Brit. Jung., tab. 2, fig. 3. Moore, Irisb Hepat, p. <S36,
(under var. minor).

McArdle — A List of Irish Hepaticce. 423

Brandon : Moore 1876. On the west side of Brandon, April 1897 :
F. W. M. & McA. Connor Hill (Moore) : Carrington 1863.

Genns VIII. Herberta Gray & Bennett.
I. Herberta sdnnca Dicks., Gr. & B.

Jungermania adunca Dicks., PI. Crypt. Brit., fasc. iii., p. 12.
J. juniperina fi adunca Hook., Brit. Jung., tab. 4. Serherta adunca
Gr. & B., Nat. Arr. Brit. PL, p. 705^ 1821. /Schisma adunea Dmnort.,
Comm., p. 1 16, 1822. Merherta adunca Gr. & B., Moore, Irish Hepat.,
p. 635. Pearson, Hepat. Brit. Isles, p. 99, plate 36.

Districts I. IV. VIII. IX. — XI. XII.

Hah. — Growing in dense tufts on damp shelving rocks and banks
at high elevations.

I. Mountains near Bantry (Miss Hutchins) : Hooker 1816. Croma-
glown, Killamey, 1845 : Spruce. Connor Hill, 1873 : Lindberg 1875.
Common at high elevations in the Dingle Peninsula : McA. 1901.

IV. Lough Bray, Co. Wicklow (Moore) : McA. 1890.

VIII. Slievemore on Achill, and on the slopes of Devil's Mother, Co.
Mayo, Sept. 1901 : Lett.

IX. Gleniff and Glenade, Co. Leitrim : Moore.

XI. Fanet: Hart 1886.

XII. On the undercliff of Fairhead, Co. Antrim (Templeton in
Herb. Belfast Museum) : Stewart 1888.

Genus IX. Mastigopliora Nees.

1. Mastigopliora Woodsii Hook., Kees.

Jungertnania Woodsti Hook., Brit. Jung., tab. 66. BUpharozia
Jfoodsii Dumort., Becueil, p. 116. Mastigophara Woodsti Nees, Nat.
Eur. Leb., iii., p. 95. Moore, Irish Hepat., p. 635. Pearson, Hepat.
Bnt. Ifiles, p. 102, plate 37.

Districts I. Vni. .

Hah. — On damp banks and on shelving rocks in mountainous
places.

I. On the ascent to Mangerton, 1809 (J. Woods): Hooker 1816.
Moore 1876. Devil's Punch-bowl on Mangerton, 1829: Taylor; and
(Stewart and Holt): Pearson 1902. Carantual: Moore. Brandon,
Oct. 1829: Taylor and Wilson. In a gorge on the N.E. side near
the summit of Brandon, June 1900: Lett & McA. Connor Hill:
Moore.

VIII. Slievemore on Achill, Sept. 1901 : Lett.

424

Ptoeeedingg of the Mot/al Irinh Academif^

GenuB X. Blepharona Dumortier.
1. Blepharo2ia oiliaris Lmu., Ihimort,
Junfffrtnamn etlmri$ Hook., Brit, Jimg.i tab, 65*

Ble

mlmris Dumort*, Recueil, p. 16. Piiiidium mimr« Hampe^ Pf
HerCt 1836» BUpharma ciltnris Diimort., Mtiore, Irisli
p, 634, Pefttson, Hepat Brit. Ifllos, p, 104, plate 38,

Districts I. — — *

Hah. — On lieaths and among rocks in &ub*alpiBe plao^.
I. Mtmgcrtan, and Ross Bay^ KiUarney: CaningtoE 18G3
MoiintaiD, 1861: Moore, Brandon, 1820: Taybr*

Genus XI. Triehoeolea Bumortier, Kees,
1. Triehoeolea tomentella Ehrhart, Bninort.

Jfingermama tomenhila Ehr., Bcitr., ii., p. 150. Hool
Jung., tab. 36. TriehQi'oiea tommUUa Ehrh., Moore, Irish
p. 634. Pearson, Hepat. Brit. Isles, p. 108, plate 39.

DistrictB I. IV. VII. XL XIL

Sah.- — On wet rocks, and damp woods in abaded pbwei.

I. Killamej Woods, abundant: Moore, Cromaglowii
Lindberg 1875, Hanging in dense tufts over the roeks wi
spray of Tore Waterfall, Killamey, Sept. 1897: Lett «
Black Stairs bridge, Eemnare Bay : Scully 1 890.

IT. Lougb Bniy: Moore. Wooden bridge, Sept, 1892; J
Septi 1894 (Miss Constance Pirn) : McA* in L Nat., toL iii.
1894,

VIL Bog near Geaabill, King's Co. : HirA, 1892 3.

XI. Wood by Mulroy Bay, and Cratleigk Wood near Hilfo
1902 : HcA. Bridge End Glen, 1889, and Linsfoot Glen, B
Marcb 1903 : Hunter. Woods about Lougb Bask, June 190S

XIL Moifit sh^dy banks, Co. Antrim, 163H: Moore* Si
Co. Down, 1897: "Waddell. By a ri?ulet in Belvoir Park^ C
(Templeton), Glenarm, Co. Antnin (Brenan); Stewart 18
tbe Sbimna Biver in Tollymore Park (Lett), Hostrevor ^
and Moygaiinon Glen (Waddell), and the Spinkwee Eirer i
more Park, very rare (Stewart) : Lett 1890.

Genus XIL Blepharostoina Dumortier,
1. Blepharo stoma trichophyllum Linn., Dumort.
Jungermania irich&phyUa Linn., Sp. PL, p. 1601* Hoai
Jung.y tab« 15. Bhpharosttmit irwhophyllnm Dumort*, Eectia

McArdlb — A List of Irish Hepaticce.

425

Moore, Irish Hepat., p. 636. Pearson, Hepat. Brit. Isles, p. 112,
plate 41.

Districts I. IV. V. VIII. XII.

Hah, — On turfy heaths, in bogs among Sphagnum, and among the
.larger Hepatics.

I. Near Bantry (Miss Hutchins): Hooker 1816. Cromaglown,
KiJlamey (Carrington) : Moore 1876. O'SuUivan's Cascade, 1873 :
Lindberg 1875. Glencar and Tore Mountains, Killamey : Scully
1890. Brandon, fertile, and Mount Eagle, 1881 : F. W. M. & McA.
Lough Adoon, Sept. 1897, rare in the Dingle Peninsula : Lett & McA.

IV. Lough Bray, Co. Wicklow, 1877 : Moore ; and 1890, fertile :
McA., I. Nat., voL ii., p. 172, 1893.

V. Omeath Glen, Co. Louth, very rare : Lett 1890.

Vni. Eylemore, Co. Galway, 1874 : Moore. Nephin : Moore
1876 ; and May 1901 : Lett & McA. Slieyemore, Achill, Sept. 1901 :
Lett.

XII. Near Belfast (Templeton) : Hooker 1816. Qlendun, Co.
Antrim, 1 836 : Moore. ToUjrmore Park and Slievenabrock, Co. Down
(Lett), Colin Glen, sparingly (Stewart and Waddell): Stewart 1888.

Sub-tribe 4. TrigonanthesB.

Genus XIII. Lepidozia Dum.

1. Lepidozia cupressina Swartz.

Jungermania eupressina Swz., Prod. Fl. Ind. Occid., p. 144.
Jttngertnania reptans var. pinnata Hook., Brit. Jung., tab. 75. Lepi-
doiia iumidula Tayl., in G. L. N., Syn. Hepat., p. 206. Lepidozia
cupreuina Pearson, Hepat. Brit. Isles, p. 116, plate 43.

DiBtricts I. IV. V. VI. — VIII. — X. XI. — .

Hah. — Growing in dense cushions on the ledges of rocks, and on
damp shaded banks.

I. Near Bantry, fertile (Miss Hutchins) : Hooker 1816. Killamey
woods, abundant: Moore 1876; Carrington 1863. Glena and 0*Sul-
livan's Cascade, Killamey, 1873: Lindberg 1876. Mount Eagle,
Coumanare Lakes, and Loughanscaul in the Dingle Peninsula, rare
Sept. 1898 : Lett & McA.

IV. Lough Bray, Co. Wicklow, plentiful (Tumer) : Hooker 1816;
McA. 1890.

V. Among rocks, Ballykill, and in the Howth Demesne, Co.
Dublin, 1895 : McA. 1897.

VI. Clare Glen, Glenstall, Co. Limerick : Hart 1886.

426 Proceedings of the Royal Irkh Acadefntf,

Tm. Nephin, plentiful, Kay 1901: Lett t McJL

Mother and Slievemore, Aehill, Sept. 1901 : Lett*

X. SborcB of Lougli Allen^ Co- Leitrim : Stewart 1885.
XL Fortlaw Hill, Bancranu, March 1 9 OB, and CtUTodomL

Lough SwiUj, March 1901 : Hunter.

2, Lepidozia reptans Lmn,^ Btun.

JunqetmanmreptamLinn,f Bp, 1^1.^ 1^99 ^ 17*S3. Lfpida^k

Bum*, Recueil, p, 19, Peareou, Hepat. Brit, Isles, p* 119, ph

Bistricta L IL III, lY* V. YI, YII. YDI, IX. X. XJ, X

Ifah. — On banks in wood9» on decayed wood, among rooks i

shaded places ; a beautiful and distinct plant,

3* Lepidona Pearsom Spruce,

L»pida%ia Pmnoni Sprtiuo, in Journ* of Bot., voL six.,
IgBl. Zepidcam Wulfih&^n Lindh^t Soc. F* & Fl, Fens., 18

Bifltrict I, — ~^ — ■,

Mah, — On damp banks among rocks with Sphagnum m
liossei.

L Lough Buff in the Brandon Yallej, May 1699 (Lett t
He A* 1901. The only locality at preecnt known in Ireland.

4, Lepidozia Betacea Web., Mitt

Junfferi^mmu 9$tac^a Web,, Spic, Fl, Goett,, p, H3, Hool
Jung., tab. 8* Bkpharoitoma ^at^eea Buiuort., Becneil, 1835.
dos^ia setucm Web., Mitt,, Joum. Linn, Soc* t., p. 103. J
Hepat. Brit, Ieles,p , 124, plate 46.

Districts L II. IIL lY. Y, YL YIL YIIL XI. XI

Ifak — On turfy bogs, ehaded banks, decayed wi>od, oni
the larger Hepaticee,

I. Bog near Bantry {BfissHutcbius): Hooker 1816, Abtti
the Killaraey Woods : Moore and others. Common m the
Fcnineula: McA. 1901.

IL Knockmeildown and Galtee Mountains, Co. Tipperar
1902: McA,

IIL Ifear Cappand, Queen's Co, : McA. 1892 it,

lY. Lough Bray, Co. Wicklow (Turner) : Hooker 1816; t!j
Lindberg 1875 ; and McA. 1890,

Y, Boggy places on the Hill o! Howth: McA, lS9d<i. A
Mountain, Co, Loutih : Lett 1 890.

McArdlb — A List of Irish MepaticcB.

427

YI. Boon bog, Clonbrock, Co. Galway : McA. 1896 h Cam Seefin,
Co. Clare, frequent: McA. 1895a.

VII. Ard Bog, King's Co., 1890 : Russell.

YIII. Nephin, May 1901 : Lett & McA. Murrisk and Croagh-
patrick, Sept. 1901 : McA. Pontoon near Foxford, May 1901 : Lett
&McA. Slievemore, Achill, Sept. 1901 : Lett.

XI. Croghan Mountain, hill above Mintiagbs, Carradoan Wood,
Rathmullan, Glenalla Hill, Rathmelton, July 1902 : Hunter. Errigal,
June 1903 : McA.

XII. Basharkin, Co. Antrim, Sept. 1838: Moore; and (Lett):
§tt;wart 1895. Hen Mountain and Tollymore Park: Lett 1890.
By the Yellow River, Co. Down (Waddell), Anabilt bog (Templeton),
Lnd Slievenamaddy (Lett): Stewart 1888. North side of Divis
fountain, 1803 (Templeton) : Stewart 1888. Glendun, Co. Antrim :
^reoan & Lett.

var. ierttUarioides Hubener.
Jungermania sertularioides Mich., Fl. Bor. Am., ii., p. 278. Lough
doon, Co. Kerry, Sept. 1897 : Lett & McA. Murrisk near West-
)rt, Sept. 1901 : McA.

5. Lepidozia trichocladoB C. Miill.

Lepidcmia irtchocladoi C. Miill., in Hedwigia, vol. xxxviii., p. 197,
99. Macricar, in Joum. of Botany, vol. xl., p. 157, 1902.

Digtricts I. IV. V. VI. — VIII. IX. .

JToB. — On moist peaty bank, shaded by rocks or trees.

I. Mount Eagle near Dingle, June 1898 : Lett & McA. Croma-
wn, Killamey, 1877 : McA. O'SuUivan's Cascade, Eillamey, with
phalozia hilemiea, I^ov. 1893 : McA.
lY. Ix>agh Bray, 1880: McA. 1890. Lugnaquilla, 1896: McA.

V. BaUjkill, Howth, Oct. 1891 : McA.

VI. Cam Seefin, Co. Clare, July 1895 : McA.

VIII. Kephin, and Pontoon on Lough Conn, May 1901 : Lett &

A. Kylemore, 1874: McA.

TX- Ben Bulben, Co. Sligo, July 1880: McA.

Genus 14. Bazzanis Gr. & Bennett.

1. Bazzania trilobata Linn., Gr.

Jung&rmania trilobata Linn., Sp. PL, 1599. Hook., Brit. Jung.,
76. 2fastigohryum trilohatum Nees, in G. L. N., Syn. Hepat.,

428

Proee«iiHjf» of the Royal Irith Aeadtmjf.

p. 230. BmEania triMuta Gt. & B., Nat AiT. Brit P
Pearson, Hepat Brit Isles, p, 12S, plate 47*

BiBtricte 1. — m. IT, w Ym. — — XI, xn

Muh. — In dump rocky placei^ iluided banks, and on deca
ofleii fotmiDg dens« patches in subalpuie sittiatiODit

I. On the summit of Magil]iciiddj'sB£»eks: Tajlor 1S31
ney Woods, abtmdaiit : Moore 1876* Ol^m and Cromaglo
Uiidberg 1875, Qlencar: Scully 1890. Conunon in V
Peninmila (Lett k McA.) : MeA, 1901,

III. Near tlie Bridge at Qraigue, Co. Carlow, among rcM
1896 a.

lY. Lough Bray and Seven Churchcts (IToore) ; IC«A. I

V, Killakee Glen, Co. Dublin. 1894: Me A.

TIIL Xylemore^ 1874: Moore. On Lettery Moimti
mara : Wade, Bar. 1804.

XI. Macamish Wood, March 1902: Hunter.

XII. Slemigh, Co, Antrim; and Bart, Co* Deny, rare, 1ft
Tolly more Park, Co. Down; Stewart 1888, Thomas Mou
rare (WaddeU & Lett): Lett 1890.

2. Bazsania triangulam SchI,, Liodb.

Jun^imtania trmti^ularii Schleicher, PL Crypt HeiT-,
Jungtrmmiia dejkx^ Mart.i Fit Crypt* Erlang., p. 125, t«l
Bmsmnia trtan^idarts Scb*, Lindbet^, Hepat. in Hibera

1875. Pearson, Hepat Brit. lales, p. 130, plate 48.

Districts I. — — YUI. IX. X. — — .

Mab, — On damp shaded banks among reeks, and amoii|

Hepatieie.

L Tore Mountain and Cromaglown, 1863 (Camngto

1876. O'SuIHtmi's Cascade : Moore, Upper Olencar r S
MacgilMcuddy's Keeks: Hart 1886. Brandon (Moore);
1875 ; and 1895 : McA. Connor Hill, 1873 : lindherg 1875
the Dingle Peninsula : MeA. 1901.

YIII. Twelve Bens: Hart 1886. J^ephin, May 1901 : I

IX. Ben Bulben: Moore.

X* Plentiful on Slievenierin, Co, Leitrim : Stewart 1 8

var. mn&vam Kees, in G, L. N., Syn. Hepat, p* *2

CarriugtOQ tk Pearson, Exsicc, no. 124.

On damp peaty banks among rocks, Mount Eagle 1
I8S8 : Lett & McA, Connor HiU, Sept, 1897 : Lett & M

MgArdlb — A Lkt of Irish Hepaticte.

42»

yar. devexum Nees, in G. L. N., Syn. Hepat., p. 232.

On damp peaty soil among rocks, Brandon : Moore. Connor Hill :
lully 1890.

3. BatsaniE tricrenata Wahlenberg.

Jungermania tricrenata Wahlenb., Fl. Carp., p. 364, no. 1207.
mania tricrenata Pearson, Hepat. Brit. Isles, p. 132, plate 49.
cA., Hepat Dingle Peninsula, p. 306. 1901.

DifitrictB I. VIII. .

JTo^.— On damp banks among rocks.

I. Killamey, June 1885 (Stewart & Holt) : Pearson 1902. Brandon
til Pleurozia, April 1897 : F. W. M. & McA. Coumanare Lakes and
Tnanaghea Lough, 1898, Lough Duff and Lough Nalachan on
ondon, May 1899 : Lett & McA.

Till. Kephin, and Pontoon near Foxford, May 1901 : Lett &
A. DeyiVs Mother and on Slievemore, Achill, Sept. 1901 : Lett.

4. BaCTftnia Fearsoni, Stephani.

Moitigohryum Pearami Stephani, Hedwigia, 1893. Banania
rsoni Steph., Pearson, Hepat. Brit. Isles, p. 133, plate 50.

Districts I. .

Sab. — On banks and rocks.

L Eagle's Nest, Horse's Glen, and Cromaglown, Killamey, June

5 (Stewart & Holt) : Pearson 1902.

Genus 15. Kantia Gr. & B.

1. Kantia Triohomanis Linn., Gr. & B.

Vnium IVichomanis, Linn., Sp. PL, p. 1579. Jungermania Tricho-
is Dicks., PI. Crypt. Brit., fasc. iii., t. 8, fig. 5. Eiantia Trieho-
'm, Gr. & B., Nat. Arr. Brit. PL, p. 706. Kantia Triehomanis
., Pearson, Hepat. Brit. Isles, p. 135, plate 51.
Hstricte I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.
Tab. — On damp clay banks, and among damp rocks, and on the
r HepaticflB, common.

2. Kantia argnta, Mont, et I^ees, Lindberg.
'algpogeia argtUa, Montague et I^ees, Nat. Eur. Leb., iii., p. 24.
malit orgutus Dum., Hepat. Eur., p. 117. Kantia arguta Lindb.,
Fepat. Hibem., p. 307, 1875. Moore, Irish Hepat., p. 632.
K>n, Hepat. Brit. Isles, p. 139, plate 53.
(istricts I. IL III. IV. V. VI. VII. VIII. XI. XII.

dbj bttniks, and amoi^g the I^rg^

L C^v^igbw*, SUkmeT : SenUy ISdO. Itoas L J
4 McAO : MeA. 19D0. O'S^tw'i Cbacade, Glam, n^
Omv HBl^ ISTS: Lttdbetf; 1875. Connor HiU, 1S8
M cA« CismB flmwigk the Bmi^ Feniimiila i MeA. 1
WmA «1 Bm Ial»i: McA. 1894.

II. IToods limit ScftftiSf Galtee Mountaijis, Jim

UL Ba&%<il • stzcan 00 Stiire Bloom, Queen- e C<
1892 «. Gndgne, Co. Culov : Mc^. 189€ a.

lY. EXbotm WiM^ Bfi«r Fovs, 1896 : Greeme,
Slaiiii^ fiiT«r near Bdamme Jimetiini, itnd neur Eimi
lal Hie Uifin BiT^r tie«r l^oe^roe, aad m EUloughrim
r» Maj 1899 : McA, Luggiekw, Co, Wtd
Mooire. 1m^ Bimy, 1S7S : He A,

V. KiUakee glen, Ca. DnWa, among Mh^ncht^d
1890: McA.

TL C^m Seefisi, Co. Qhae, IS95 : Ph)i. T. Joh^
Lon^ Ckmrib. Co. Qf^wt^y : M«!A. 1^95 0^

YII. Ani bog, KingV Co,, 1891 : RiiSiell.

Till. KepbiB, and Pontoom near Foxjord^ Ha}
Me A. BeTil's Mother juid Acbill Igland. Sept. 190!

XI. Damp clay banks at Mount Charles, and G
Longb £aak, June 190S: McA.

XII. Mountains abore Newcastle and in BaU;
Stewart 1888. Slieve Donaxd, Hen Mountain, Ww
Btairs on Slieve Donand, Slieve namaddj^ : Lett 1 69(>,
by the Yellow Water, Co, Down (WaddeU): Lett 18
HOI, May 190S: Htmter,

Genus 16. Caphalona D amort.

Bnb-geEUua 1. Eucephalozia Spruce*

1 . Cephalona catenulata Huben.

Jtm^m^mama eaimulaia Hiiben., Hepai» Germ-, ]
mania ndma TayL, Lond. Joum. Bot, 1846, p. 1
mUnulaia Hiiben., Spruce on Cephehsia^ p. 3S,
Brit. Islea, p. 144, plate 44.

Dietrietfl I, IL IlL IT. V, VI, YII. VHL IX. J
Ilah, — On dompi shaded banka among rocks, and c

McArdle — A List of Irish Hepaticce,

431

I. Cromaglown and other places in the S.W. (Taylor, Moore &
)pace) : Spmce 1882. Finnahay River near Kenmare (Spruce &
lylor): Carrington 1863. O'Snllivan's Cascade, 1873: Lindberg
»75. Upper Olencar : Scully 1890. Brandon: Moore. Lough Adoon
arClobane, 1897 : Lett & McA. Bamanaghea Lough near LispoU,
d Mount Eagle, 1898 : Lett & McA. Derrymore Olen near Tralee,
99 (Lett & McA.) : McA. 1901. Caha Mountain and Dnnboy Wood,
.Cork: McA. 1894.

II. Enockmeildown Mountains, Co. Tipperary, June 1902 : McA.
IIL Graigue, Co. Carlow: McA. 1896 a.
IT. Lough Bray, Co. Wicklow : Moore 1876.

V. On damp banks between rocks with Kantia TriehomaniSj Baily
rhthouse, Howth : McA. 1893 a. KillakeeOlen, Co. Dublin : Moore
6. Kelly's Glen, Co. Dublin, Aug. 1896 : McA.

VI. Doon bog, Clonbrock, Co. Gal way : McA. 1896 3. On peat,
Q Seefin, and Ballyvaughan, Co. Clare, very fine: McA.
5 a.

VII. Aid bog and Killoigh bogs. King's Co., Nov. 1891 (Eussell) :
L 1892 a.

nil. Kylexnore, Co. Galway, 1874 : Moore. Diamond Mountain,
I : McWeeney. Lachan bay, Co. Mayo : Moore. I^ephin, May
f : Lett & McA. Murrisk, Sept. 1901 : McA. Pontoon by Lough
1, May 1901 : Lett & McA. Achill, Sept. 1901 : Lett.
X. Gleniff, Co. Leitrim : Moore.

I. On decayed wood on shores of Lough Cultra, on Slieve Glah,
it Ballyfaaise, Co. Cavan, amoug JDiplophyllum, 1893 : McA. 1898.
LI. CroghanMountain8,Bathmullan, July 1902 : Hunter. Gartan
d and Bunlin Waterfall near Milford, Sept. 1902 : McA.
Ill, Bocks near the head of Glenbush, Co. Antrim, 1888 : Moore
&r C. recluaa Tayl.). Bog at Ballygenan, Co. Down, plentiful,
: Lett & Waddell.

2. Cephaloaa pallida Spruce.

ephalo%ia caUnulala var. pallida ( C, pallida nobis in hb.), Spruce
phalaiiaf p. 34, 1882. Pearson, Hepat. Brit. Isles, p. 146,

45.

IstricU I. IV. V. VIII. XI. — .

ah. — On turfy bunks among rocks.

Brandon, 1875 : McA. Mount Eagle, June 1898 ; Lough Adoon
Clohane, Sept. 1897; Derrymore Glen near Tralee, 1898:
tMcA.

433

^OtM^gmtTrfAArnkmi

IT. I^fl^Bof^Oi. Wkk^v» 1979: MeA, 1890.

T. T!i iiM| I iiiif iMimt l^^te jnTmYi if Ml
7^1^ 18M: XcA.tin.

YIIL lAC^« ^r«0^1i^{lfMle):fte9lNl W

¥T rkmiiMTrini, TTiTVir -. f'j "^^ Ha

, 1^ la^ IS35. C. Htf^M Li

^ t7. e riirfa^fti Wmam, Mef^V BnL Uai, p. I
HigMlftl^n.^ v. — TILVIIL— —XI.

<•

1875 (o^r a wdifjflyi), fllwwr; Sesify 1890,
side of iMiia^ Imi liOO, at To<e^>"g<^"^ «>^
bfcf» Bcflr OouM^ HiCl nre. Srpt IBM: Lett & M<
rL 6knsuimWooi,Gd!toeM£Bs^uiliK«rB«jlo«|
duvmlftM^ Juel9M: lUA.

T. :rar Ite CtmsM, BMa^ Oow BnUin : IfcA. )

VIL <MtbeGiiElbof aMrTttlknm, Xing^sCe.:

Tm. ITepkiB^ Ibj 1901 : Lett It McA. Boobai

Utt. Aehill, Sept. 1901 : L«tt JNsUmh Be«r Foxlk

Lett 4 McA. Wff^Mt JwiMB, 8ipC. 1901 1 ICcA

XL SiU^aM W«od, Rithtalliii, Jolj 1902 : Q
Wood near lfilf»ii, SeyL 190S : McA.

4, Cepialimft liiciispidAti Umul^ P , :.

Ai^vfiMHM ^MJijpt;M# LiaiL, Sp. PL, lSt9, 17M
Jiaf., tak IK CSqpiWMi HmnpiimU Btui., Bmial, |
HepsL Bht. Iflks, |h. I M, fl«fe» 47 .

Dlitricfei L IL IIL IV. V, VL TU. YIU. IX X.

A). — On iknip twnks, bogs, mi on dec&jea wood,
lo tii^ ^errstioiis*

Ttr. M;p>tr Cftmfegton.

IftpluB, Cow Uayo (Mocm) : Caitii^gtsii ISil^

McArdlk — A List of Irish Hepaticce. 433

yar. rigidula Garrington.
Cromaglown, Eillarney (Carrington) : Moore 1876.

yar. setulosa Spruce, on Ceph., p. 42.

Caha Mtn., Co. Cork : McA. 1894. Doon bog, Clonbrock, Co.
}alwaj: McA. 1896 3. Ard bog, King's Co., 1891 (RusseU) : McA.
892 a. Killakeen, Co. Cayan, 1893 : McA. 1898.

yar. ienuirama Carrington & Pearson, Exsicc, no. 252.

On a peaty bank between Emalongb and Incb, Co. Kerry, April
898: Lett&MoA.

yar. (?). A minute reddisb plant found among the rocks near the
>aily Lighthouse, Howth : McA. 1893. Ireland's Eye, Co. Dublin,
S93 : McA. Summit of Slieye Glah, Co. Cavan ,1893 0 : McA. Connor
[ill on the hard ground, 1898 : Lett & McA. On the pathway in
Qloughrim Oak Forest, Co. Wexford, May 1899 : McA.

Always sterile. This may proye to be a distinct species.

5. Cephalozia Lammersiaiia Hiiben., Spruce.

Jun^ermania Zammertiana Hiibener, Hepat. Germ., p. 165. J.
TuspidaiaYai. uliginoM Nees, Eur. Leberm., ii., 253. J, hictupidata,
Lg. Bot., yol. xzxii., pi. 2239. Cephalozia Lammersiana Hiiben.,
irace on Ceph., p. 43. Pearson, Hepat. Brit. Isles, p. 153, plate 48.

Districts I. II. III. IV. V. VI. VII. VIIL — X. XI. XXL

Hah. — Damp boggyjplaces.

I. O'Sulliyan's Cascade, Killarney, 1869 : Moore; 1880: McA. ; and
iilly 1 890. Bog on Connor HiU : Moore; and Sept. 1898 : Lett & McA.
tar Brandon Head, Anascaul, Bamanaghea Lough, Mount Eagle, and
i bogs about the Coumanare Lakes, in the Dingle Peninsula, 1897-8

tt & McA.) : McA. 1901. Dunboy Wood, Co. Cork : McA. 1894.

II. Oaltees, Co. Tipperary, June 1902 : McA.

III. Slieye Bloom, Queen's Co. : McA. 1892 a.

IV. I«oiigh Bray, Sept. 1897; Glenmalure, May 1896 ; and wood by
Slancy Biver near Enniscorthy, Co. Wexford, May 1899: McA.

V. Killakee Glen, Co. Dublin, 1874: McA. Boggy places on
Hill of Howth: McA. 1898 a. Kelly's Glen, Co. Dublin, Aug.

>6 : McA.

TI. Cam Seefin, Co. Clare: McA. 1895 a.

VII. Ard bog. King's Co., 1891 (Eussell) : McA. 1892 a.

!•

434 Proceedings of the Royal Irish Academy.

VTII. Kylemore, July 1869 (Moore) : McA. 1880 (under C. iwa-
pidatay&r. uliginosa). Doonbog, Clonbrock, Co. GkJway: McA. 1896 J.
Nephin, and Pontoon near Foxford, May 1901 : Lett & McA. Doo-
lougli, Co. Mayo, Sept. 1901 : Lett.

X. Boggy places about Ballyhaise woods and on Slieve GIbIi, Co.
Cavan, October 1893 : McA. 1898. Slieve Gullion, Co. Armagb
McA., Ir. Nat., vol. 3, March 1894.

XI. Carradoan Wood, Rathmullan, July 1902: Hunter. Garta:
Wood and Cratleagb Woods near Milford, 1902 : McA.

XII. Abundant on Slieve Croob : Waddell in Guide to Belfast.
1902.

6. Cephalozia hibemica Spruce MS.

Pearson, in Irisb Naturalist, vol. iii., p. 245, plate 6, 18^^
Pearson, Hepat. Brit. Isles, p. 155, plate 49.

Districts I. .

Mah. — On damp banks among Mosses.

I. Among Plagiotheeium Borrerianum, Xillamey, 1865: ^^^'
Killamey : Scully 1890. O'Sullivan's Cascade, Nov. 1893: McA

7. Cephalozia connivens Dicks.

Jungermania connivens Dickson, PI. Crypt. Brit., hac, iv., p. ^'-^
tab. 2, fig. 15. Hook., Brit. Jung., tab. 15. Eng. Bot, Ub.a4>t'
Cephalozia connivens Dicks., Spruce on Ceph., p. 46. Moore, Irij*^
Hepat., p. 626. Pearson, Hepat. British Isles, p. 157, plate 50.

Districts I. II. — IV. V. VI. VII. VIII. — X. XI. XH.

Mah, — On wet peaty banks, bogs, and on decayed wood.

I. Mount Eagle, July 1881 : F. W. M. & McA. Derrymore Glrc
near Tralee, May 1899 (Lett & McA.), rare or overlooked: Mci.
1901.

II. Galtee Mtns., Co. Tipperary, June 1902 : McA.

IV. Lough Bray: McA. 1890. Altadore Glen, Co. Vicklo^
1887: McA.

V. Bog on the Sutton side of Howth, very rare : McA . 1893i

VI. Bogs about Clonbrock, Co. Gal way : McA. 1896*.

VII. Killeigh Bog, King's Co., 1891 : Kuasell. Welsh Island ne^z
GeashUl: McA. 1892 a.

VIII. Marsh in Connemara (Mackay): Hooker 1816. Kjlemore.
1891 : McWeeney. Pontoon near Foxford and on Nephin, May l^^'" -
Lett & McA.

X. Derrycrow bog, Montiaghs, Co. Armagh, 1885 : Lett. Top c.:
Camlough Mtn., very rare : Lett.

McArdle — A List of Irish Hepattcm. 435

XI. On Sphagnumy Garradoan Wood, BathmuUan, July 1902:
Hunter.

XII. Near Belfast (Templeton) : Hooker 1816. Bogs between
Swatragh and Eilrea, 1834 (Moore): Stewart 1888. Slieven^-
maddy, Co. Down (Lett) : Stewart 1895 ; Lett 1890.

8. Cephalozia onrvifolia Dicks., Dumort.

Jungermanta curvifolia Dicks., Pi. Crypt. Brit., fasc. ii., 15, tab. 5.
Hook., Brit. Jung., tab. 15, et Suppl., tab. 1, ex parte. Jungermania
-fffliwri Mart., Fl. Crypt. Erlang., p. 172, tab. 6, fig. 46. Cephdma
curpi/olia Dum., Recueil, p. 18. McArdle, Hepat. Howth, p. 115,
plate 4, figs. 7 to 13. Pearson, Hepat. Brit. Isles, p. 159, plate 51.

Districts I. IV. V. VIII. IX. — XI. XII.

Mab. — On decayed wood, damp banks among heather, sheep-
pntha, &c.

I. On decayed wood by a mountain lake near Bantry (Mis&
Hutchins) : Hooker 1816. Dunboy Wood and banks of the Pulleen
River, Castletownbere : McA. 1894. O'Sullivan's Cascade, Killamey^
1873 : Lindberg 1875. Glencar and Killamey, frequent : Moore ; an^l
Really 1890. Connor Hill: Moore; and Sept. 1897 : Lett & McA,
LoQgh Nalachan on Brandon, 1899 ; and Mt. Eagle, Sept. 1898 : Lett &
tfcA. Rare in the Dingle Peninsula: McA. 1901.

IV. Altadore Glen : McA. 1889. On decayed wood, Lough Bray^
879 (F. W. M.): McA. 1890. Douce, Co. Wicklow, fertilis
897: McA.

V. On shallow peat under the branches of Erica, BaUykill, and
ther places on Howth : McA. 1893 a (with fig.).

VIII. Kylemore, 1874 : Moore. Pontoon near Foxford and on
^ephin. May 1901 : Lett & McA. Doolough, and Slievemore on
L'hill, Sept. 1901 : Lett.

IX. Qlenade, Co. Leitrim : Moore.

XI. Qartan Wood near Milford, Sept. 1902 : McA. Very fine on
unk at waterfall on Errigal and Goat Island near Lough Eask, June
903 : McA. Ned's Point, Buncrana, March 1902 : Hunter.

XII. Crannies of rocks, Moume Mtns. (Templeton) : Hooker 1816.
lieve Bing:ian, Shanslieve, and Hen Mountain : Waddell in Guide to
elfast, 1902.

9. Cephalozia Franoiaoi Hook., Dumort.

Jungermania Franeisei, Hook., Brit. Jung., tab. 49. Cephalozia
Vancisci Dum., Recueil, p. 18. Moore, Irish Hepat., p. 624.

B.X.A. F&OC., TOL. XXIT., SEC. B.] 2 N

i36

Proceedings of the Royal Irish Academy.

McAidle, Hepat. Howth, p. 112, plate 3. Pearson, Hepat Bnt.
Isles, p. 163, plate 52.

Diatricts I. V. VIIL Xn.

JEM. — Damp peaty banks among rocks.

I. Near Bantiy (Miss Hntchins) : Hooker 1816.

y . Side of a shallow channel on a small bog near Ballykill planta-
tion, Howth, fertile : McA. 1893 a.

Vin. Boggy place among rocka at Pontoon on the shore of Loogti
Conn, May 1901 : Lett & McA.

XII. Kinahalla, June 1893, and Deer's Meadow in the Moumr
Mtns., Co. Down, 1886: LeU.

10. Cephaloiia fluitana Nees, Spruce^

Jtmyermania Jiuitans Nees, in Syll. Ratisb. /. inJUUa Ysi.jUit^
Nees, Eur. Leber. Cephalazia ohtusiloha Lindb., Bot. Not., 1872.
Cephahaia eladorhtsam Spruce, McArdle, new or rare Irish Hepat^
Sci. Proc. R. Da Soc., vol. iii., p. 3, fig. 6, 1880. Cephalma JluitiM
Nees, Spruce on Ceph., p. 50. Pearson, Hepat. Brit Isles, p. 16-3,
plate 53.

Districts I. V. — VII. VUI. — X. — XU.

Mob, — On Sphaynum in wet bogs, often floating.

I. Amongst Sphaynum ewtpidatum yar. pUunosum on Knockiiageebi>
bog near EiUamey, Jan. 1897 : Praeger.

V. Wet bog, Ballykill, Howth : McA. 1893«.

VII. Bracklin bog near Killucan, Co. Westmeath, on Sphap^^
'ubellum-, McA. 1880.

VIII. Floating in bog-holes near Kylemore with CephaUxia multi-
flora and C. Zamm&rsiij 1869 (Moore) : McA. 1880.

X. Camlough Mtn., Co. Armagh, 1887 : Lett,
xn. Giant's Causeway, 1836 (Moore): Ord. Surv. CollectioE
(under Junyertnania inflata^ aquatic form).

Sub-genus 2. Odontoaohiama Dumort

11. Cephalozia Sphagni Dicks., Spruce.

Jungermania Sphayni Dicks., Crypt. Brit. Hook., Brit. Jm-
tab. 33. Odontoschisma Sphayni Dum., Kecueil, p. 19. CephaJm^
{Odontoschistna) Sphayni Dicks., Spruce on Ceph., p. 60. Fear^''.
Hepat. Brit. Isles, p. 171, plate 55.

Districts I. II. III. IV. V, VI. Vii. viii. IX. X. XI. xn.

McArdlb — A List of Irish Hepaiicw. 437

ffah. — On bogB in dense patches, and more frequently among
Sphagnum, Probably not so common in the northern counties as
elsewhere.

12. Cephalozia denudata Nees, Spruce.

Jungirmania denudata Nees, in Mart. Crypt. Erlang., p. 14. /.
Sphagni Hook., Brit. Jung., Suppl., tab. 2. Cephaloiia denudata
Spruce on Oeph. Pearson, Hepat. Brit. Isles, p. 174, plate 56.

Districts I. V. VIII. XI. — .

Hob. — On decayed wood, more frequent on damp peat on worked-
out bogs ; and among rocks.

I. Bere Island: McA. 1894. Bog between Emalough and Inch,
Co. Kerry, May 1899 : Lett & McA. Extremely rare in the Dingle
Peninsula: McA. 1901.

Y. Plentiful on peat among rocks, fertile, Ballykill, Howth, 1891 :
MoA. 1893 a.

VIII. Corslieve Mtn. nearBangore, Co. Mayo, 1859 : Moore 1876.
Pontoon near Foxford, May 1901 : Lett & McA.

XI. Near Buncrana, 1903: Hunter.

Sub-genus 3. Cephaloziella Spruce.
13. Cephalozia divaricata Smith, Dumort.

Jungermania divarteata Sm., Eng. Bot., tab. 719. Jungermania
guacsa Roth, Trent. Fl. Germ., i., p. 387. Hook., Brit. Jung., tab.
2. J. Starkii Herb. Funck, Nees, Nat. Eur. Leb., 1 1, 225. Cepha-
nia divarteata Sm., Dum., Uecueil, 1835. Pearson, Hepat. Brit,
sles, p. 177, plate 57.

Districts I. IV. V. VI. VII. VIII. — X. XI. XII.

Hob. — On wet rocks, banks, decayed wood, and on the larger
[epaticce.

I. Cromaglown, Killamey : Moore. Waterville and Glencar: Scully
590. Tore Waterfall, Sept. 1898 : McA. and Lett 1899. Ross I.,
^99 (Lett & McA.): McA. 1900. Brandon: Moore; and 1897:
ett & McA. Hickson*s Wood near Anascaul, 1898, and Connor Hill,
i97 : Lett & McA. Derrymore Glen and bank between Emalough
id Inch, May 1899 (Lett & McA.) : McA. 1901. Near Bantry (Miss
utchins) : Hooker 1816 (under J, hyseaeea).

IV. Woodenbridge, Co. Wicklow, 1894 : McA, Killoughrim Oak
>re8t, Co. Wexford, 1899 : McA. 1903.

V. Howth, Ireland's Eye and Dalkey Island : McA. 1893 a.

VI. Gam Seefin, Co. Clare : McA. 1895 a.

2N2

438

Proceedings of the Royal Irish Academy.

i \ t

VII. KiUeigh and Ard bogs, King's Co., 1891 (RnBseU): McA.
1892 a.

VIII. Doon Bog, Clonbrock, Co. Galway : McA. 1896 J. Pontoon
near Foxford and Nephin, May 1901 : Lett & McA. Mumsk, Sept.
1901,: McA. Slievemore on AchiU, and Devil's Mother, Co. Mayo,
Sept.' 1901: Lett.

X. Lakelet near the top of Camlongb Mtn., very rare : Lett 1890.
Slieve Glah, Co. Cavan, 1893 : McA.

XI. Bathmullan, July 1902 : Hunter, Gartan Wood and Columbkil
Lake near Milford, Sept. 1902 : McA.

XII. About Belfast (Templeton) : Hooker 1816. Slemish and
Fairhead, Co. Antrim : Moore 1876. Glenarm (Dickie) : Moore 1876.
Claggan Woods, Co. Antrim, 1834: Moore (under /. byMoeea). Stones
on Slemish: Stewart 1888. Colin Glen (Waddell): Stewart 1888.
Hen Mtn. and bog between Hill town and Rathfriland: Lett 1890.
Tollymore Park, Co. Down (Waddell) : Stewart 1895.

var. Starkii Spruce.

Jungermania Starkii Funck, Nees, Hepat. Eur., ii., p. 223 ; Syn.
Hep., p. 134. J, Ghrimsulana Jack., in G. et R., Hepat. Eur.

Hah. — On rocks and damp banks.

Mt. Eagle and Coumanare Lake, 1898: Lett & McA. Lou^
Nalachan, Brandon, 1899: Lett & McA. Bere Island; McA. 1894.
Ballyvaughan and Cam Seefin, Co. Clare: McA. 1895a. Mournc
Mtns. and Carlingford Mtn. : Lett 1890. Sallagh Braes, Co. Antrin,
1837 : Moore, Baily Lighthouse, Howth, on a damp bank, 1893:
McA. 1897. On rocks, Bathmullan, July 1902, and behind Bath-
mullan Church, Sept. 1902 : Hunter.

14. Cephalozia stellnlifera Tayl. MS.

Jungermania gtellulifera Tayl. MS., G.L.N., Syn. Hepat., p. 13i
J, Starkii var. proeerior Nees, G. & K., Hepat. Eur., no. 625. Cefha-
loziadivaricataSm.YeiT.stellulifera Spruce, on Ceph., 1882. Ctphd^
stelluUfera Pearson,. Hepat. Brit. Isles, p. 179, plate 58.

Districts I. VIII. XII.

Hah, — On the ground among damp rocks.

I. Cromaglown,^Ejllamey, June 1889 : Scully.

VIII. Pontoon on Lough Conn, May 1901 : Lett & Mci.

XII. Tollymore Park, Co. Down : Waddell in Guide to Belia^^^
1902.

McAbdlb — A List of Irish HepaticcB.

439

Cephalozia elachista Jack., Lindberg.

Jungermania elachista Jack., in G. & K., Hepat. £ur., no. 574.
Vephakvia elachista Lindb., Hepat. in Hib., p. '502. Moore, Irish
Hepat, p. 625. Pearson, Hepat. Brit. Isles, p. 184, plate 62.

Districts I. IV. V. ^ .

Md^. — On damp peaty banks, and on moist rooks.

I. Brandon 1864 (Moore): 8prac€t 1882. . Connor Hill, Sept.
1875: McA. Coumanare lakes, Sept. 1898: Lett & McA. Very
rare in the Din^e district : McA. 1901. Waterville, Upper Glencar,
MacGillicnddy's Eeeks to 2,500 feet : Scully 1890.

IV. Longh Bray, Co. Wicklow, among Sphagnum euspidatum^ 1873 :
landberg 1875; Moore 1876.

Y. Among rocks near Sutton, Howth, and clifis on Ireland's Eye:
McA. 1893a.

Cephalozia leucantha Spruce.

Cephaloua leucantha Spruce, on Ceph., p. 68. Pearson, Hepat.
Brit. Isles, p. 186, plate 63.

Districts I. VIII, .

ffab. — On decayed wood, and on damp peat among rocks.
I. Mount Eagle, Co. Kerry, near the lake, June, 1898 : Lett &
ftcA.

VIII. On the rocky shore of Lough Conn at Pontoon, and on
^ephiiiy May 1901 : Lett & McA. Devil's Mother and Slieyemore,
Lchill, Sept. 1901 : Lett.

Oenus XVII. PrionolobnB , Spruce.
I. Prionolobns Tomeri Hook., Spruce.

Jungermania Tumcri, Hook., Brit. Jung., tab. 29. Anthelia Tur-
fri Duxn.y Becueil, p. 18. CepKalozia Tumeri Lindberg, in Hepat. in
ib., p. 502. Moore, Irish Hepat., p. 627. Frumohhus Tumeric
praoe, Hepat. Amaz. et And., p. 507. Pearson, Hepat. Brit. Isles,

190, plate 65.

Districts I. .

Siah. — On shady damp banks.

I. Shady bank of a rivulet mountain near Bantry, fertile, March
1 1 (MiBB Hutchins) : Hooker 1816 ; Moore 1876. On a wet sandy
jik at Cromaglown, Eillamey, bearing perianths, July 1873 ; lind-
rg 1876; Moore 1876.

440 Proceedings of the Royal Irish Acad^iy.

Genus XVIII. Hygrobiella Spruce.
I. Hygrobiella lazifolia 8pruce.

Jungermania laxifolia Book., Brit. Jung., tab. 69. Gymnocoki
laxifolia, Dum., Recueil, p. 17. Cephalozia laxifolia, Lindberg, Muse.
Scand., p. 76, 1879. Hygrobiella laxifolia Spruce, on Cepb., p. "-•
Pearson, Hepat. Brit^ Isles, p. 197, plate 67.

Districts I. V. VIII. .

Hab» — In tbe crevices of wet rocks.

I, — ^Mountain rivulet near Bantry (Miss-Hutcbins): Hooker 1816:
Moore 1876. Aooreagb River near Sneem (Taylor) : Moore 1876.
Magbanabo Glen and Brandon: Moore. Connor HiU, 1875: HrA.
Cromaglown and Mangerton (Carrington) : Moore 1876. 0'SiilliTaD'>
Cascade, and Slieve Misb to 1500 ft, : Scully 1890.

V. Castlekelly Mountains, Co. Dublin (Taylor) : Hooker 1816.

VIII. Leenane, Co. Galway, Sept. 1901 : Lett.

Genus XVIII« Adelantbus Mitten.
I. Adelianthus deoipienB Hook., Mitt.

Jungermania decipiene Hook., in Eng. Bot., vol. xxxvi., t. 2567 ;
Brit. Jung., tab. 60. Adelanthus dedpiens Mitt., in Joum. Linn. Soc.,
vol. iii., p. 264, 1864. Moore, Irish Hepat., p. 658. Pearson, Hepat.
Brit. Isles, p. 204, plate 69«

Districts I. — VIII. IX. — XI. XII.

Hah. — On damp rocks and on peaty banks.

I. Near Bantry (Miss Hutcbins): Hooker 1816. West Cork
(Carroll) and Glengariff : Carrington 1863. Killamey : Moore. A:
tbe Upper Lake : Scully 1890. Connor Hill, and at tbe foot cO
Brandon, 1881: F. W. M. & McA. Lougb Duff in the Branlr
valley, very fine on rocks between Emalougbjand Incb, and Deny
more Glen near Tralee, May 1899 : Lett k McA.

VIII. Nepbin, and Pontoon near Foxford, May 1901 : Lett i
McA.

IX. Glenade, Co. Leitrim : Moore. Slish Wood, Co. Slig
Waddell 1893.

XI. Lougb Eask Woods, Goat Island, June 1903 : McA. Ka^b
muUan Wood, 1908 : Hunter.

XII. Moist woods at tbe bead of Glenariff, July 1836: Mc-m
Sallagb Braes: Waddell in Guide to Belfast, 1902.

McArdlk — A List of Irish HepaticcB.

441

Sab-tribe 5. SCAPAHIOIDEiB.

Genas 19. Soapania Damort.

I. Soapania oompaota Roth., Damort.

Jun00rmania eampada Botb., Tent. Fl. Germ., iii., p, 375. Junker-
mania resupinata Hook., Brit. Jung., tab. 23. Soapania eompaeta
Dam., Beoueil, p. 14. Moore, Irish Hepat., p. 637. Pearson, Hepat.
Brit. Isles, p. 207, plate 70.

Districts I. — III. IV. V. Vin. — X. — XII.

Hah. — ^Banks among moist rocks, and in the crevices of rocks.

I. Brandon : Moore 1876. Connor Bill, very rare, April 1897 :
F. W. M. ft McA. Near Waterville and Killamey : Scully 1890.
Ardrone Hill and Musberagh Mountains, Go. Cork (Carroll) : Carring-

tonises.

III. Amongst granite rocks, bank of the Biver Barrow near the
bridge, Oraigue, Co. Carlow, rare: McA. 1896a.

IV. Bocks between Woodenbridge and Arklow, Co. Wicklow :
ITcA.

V. Anglesey Mountain, Co. Louth: Lett 1890.
Vin. Nephin, May 1901 : Lett & McA.

X. Camlough Mountain, Co. Armagh: Lett 1890.

Xll. Slieyenabrock, Spaltha and Deer's Meadow, Co. Down
Lett and Waddell): Lett 1890. Slieve Donard (J. J. Andrew):
jBtt 1890. Holywood Hill, 1903 : Hunter.

2. Soapania snbalplna Nees, Dumort.

Jun^&rmania iuhalpina Nees, in Syn. Hepat. Eur., p. 55. Seapania
ihalpina Dumort., Becueil, p. 14. Moore, Irish Hepat., p. 638.
earson, Sepat. Brit. Isles, p. 211, plate 72.

BiBtricta I. IV. VIII. XII.

Mob. — On rocks in mountain rivulets.

I. Magillicuddy's Beeks and Slieve Mish, to 2500 ft. : Scully 1890.
ank near Inch, and Lough Duff in the Brandon Valley, 1899 :
f-tt ft McA .

IV. Ijugnaquilla, 1864: Moore; and May 1896: McA. Lough
ray (Moore) : McA. 1890.

VIII. Nephin, 1862 : Moore.

XII. Mountains near Camlough, Co. Antrim, 1837 : Moore.

442

Fraceedinya qfthe Uoyal Iiiah Academy.

yar. unduHfoUa^ G. L. N., Syn. Hepat., p. 64.

Hah, — On wet rocks, often submerged.
Lugnaquilla, 1864: Moore; and May 1896: McA. Lough Bray
(Moore): McA. 1890. l^elly's Glen, Co. Dublin: Moore.

I r

i|,i

8. Soapania flDquiloba Scbw»gr., Dumort.

Jungermania aquiloha, Schwaegr., Prodr. Hepat., p. 214. Seamu
aquiloha Dumort., Recueil, p. 14. Martinellia aquiloha Lindb., Hepat
in Hib., p. 521. Soapania aquiloha^ Moore, Irish Hepat, p. 639.
Pearson, Hepat. Brit. Isles, p. 212, plate 73.

Districts I. — III. IV. V. VI. IX. — XI. — .

Hah. — On rocks and on rocky banks among Mosses.

I, Maugerton and Ross Bay, KiUamey : Oarrington. Near the
Hunting Tower : Scully 1890. Maghanabo Glen near Glohane, April
1897 : F. W. M. & McA. Anascaul, 1898, Lough Duff in the Brandon
valley. May 1899 : Lett & McA. Caha Mountains, Co. Cork:
McA. 1894.

III. On rocks by the River Barrow near the bridge, Graigae, Co.
Carlow: McA. 1896 a.

IV. Douce Mtn., Co. Wicklow, 1897 : McA.

V. Ballykill, and in Howth Demesne, 1894 : McA. 1897.

VI. Cappanwalla Mountain, Co. Clare : McA. 1895 a.
IX. Near the head of Gleniff, Co. Leitrim, 1875 : Moore.

Slieveanierin, Co. Leitrim, rare : Stewart 1885.

XI. Near the lake, Macamish, July 1902 : Hunter.

var. inermis Gottsche.

Gott. et Rab., Exsicc, 80, 404, 408.

On banks in rocky places. Mount Eagle, Co. Kerry, near the lake,
Sept. 1897, and at Loughanascaul, Sept. 1898 : Lett & McA.

4. Scapania aspera Miiller and Bemet.

Seapania asptra M. & B., Henri Bemet, Cat. Hepat. du Sud-Oue$t
de la Suisse et de la Haute-Savoie. Pearson in Joum. of Bot., vol. xxx^
p. 353, plate 829. McArdle, Musci and Hepat. Co. Cavan, p. 613s
plate 21. Pearson, Hepat. Brit. Isles, p. 214, plate 74.

Districts I. — in. — V. VI. — VIII. - X. XI. -.

Hah, — On limestone rocks and rocky banks, among Mossea.

I. Tore Waterfall, KiUamey, Sept. 1897 : Lett & McA.

McArdlb — A lAat of Irish Hepaticce, 448

III. Among rocks, side of the River Barrow, Graigue, Co. Carlow :
McA. 18960.

Y. On rocks among Mosses in Howth Demesne : KcA. 1897.

YI. Ballyyaaghan and yery fine on Cappanwalla Ktn., Co. Clare :
He A. 18950.

YIII. Nephin and Pontoon near Foxford, Co. Mayo, May 1901 :
Lett & McA. Salthill near Galway : McA. 1895 a.

X. On rocks, fertile, Bally haise Wood, Co. Cavan, 1893 : McA.

XI. Behind Rathmnllan Church, and near the lake, Macamish,
July 1902 : Hunter.

5. Scapania resnpinata Linn., Dumort.

Jungermania resupinata Linn., Sp. PL, p. 1599, 1753. Eng. Bot.,
tab. 2487. Seapania resupinata Dumort., Kecueil, p. 14. MarUn$llia
gracilU Lindb., Manip. Muse. Secund., p. 365. Seapania retupinata
Moore, Irish Hepat., p. 639. Pearson, Hepat. Brit. Isles, p. 217,
plate 75.

Districts I. II. III. lY. Y. YIII. — X. XI. XII.

Sah, — On rocks, banks, and trunks of trees in mountainous places.

I. O'SuUivan's Cascade, Killamey, 1873: Lindberg 1875. Tore
Waterfall, 1897: McA. & Lett 1899. Common at Boss I., 1900:
Lett & McA. Common through the Dingle Peninsula (Lett & McA.) :
McA. 1901.

II. Knockmeildown and Galtee Mtns., June 1902: McA.

III. Bocks by the Biver Barrow, Graigue, Co. Carlow: McA.
1896ff.

lY. Wood by the Slaney Ri^er, and Xilloughrim Oak Forest near
Enniscorthy, May 1899: McA. Near Ferns, Oct. 1896 (Greene):
McA. 1903. Lough Bray, 1873: Lindberg 1875; Moore 1876; and
1889 : Scully & McA. Lugnaquilla, 1896, and Douce, Co. Wicklow,
May 1897: McA

Y. Kelly*s Glen, Co. Dublin : Moore 1876. Anglesey Mtn., Co.
Louth: Lett 1890. Howth, April 1895: McA. 1897.

YIII. DeviPs Mother, Doolough, and Slievemore on Achill, Sept.
1901 : Lett. Croaghpatrick, about Murrisk, by the Moyire River,
and in Westport demesne, Sept. 1901: McA. Nephin, and Pontoon
near Foxford, May 1901 : Lett & McA.

X. Camlough Mtn., Co. Armagh: Lett 1890. Ballyhaise, Co.
Cavan, 1898 : McA.

XI. Muckish: Moore 1876. Bathmullan Wood, Macamish, Cro-
ghan Mtn., Carradoan Wood, Glenalla Hill, Bathmelton, July 1902 :

444 Pivceedings of the Royal Irish Academy.

Hunter, Columbkil Lake, on rocks near Kilfoid, Gartan Wood, 8epL
1902 \ Errigal, and Lough Eaak woods, 1903 : McA.

XIL ThomfiB Mtn., the BaUagh, and SUeve Donaid: Lett 1890
Eagle Mtn., vtry rare (Waddell) : Lett 1890. Cave Hill and Colin
Glen (Stewart); Sallagh Braes (Lett); Ballybarrigan Glen, 0). Deny
(Moore): Stewart 1888. Fairhead and Glendon (Templeton
Stewart 1895.

6. Soapania nemorosa Linn., Dumort.

Junyermama «^moro«flLinn., 8p. PI., Srded., p. 1598. M»ti^^'^
nemorom Gr. & B., Nat. Arr. Brit. PI., i., p. 692. Seap^U mwu^
Dum,, Secueil, p, 14. Moore, Irish Hepat., p. 640. Peaison, Htpai
Brit Isles, p. 222, plate 78.

Districts L II. III. IV. V. VI — VIII. — X. XI. xn.

Mah^ — On damp shaded banks among rocks, on tbe trunb ci
trees, &c,

I. KiUamejr: Moore; Carrington ; McA. Cromaglown and Tor
Waterfall, 1873: Lindberg 1875 (under ifar<ffk?//ttf). Tore Wat^rfsB.
1B97 ; Lett & McA. MagiUicuddy's Reeks at 2500 ft. : Scully 18^'
Frequent in the Dingle Peninsula (Lett & McA.) : McA li^^^
Mountains near Bantry (Miss Hutchins): Hooker 1816. Dnnboy
Wood: McA, 1894.

II. Woods at Scarrijff in Galtee Mtns., and among rocks, Kno-ck
meildown MtnH., Co. Tipperary, June 1902 : McA.

IIL Rocks by the River Barrow at Graigue, Co. Carlow : McA
1896(7.

IV. EUloughrim Oak Forest, Co. Wexford, 1899: McA. m^
(flemnalnrc^ May 1896 : McA.

Y, Anglesey Mtn., Co. Louth : Lett 1890. Howth : McA. 189'>
Kelly'a Glen, Co. Dublin, Aug. 1896: McA.

VI. Cam Seefin and Ballyvaughan, Co. Clare: McA. 1895 «.

Yin. Woods about Kylemore, 1874: Moore; andl891: McWeent;
Near the Cottage, Ballinahinch, Connemara: Wade Rar. 1804. P^
toon near Fox ford, May 1901 : Lett & McA. Slievemore on Ach-
Sept 1901 : Lett

X, Eollyhaise Woods, Co. Cavan, 1893: McA. Strabane Gle
Co. Tyrone, 1882: Lett

XI. Saltpiins Wood, Rathmullan, and very fine on hill sb^
Mintiaghs, July 1902 : Hunter. Cratleagh Wood and Gartan Vo»
near Milford, Sept 1902: McA.

XII fVanraore, Co. Antrim (Templeton), Sallagh Braes (Let

McArdlb — A List oflnnh Hepaticw, 446

Stewart 1888. Glendun : Lett & Brenan. Co. Antrim, abundant,
1838: Moore. Near Lignapieste, Co. Berry, 1834 (Moore): Stewart
1888. Eostrevor Wood and ToUymore Park, Co. Down, very rare
(Waddell), Slieve Donard (Lett) : Stewart 1888. Saintfield, Co.
Down (Waddell) : Stewart 1895. Purdysbnm and Castlereagh Glen,
1803 (Templeton) : Stewart 1888.

var. purpurea Hook.

Jungermania purpurea Hook., Eng. Bot., tab. 1023.

Hah. — Among rocks in wet boggy places.

Prequent in tbe Dingle Peninsula: McA. 1901. Lough Bray:
McA. 1890. Croaghpatrick, Sept. 1891 : McA. Nephin, and Pontoon
near Pozford, May 1901 : Lett & McA. Devil's Motber, and
Slievemore on Achill, Sept. 1901 : Lett. Co. Antrim, 1838 : Moore.
Garlingford Mtn., Co. Loutb, 1894: McA. RathmuUan Wood, Co.
Donegal, July 1902 : Hunter.

7. Scapania nimbosa Taylor.

Seapania nimhoea Tayl., in Lehm., Pugill. Plant., viii., 1844.
Trans. Bot. Soc. Edinb., ii., p. 115. Carrington, Brit. Hepat., part ii.,
plate 7, fig. 21. McArdle, Hepat. Dingle Peninsula, p. 311, plate
2, figs. 1, 2, 3. Pearson. Hepat. Brit. Isles, p. 220, plate 77.

Districts I. .

^o^.— On moist rocky ledges among Mosses, &c.

I. Near the summit of Brandon, Co. Kerry, 1813 : Taylor. On
the same mountain more recently, 18 ? : Mitten.

8. Scapaiiia omithopodioides Dill., Withering.

Jungemumia omithopodioides Withering, Bot. Arrang., vol. ii.,
p. 695, no. 14, 1776. Jungermania planifolia Hook., Brit. Jung., t. 67,
1816. Scapania planifolia Hook., Dum., Becueil, p. 14, 1835.
McArdle, Hepat. Dingle Peninsula, p. 311, plate 2, figs. 4-9, 1901.
Pearson, Hepat. Brit. Isles, p. 219, plate 76.

Districts I. VUI. .

Hah. — On damp rocky banks, and on shelving rocks among Mosses
and large Hepatic®.

I. Brandon 1873: Taylor and William Wilson. In the same
station more recently, 18—?: Mitten. In a rocky gorge on the
N.E. side of Brandon at 2800 ft. on shelving rocks SLUiong Mastigophora
Woodsii, &c., June, 1900 : Lett & McA.

VIII. On Slievemore, Achill, Sept. 1901 : Lett.

446 Proceedings of the Boyal Irish Academy.

9. Soapania nndulata Liim., Dmnort

Jungermania undulata Linn., Sp. PI., 1598. Hook., Brit. JTmgi
tab. 22. Seapania undulata Dum., Recueil, p. 14. Pearson, Hepat.
Brit. Isles, p. 224, plate 79.

Districts I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.
Sah, — On wet banks and rocks, and on stones in strwn*
Common.

var. purpurasems Hiiben., Hepat. Germ.

Moore, Irish Hepat., p. 638. G. L. N;, 8yn. Hepat, p. 6^
Seapania purpuroicens Pearson, Hepat. Brit. Isles, p. 225, pbtedO
Hah. — On wet rocks, and among Mosses in mountainous pkco*
Common in Co. Kerry. Well distributed among the moimtaiu it
the Dingle Peninsula: McA. 1901. Frequent in the mountaisf-if
the northern counties : Lett 1890, Stewart, and others. Rathmnl]^
Co. Donegal : Hart 1886. Goat Island near Lough Eask, Co. Dont^
June 1901 : McA. Lough Bray, Co. Wicklow, 1873 : Lindberg 18'5
(under MartineUia) ; and McA. 1890. Bog at Ballykill, Howth, Co.
Dublin: McA. 1893 a. Stream at Knockroe, Co. Wexford, 1899: KcA-
1903.

var. speciosa N. ab E«, Hepat* Eur., i., p. 185.

G. L. N., Syn. Hepat., p. 66. Moore, Irish Hepat., p. 638.

Mob. — On wet rocks in mountainous places.

Lugnaquilla: Moore. Nephin, May 1901: Lett & McA. l--^-
gore and Devil's Mother, Co. Mayo, Sept. 1901 : Lett. Pontoon w^
Foxford, May 1901 : Lett & McA. Knockroe, Co. Wexford, M«t
1899 : McA. 1908.

var. isoloha Nees, Syn. Hepat,, p. 66.

N. ab E., Hepat. Eur., iii., p. 421. Seapania isoloba Duinort-*
Hepat. Eur., p. 33.

Hah, — Floating in moimtain streams.

Stream on the Clogreen side of Brandon ; Moore ; and Sept 189* :
Lett & McA. Lough Doon on Connor Hill : McA. 1894. Coummne
Lakes, 1898 (Lett & McA.) : McA. 1901.

var. major Carrington, N. ab E.
Jungermania rempinata, Eng. Bot., tab. 243. Seapai^i^ ^mdtdsfa
var. mqfor G. L. N., Syn. Hepat., p. 65.
Hab, — On wet rocks.

McArdlk— ^ List of Irish Hepaticw, 447

Cromaglown, Killamey, and Connor Hill, Co. Kerry ; Maara Turc,
Co. Galway; Lough Bray, Co. Wicklow (Moore): Carrington 1863.
Pontoon near Foxford, May 1901 : Lett & McA.

var. laxifoUa Dumort.

Seapania resupinata var. laxifolia Dumort., Hepat. Eur., p. 34.

Bab. — On heathy banks.

Cromaglown, Killamey, 1899: Lett. Connor Hill, 1894: McA.
Muckish, Co. Donegal : Moore 1876. Shanslieve, Co. Down, 1898:
Lett.

var. dentata N. ab E.

Rob. — On wet rocks.

Pontoon near Foxford, May 1901 : Lett & McA. Gartan Wood
near Milford, Sept. 1902: McA. Gkdtee Mtns., Co. Tipperary, June
1902: McA.

10. Seapania intermedia Husnot.

Seapania intermedia Husnot., Hepat. Gall., plate 3, fig. 23, 1875.
8, nemorosa var. intermedia Husnot, Hepat. Gall., p. 22, 1876.
Pearson, Hepat. British Isles, p. 227, plate 91.

Districts — II. XI. XII.

Mah, — On moist shaded rocks.

II. On sandstone rocks south of Lough Muskry, Galtees, July
1902: Lett.

XI. Glen Columbkil, Milford, 1902 : McA.

XII. Colin Glen, Co. Antrim, July 1887 : Waddell, Joum. Bot.,
vol. xli., p. 286, 1903.

11. Seapania irrigna Nees, Dumort.

Jungermania irrigua Nees, Nat. Eur. Leb., i., p. 193. Seapania
irrigua'Dnm,, Recueil, p. 13. Moore, Irish Hepat., p. 639. Pearson,
Hepat. Brit. Isles, p. 230, plate 92.

Districts I. IV. VIII. IX. XII.

Bab. — Marshy banks among rocks.

I. Cromaglown, Killamey: Moore. KnockavohiU, Co. Kerry
(Taylor) : Moore 1876.

IV. Lough Bray : Moore; and 1889 (F. W. M.) : McA. 1890.

VIII. Nephin, May 1901 : Lett & McA.

IX. Benbulben: Moore.

XII. Moist bank, Holywood, Febr. 1902 : Hunter.

448 Proceedings of the Royal Irish Academy.

12. Soapania uliginosa Swartz, Dmnort.

Jungermania uHgtnosa Sw., in Lindb., Syn. Hepat., p. 59. 8captmi4i
uliginosa Dumort., Eecueil, p. 14. Moore, Irish Hepat., p. 639.
Pearson, Hepat. Brit. Isles, p. 281, plate 98.

Districts I. IV. V. VIII. XII.

Mah, — In swampy places, on wet rocks, &c., in mountainous
districts.

I. KnockavohiU, Co. Kerry : Taylor. Near the Hunting Tower,
Cromaglown, 1875: Moore. Near Waterville : Scully 1890. Connor
Hill, 1878 : Moore; and 1894 : McA. Brandon, Sept. 1898-1900:
Lett & McA. Loughanscaul, 1894 : McA. Bare in the Dingle
Peninsula: McA. 1901.

rV. Lough Bray : Moore. On old wood at Lough Bray, April
1 879 : F. W. M. & McA. Banks of a stream, Knockroe, Co. Wexford,
May 1899: McA. 1908.

V. Anglesey Mtn., Co. Louth, rare (Waddell) : Lett 1890.

VIII. Pontoon near Foxford and on Nephin, May 1901:
Lett & McA.

Xn. Bog between Hilltown and Rathfriland, and in the Brown
Bog near Loughbrickland : Lett 1890.

13. Soapania cnrta Mart., Dumort.

Jungermania ettrta, Mart., Fl. Crypt. Germ., i., p. 148, tab. 4,
fig. 24. J, nemorosa var. denudata. Hook., Brit. Jung., tab. 21,
fig. 17-19. Soapania eurta Dumort., Recueil, p. 14. Moore, Irish
Hopat., p. 641. Pearson, Hepat. Brit. Isles, p. 235, plate 95.

Districts I. IV. Vni. IX. X. XI. XII.

Hah. — On moist banks among rocks, decayed wood, and among the
larger Hepatic®.

I. Cromaglown, 1873 (Moore) : Lindberg 1875. On the east side
of Brandon and at Lough Adoon near Clohane, 1897 ; Bamanaghea
Lough near LispoU and Connor Hill, 1898 ; Lough Duff in the
Brandon Valley, 1899 : Lett & McA.

IV. Lough Bray, on the stems of Ulex, 1879 : McA. 1890.

VIII. Ballinakill Harbour near Letterfrack, 1874: Moore. Pontoon
near Foxford, May 1901 : Lett & McA. Devil's Mother, Doolough,
and Slievemore on Achill, Sept. 1901 : Lett.

IX. GlenifP, Co. Leitrim : Moore. Benbulben range : Moore 1876.

X. Slieve Glah, Co. Cavan, rare : McA. 1898.

XI. Bunlin Waterfall near Milford, Sept. 1902 : McA,

McArdlb — A List of Irish HepaticcB. 449

XII. Sallagh Braes and Slemish, Co. Antrim: Moore 1876
Benevenagh, Co. Deny, 1900 : Lett & Waddell,

14. Scapania nmbrosa Schrader, Dumort.

Jungermania convexa Scopoli, Flora Camiolica, 2nd Ed., p. 349.
J, umbroia Scbrad., Syst. Samml. Krypt. Gew., ii., p. 5. Hook.,
Brit. Jnng., tab. 24. Scapania umbrota Dum., B«caeil, p. 14. Moore,
Irish Hepat., p. 641. Pearson, Hepat. Brit. Isles, p. 236, plate 95.

Districts I. IV. V. VIII. — X. XI. Xn.

Mah, — On moist rocks and on decayed wood.

I. Frequent in the Killamey woods : Moore. O'SuUivan's Cascade,
1873: lindberg 1875. Carantual (Miss Hutchins) : Carrington 1863.
Magillicuddy's Eeeks at 2500 feet: Scully 1890. Brandon, 1873:
Moore. Connor Hill, 1874 : McA. Kocks near Mt. Eagle Lake and
Coumanare Lakes, 1898 ; Derrymore Glen near Tralee, 1899 ;
Lett & McA.

IV. Lough Bray, Co. Wicklow, 1873 (Moore): Lindberg 1875;
1887-89 : McA. 1890. Boulacross Mtns., Co. Wicklow (Taylor &
Mackay) : Hooker 1816, Mountain near Powerscourt Waterfall
(Mackay): Hooker 1816.

V. EiUakee Glen, Co. Dublin: Moore 1878. On rocks near
Dublin, 1836 (Taylor) : Moore 1876. Kelly's Glen, Co. Dublin, Aug.
1896: McA.

VIII. Kylemore, 1874: Moore. Murrisk, Co. Mayo, Sept. 1901 :
McA. Nephin, May, 1901: Lett & McA. Devil's Mother, Doolough,
Slievemore on AchiU, Sept. 1901 : Lett.

X. Slieve Gullion, Co. Armagh, 1894: McA.

XI. Carradoan Woody^Bathmullan, and near the lake, Macamish,
July 1902 : Hunter. Gartan Wood and Cratleagh Wood near Milford,
Sept. 1902 : McA.

XII. Glenarm, Co. Antrim, Sept. 1836 : Moore. Hen Mountain,
Co. Down (Lett) : Stewart 1895.

Genus 20. Diplophyllum Dumort.
1. Diplophyllum albicans Linn,, Dumort.
Jungermania albicans Linn., Sp. PL, p. 1599. Hook., Brit. Jung.,
tab. 25. Diplophyllum albicans Dum., Recueil, p. 14. Moore, Irish
Hepat., p. 642. Pearson, Hepat. Brit. Isles, p. 238, plate 97.

Districts I. II. m. IV. V. VI. VII. VIII. IX. X. XI. xn.

Hob, — Moist banks in woods, and by roadsides, old walls, and
rocks, and on the trunks of treesy from sea-level to high elevations

460 Proceedings of the Royal Irish Acadetny.

2. Diplophyllnm obtasifolium Hook., Dumort

Jungirmania ohiusi/olia Hook., Brit. Jung., tab. 26. l>iphpk/llim
ohttmfolium Dnm., Recueil, p. 16. Moore, Irisli Hepat, p. 642.
Pearson, Hepat. Brit. Isles, p. 241, plate 99.

Districts I. V. .

Sab, — On moist clay banks in shaded places.

I. Near Dnnkerron, Co. Kerry, 1836 : Taylor. Kear Bantiy, 18U
(Miss Hutchins) : Hooker 1816. Dunscome's Wood near C^^rk.
1829 (W. Wilson): Moore 1876.

V. Glendhu, Co. Dublin, 1890: McA.

3. Diplophyllum Dioksoni Hook., Dumort.

Jungermanta Dicksoni Hook., Brit. Jung., tab. 48. D^^lophfQwm
Dicksoni Dum., Eecueil, p. 16. Moore, Irish Hepat., p. 649 (miuer
Jungermanta {Sphenolohum) Dicksani), Pearson, Hepat. British Is c^.
p. 243, plate 100.

Districts I. IV. V. VIII. IX. XII.

Mah, — In the crevices of rocks, on moist banks, and among thel&i^
Mosses and Hepaticse.

I. On the north side of Connor Hill growing among Mosses, 1873 :
Moore. Lough l^alachan on Brandon, May 1899 : Lett & McA.

IV. Lough Bray, very scarce : Moore.

V. Mountains near Dublin (Taylor) : Hooker 1816.

VIII. Pontoon near Foxford, very rare, and on Kephin, Hsy
1901 : Lett & McA. Slievemore, Achill, Sept. 1901 : Lett

IX. Glenifl, Co. Leitrim, single stems : Moore.

XII. Shanslieve, Moume Mountains, 1898 (Lett): Pearson 19C»i,
Extremely rare in all these localities.

Sub-tribe 6. EPIOONEAHTHKfi.

Genus 21. Lophooolea Dumort.
1. Lophooolea bidentata Linn., Dum.

Jungermanta bidentata Linn., Sp. PL, p. 1508. Hook., Brit Jii2i^.>
tab. 30. Lophooolea bidentata J)\xm., Eecueil, p. 17. ZcjvAocoJIm ^Mr»<
tata Moore, Irish Hepat., p. 628. Pearson, Hepat. Brit, lales, p. :f 4T.
plate 101.

Districts I. II. III. IV. V. VI. VIL VIII. IX. X. XI. XD.

Sab. — On moist shady banks, among rocks, in woods, on pl^ai -.
from sea-level to the tops of the highest mountains.

McArdlb — A List of Irish Sepaticae.

451

var. Mookiriana l^ees.

Zophoeolea Hooksriana Nees, Nat. Eur. Leb., ii*, p. 336,

Eah, — On moist shaded rocks.

Nephin, Co. Mayo, May 1901 : Lett & JffoA.

2. Lophocolea onspidata Limpn

LophocoUa hidentata var. euspidata Nees, Nat. Eur. Leb., ii.,
). 827. Lophocolea euspidata limprioht, in Cohn, Krypt. EL Sohles.^
}. 303. Peairon, Hepat. Brit. Isles, p. 249, plate 102.

Districts I. II. VIII. XI. — .

Sal, — On the tmnks of trees among Mosses, and on decayed wood,
ianks, &c. Probably widely distributed, but confounded with the
^receding species.

I. On dead trees, Eillamey (Carrington) : Moore 1876. Olena
nd Cromaglown, Killamey, 1873: Lindberg 1876.

IL Glengarra Wood and Scarrifl Woods, Galtees, June 1902 :
[cA.

VIII. Pontoon near Foxford, May 1901 : Lett & McA.

XI. Gartan Wood, and wood by Mulroy Bay near Milford, Sept.
902; Lough Eask Woods and Bamesmore Gap, June 1903 : McA.

3. Lophooolea heterophylla Schrad., Dumort.

Jungermania heterophyUa Schrad., Diar. Bot, p. 66, Hook., Brit.
tug., tab. 31. Lophocolea heterophylla Dum., Eecueil, p. 17. Moore,
ish Hepat, p. 628. Pearson, Hepat. Brit. Isles, p. 250, plate 103.

Districts I. IL — IV. V. VIIL — X. XL XII.

Sah. — On the trunks of trees near the base ; more frequently on
cayed wood.

I. Tore Mountain, Xniamey : Carrington 1863. Eermoy (Carroll) :
x)re 1876.

II. Wood at Scairiff, Galtees, rare, June 1902 : McA.

IV. Altadore Glen, Co. Wicklow, 1887-8 : McA. 1889.

V. fiallykill, 1893, and in Howth Demesne, 1896 : McA. 1897.
VUI. Near Cong, Co. Galway : Moore.

X. On decayed wood, shores of Lough Cultra, Go. Cavan, 1893 :
A. 1898.

XI. Gartan Wood near Milfoid, rare, Sept. 1902 : MoA. Bun-
Da, March 1903 : Hunter.

XII. On decayed wood, Holywood, March 1903 : Hunter.

■• I. A. FBOC., TOIm XZIV., SBC. B.l 2 0

452 Proceedings of the Royal Irish Aeculemy.

4. Lophooolea spicata Taylor.

Lophoeolea spteata TayL, in G. L. N., Syn., Hepat., 167. Moore,
Irish Hepat., p. 629. Pearson, Hepat. Brit. Isles, p. 252, plate 104.

Districts I. — III. IV. .

Sah, — On damp shaded rocks and among the larger Hepatics and
Mosses.

I. Dunkerron, Co. Kerry (Taylor) : Moore 1876. On rocks belov
Tore Waterfall : Carrington ; fertile there, 1897 : Lett & McA. By
the side of the Upper Lake, Killamey, fertile, June 1869: Moore,
O'Sollivan's Cascade, Glena, and Tore Cascade, 1873: Lrndbeif
1875. Glencar and Mangerton: Scully 1890. Connor HiU, July
1881 : F. W. M. & McA. Near Brandon Head, Loughanscaul, near
Dingle, and Mount Eagle, 1898: Lett & McA. Near Bantry(HiF$
Hutchins): Moore 1876. Glensiskin, Co. Cork (T. Chindke)*
Moore 1876; and (the female plant) (I. Carroll): Carrington 186^.
Dunhoy Wood, Co. Cork : MoA. 1894.

III. Wood near Goresbridge, Co. Carlow, very scarce: McA.
1896 a.

IV. Altadore GIod, 1873 (Moore) : McA. 1889.

Genus 22. Clasmatocolea Spruce.
1. Clasniatocolea ouneifolia Hook., Spruce.

Clasmatoeolea cuneifolia Spruce, Hepat. Amaz. et And., p. 440.
Jungermania (Aplozia) cuneifolia Moore, Irish Hepat., p. 646. J-
cuneifolia Hook., Brit. Jung., tab. 64. Eng. Bot., SuppL, tab. 2700.
Pearson, Hepat. Brit. Isles, p. 254, plate 105 (under ClasfnaUcoUa).

Districts I. VIII.

ffah. — On damp peat among rocks, on Fndlania^ and on tbt
trunks of trees.

I. Tore Mtn. (Carrington) : Moore 1876. On the stems of tree<«
creeping over Fndlania Tamariaeij between the police-barrack and
Upper Lake, Killamey : Moore. Waterville, Glena, and Upper Lak^*
Killamey : Scully 1890. O'Sulliyan's Cascade, 1894 : McA. Cocsor
Hill, 1899 : McA. ; and 1898 : Lett & McA. Brandon 1877 : McA.
1880 ; and 1881 (F, W. M. & McA.): McA. 1901. Loughanseaul neir
Dingle, 1898 : Lett & McA . Lough Duff in the Brandon valley, Ttrr
fine on damp rocks, on Frullania Tamarisei, June 1900 : Lett & Me A.

VIII. On the slopes of Devil's Mother and on Slievemore, Achill.
Sept. 1901 : Lett.

McArdle — A List of Irish HepatiecB. 468

Genus 23. ChiloscyphnB Dumort.

1. Chilosoyphos poljr&nthiu Linn., Dam.

Jungermania polyanthua Linn., Sp. PI., no. 1597. Hook., Brit.
Jung., tab. 62. Chiloseyphw polyanthus Bum., Syll. Jung. Eur.,
p. 67, tab. 1, fig. 9. Moore, Irish Hepat., p. 630. Pearson, Hepat*
Brit. Isles, p. 256, plate 106.

Districts I. — in. IV, V. vni. XI. xn.

Hah, — Damp shaded places on rocks, and among stones.

I. Killamey : Moore. O'Sullivan's Cascade, 1873 : Lindberg
1875. Common in the Dingle Peninsula : McA. 1901.

III. Stream on Slieve Bloom, Queen's Co. : McA. 1892 a,

lY. On wet rocks, Co. Wicklow: Moore 1876.

V. Kelly's Glen, Co. Dublin (Moore) : Carrington 1863. Howth,
rare, April 1895: McA. Anglesey Mtn., Co. Louth: Lett Omeath
Waterfall, Co. Louth, very rare (Waddell) : Lett 1890.

VIII. Bangore, Co. Mayo (Moore) : Carrington 1863. Nephin,
May 1901: Lett & McA.

XI. Gartan Wood, fertile, Sept. 1902 ; Lough Eask Woods and
Bamesmore Gap, June 1903: McA.

XII. Rostrevor Mtn. (Waddell), Glenmackan, Co. Down : Stewart
1888. Rocky Mtn. near Hilltown : Lett 1890. Aghaderg Glebe
and Bally varley bog (Lett) : Stewart 1895. Eocks in the River
Bann, Corbett, Co. Down, 1895 : Lett. Colin Glen, Carr*s Glen, and
Eathlin Island, Co. Antrim: Stewart 1888. Slemish (Templeton),
Glendun (Brenan & Lett) : Stewart 1895. On the Derry side of the
Faughan River below Claddy (Moore) : Stewart 1888. Glenariff, Co.
Antrim, 1836 : Moore. Holy wood, Febr. 1903 : Himter. Annahilt, Co.
Down, June 1901 : Waddell.

var. pdllescens Lindenberg.

Jungermania palleseens Schrad., Syst. Samml. Krypt. Gew., ii.,
p. 7. Chiloscyphus palleseens Dumort., Syll. Jung., p. 67.

Sah, — On wet banks and on rocks, often submerged.

Mountain streams, Killamey (Carrington) : Moore 1876. Connor
Hill, 1894 : McA. Fermoy (T. Chandlee): Moore 1876.

var. rivularis Nees.

Bab. — Wet places on rocks and stones, often in mountain streams.
Frequent about Killamey: Carrington 1863. In oaves, Connor

702

454 Proceedings of the Royal Irish Aaidetny.

Hill : Lindberg 1 875. Stream on the west side of Brandon, April 1897 :
F. W. M. & McA. Mt. Eagle Lake and Coumanare Lakes near Connor
Hill, Sept. 1898 : Lett & McA. In a spring near Clongb, Go. Antrim
(Lett) : Stewart 1888.

« Genus 24. Harpanthns Nees.

1. Harpanihnli seutatos W. et M., Spmce.

Jungermania sctUatus Web. et Mohr., Bot. Taschenb., p. 408. TajL
in Fl. Hib., part ii., p. 64. Moore, Irish Hepat., p. 631 (under Ew-
panthus). Carrington, Brit. Hepat., p. 49, plate 7, fig. 52. Dnmoit,
Hepat. Europ., p. 67. Sarpanthus scuiattu Pearson, Hepat Brit-
Isles, p. 259, plate 107.

Districts I. IV. VIII. XL — .

Sab, — On moist banks, rocks, and stones, and on decayed wooi

L Gal way River, Killamoy, fertile, August 1829 (W. Wilson]:
Carrington 1874. Tore Waterfall, Killamey : Taylor 1836. C»-
.maglown and Glena: Moore. O'Sulli van's Cascade, 1873 (fenule.
sterile): Lindberg 1875; Scully 1890. Tomies Mtn. : Carriiigto!i
1 863. Cromaglown (Stewart & Holt) : Pearson 1 902. Gap of Dusloi^
June 1900: Lett. Mount Eagle, 1881 : F. W. M. & McA. Cocsor
HiU, 1877 : McA. Hare in the Dingle Peninsula : McA. 1901. Near
Biintry, 1812 (Miss Hutchins) : Moore 1876.

IV. Lough Bray, Co. Wicklow : Taylor ; Moore 1876 ; McA. 1S»>

VIII, Nephin, May 1901 : Lett & McA. Slievemore, Achill-
Sept. 1901 : Lett.

XI. By the Biver Leenan near Ramelton, Dec. 1878 : Hart ISS6.
On rocks, Columbkil Lake near Milford, Sept. 1902 : McA.

Genus 25. Mylia Gray & Bennett.

1. Mylia Taylori Hook., Gr. & B. I

Jungemiania Taglori, Hook., Brit. Jung., tab. 57. MyUs J'Tj^*
Or. & B., Nat. Arrang. Brit. PL, i., p. 695. Moore, Irish Hepat-
p. 645. Pearson, Hepat. Brit. Isles, p. 264, plate 109.

Districts 1. IV. V. VIII. — X. XI. XIL

Jlah, — On open heaths, boggy places, and moist rocks, in ^^^
patches.

I. Killamey, Devil's Punch Bowl, Mangerton, Brandon : Moom
Connor Hill, 1873: Lindberg 1875. Common in the Dingle Peniitfoi*!
McA. 1901. Near Bantry (Miss Hutchins) : Hooker 1816.

IV. Tonlagee, Co. Wicklow (Taylor) : Hooker 1816. On *^1
banks, Lough Bray : Moore 1878 ; McA. 1889.

. McArdlk — A List of Imh Hepatic(B. 466

Y. Omeath Glen and Anglesey Mtn., Co. Louth : Lett 1890.

YIII. Mnrrisk, and on the slopes of Croaghpatrick, Sept. 1901 :
ICcA. Nephin, May 1901 : Lett & McA. Bangore, and on Slievemore,
AchiU, Sept. 1901 : Lett.

X. Camlough Mtn., Co. Armagh, 1887 : Lett.

XI. Lough Belshade, Errigal, June 1908 : McA.

XII. Olenandra, and near Lignapieste, Co. Derry, 1835 : Moore.
Top of Divis, Co. Antrim (Templeton) : Stewart 1888. Moist moun-
tain near Orra, parish of Loughguile, Co. Antrim, 1838: Moore.
Shanslieve and Deer's Meadow (Waddell), Slieve Donard: Stewart
1888. White River, Spinkwee River, Moygannon Glen: Stewail
1888; Lett 1890. Shanslieve: Lett 1890.

2. Mylia anomala Hook., Gray & Bennett.

Jungermania anomala Hook., Brit. Jung., tab. 34. J. Tayhri vai*.
anamdla Nees, Nat. Eur. Leb., ii., p. 455. Mylia anotnala Moore,
Irish Hepat., p. 646. Pearson, Hepat. Brit. Isles, p. 267, plate 110.

Districts I. IV. X. — XH.

Hiab, — On damp heaths on mountains^ often creeping among
'Sphapnum.

I. Lough Guitane, among Sphagnum : Carrington. Connor Hill,
1873: Lindberg 1875. On the ascent of Brandon from Cloghane,
July 1881 : F. W. M. & McA. Near Bantry (Miss Hutohins) : Hook^
1816.

IV. Lough Bray, Co. Wicklow, 1889 (F. W. M.) : McA. 1890.

X. Camlough Mtn., very rare (Lett & Waddell) : Lett 1890.

XII. Divis, Co. Antrim, and Annahilt Bog (now extinct), Co.
Down (Templeton) : Hooker 1816.

Genus 26. Pedinophyllum Lindberg.
1. Pedinophyllum intermptnm Nees, Lindb.
Jung&rmania interrupia Nees, Nat. Eur. Leb., i., p. 165. Junger-
mania Dwmrii&ri Lib., PI. Crypt. Ard., iv., no. 311. PedinophyUam
pyrenaicum Lindb., Hepat. in Hib., p. 505. Plagioehila interrupta var.
pyrenaiea Carr., Brit. Hepat., plate 3, figs. 2-9. Pedinophyllum pyre-
naicum Moore, Irish Hepat., p. 629. P, interruptum Pearson, Hepat,
Brit. Isles, p. 269, plate 161.

Districts I. VI. — vin. IX. xn.

Mah. — On shady rocks and banks.

I. Bumham Wood near Ventry, 1898, rare (Lett & McA.) : McA
1901 (under Plagioehila).

456 Proceedings of the Royal Irinh Academy.

YI. Ballyraughan, Co. Clare : McA. 1895 a.

Vni. Bangore, Co. Mayo, Sept. 1901 : Lett.

IX. Tbrough the Benbulben range, 1871 ; Oleniff, Co.Leifarim,
1875 : Moore.

XII. Roadside wall, Rostre vor Quay (Lett) : Stewart 1895 (under
Lepto9eyphu»\

Oenus 27. Plagiochila Dumort.
1. Plagiochila asplenioides Linn., Bum.

Jungennania asplenioidss Linn., Sp. PL, p. 1597. HooL, Biit
Jung., tab. 13. Eng. Bot., tab. 1061. Plagiochila oiplenundei'Dm.^
Becueil, p. 14. Moore, Irish Hepat., p. 643. Pearson, Hepai Biit
Isles, p. 274, plate 113.

Districts I. II. III. IV. V. 71. VII. VIII. IX. X. XI. HI

Mob. — Banks and rocks in woods ; on the moss-coyered tranksof
trees.

Tar. minor Lindenberg, Plagiochila, p. 111.

Carr., Brit. Hepat., part, iii., p. 56.

Eillamey woods, plentiful : Moore. Ross Bay : Carrington 1S6S
Ross I., 1899 (Lett & McA.) : McA. 1900. Tore Waterfall, rare, W''"
Lett & McA. Frequent in the Dingle Peninsula: McA. 1901. ^^
rocks, Gndgue, Co. Carlo w : McA. 1896 0. On Fhdlania, shon^ ^f
Lough Conn at Pontoon, May 1901 : Lett & Mo A.

var. devexa Carr., Brit. Hepat., part iii., p. 56.
On rocks, Dingle Bay, and at Ross Bay, Eillamey, July, l^^l '
Carrington. On shaded banks close to the lake, Ross I., 1893: M«A-
1900. On damp rocks. Tore Waterfall. Sept. 1897 : Lett & McA.

var. humilii Lindenberg, Plagiochila, p. 111.
Garr., Brit. Hepat., part iii., p. 56.

On damp rocks, Derrymore Glen near Tralee, May 1899 : Lett 4
McA. Damp bank, Ross I., Killamey, 1893 : McA. 1900. Caif^
Wood between Scarva and Tanderagee, Co. Armagh, July 1^^-
Lett.

2. Plagiochila ambagiosa Mitten.

Plagiochila ambagiosa Mitten, Trans. Linn. Soc., vol. iii., p. 1^^
1891. Steph., Bull. Herb. Boissier, voL v., p. 83, 1897, Peai**.
Hepat. Brit. Isles, p. 275, plate 115.

Districts I.

Hah. — In damp rocky places.

McArdlb — A List oflrwh Hepatuue.

457

I. Bantry : Miss Hntchins.

" Mr. Mitten detected this fine species in a collection made many
yean ago by the late Miss Hntchins in the sonth of Ireland. It is
abundantly distinct from &i\^eTF, punctata Tayl., or P. spinulosa Dicks.,
to which it is most nearly allied " : Pearson 1902. My search for this
fine Plagiochila has not met with success, though I am well acquainted
with the plant, haying studied a specimen sent to me by Mr. Mitten.

3. Plagiochila spinnlosa Dicks., Dumort.

Jwngermania apinulo$a Dicks., PL Crypt. Brit., fasc. ii., p. 14.
Hook., Brit. Jung., tab. 14. Pktgioehila spinidosaf Dum., Siecueil,
p. 5. Moore, Irish Hepat., p. 643. Pearson, Hepat. Brit. Isles, p. 276,
plate 116.

Districts I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.

Sah. — In woods, on moist banks, and on rocks.

yBX.flagellifera Garr., Brit. Hepat., p. 60.

MoorSy Irish Hepat., p. 644.

Glengariff, Sept. 1869, and Gromaglown, Xillarney: Carrington.
Chores of Lough Duff in the Brandon valley, and at Derrymore Olen
lear Tralee, rare, 1899 : Lett & McA.

yar. inermia Carr., Brit. Hepat., p. 60.
On moist rocks. Tore Waterfall, Sept. 1897 : McA. & Lett
899. Coumanare Lakes near Dingle, 1898 : and on bank near Inch,
899: Lett & McA. On rocks in the Brandon Valley, June 1900
Lett & McA.) : McA. 1901.

4. Plagioohila punctata Tayl.

Ploffiachila punctata Tayl., Hepat. Antarct., in Lend. Joum. Bot.,
. 371, 1844. G. L. K, Syn. Hepat., Suppl., p. 626. Plagiochila
nnulosa var. punctata Carr., Brit. Hepat., part iii., p. 60. Garr., Irish
rypt., p. 19, tab. 2, fig. 3, 1863. Plagiochila punctata Moore, Irish
[epat., p. 644. Pearson, Hepat., Brit. Isles, p. 278, plate 117.

Districts I. - ni. IV. Vni. XI. XII.

Hah. — On shady banks and on rocks.

I. Killamey woods(Moore) : Carrington 1863. O'Sulliyan's Cascade,
iena, and Gromaglown: Lindberg 1875. Olena and Blackwater
ridge, Kenmare River: Scully 1890. On rocks. Tore Waterfall,
197: McA. & Lett 1B99. Common in the Dingle Peninsula:
cA. 1901. Caha Mtn., Co. Cork : McA. 1894.

III. CJraigue, Go. Garlow : McA. 1896 a.

458 Proceedings of the Royal Irish Academy.

lY. Altadore Olen and Seyen ChuroheB, Co. Wicklow : Moore.

Yin. Croaghpatrick and bank by the Mojire Biver, Go. Mayo,
Sept. 1901 : MoA. Kephin, May 1901 : Lett ft McA. Bugoie,
Devil's Mother, and Slievemore on Aohill, Sept. 1901 : Lett

XL Rathmnllen Wood and Saltpans Wood, Rathmnllan, Jalj 1902:
Hunter. Gratleagh Wood and wood by Mnlroy Bay near IGlford,
Sept. 1902 ; Goat Island, Lough Eask, June 1903 : McA.

XII. Colin Glen and Loughmoume, Co. Antrim, very rare (Lett) :
Stewart 1888.

5. Plagiochila tridenticnlata Tayl., Domort

Jtmgermania spinuloia yar. tridentietdata Hook., Brit. Jong.,
p. 9, tab. 14, figs. 9-10. Plagiochila tridentieulata Bum., Becoeil
p. 15. Tayl., in G. L. N., Syn. Hepat., p. 26. Moore, Irish Hepat,
p. 644. Pearson, Hepat. Brit. Isles, p. 280, plate 118.

Districts I. vin. XI. xn.

Hob. — On damp peat, rocks, and on the larger Hepaticae.

I. Cromaglown, Xillamey: Taylor 1836; Carrington 1S6S.
Glena ; Scully 1890. Tore Waterfall, Sept. 1897 : McA. & Lett
1899. Brandon and Connor Hill, 1873 : Moore & Lindberg. On
FruUania Tamarisei on Brandon, July 1881 : F. W. M. ft McA.
On Radula aquilegia, Loughanscaul, 1898 : Lett & McA. Mt Eagle,
1898 : Lett & McA. Near Bantry (Miss Hutchins, Mackay) : Taybr
1836. Glengariff: Carrington.

YIII. Devil's Mother, Slievemore on Achill, and Bangore, Oo.Mayo*
Sept. 1901 : Lett.

XI. By the Leenan River near Ramelton : Hart 1886.

XII. Drumnasole, Co. Ajitrim (Brenan): Stewart 1895.

6. Plagiochila ezig^ Tayl.

Jungertnania exigua Tayl., in Trans. Bot. Soc. Edinb., i., p. 1<^-
1843. Plagiochila exigua Tayl., in Lend. Joum. Bot., voL ▼., p. 264.
Moore, Irish Hepat., p. 645. Pearson, Hepat. Brit. Islee, p. 28S.
plate 119.

Districts I. VIII. .

Sah, — On the trunks of trees near the ground, and on FhdleMts
nnd Badula.

1. Enockavohila, Dunkerron, and Tore Mtn. on Ihdkmim Tamaruct^
1840: Taylor. Cromaglown, Eillaroey, 1865 : Moore. O^SolliTans
Cascade and Glena, 1875 : Moore. Mount Eagle near the lake, anc
Loughanscaul, 1898; Lough Duff in the Brandon Valley, IS99:

McArdlk — A List of hinh BLeimticm.

459

Lett&McA. Olengariff, Sept. 1869: Carrington 1874. Adrigole
near Glengariff, Rev. C. H. Binstead: Pearson 1902.
Vni. SUevemore, Achill, Sept. 1901 : Lett.

Genus 28. Jungermania Linn.
Sub-genus 1. Aplosia Dumort.

1. Jungermania cordifolia Hooker.

Hook., Brit. Jung., tab. 32. Jungermania tersa Nees, Nat. Eur.
Leb., i., p. 329. Aplozia cordifolia Dum., Hepat. Eur., p. 59. Moore,
Irish Hepat. » p. 647. Jungermania cordifolia Pearson, Hepat. Brit.
Isles, p. 290, plate 122.

District I. XII.

Hab. — On moist banks and on rocks in mountain streams.

I. On Mangerton in the stream from the Punch Bowl: Taylor
1836. Coomashana Lake: Carrington 1863 ; Moore 1876. Near the
Hunting Tower, Killamej : Scully 1890. Brandon, 1864 : Moore.
Maghanabo Glen, 1875 (McA.) : Moore 1876. By the stream which
flows from Lough Doon into the Brandon valley, 1881 : F. W.M.&McA.
Loughanscaul and Bamanaghea Lough near Anascaul, Sept. 1898 :
Lett & McA.

XII. Cushendun, Co. Antrim, 1836: Moore. On Sawell, Co. Derry,
%i 1600 feet, 1838 (Moore) : Stewart 1888.

2. Jungermania pumila Withering.

With., Brit. Flora, ed. 3, p. 866. Hook., Brit. Jung., tab. 17.
Aplozia pumila Dumort., Hepat. Europ., p. 59. Jungermania pumila
Onmort., Moore, Irish Hepat., p. 647. Pearson, Hepat. Brit. Isles,
>. 292, plate 123.

Districts I. IV. VIII. IX. X. XI. XII.

Hah. — On rocks and banks of streams.

I. Glen near the Hunting Tower, Killamey: Carrington 1863;
loore 1876. Tore Waterfall, 1873: Lindbergl875. Near Waterville
LiidBlackwaterBridge: Scully 1890. Brandon: Moore. Loughanscaul,
894 : McA. Margin of Mt. Eagle Lake, Sept. 1898 ; bank between
Smalougli and Inch, and in Derrymore Glen near Tralee, 1899 :
.ett & McA. Mtn. near Bantry, Miss Hutchins: Hooker 1816.

IV. Lough Bray: Moore 1878. Dargle (Stokes & Taylor):
looker 1816.

VIII. On the shore of Lough Conn at Pontoon, and on Nephin,
faj 1901 : Lett & McA. Doolough, Co. Mayo, Sept 1901 : Lett.

460 Proeeedinga of the Royal Irish Academy,

IX. Glenade, Go. Leitrim : Moore 1876.

X. On wet rocks, Ballyhaise Wood, Co. Cuvan, 1893: McA.
1898.

XI. Seashore on rocks, Macamish, July 1902 : Huoter. Bines-
more Gap, Goat Island, Lough Eask Woods, June 1908 : HcA.

XII. Near Belfast (Templeton) : Hooker 1816. Glendun, Co.
Antrim (Brenan), Slievenanee (Lett) : Stewart 1895. Hen Mtn.,
Co. Down (Waddell) : Lett 1890.

3. Jungennania riparia Taylor.

Tayl., in Ann. and Mag. Nat. Hist., Aug. 1843. Spruce, inPhjt,
March, 1843. Jpola%ia riparia Bum., Hepat. Europ.,p. 65. Moore,
Irish Hepat., p. 648. Jungermania riparia Pearson, Hepat Brit.
Idles, p. 294, plate 124.

Districts I. IV. — VI. IX. — XI. xn.

Mah. — In damp places, sides of streams, and on wet rocks.

I. Co. Kerry (Taylor) : Moore 1876. Tore Waterfall : CarringtoiL
Brandon : Moore. Wet rocks on Connor Hill, 1873 : lindherg 1875.
Maghanabo Glen, 1 875 : McA. Loughanscaul near Dingle, Sept 1B98 :
Lett & McA. Bare in the Dingle Peninsula: McA. 1901. Eniu^
cona, Co. Cork (I. Carroll) : Moore 1876.

IV. Lough Bray : Moore 1876 ; McA. 1890. Woodenbridge, Co.
Wicklow: Moore.

VI. Moist bank near Xilronan, Aran Isles : McA. 1895ff.

IX. Benbulben : Moore.

XI. Saltpans Wood, RathmuUan, July 1902 : Hunter.

XII. Glens in Co. Antrim, 1836: Moore (under JlinmitZtf). Rathhn
Island, rare: Stewart 1888. Colin Glen near Belfast (Waddeli;:
Drumnasole and Glendun (Brenan); Glenariff and Rasharkin Bog
(Lett): Stewart 1895. Cove Mountain: Lett 1890. Moygannon
Glen and ToUymore Park, Co, Down, very rare (Waddell) : Lett
1890.

4. Jungermania spbsBrocarpa Hooker.

Hook., Brit. Jung., tab. 74. Jungermania Ooulardii ^msh,^ Hepat>
Gall., no. 68, et Hepaticol. Gall., p. 29. Aphtia ephmrecarpa Bon.,
Hepat. Eur., p. 71. Moore, Irish Hepat., p. 648. 7. ephmroeer^.
Pearson, Hepat. Brit. Isles, p. 296, plate 125.

Districts I. IV. V. VIII. IX. X. — Xn.

J3[dh. — On stones near streams, and damp banks.

I. Tore Waterfall: Taylor 1836. Waterville, Glencar, sod
Mangerton : Scully 1890. Connor Hill, 1875 : McA. ; and 189T :

MoAkdlb— ^ Lint of Irish Hepaticce.

461

Lett & McA. Derryxnore Glen near Tralee, rare. May 1899 : Lett &
McA. Temple Michael Glen, Go. Cork (I. Carroll) : Carrington
1863.

IV. Lough Bray : Moore ; on stones at the margin of the lake :
MoA. 1890. Bargle Biver : McA. 1889.

V. Near Dublin (Taylor): Hooker 1816. Kelly's Glen, Co.
Dublin: Moore. The Quarries, Sutton, Co. Dublin, rare: McA.
1898 a. Omeath, Co. Louth : Lett 1890.

Yin. Nephin, May 1901 : Lett & McA.

IX. Olenade, Co. Leitrim : Moore.

X. Camlough Mountain, Co. Armagh : Lett 1890.

Xn. Glenarifi and Glenarm, Co. Antrim, rare, July and Sept.
1836 : Moore. CoHn Glen, Co. Antrim, 1816 (Templeton), Park-
more (Lett) : Stewart 1895. Carr's Glen : Stewart 1888. Mountain
liYulets above Mahera, 1886 (Moore) : Stewart 1888. Pigeon Bock
Mountain, and Tollymore Park, Co. Down (Lett) ; Eagle Mountain,
Co. Down, rare (Waddell) : Stewart 1888. Slievenamaddy and Hen
Mountain : Lett 1890.

Tar. lurida Dumort.

Jungemumia lurida Dum., Syll. Jung., p. 50. J, nana Nees, Nat.
Eur. Leb., i., p. 817. Moore, Irish Hepat., p. 648. Jungermania
ipharocarpa Hook., var. lurida, Pearson, Hepat. Brit. Isles, p. 298,
plate 126.

Sab. — On wet rocks and by the sides of streams in mountainous
districts.

OlengarifP (Miss Hutchins) : Moore 1876. Blackwater Bridge,
Kenmare: Scully 1890. Connor Hill, rare, 1875: McA. 1901.
Near the Seven Churches, Co. Wicklow : Moore. Lough Bray,
fertile, 1879 : McA. Kelly's Glen, Co. Dublin : Moore 1876.
Glenasmole, 1875 : McA.

The late Dr. Spruce and Mr. Pearson have concluded that there is
no specific difference between Jungermania spharoearpa Hook., and J.
lurida Dum. = •/] nana Nees ; and, as the result of careful comparison
of a series of specimens of both plants, I feel bound to agree with them.
The oell-0tructure of the two is the same ; var. lurida is a smaller
plant, darker-coloured, and having a more compressed mode of growth.

5. Jungermania crenulata Smith.

8m., Eng. Bot., tab. 1463. Hook., Brit. Jung., tab. 37. Aplozia
crenulata Dam., Hepat. Eur., p. 57. Nardia crenulata Lindberg,

462 Proceedings of the Royal Irish Academy.

Hepat. in Hibemia, p. ^29. McArdle, Hepat. Dmgle PeniiuaLi,
p. 321. Jung^rmania erenulata Pearson, Hepat. Brit. Isles, p. 300,
plate 127.

Districts I. II. III. IV, V. VI. — Vin. — X. XL XII.

JIah, — On moist banks, bogs, often on haid roads, in damp plae«6.

I. Kear Bantry (Miss Hutchins) : Hooker 1816. Dnnboy, Go.
Cork : McA. 1894. Koss Bay, Eillamey : Cairington 1868. Water-
ville and Glencar : Scully 1890. Connor Hill, 1878 : Lindberg 1875.
Common in the Dingle Peninsula : McA. 1901.

II. Bay Lough, Enockmeildown Mtns., Co. Tipperary, June 1903:
McA.

III. Roadside, Slieve Bloom, Queen's County: McA. 1892a.

IV. Powerscourt (W. Stokes) : Hooker 1816. Sideof the riTcrat
the Seven Churches: Moore. Luggielaw, 1878: Lindberg 1875.
Woodenbridge, Sept. 1894 : McA. By the roadside at Loag^Bny,
1889 : Scully & McA. Glencullen, Co. Wicklow, 1887 : McA Bank
of the Urrin Eiver, near Knockroe, Co. Wexford, May 1899 : KcA.
1903.

V. Howth: McA. 1893a. Kelly's Glen, Co. Dublin: M<»re.
Golden River on Carlingford Mountain, Co. Louth : Lett 1890.

VI. Ballyvaughan, Co. Clare, and Gentian Hill near Galway:
McA. 1895 0.

VIII. Connemara : Moore 1876. Pontoon near Fozford, and on
Nephin, May 1901 : Lett & McA. Doolough and Achill, Sept 1901 :
Lett.

X. Slieve Glah, Co. Cavan, sparingly, 1893: McA. 1898.

XL Rathmullan, seashore at Macamish Hill, above Mintiagbs,
July 1902: Hunter. Gartan Wood, Columbkil Lake, Cratl^^
Wood near Milford, Sept. 1902; Goat Island, Lough Eask Woods.
Errigal, June 1903 : MoA.

XII. Warrenpoint, Tollymore Park, Moume Mountains, frequiettt
( Waddell) ; Slieve Commedagh (Lett) : Stewart 1888. Hen Mountain :
Lett 1890. Narrow Water demesne, 1894 : McA.

var. qraeiUima Smith.

Jungermania erenulata var., Hook., Brit. Jung., tab. 37. J. grttH-
lima Smith, Eng. Bot., tab. 2238. «/. genthiana Hueben., Hepat
Germ., p. 107. Aplozia gracillima Dura., Hepat. Eur., p. 57. 3Vrfia
gracillima Lindberg, Hepat. in Hib., p. 530, 1875. JunganMU*
{Aploiia) erenulata var. gradUima Moore, Irish Hepat., p. 647. Pearson,
Hepat. Brit. Isles, p. 302, plate 128.

McArdlb — A Lisi of Irish Hepatieee.

468

Eab. — In shallow patches on moist sandy banks, often on hard
damp gronnd by roadsides.

JOitch bank, Folleen, Berehaven: McA. 1894. Killarney: Car-

rington 1863. Waterrille and Olencar : Sonlly 1890. Freqnent in

the Dingle Peninsula: McA. 1901. Galtee Mtns., Go. Tipperary,

Jane 1902 : McA. KiUoughrim Oak Forest near Enniscorthy, May

1899 : McA. Westaston, Co. Wicklow : Moore 1876. Lough Bray>

1873: Lindberg 1875. By the roadside at Lough Bray, 1889:

Scully ft McA. Woodenbridge, Co. Wicklow, Sept. 1894: McA.

Ditch bank near the Daily Lighthouse, and on hard ground among

Tocks near Sutton, Co. Dublin: McA. 1893a. Golden Eiver on

Carlingford Mtn., Co. Louth, rare: Lett 1890. On the shores of

lough Conn at Pontoon, and on I^ephin, May 1901 : Lett & McA»

Bangore, Co. Mayo, Sept. 1901 : Lett. On diunp ground, Killakeen,

Slieve Glah, and on the shores of Lough Cultra, Co. Cavan, 1893 :

McA. 1898. Wood by Mulroy Bay, Sept. 1902 : McA. On damp

peat, Lough Belshade/ Errigal, June 1903 : McA. ToUymore Park

ui<l Sliere Commedagh : Lett 1890. Aghaderg : Lett. Moygannon

(lien, Co. Down (Lett): Stewart 1888.

Sub-genus 2. Oymnocolea Dumort.
6. Jungermania inflata Hudson. s

J, inflata Huds., PI. Angl., p. 511. Hook., Brit. Jung., tab. 38.
GymnocoUa inflata Dumort., Kecueil, p. 17. Moore, Irish Hepat.
). 654. Pearson, Hepat. Brit. Isles, p. 307, plate 131.

Districts I. II. — IV. V. — VU. Vin — X. XI. XII.

ffah. — On bogs and heathy places, on stones by streams.

I. Near Bantry (Miss Hutchins) : Moore 1876. Upper Olencar,
M, Kerry : Scully 1890. Bog between Emalough and Inch, Co. Kerry,,
are, 1899 (Lett & MoA.): McA. 1901.

II. Ghdtee Mtns., Co. Tipperary, June 1902 : McA.

IV. On stoDos, Powerscourt Waterfall: Taylor 1836. Lough
{ray : Moore 1878; fertile there, 1880 (McA. & P. W. M.) : McA.
890.

V. Featherbed Mtn., Co. Dublin : Moore. Finglas Quarries: McA*
'ommon on Howth : McA. 1893 a,

VU. Bog near Oeashni, King's Co., 1894: EusseU.

VIII. Diamond Mtn., Co. Galway, 1891 : McWeeney. Croagh-
atrick. Sept 1901 : McA. Shores of Lough Conn at Pontoon and
D Nephint May 1901 : Lett ft McA.

464 Proceedings of the Royal Irish Academy.

X. Moist "bank near the summit of Slieve Glah, Co. Cavan, 1893 :
McA. 1898.

XI. On schist rock behind Rathmullan, 1903 : Hunter.

XII. Moist banks, Giant's Causeway, 1836 : Moore. Seashore at
Holywood, Febr. 1903 : Hunter.

var. eompaeta Carrington.
On dry rocks, top of the Hill of Howth (Moore) : Carrington 1863.

var. laxa Carrington.

Lough Guitane, Dean's Bridge, and Ross Bay, Eillarney (Ctr-
rington) : Moore 1876. RathmuUan, July 1902 : Hunter. On
Sphagnum^ Holywood Hill, Febr. 1903 : Hunter.

var. heterostipa Lindberg.

Lindb. in Am. and Lindb., Muse. Asm Bor., p. 47, 1888.
Kaalaas, Leverm. Norge, p. 290, 1893. Cephakmia heUroit^ ^!^6ii'
et Spruce, Spruce on Cephalozia, p. 55, 1882.

Hob. — On wet rocks in mountainous districts.

On schist rocks, Eathmullen, Co. Donegal, 1903 : Hunter. 5e«
to the Iri^ cryptogamic flora.

7. Jungermania torbinata Raddi.

e/. turhinata Eaddi, in Act. Soc. Sc. Modena, zviii., p. 29, tab. 1 U.
figs. 2, 3. 7. affinis Wils., in Hook., Brit. Fl., ii., p. 128. J. eorcf-
racea Nees, Nat. Eur. Leb., ii., p. 89. J. Wihmiana Nees, Nat Eur.
Leb., iii., p. 548. J. affints Wils., Moore, Irish Hepat., p. 664.
£. turhinata Pearson, Hepat. Brit. Isles, p. 309, plate 132.

Districts I. — III. IV. V. VI. VII. XI. XII.

Rah. — On damp shady banks or on damp rocks, in the limestone
districts chiefly.

I. Eillarney: Carrington 1863. Tore Cascade and CarantuaL:
Moore.

in. Quarry bank near Qoresbridge, Co. Carlo w: McA. 1896^.
Bank by roadside, Slieve Bloom, Queen's Co. : McA. 1892«.

IV. Bank by the Dargle River: McA. 1889.

V. Leizlip, Co. Kildare {aeutiloha) i McA. 1893 i. Woodlands
near Dublin, 1830 (W. Wilson) : Taylor 1836. Finglas qoaniea m-ar

Mc Akdle — A List of Irish Hepaticm. 465

Dublin (McA.) : Moore 1876. Kelly's Glen, Co. Dublin, Aug. 1896 :
McA.

VI. Oak Wood and Doon Bog, Clonbrock, Co. Galway : McA. 1 896 h

VII. Bog at Geashill, King's Co., 189a; Emo Park, Portarlington,
1898: Russell.

XI. Clay banks near the quay at Donegal, Mount Charles, Barnes-
more Ghip, Goat Island near Lough Eask, June 1903 : McA. Bundoran,
1894 : Professor T. Johnson. (On these specimens, Spharospora hi-
nominaia, Massee, was growing, an addition to the Irish fungus flora.)
On rocks, seashore, Macamish, July 1902 : Hunter.

XII. On white limestone, Glenarm: Moore. Colin Glen near
Belfast, 1837: Moore; and (Waddell) : Stewart 1888. Giant's Cause-
way, 1836 : Moore. Carr's Glen and Springfield Glen : Stewart 1888.
White rocks, Portrush, 1893 : Russell. Narrow Water demesne, Co.
Down, 1894: McA.

Sub-genus 3. Lophozia Dumortier.
8. Jungermania bantriensis Hooker.

J. hantrtensu Hook., Brit. Jung., in note under J. stiptUaeea^
no. 41. tT. hidentata var., Hook., Brit. Jung., Suppl., t. 111.
Jungermania hygrophylla Spruce, Trans. Bot. Soc. Edinb., ii., 1846.
J, eukaris Wils. MS., in Spruce, Hepat. Pyr., Trans. Bot. Soc. Edinb..
iii., p. 20. J, hantriensU Hook., Moore, Irish Hepat., p. 649. Pearson^
Hepat. Brit. Isles, p. 312, plate 133.

Districts I. — in. IV. IX. XII.

Hah, — ^In loose patches on rocks and damp banks.

I. Brandon : Moore. Laham Wood near Bantry (Miss Hutchins) :
Hooker 1816 (under J, stipulacea^ tab. 41). GlengarifE (Carrington) :
Moore 1876.

III. Among Campylopw in a sandy deposit beside a stream on
Slieve Bloom, near Clonaslea, Queen's Co., August 1891 : McA.
1892 a.

IV. Lough Bray (Taylor) : Hooker 1816.
IX. Benbulben: Moore 1876.

XII. Annahilt Bog, Co. Down (Herb. Belfast Mus.) ; glen on the
shore of Belfast Lough (Templeton in Herb. Belfast Mus.); Colin
Glen near Belfast : Stewart 1888.

yar. MuelUri Nees.
Jungirmania MueU&ri Nees, in Lindenb., Syn. Hepat. Eur., p. 39.

466 Proceediugs of the Royal Irish Academy.

Lophozta Muelleri Dum., Recueil, p. 17. J. hantrienm Hook., tar.
Muelleri Nees, PearBon, Hepat. Brit. Isles, p. 314, plate 3S4.

.fia^.^- Among damp rocks.

Glenade, Co. Leitrim, May 1875 : Moore.

var. acuia Lindenberg.

Jungermania acuta lindenb., Syn. Hepat Eur., p. 88. hiflssa
acuta Dumort., Eecueil, p. 17. J. hantrienne Hook., var. ockiA
Lindenb., Pearson, Hepat. Brit. Isles, p. 316, plate 135.

jy^. — On damp rocks.

Connor Hill, emong Plagioiheeium dentieulatum^ 1873; Lindber?

1875.

var. Momeehuchiana Nees.

Jungermania JSormehuchiana Nees, Enrop. Leberm., ii., p. l^'
Moore, Irish Hepat., p. 660.

ffcA, — Among damp rocks in mountainous places.

Cromaglown, Killamey, and among rocks near Tore Mountaia,
July 1869: Moore. Stream near Woodenbridge, Co. Wicklov:
Moore 1876.

Note.— Dr. Moore states* that *' this form might be easily paaed
over for J. riparia, both in a fresh and a dried state. It isonly w^«-^
the peculiarly notched subvertical leaves, with their amphigistna
(stipules), are examined, that its distinguishing characterB are ob-
served. Dr. Lindberg considers J. Somsehuchianay J. banirisntit, and
J, Muelleri to be forms of one species."

9. Jungermania capitata Hook.

J. exeisa Dicks., PI. Crypt. Brit., fasc. iii., p. 11. /. tapi^^^
Hook., Brit. Jung., tab. 80. J. intermedia var. eapitata Nees, Enrop,
Leberm., ii., p. 125. J, capitata Hook., Moore, Irish Hepat., p. ^5-
Pearson, Hepat. Brit. Isles, p. 822, plate 138.

Districts I. II. VIII. IX. — XI. Xn.

Jlab» — On mountain rocks in damp places, often epiphytic <m tb
larger Hepatics, such as PruUania,

I. Near Bantry (Miss Hutehins) : Hooker 1816. Bog betwea
Emalough and Inch, Co. Kerry, rare, 1899 : Lett & MoA.

II. Glengarra Wood, and among Campylopus fragiUe near Loush
Muskry, Galtee Mtns., Co. Tipperary, Jime 1902 : McA.

YIII. Connemaro, 1891 : McWeeney. Murrisk near Westport.

1 Report on Iziah Hepatics, Proe. R. I. A., 8er. 2, vol. 2»

McArdlb— 2* List of Irish EepaticcB. 467

Sept. 1901 : McA. On the slopes of Devfl's Mother, Co. Mayo, Sept.
1901 : Lett.

IX. Benbulben, 1880: McA.

XI. Eathmullan, Saltpans Wood, Eathmelton, July 1902 : Hunter.
On Drullania, Qoat Island, Lough Eask, June 1902 : McA,

Xn. On rocks by roadside, Tonaghmore near Saintfield, Co. Down,
1900 : Waddell, Ir. Nat., vol, xii., p. 219, 1908.

11, Jungermania biorenata Schmidel.

J, hierenata Schmid., Anal., p. 347, t. 64, fig. 1. Lophozia hicrenata
Dum., Kecueil, p. 17. Jungennania hiorenatalAn^enh,^ Synop. Hepat.,
p. 82. Moore, Irish Hepat., p. 652. Pearson, Hepat. Brit. Isles,
p. 324, plate 139.

Districts I. II. — lY. V. Vm. XIL

Hah, — On exposed rocks on the ground among heaths, and on damp
shaded banks.

I. Temple Michael, Co. Cork (I. Carroll) : Moore 1876. Kinnordy,
Co. Kerry (Taylor) : Carrington 1863. M*Gillicuddy's Keeks at 2000
feet: Scully 1890.

II. Knockmeildown Mtns., Co. Tipperary, June 1902 : McA.

IV. Lough Bray, Co. Wicklow: Moore 1878 ; McA. 1890.

V. On hard peaty soil, Ballykill, Howth, 1893-4 : McA. 1897. In
Howth demesne among rocks, April 1895, yery rare : McA. 1897.

VIII. Near Letterfrack and Eylemore, 1874 : Moore. Kylemore,
1891 : McWeeney.

xn. On rocks at roadside between Banbridge and Scarva :
Waddell in Guide to Belfast, 1902.

12. Jungermania yentrioosa Dicks.

J, ventrieosa Dicks., PL Crypt. Brit., faso. ii., p. 14. Hook., Brit.
Jung., tab. 28. Lophozia ventrieosa Dum., B-ecueil, p. 17. Junger-
mania ventricosa Moore, Irish Hepat., p. 652. Pearson, Hepat. Brit.
Isles, p. 327, plate 140.

Districts I. U. — IV. V. VI. — VIII. IX. X. XI. XII.

Sdh. — On banks and rocks in mountainous places.

I. I^ear Bantry, fertile in November (Miss Hutchins) : Hooker
1816. KiUamey, rare : Carrington 1863, Slieve Mish and MagiUi-
cuddy's Reeks : Scully 1890. Connor Hill, 1873: Lindberg 1875;
1880 and 1897 : McA. Brandon : McA. 18.80 ; and Sept 1897 : Lett &
McA. Derrymore Glen near Tralee, and on bank between Emalough
and Inch, 1899 (Lett & McA.) : McA. 1901.

a.I.A. PBOC, VOL. XXIV., BBC. B.] 2 P

468 Proceedings of the Royal Irish Acctdemy.

n. G^tee and Enockmeildown Mtns., Co.Tipperary: Moore 1876;
and June 1902 : McA.

IV. Co. Wicklow: Taylor 1836. Lough Bray: Moore 1878;
Lindberg 1875 ; McA. 1890. Lugnaquilla, May 1896 : McA.

V. Boulacross Mtn., fertile in May: Hooker 1816; Taylor 1836.
Near Dublin (Mackay) : Hooker 1816. Anglesey Mtn., Co. Louth:
Lett 1890. Howthj^ 1894 and 1895 : McA. 1897.

VI. Shore of Lough Corrib, Co. Gal way : McA. 1895 a.

VIII. Kylemore, 1891 : McWeeney. Slopes of DeviPB Mother,
and Slieveraore on Achill, Sept. 1901 : Lett. Pontoon near Foxford,
and on Nephin, May 1901 : Lett & McA. Murrisk near Westport,
Sept. 1901 : McA.

IX. Benbulben : Moofe 1876.

X. Top of Camlough Mtn., and Slieve Gullion, Co. Armagh : Lett
1890. Slieve Glah, Co. Cavan, 1893: McA. 1898.

XI. Saltpans Wood, RathmuUan; Croghan Mtn. ; Carradoan Wood,
Bathmnllao, July 1902 : Hunter. Bunlin Waterfall near Milford,
Sept. 1902 : McA.

XII. Sallagh Braes, Co. Antrim, 1837 : Moore. Moygannon Glen
and Slieve Martin: Lett 1890. Summit of Slieve Commedagh, and
Eostrevor Wood, Co. Down (Waddell) : Stewart 1888.

var. pwrphyroUuca Limpr., in Cohn, Krypt. Fl. Schles., i., p. 280.

Jung&rmania parphyroleuea Nees, Nat. Eur. Leb., ii.,p. 78.

Sah, — Growing in patches, in subalpine situations, on damp peat,
and on decayed wood.

Lough Bray: McA. 1889. Clermont Mountain and Carling-
ford, Co. Louth (Waddell), Slieve Gullion: Lett 1890. Pontoon
near Foxford, and on Nephin, May 1901 : Lett & McA. Among
rocks, Columbkil Lake near Milford, Sept. 1902 : McA. BaUyvalley
near Eostrevor, very rare (WaddeU) : Stewart 1888.

13. Jungermania alpestris Schleich.

J. alpestris Schleicher, Exsicc., oent. ii., no. 59. Web., Prwl.,
p. 81. Jungermania sudetica Nees, in Hueben., Hepat. Germ., p. 142.
J, Oospertiana Hueben., Hepat. Germ., p. 254. Csphalotia alpestris
Cogn., Hepat. Belg., p. 35» Jungermania alpestris Schl., Pearson,
Hepat. Brit. Isles, p. 331, plate 142.

Districts L — IIL IV. Vv — -^ IX, X, — XIL

J3M» — On banks and rooks in mountainous places.

I. Kinnordy, Co. Keny (Taylor) : Carrington 1863. Connor Hill,

McArdle — A Lint of IrUh Hepaticcs. 469

June 1894: McA. Mount Eagle, Sept. 1898; Derrymore Glen near
Tralee, May 1899: Lett & McA. Bare in the Dingle Peninsula:
McA. 1901.

III. Among rocks near the River Barrow at Graigue, Co. Carlow,
yery rare, 1895 : McA. 1896 a.

rV. Lough Bray, March 1892 : McA.

V. Bank at the Rabbit Warren, Howth, 1893, rare : McA. 1897.

IX. Benbulben, 1880 : McA.

X. Damp bank, Slieve Glah, Co. Cavan, 1893 : McA* 1898.
Xn. Qlenariff, Co. Antrim, 1836 : Moore. Near Saintfielcl, Co.

Down: Waddell in Guide to Belfast, 1902.

14. Jungermania inoisa Schrader.

J, incisa Schrad., Syst. Samml. Krypt. Gew., ii., p. 5. Hook., Brit
Jung., tab. 10. ZopJiotia incisa Dum., Recueil, p. 17. J. ineisa
Schrad., Moore, Irish Hepat., p. 653. Pearson, Hepat. Brit. Isles,
p. 335, plate 144.

Districts I. — III. IV. V. VI. VII. VIII. IX. X. XI. XII.

JS^ah. — On bogs and heaths, and among rocks.

I. Near Bantry (Miss Hutchins) : Hooker 1816. Near Coonea-
«hana Lake : Carrington 1863 ; Moore 1876. Slieve Mish and near
*the Hunting Tower, Killamey : Scully 1890. Bogs about Connor
Hill : McA. Coumanare Lakes, Sept. 1 898 ; Derrymore Glen, May
1899 ; bog between Emalough and Inch, Co. Kerry, 1899 : Lett & McA.

IIL Slieve Bloom, Queen's Co., 1891 : McA. 1892 a.

IV. Lough Bray : McA. 1890. Douce Mountain, 1897 : McA.

V. Damp peaty banks, Howth : McA. 1893 a.

VI. Bogs about Clonbrock, Co. Gal way, common: McA. 1896 5.

VII. Bog near GeashiU, King's Co., 1890 : Russell.

VIII. Corslieve and Bangore mountains, Co. Mayo: Moore 1876.
Kylemore, and abundant on the top of Mweelrea, 1874 : Moore.
Nephin, May 1901 : Lett&McA. Slievemore, Achill, Sept. 1901 : Lett.

IX. Slish Wood, Co. Sligo : Waddell 1892.

X. Derrytagh Bog, Mintiaghs, Co. Armagh, 1882 : Lett.

XI. Rathmullan, July 1902 : Hunter.

XII. Aughalogan Bog, Parish of Duneane, Co. Antrim, rare, 1888 :
Moore. Lisbane near Saintfield, and Ballygowan : Waddell in Chiide
4)0 Belfast, 1902«

15. Jungermania ezseota SchmideL

•T. essecta Schmid., Ic. et Anal., p. 241, tab. 62, fig. 2. Hook.,

2P2

470 Proceedings of the Royal Irish Academy.

Brit. Jung., tab. 14, et suppL, tab. 1. Lophona exuda DanLt
Becueil, p. 17. J. exseeta Scbm., Moore, Irish Hepat, p. 661.
Pearson, Hepat. Brit. Isles, p. 837, plate 145.

Districts I. IV. — VI. — VIII. IX. XII.

Mfib, — On damp banks, on decayed wood, and among locb io
woods and heaths.

I. Near Bantry (Miss Hutchins) : Hooker 1816. Ballinhassigind
Kildowry, Co. Cork (I. Carroll) : Carrington 1863. On rotten logs,
Cromaglown, Killamey: Carrington 1863. Olencar: ScuUj 1890.
Bank near Mount Eagle Lake, July 1881 : F. W. M. and McA. Bar
in the Dingle Peninsula: McA. 1901.

IV. Lough Bray, 1887, female, very rare: McA. 1890.

VI. In the oak wood, on 8heep-poolbog, and Doon bog, Qonbrod,
Co. Galway, 1896, very rare : McA. 1896 h.

Vin. Slievemore, Achill, Sept. 1901 : Lett

IX. Gleniff, Co. Leitrim : Moore.

XII. SaUagh Braes, Co. Antrim : Moore. The Warren, Holy-
wood (Templeton) : Herb. Belfast Museum.

16. Jungermania ezseotsBformis Breidler.
J. exseotaformis Breidler, Leberm. Steiermarks, in Mitth.d.natarw.

Ver. fiir Steiermarks, Jahrg. 1893, p. 321.

Districts IV. — VI.

Mah, — On bogs, closely adhering to peat, and on old wood.
IV. On decayed wood. Lough Bray, Co. Wicklow, July 1887: McJL
VI. On peat, Doon bog, Clonbrock, Co. Galway, June 1896 : McA.

New to the Irish cryptogamic flora.

17. Jnngermania Lyoni Taylor.

Jung&rmania quinquedentata Web. et Mohr., Bot. Tasch^b., p. 43 )
Eng. Bot., tab. 2517. Jungermania Lyoni Tayl., in Trans. Bot- S*:.
Edinb., i., p. 116, tab. 7. Dum., Hepat. Europ., p. 73. Moore, Irifi
Hepat., p. 651. Pearson, Hepat. Brit. Isles, p. 339, plate 146.

Districts I. IV. XII.

Rah, — On rocky banks among Mosses, in patches interwoTen.

I. On rocky banks, Coumanare Lakes, Co. Kerry, 1 898 : Lett & MoA
Very rare in the Dingle Peninsula : McA. 1901.

IV. Glenmalure, Co. Wicklow, among tufts of .Sct^jMtnia nw-
nata, very rare, single stems only found : Moore.

XII. Rostrevor Mountain, Moygannon Glen, and SLieve Dobju.
Co, Down, rare (Waddell): Stewart 1888; ^d Lett 1890 {\^k:

McArdlk — A List oj In'sh Hepaticce.

471

/. qumquedentata), Divis (Thompson), Caye Hill, and Sallagh Braes,
Co. Antrim : Stewart 1888 (under J. quinquedentata).

18. Jnngermania gracilis Schleicher.

/. graeil%$ Schleicher, PI. Crypt. Helv., cent. iii«, no. 60. J. harhata
w. mimr Hook., Brit. Jung., tab. 70, figs. 18-20. J. harhata var.
\Uenmta Mart., Fl. Crypt. Erl., p. 177, tab. 6, fig. 50 e. J, attetiuaia
[indenb., Hepat, Enr., p. 44, no. 40. McArdle, Proc. R. I. Acad.,
Wser., vol. iv., no. 1, 1897.

UiBtricts I. II. V. r — — XI. — .

Bab. — Among rocks in heathy places.

1. Near Bantry, so embedded among Mosses and tufts of Seapania
^^morosa that only the tops of the shoots appear (Miss Hutching) :
looker 1816.

IL Baylough, Knockmeildown Mtns., Co. Tipperary, rare^ June
902 : McA.

V. Near Dublin (Taylor) : Hooker 1816. Among rocks in tufts
f Leueobryum glaueum and Telraphis pellucida at Ballykill, Howth,
leotiful, June 1893, and Eebr. 1894, and very fine in Howth demesne,
895: McA. 1897.

XI. Among Diplophyllum Micam^ Columbkil Lake near Milford,
ept. 1902 : McA.

19. Jnngermania barbata Schreber.

J. harhata Schreber, PI. Lip., p. 107. Hook., Brit. Jung., tab. 70,
I' 7-8. Laphozia harhata Dum., Recueil, p. 17. Moore, Irish
<;pat., p. 344. Jungermania harhata Schr., Pearson, Hepat. Brit,
les, p. 344, plate 148.

Districts L IV. V. YIII. — X. XI. XIL

Hob. — On damp shady rocks and banks.

I. Connor Hill, 1877 : McA. 1880. Frequent in the Dingle
niinsula : McA. 1901.

IV. Co. Wicklow : Moore 1876. Douce Mtn., 1897 : MoA.

V. Co. Dublin, frequent: Moore 1876. In Howth demesne,
prill 885: McA. 1897.

VIII. Kephin, May 1901 : Lett ft McA.

X. Bank among Diplophyllwn albicans, Slieye Glah, Co. Cavan,
ry scarce, 1893: McA. 1898.

XI. Co. Donegal, abundant: Moore 1876. Near the lake at
Bcamiah and behind the church at BathmuUen, July 1902 : Hunter.

Xn. Co. Antrim, frequent, 1838: Moore. Near Claggan, and on

472 JProceedtngs of the Royal Irkh Academy.

Slemish, Co. Antrim, rare, 1836 : Moore (under /. incUa), Basharkiii
bog, Co. Antrim (Lett) : Stewart 1895.

20. Jnngermania lyoopodioides Wallroth.

J. lycopodxoides •"Wallr., Fl. Crypt. Germ., iii., p. 76. J. J«tW«
var. lycopodiotdes Nees, Nat. Eur. Leb., ii., p. 1 85. Laphmia lywj^i»^
Cogn., Hepat. Belg., p. 31. Pearson, Hepat. Brit. Isles, p. 346,
plate 149.

var. FloerJcii Web. et Mohr.

J. Floerkii Web. et M., Bot. Taschenb., p. 410. /. harhftt tst.
Floerhii, G. L. K, Syn. Hepat., p. 123. Moore, Irish Hepat, p. 651.
J, lyeopodioides var. Floerkii Lindb., Muse. Scand., p. 7. PeaiwB,
Hepat. Brit. Isles, p. 348, plate 150.

Districts VIII. r XI. -T.

Hah, — On damp rocks in subalpine situations.

VIII. Croaghpatrick, Sept. 1901 : McA.

XI. Muckisb: Moore 1876.

Sub-genus 4. Sphenolobns Lindb.
21. Jnngermania minuta Crantz.

/. minuta Crantz ex Dicks., PI. Crypt. Brit., fasc. xi., p. 18. H«i.
Brit. Jung., t. 44. Diplophyllum minutum Dum., Recueil, p. 1^
Junyermania {Sphenolohum) minuta Moore, Irish Hepat., p. 649.
Pearson, Hepat. Brit. Isles, p. 353, plate 153.

Districts I. II. — IV. V. VIII. .

Had. — On heaths aud rocky banks in subalpine districts.

I. About Bantry (Miss Hutchins) ; Hooker 1816. Mangeiton-
Carrington 1863. Connor Hill, and banks about the Coumanare Lik^
near Dingle, 1898 ; shores of Lough Duff in the Brandon Valief,
1899; Bamanaghea Lough near Anascaul, 1898 : Lett & McA.

II. Knockmeildown Mtns., Co. Tipperary, June 1902: MeA.
Book Lough, Coomeraghs, Co. Waterford, July 1902: Waddell.

IV. Lough Bray: Taylor 1836. Fertile there, 1889: ScuBt 4
McA. Seven Churches : Moore 1876.

V. Glencullen, Co. Dublin, 1890 : Scully & McA. On a peaty
bank among rocks at Ballykill, Howth : McA. 1893 a. Hearth demeazu*.
very fine, 1896 : McA. 1897.

VIII. Nephin, May 1901 : Lett & McA. Slieremore, Adat
Sept. 1901 : Lett.

McArdle — A List of Irish Hepaticce.

473

Sub-genns 5. Anastrepta Lindb.
22. Jnngermania oroadensis Hook.

/. oreadensis Hook., JBrit. Jung., tab. 71. Mesophylla oreadensis
DniQ., Hepat. Eur., p. 130. Jungirmania aread&nsis Hook., Moore,
Irish Hepat., p. 650. Pearson, Hepat. Brit. Isles, p. 357, plate 155.

Districts I. VIII. >^ .

jffah. — Growing in loose patches on rock-ledges, or mixed with
)thvT Hepaticfle or Mosses, in alpine and subalpine situations.

I. On the summit of Brandon among Sypnum hreum, 1813
Taylor) : Hooker 1816. On the east side of Brandon, near the sum-
nit, among Scapania ornithopodioides, Jxme 1900: Lett & McA.
^DDor Hill, among JSinrberta adunea, 1873 : Lindberg 1875. Glencar,
md at 2500 ft. on the Reeks : Scully 1890.

VIII. On Slievemore, Achill, Sept. 1901 : Lett.

Note. — An interesting Jungermania was found by Mr. Hunter on
thist rocks at Eathmullen, Co. Donegal, in 1893, unfortunately
rithout fruit. Prof. Kaalaas, to whom specimens were sent for
ientification, writes : — ** The most striking character in the plant is
be almost rectilinear truncate leaves, which, with regard to their
mn, are very like those of Jungermania alpestris ; but it cannot for
lany reasons be referred to that species. In most respects, it seems
) come near J. Wentelii^ which, however, is somewhat larger, and
ioreover an alpine species that is not likely to be met with in Ireland,
do not know any European species to which your plant with cer-
inty can be referred. Until fertile specimens shall be found it will
t difficnlt to settle this question."

Genus 29. Hardia Ghray and Bennett.

Subgenus 1. Enoalyz Lindberg.
1. Hardia hyalina Lyell, Garrington.

Jungermania hyalina Hook., Brit. Jung., tab. 63. Aphnia hyalina
amort., Hepat. Eur., p. 58. Nardia hyalina Garrington, Brit. Hepat.,
35, pi. 11, fig. 36. Moore, Irish Hepat., p. 657. Pearson, Hepat.
it. lales, p. 364, plate 158.

Districts L — III. IV. V. VUI. — X. XI. XII.

Hob. — Moist banks and by the sides of streams in rocky places*

I. Aooreagh Eiver near Sneem, rare: Taylor 1836. Brandon,
23 (W. Wilson): Moore 1876 ; and in 1881 : F. W,M.&McA. Gonnor
ill, 1881 : P. W. M. & McA. Between Emalough and Inch, 1899 :

474 Proceedwg8 of the Royal LHsh Academy,

Lett & McA. Olencar and the Paps : Scully 1890. Bimboy Wood,
Bere Island, and Dursey Island : McA. 1894.

III. Graigue, Co. Carlo w, rare: McA. 1896 n.

lY. Luggielaw and Seven Churches : Moore. Longji Bny, 1B87
(F. W. M.): McA. 1890. GlencuUen, Co. Wicklow, 1887: McJL
Near Arklow, 1895 : McA.

V. Seefin Mtn. near Dublin : Hooker 1816 ; Taylor 1836.
Anglesey Mtn., Go. Louth, rare : Lett 1890.

VIII. Co. Galway, 1891: McWeeney. Nephin, very rare, MsT
1901; Lett & McA.

X. Slieve Glah, Co. Cavan : McA. 1898. Slieve GulHon, Co.
Armagh, 1894 : McA.

XI. Among rocks, Columbkil Lake, and wet bank near Milioii
Sept. 1902 : McA.

XII. Co. Antrim (Moore) : Carrington 1863. Kostrevor Mtn.,
ToUymore Park, and Spinkwee Glen, Co. Down (Waddell): Stewart
1888. Slievenabrock and Slievenamaddy, Cove Mtn. and HenUtu^
Co. Down: Lett 1890. Parkmore, Co. Antrim (Lett): Stewat
1895. Narrow-water demesne, Co. Down, 1894: McA.

2. Nardia oboyata Nees, Carrington.

Junysrmania ohovata Nees. Nat. Eur. Leb., i., p. 332. ImuI^^
ohovata Lindb., Bot. Not., 1 872. Southhya ohovata Dum., Hepat EoTn
p. 133. Nardia ohovata Carr., Brit. Hepat, p. 32, pi. 11, % ^^
Moore, Irish Hepat., p. 657. Pearson, Hepat. Brit. Mes, p. 366,
plate 159.

Districts I. n. — IV. — VI. — VIII. — X. XI. XIL

Hah, — On moist rocks and by the sides of streams.
I. Tore Mtn., Killamey, 1829 (W. Wilson) : Moore 1876. BUtf-
ney: Scully 1890. Cromaglown, fertile, June 1861: Caningtoi
1874. Brandon, 1 823 (W.Wilson) : Carrington 1 874 ; and 1895 : McA.
Connor Hill (Moore) : Pearson 1902 ; 1881 : F. W. M. & Mci.:
1898 : Lett & McA. Rocky shores of Lough Duff, 1899 ; Coumsnare
Lakes near Connor Hill, 1898: Lett & McA. Mt. Eagle, ISbl:
F. W. M. & McA.

II. Glengarra Wood and among rocks over Lough Moskry, GjJtee
Mtns., Co. Tipperary, June 1902 : McA.

IV. Lough Bray (Moore) : McA. 1890.

VI. Cam Seefin, Co. Clare : McA. 1895 a.

VIII. On the slopes of Devil's Mother, Co. Mayo, Sept 1901 :
Lett.

McAkdlb— ^ List oflrUh Hepaticm. 475

X. Moist bank among stones, Slieve Glab, Co. Cavan : McA. 1898.

XI. Hill above Mintiaghs, Ratbmullen, July 1902 : Hunter.
Gartan Wood, Columbkil Lake near Milford, Sept. 1902 ; Goat
Island near Lougb Eask, June 1903 : McA.

XII. Bocks in a stream on Slieve Donard, Co. Down, very rare :
Lett 1890. Tollymore Park : Waddell 1892.

Sub-genus 2. Eonardia Lindberg.
3. Nardia oompressa Hook., Gr. & Benn.

Jungermmia eampressa Hook., Brit. Jung., tab. 58. Nardia
empresM Gr. & B., Nat. Arr. Brit. PL, p. 694. Altctdaria compresia
Book., Moore, Irish Hepat., p. 656. Pearson, Hepat. Brit. Isles,
p. 368, plate 160.

Districts I. IV. V. VIII. XI. XII.

Bab. — On wet rocks and stones by sub-alpine rivulets.

I. Near Bantry (Miss Hutcbins): Hooker 1816. Aooreagb Biver
near Sneem, 1831 (Taylor) : Moore 1876. Near the Hunting Tower,
XiUamey : Scully 1890. Elillamey, rare (Moore) : Carrington 1863.
Near the summit of Bamanaghea Mtn. near Anascaul, plentiful,
Sept. 1898 : Lett & McA. Bare in the Dingle Peninsula : McA. 1901.

IV. Lough Bray (Taylor) : Hooker 1816. Abundant at Upper
Lough Bray, more sparingly at Luggielaw and Seven Churches
(Ifoore) : McA. 1890.

V. Kelly's Glen, Co. Dublin : Moore.

VIII. Kylemore, 1874: Moore. Nephin, May 1901: Lett & McA.

XI. On rocks in river at Errigal and at Lough Belshade, June
1903: McA.

XII. Slieve Donard, June 1903 : Hunter. On stones in the Blue
Lake, Slieve Lamagan, Spinkwee Biver glen, stream west of Slieve
tfeel More, and Diamond Mountain, Co. Down ; Lett 1890. Eagle
Mtn. and in the Windy Gap, rare (Waddell) : Lett 1890.

var. rigtda Lindberg, Moore.

Lough Bray, 1873 : Lindberg 1875. On boggy land near Seven
[lurches : Moore 1876.

NoTB^ — '' Near N, sphaeelata, stems shorter, narrower, and more
igid, more branched, more densely foliaceous, and here and there
lexuose. Leaves more spreading and rigid. Cells twice the size and
hickened, generally highly coloured. Dr. Lindberg states that this
brm ia intermediate between the typical form of the species and its

476 Proceedings of the Etoyal Irish Academy.

yar. Carringtonii {Adelanthua CarringUmii Balfour MS., Nmrik Cm-
ringtonii Lindberg). The first plants of this which I collected were
sent to Dr. Lindberg, who named them N. CarringUmii without my
reservation." — ^Moobe.

4. Nardia scalariB Schrad., Or. & B.

Jungermania sealarii Schrad., Syst. Samml. Krypt. Gew., iL, p. 4.
Nardia scalaris Gr. & B., Nat. Air. Brit. PL, 694. Alieularia tedgrif
Corda in Opiz., Nat., p. 653. Nardia sealarts Sch., Moore, liisb
Hepat., p. 656. Pearson, Hepat. Brit. Isles, p. 370, plate 161.

Districts I. II. III. IV. V. VI. VII. vin. IX. X. XI. in.

JETah. — On moist clay banks and rocks, heaths, &c., common.

var. compressa Carrington, Brit. Hepat., p. 24.

Jungermania Walrothiana Hiiben., Hepat. Oerm, p. 85, no. 30.
Among wet stones, Connor Hill, Sept. 1897 : Lett & McA.

yer. distans Cairington, Brit. Hepat., p. 24.
In wet places on Brandon, June 1900 : Lett & McA.

var. rtvularit Lindberg.

In wet places, often in running water, and among 8phapi»
cuspidatum and on submerged rocks in mountain streams. Lon^
Bray, Co. Wicklow, 1873 : Lindberg 1875 ; Moore 1876 ; McA. 1890.

var. rohusta Lindberg.
On wet rocks at Lough Bray, 1872 : Mooie ; McA. 1890.

Genus 30. Marsnpella Dumort.
1. Manmpella emarginata Ehrhart, Dum.

Jungermania emarginata Ehrh., Beitr., iii., p. 80. Mtrtvpitt*
emarginata Dumort., Comm. Bot., p. 114. Sareoscgpkm Ekrktt
Corda, in Opiz., Nat., p. 632. Nardia emarginata Ehrh., Moore, Iris^
Hepat., p. 655. Ifarsupella emarginata Pearson, Hepat. Brit. \^^
p. 375, plate 163.

Districts I. II. in. IV. V. VI. Vn. VII. IX. X. XI. XII.

Sah. — On wet rocks and stones from the plains to high eleTitiofis
by mountain rivulets.

I. Common in the Dingle Peninsula: McA. 1901. FxeqiNnt
about XiQaniey : Carrington 1863.

McArdle — A List of Irish HepaticcB.

477

Co. Carlow: McA.
Queen^s Co. : McA.

II. Oaltee and Enockmeildown Mtns., Co. Tipperary, June
1902: McA.

III. Rocks by the River Barrow, Gndgae,
1896 a. On stones in a stream on Slieve Bloom,
1892 a.

rV. Rocks by the Urrin River at Knockroe, Co. Wexford, May
1899 : McA. 1903. Lough Bray, 1873 : Lindberg 1875 ; McA. 1890.

V. Anglesey Mtn., Co. Louth: Lett 1890. Co. Dublin, common :
Moore 1878.

VI. Cam Seefin, Co. Clare: McA. 1895 a,
YU. King's Co.: McA. 1892 a.
Yin. MaamTurk, Connemara (Moore) : Carrington 1863

and shores of Lough Conn at Pontoon, 1891 : Lett & McA.
Hm near Oalway : McA. 1895 a.

IX. Benbulben, 1880: McA.

X. Wet rocks, Ballyhaise Wood, and in Eamham demesne, Co.
Cavan: McA. 1898. Camlough Mtn. and Slieve Gullion : Lett 1890.

XI. Rathmullen, Febr. 1903; Buncrana, March 1902: Hunter.
Errigal, Lough Belshade, Goat Idand near Lough Eask, June 1903 :
McA.

Xn. Frequent in the Moume Mts., Co. Down: Lett 1890. Co.
Antrim, frequent, 1836-7 : Moore.

Nephin
Gentian

var minor Carrington, Brit. Hepat., p. 14.

Hah. — On wet rocks and stones, rare.

Cromaglown, Killamey: Moore 1876. On exposed rocks. Lough
Adoon, Co. Kerry, rare : McA. 1901. On stones near the summit of
Slieve Olah, Co. Cavan : McA. 1898. On Slievenabrock above New-
castle, Co. Down, among Andraa alpina^ 1884 : Lett. Rare in Co.
Down : Lett 1890; Stewart 1888. On stones in the Ring Wood by
the Slaney River, Co. Wexford, May 1899 : McA. 1903. Nephin and
Devil's Mother, 1901 : Lett. Cratleagh Woods, Co. Donegal, Sept.
1902: McA.

y^r.pieea Carrington, Brit. Hepat., p. 14.

On wet locks, Connor Hill, Co. Kerry, Sept. 1898 (Lett & McA.) :
McA. 1901.

var. fM^or Carrington, Brit. Hepat., p. 14.

On exposed rocks, Lough Adoon, Co. Kerry, rare, Sept. 1897:
Lett & McA.

478 Proceedings of the Eopal Irish Academy.

2. Mannpella sphacelata Gieseke, Dumort.

Jungermania sphacelata Gieseke, in Lindenberg, Syn. Hepat.,p.76,
tab. 1, fig. 9. Nardia sphacelata Carr., Brit. Hepat., p. U, pl«}
fig. 5. Moore, Irish Kepat., p. 655. MarsupeUa iphaeelaU Dunt,
Eecueil, p. 24. Pearson, Hepat. Brit. Isles, p. 377, plate 165.

Districts I. IV. Vin. XII.

JEM. — On wet rocks by the side of mountain zivolets, wet roc^
boulders in glens at high elevations.

I. Horse's Glen, Mangerton : Scully 1890. Loughanscaul near
Dingle, May 1894: McA. Derrymore Glen near Tralee, 1899:
Lett & McA.

IV. Lough Bray, 1869 (Moore) : Lindberg 1875 ; Canington 18H
Station verified by F. W. M., July 1887 : McA. 1890, and (fertile),
1889 : ScuUy & McA.

VIII. Nephin, May 1901 : Lett & McA. Bangore, Co. Mayo, Sipt
1901 : Lett.

XII. Slieve Donaid, Co. Down (Andrew & Lett) : Stewart 1895,

3. MarsupeUa Funokii Web. et Mohr., Dumort
Jungermania Funckii Web. et Mohr., Deutsch. Erypt., p. -i*--

Marsupella Funckii Dumort., Eecueil, p. 24. Moore, Irish Hepat,

p. 655. Pearson, Hepat. Brit. Isles, p. 379, plate 166.

Districts Vni. — X. — XII.

JSiab, — On rocks and sheep paths at high elevations.

VIII. Mountains above Eylemore Lake, 1874: Moore. Mass

Turc Mtns. (Moore) : Carrington 1863. On Slievemoie, Achill, Srpt

1901 : Lett. Nephin, May 1901 : Lett & McA.

X. Camlough Mtn., Co. Armagh, June 1902 : Lett
XII. Black Mountain near Belfast, 1837 : Moore.

Genus 31. Ceaia Gr. & Bonn.
1. Cesia oorallioides Nees.

Qgmnomitrium coraUioides Nees, Nat. Eur. Leb., i., p. U^*
McArdle, New or Bare Irish Hepat., Sci. Proc. K. D. S., vol. ii^ •
plate 5, fig. 1, 1880. AcoUa coraUioides Dum., Becueil, p. 25. O*^
eorallioides Carruth., in Joum. Bot, iii., p. 300. Pearson. Hepat Bnt.
Isles, p. 401, plate 177.

Districts 1. 11. — IV. .

Sab. — On mountain rocks.

I. On rocks near the Tunnel, Cromaglown, Killamey :. CaningtoD-

McArdle — A List of Irish Hepatic<B.

479

Dankerron and Enockayohil Mtn. : Taylor. Carantnal : Moore. Bran-
dos, 1840 (Moore): McA. 1880.

II. Galtee Mtns., Co. Tipperary (I. Carroll) : Carrington 1863.

lY. Lugnaqnilla (I. Carroll) : Carrington 1863.

2. Cesia obtusa Lindberg.

Cesia obtusa Lindberg, Muse. Scand., p. 9. Gymrumitrium eon-
cinatum var. ermtdatum limpr., in Cohn, Krypt. Scbles., i., p. 246.
Cc9%a ohiusay McArdle, New or Bare Irish Hepat., in Sci. Froc. B.D.S.^
Tol. iii., plate 5, fig. 9, 1880. Pearson, Hepat. Brit. Isles, p. 405^
plate 179.

Difltricts VIH. XH.

Bah, — On exposed mountain rocks.

VIII. Mweelrea, 1874 (Moore): McA. 1880.

XII. On Slieve Donard : Lett 1890 ; and 1903 : Hunter. Slieve
Commedagh at 1800 ft., and Slievenamady : Lett 1890. Thomas Mtn.^
Hen Mtn., and Hare's Gap (Waddell) : Stewart 1888.

3. Cesia orennlata Gottsche.

Oymnomitrium erenulatum Carrington, Irish Crypt, p. 18, pi. i.,
fig. v., 1863. Gottsche et Babenhorst, Hepat. Europ. Exsicc, no. 478.
Jungermania etmeinnata Tayl., in Fl. Hib., part ii., p. 59. Cesia
erenulaia Moore, Irish Hepat., p. 659. Pearson, Hepat. Brit. Isles,,
p. 407, plate 180.

Districts I. II. — IV. V. XII.

Mob, — On exposed rocks and stones, in alpine and sub-alpine situa-
tions, rarely descending to sea-level.

I. Cromaglown and Carantual, 1851 : Moore. Near the Tunnel,
Cromaglown: Carrington. Connor Hill, 1873: Lindberg 1875; and
1881 : P. W. M. & McA. Dunkerron and Knockavohil Mountains,
Co. Kerry (Taylor) : Carrington 1874. Bocks on the west side of
Brandon, April 1897 : F. W. M. & McA. ; and 1900 (Lett & McA.) :
UcA. 1901. Glengariff, May 1851 (Moore): Carrington 1874.

II. Galtee Mountains, Co. Tipperary (I. Carroll) : Carrington 1874.
rV. Lugnaqnilla (I. CarroU) : Carrington 1874.

V. Bocks on Howth (Moore) : McA. 1893 a. Golden Birer, Co.
Louth (Lett): Pearson 1901.

XU. Mountain above Camlough, Co. Antrim, Sept. 1836 : Moore.
Yianite rocks on Slieve Donard (Stewart), Slieve Commedagh and
Slievenamady (Lett): Stewart 1888. West side of Clontygeragh,
836 : Moore. Metamorphic rocks, Mullaghmore, Co. Derry : Stewart

.Hill

fifl

480 Proceedings of the Boyal Irish Academy,

1888. Hen and Pigeon Kock MoimtainB, lare : Lett 1890. Eocki,
north-east side of Slemish, Co. Antrim (Lett & Waddell) : Stewut
1895.

Genus 32. Aorobolbns Nees.

Aorobolbus Wilsoni Taylor, Nees.

Oymnanthe WiUoni TayL, G. L. N., Synop. Hepat, p. 192.
Aorobolbus Wilsoni Nees, G. L. N., Synop. Hepat., p. 5. Moore, Imb
Hepat., p. 659. Pearson, Hepat. Brit. Isles, p. 410, plate 181.

Districts I. VIIL .

Mob, — In scattered tufts in moist shaded places, mostly epiphytic^
on FruUaniaf JRadtUa, &c.

I. Near Bantry with young fruit, Nov. 1812 (Miss Hutclun<:
Carrington 1874 (with excellent fig.). Glengariff, on IhUM
^ermana, 1869 (Carrington): Moore 1876. Banks of a ratine nt?«r
the Hunting Tower, Cromaglown, in fruit, Nov. 1829 (W. Wilson :
Carrington 1874. Tore Mountain, Killamey, Sept. 1841 (Taylor,:
Carrington 1874 ; Moore 1876.

VIII. On the slopes of Devil's Mother, Co. Mayo, Sept 1901 :
Lett.

Genus 33. Sacoogyna Dumort.

Sacoogyna yitioulosa Linn., Dumort.

Jungsrmania viticulosaJJjm.j Sp. PI., p. 1597. Hook., Biit Jusg.,
tab. 60. Sacoogyna vitioulosa Dum., Comm. Bot., p. 113. Moore.
Irish Hepat., p. 633. Pearson, Hepat. Brit. Isles, p. 413, plate l^-

Districts I. II. III. IV. V. VI. VII. VIII. IX. X XL XH

JETab. — On damp ground, on rocks, &c.

I. Near Bantry, in fruit, 1813 (Miss Hutchins) : Hooker 1S16.
Xillamey woods: Moore 1876. O'Sullivan's Cascade, and Cromi*
glown, in fruit, 1873: Lindberg 1875. Tore Waterfall, KilUnyty.
Sept. 1898 : McA. & Lett 1899. Connor Hill, 1873 : Lindheig IST^.
Common in the Dingle Peninsula : McA. 1901.

II. Glengarra Wood, Galtee Mountains, Co. Tippemy, June W-'
McA.

III. Wood by the roadside at Graigue, Co. Garlow : McA. 1S96 \
Slieve Bloom, Queen's Co., 1891 : McA. 1892 a.

IV. Lough Bray: Moore 1876; McA. 1890. Eillou^ixim For^t
near Enniscorthy, Co. Wexford, May 1899 : McA. 1903.

V. EiUakee Glen, Co. Dublin : Moore 1878. Bank in Howib
demesne, 1895 : McA. 1897. Clermont Cam, Go. Louth (Waddeli. :
Lett 1890.

McArdlb — A List of Irish Sepatica. 481

VI. Cam Seefin, Co. Clare : McA. 1895 a.

VII. Aid Bog, King's Co. (Russell) : McA. 1892 a.

VIII. Co. Mayo: Moore 1876. Pontoon on Lough Conn, and
Nephin, May 1901 : Lett & McA. Bangore, and Slievemore on Achill,
Sept. 1901 : Lett. Croaghpatrick, Sept. 1901 : McA. Letterfrack,
^0. Galway, 1891 : McWeeney.

IX. Slish Wood, Co. Sligo : Waddell 1892.

X. Ballyhaise Woods. Co. Cavan, 1893 : McA. 1898.

XI. Co. Donegal: Moore 1876. Rathmullen Wood, and Saltpans
^ood, Rathmelton, July 1902: Hunter. Columbkil Lake and Crat-
eagh Wood near Milford, Sept. 1902 ; Lough Eask, June 1903 : McA.

XII. Rostreyor Mountain, Donard demesne, White River Glen,
pinkwee River, and Tollymore Park, Co. Down: Lett 1890.
lenariff, 1836 (Moore); Woodbum and rocks at the summit of
'allygally Head (Stewart) : Sallagh Braes, Co. Antrim (Lett) :
tewart 1888.

Subtribe 7. FOSSOMBEONIiE.

Genus 34. Soalia Gray & Bennett.

1. Scalia Hookeri LyeU, Gray & Bennett.

Junpermania Hooheri Lyell, Eng. Bot., tab. 2555. Hook., Brit.
ing., tab. 54. Scalia Hookeri Gr. & Benn., Nat. Arr. Brit. PL,

704. Moore, Irish Hepat., p. 660. Pearson, Hepat. Brit. Isles,
427, plate 189.

Districts I. V. .

Hah, — Heaths and damp sandy places.

I. Connor Hill, one female plant, among Riccardia pinguis var.,
73: Lindberg 1875; Moore 1876.

V. On damp sandy flats between sand-hills, Malahide, 1 878 (Moore)
single plant, shown to F. W. M. and D. McA.) : McA. 1880.

Genus 35. Fossombronia Raddi.

1. Fossombronia pnsilla Dill., Linn., Dumort.

Jungermania pusiUa Linn., Sp. PL, p. 1603. Foisombronia pusilla
mort., Bev. Jung., p. 11. Lindberg, Manipulus Muse. Secund.,
(84, tab. 1, fig. 5. Moore, Irish Hepat., p. 661. Pearson, Hepat.
tish Isles, p. 418, plate 183.

Districts I. XI. XII.

Hah. — On moist clay or sandy ditch banks and fallow fields.

L On clayey soil near Dingle, 1875 : Moore. Bank in Bumham

482 Praeeedinga of the Royal Irish Academy.

Wood, Ventry, near the sea, Sept. 1898 ; wet bank between Emalough
and Inch, May 1899 : Lett & McA. In the fissnrea of rocks between
Dingle and Yentry with Anthoeeros Uwis, 1873: Lindberg 1875.
Bank of Finglas Kiver, Waterville : Scully 1890. Near the Hnntang
Tower, Slillamey: Scully 1890. Among rocks near the sea, Bere
Island, and banks of the Pulleen River near the sea, Berehareii:
McA. 1894.

XI. Field near Eathmullen with Anthoeeros punetaUu and i{*«M
fflauca, 1903 : Hunter.

XII. Island of Bathlin, Co. Antrim: Moore. Saintdeld, Co.
Down: Waddell.

var, ochrospora lindb., Not. pro F. et. Fl. Fenn., p. 887.
Co. Kerry (Moore) : Pearson 1901.

2. Fossombronia cristata Lindb.

Fo8»ombr<m%a eristata Lindb., apud Soc. pro F. et Fl. Fenn. (1878',
Not. pro F. et Fl. Fenn., 382 (1874). Pearson, Hepat Brit. Hea,
p. 420, plate 184.

Districts XH.

JETah. — On bare damp soil, sides of ditches, margin of ponds, &c.

XII. On whitish clay, shore of Loughbrickland, Co. Down, Oct.
1890 (Lett): Pearson 1901. Holywood, 1902: Hunter.

3. Fossombronia angulosa Mich., Dicks., Raddi.

Jungermania anguhsa Dicks., Fasc. PI. Crypt. Brit., i., 7, ITSo.
J, pusilla Sm., I^ng. Bot., xxv., tab. 1775. Fonambrfmia anpdou
Baddi, in Att. Soc. Sc. Modena, xyiii., p. 40. Moore, Irish Hepfit,
p. 661. Pearson, Hepat. Brit. Isles, p. 421, plate 185.

Districts I. •

Sah, — On ditch banks and fissures of rocks, near the sea.

I. Very fine at Dingle Bay and on clifh near the sea betrcei:
Dingle and Yentry : Moore. Abundant on a ditch bank near the ^-^
by the entrance of Dingle Harbour, April 1897 (F. W. M.): McA.
1901. Dingle Bay opposite Ventry, male et c. fr., 1873: Lindbtit
1875. Bumham Wood, Ventry, Sept. 1898: Lett & HcA. Bc^«
Bay, Xillamey : Carrington 1863 ; Moore 1876. In fisames of rocb
near the sea, Dursey Island : McA. 1894. Rocks in the Pulleen Ri^r
and on Bere Island, Co. Cork : McA, 1894.

McArdlb— ^ lAat of Irish Hepatiece. 483

4. Fossombronia Dnmortieri Hiib. et Oenth., Lindberg.

Codonia Dumortieri Hiib. et Genth., Deutchl. Leberm., no. 80.
FoumWmia foveolata Lindb., Not. pro F. et Fl. Fenn., p. 382.
Fmomhronia Dumortieri Hiib. et G., Lindb., Not. pro F. et Fl. Fexm.,
xiii., p. 380.

Districts I. .

Hah, — On damp moorlands and banks of ditches.

1. Damp field near Farranfore, Co. Kerry, August 1901 : Scully.

5. Fossombronia osBspitiformis De Not.

FoBsomhronia angulaa yar. easpitiformis Raddi, in Att. Soc. Sc.
tfodena, xviii., p. 41. F, easpitiformiit De N., in G. R., Hepat. Eur.,
lee. xiii. et xiv., no. 123. Pearson, Hepat. Brit. Isles, p. 424, plate
187.

Districts — XI. — .

Hah. — On bare damp soil.

XI. Boggy ground, Ned's Point, Buncrana, 1903 : Hunter.

Genus 36. Petalophyllnm Gottscbe.
Petalophyllum Balfsii Wils., Oott.

Jungtrmania hihetmea var., Wils., £ng. Bot., tab. 2750. J, Ralfiii
Vils., Eng. Bot., Suppl., tab. 2874. Petalophyllum Ralfiii Wils.,
'ott. in Lehm., Pug. PL Nov. et Min. Cogn., viii., p. 29. Moore,
rish Hepat., p. 663. Pearson, Hepat. Brit. Isles, p. 430, plate 190.

Districts I. V. .

Hob, — On damp sandy ground near the sea.

I. Damp hollows in sand-hills on both sides of Inny ferry, Water-
lUe : Scully 1890. Abundant a mile west of the ferry on the north
de: Scully 1890.

y. On damp sandy flats among sand-hills, Malahide : Moore 1876 ;
Eld sparingly, January 1902 : Lett & McA. On the North Bull near
liblin, plentiful, 1880 : F. W. M. & McA.

Genus 37. Pallavioinia Gr. & Bonn.

1. Pallavioinia Lyellii Gr. & Benn.

Jungermania Lyellii Hook., Brit. Jung., tab, 77. Blyttia Lyellii
ndl., Gen. PL, 1840. Pallavioinia Lyellii Gr. & Benn., Nat. Arr.
rit PL, p. 775. Moore, Irish Hepat., p^ 662. Pearson, Hepat.
rit Isles, p. 482, plate 191.

K. t. A« PHOC, TOL. XXnr., 8SC. B.] 2 Q

484 Proceedings of the Royal Irish Academy,

Districts I. IV. V. .

Miib. — Boggy places among Sphagnum and on wet dripping rocks.

I. Near Bantry (Miss Hutchins) : Hooker 1816. Woods at
Killamey: Taylor 1886. Maghanabo Glen near Fermoyle and bj the
lakes between the glen and Connor Hill, 1865 : Moore.

IV. Lough Bray: Taylor 1836; Moore 1878. Same station,
among Sphagnum^ 1887 : F. W. M. & McA.

V. Piailway bank near the Eyewater River at Leizlip railwif
station, Co. Eildare, very scarce, 1890 : ScnUy & McA.

2. Pallayioinia hibemioa Hook., Gr. & Benn.

Jungermania hibemica Hook., Brit. Jung., tab. 78, suppL, Ub. 4.
Moerekia hiherniea Gottsche, in G. & R., Hepat. Eur. Exsicc., now 295.
Pallavioinia hibemica Gr. & Benn., Nat. Arr. Brit. PL, p. 6S4.
Moore, Irish Hepat., p. 662. Pearson, Hepat. Brit. Isles, p. 454.
plates 192-193.

Districts IV. ,

Mab, — In marshy places among Sphagnum, &c., at high dera-
tions.

rV. Among Sphagnum cuspidatum and Jungermania emargindU os
the shores of Lough Bray : Taylor; Moore ; and July 1878 : McA

var. Uptodesma Tansley, Pearson.

Pallavioinia Uptodesma Tansley and Childs, in Annals of Bot, tqL
XT., March 1901.

JSdb, — In marshy places on the coast.

Among the sand-hills, Malahide : Moore; McA.; and January 1902:
Lett & McA. Between Malahide and Portrane : Moore. On th^
North Bull near Dublin : Moore ; rare there : McA. Cushendun, Co.
Antrim: Lett 1890. Sand-hills north of Newcastle, Co. Down:
Stewart 1888.

Genus 38. Blasia Micheli.
1. Blasia pnsilla Linn.

Blasia pustUa Linn., Sp. PL, p. 1605, 1753. Jungmrmmim Bk^
Hook., Brit. Jung., tab. 82-84. TayL, in Fl. Hib., pt. ii., p. 56,
Blasia pusdla Linn., Moore, Irish Hepat, p. 668. Pearson, Heptt
Brit. Isles, p. 440, plate 196. .

Districts I. IL — IV. V. VIII. — X. XI. XII.

Eab. — On wet clayey banks in woods, by the side of streams, &c.

I. At the foot of Brandon : Moore 1876 ; and 1881 : F. W, M. 4

MgAkdlb — A List of Iruh Hepaticce.

485

tfcA. Connor Hill, Sept. 1898, and Lough Duff in the Brandon
Valley, 1899 : Lett & McA. Near Waterville and Ballybunion, Co.
Xeiry: Scully 1890.

II. By the River Suir near Carrick, Co. Tipperary : Hart 1886.

IV. Woodenbridge : Moore 1876. On a bank by a stream in
Altadore Glen, Co. V^icklow : McA. 1889. Eilloughrim Forest near
Knniscorthy, May 1899: McA. 1903.

V. Castlekelly Mtn., Co. Dublin, fertile, March 1836: Taylor.
Anglesey Mtn., Co. Louth (Waddell) : Lett 1890. Boggy place at
Ballykill, Howth, fertile, March 1894 : McA. 1897. Railway bank
nearLeixlip, Co. Kildare: McA. 1893 &.

VIII. Slieyemore on Achill, Sept. 1901 : Lett.

X. Southern end of Lough Allen : Stewart 1885. Side of a stream
at Eillakeen, Co. Cavan, 1893: McA. 1898. Camlough Mtn., Co.
Armagh, Oct. 1898: Lett. On Slieve Gullion, Co. Armagh, 1894:
lIcA.

XL Saltpans V^ood, Rathmullen, July 1902 : Hunter.

XII. ToUymore Park and Victoria Park. Co. Down : Stewart
1888. Moygannon Glen and Narrow Water (Waddell) : Stewart 1888.
SlieTenamaddy, Omeath Glen, Brown Bog near Loughbrickland :
Lett 1890. Bank of a stream on Knockagh, Co. Antrim: Stewart
1888. Glendun, Co. Antrim, 1836: Moore.

Genus 39. Pellia Raddi.
] . Pellia epiphylla Linn., Lindberg.

Jwigermania epiphylla Linn., Sp. PL, i., ed. 2, p. 1135. Hook.,
Brit. Jung., tab. 47, figs. 1, 4, 8, 17. Pellia Fabroniana Raddi, in
Att. Soc. Sci. Modena, xviii., p. 49. Fellia epiphylla Linn., Moore,
Irish Hepat., p. 664. Pearson, Hepat. Brit. Isles, p. 443, plate 197.

Districts I. II. ni. IV. V. VI. VII. VIII. IX. X. XI. XII.
Hah. — Growing in large patches on clay banks, on rocks and bogs,
from sea-level to high elevations.

2. Pellia oalyoina Tayl.

Jungermania ealyeina Tayl. in Fl. Hib., part ii., p. 55. Jungermania
mdiva/olia Dicks., PL Crypt. Brit., fasc. iv., p. 19. Jungermania
fpiphyUa ^91, fureigera Hook., Brit Jung., tab. 47, fig. 18. Pellia
ToJycina Tayl., Moore, Irish Hepat., p. 664. Pearson, Hepat. Brit,
[ales, p. 447, plate 198.

Districts I. — III. IV. V. VI. VII. — IX. — XL XIL

2 Q2

486 Proceedings of the Royal Irish Acadenty,

Hob. — In wet boggy places, dripping rooks, often submerged.

I. Dunkerron: Taylor 1836. Tore Cascade and Croma^owHf
1 863 (Carrington) : Moore 1876. On wet rocks, Tore Cascade, 1897 :
Lett & McA. Stream on Brandon, Sept. 1897 : F. W. M. & Mel
Longhanscaul near Bingle, Sept. 1898 (Lett & ICcA.) : McA. 1901.

III. Stream on Slieve Bloom, Qneen's Co. : McA. 18920.

rv. Altadore Glen and Lough Bray: Moore (nnder FtliiA
endivafoiia) ; Lindberg 1875; McA. 1890. Dargle River: McA.
1889. Wood by the Slaney Kiyer near Enniscorthy, May 1B99:
McA. 1903.

y . In a marshy place at Eilrock qnarries, and on a shsUoir bog
at Ballykill, Howth: McA. 1893 a. Beauparc, Co. Meath, 1893:
McA. Leizlip, Co. Kildare : McA. 1893 h.

VI. Cam Seefin, Co. Clare: McA. 1895*.

VII. Geashill, King's Co., 1890: Russell.
IX. Glencar, Co. Sligo : Moore 1876.

XI. Gartan Wood, Bunlin Waterfall, wet rocks at Columbkil Like
near Milford, Sept. 1902 ; Goat Island, Lough Belshade, Errigal, Jnne
1903: McA.

XII. Glenballyemon, Co. Antrim : Moore. Dundonald Glen :
Stewart 1888. Moygannon Glen and Tollymore Park: Lett 1890.
Colin Glen and Carr's Glen, Co. Antrim : Stewart 1888. Agbaderg
Glebe (Lett): Stewart 1895. Drumnasole, Co. Antrim: Brenan
Parkmore, Co. Antrim : Lett.

3. PeUia Neesiana Gottsche, Limpr.

Felliti epiphylla p, forma Neeiiana Gottsche in Hedwigia, p. 69,
1867. P$llia Neesiana Limpr., in Cohn, Erypt. Fl. Scbles., p. 329.
Macvicar, Joum. of Bot., toI. xxxviii., p. 275, 1900. Pearson, H^t.
Brit. Isles, p. 445.

District I. .

Hah, — In moist rocky places, and on wet grassy ground.
I. By the banks of the Pulleen River in a marsh among rocks ne:J
Pulleen Coye, Castletown Berehaven, Co. Cork {fide Slater) : McA.
1894.

Sub.tribe 8. HETZGEBIE2.

Genus 40. Aneura Dumort.

1. Aneura palmata Hedw., Dum.

Jungermania palmata Hedw., Theor. Gen., ed. i., p. 87. Antur4
palmata Dum., Comm. Bot., p. 115. Riccardia palmata Carrath., in

McAbdlk — A List of Irish Hepaticat.

487

Journ. of Bot., iii., p. 302, 1865. Moore, Irish Hepat., p. 667. Aneura
palmata Pearson, Hepat. Brit. Isles, p. 450, plate 199.

Districts I. IV. VIII. — X. .

Jliah. — On decayed wood.

I. Cromaglown, KiUamey: Moore. Tore Mtn. and Eagle's Nest:
Carrington 1863. Tore Waterfall, Sept. 1897 : McA. & Lett 1899.
Ifangerton andMuckross, KiUamey : Scully 1890. Brandon, 1881 :
F. W. M. & McA. Connor Hill, 1875 : McA. Lough Adoon and
Loughanscaul, 1897; Bamanagbea Lough and Mt. Eagle, 1898 (Lett
4tMcA.): McA. 1901. Glengariff : Carrington 1863.

IV. Altadore Glen, Co. "Wicklow : Moore.

VIII. Abundant on a small island off Ballykill Harbour, Co. Gal-
way, 1874: Moore. MurriBk and Croaghpatrick, Sept. 1901 : McA.
Kephin, May 1901 : Lett & McA.

X. Shores of Lough Cultra, Co. Cavan, rare : McA. 1898.

2. Aneura mnltiflda Linn., Dumort.

Jung$rmania muUifida Linn., Sp. PI., p. 1602, 1753. Bieeardia mul-
tifida. Or. & Benn., Nat. Arr. Brit. PI., i., p. 683. Aneura multifida
Dumort., Comm. Bot., p. 115. Riccardia multtfida Moore, Irish
Hepat., p. 667. Aneura multtfida Pearson, Hepat. Brit. Isles, p. 451,
plate 200.

Difltricts I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII.

Mud, — In marshy places, damp woods, bogs, sides of ditches, &c.

3. Aneura ambrosioides Nees.

Am0ura multifida var. anibroaioides Noes, Nat. Eur. I^eb-, iii., p. 450.
\» L. N.y Syn. Hepat., p. 497. Riccardia multijida var. ambrosioideB
rees, Lindberg, in Acta Soc. Sci. Fenn., x., p. 511. Moore, Irish
lepat., p. 668. Carrington, Irish Crypt., 1863. Aneura ambraeioidee
'earsoiiy Hepat. Brit. Isles, p. 453, plate 201.

Districta I. .

JETab, — Among wet Mosses and shaded wet rocks,

I. Cromaglown and Glena : Carrington 1863. Near Ventiy,
onnor Hill, and O'SuUivan's Cascade, 1873 : Lindberg 1875. Glen-
skin, Co. Cork (T. Chandlee) : Carrington 1863.

4. Aneura latifrons Lindberg.
Jumg^rmania multijida Schmidt, Icon. PI., iii., pp. 213-216 (excl.

^non. at pp-} 1797. Hook., Brit. Jung , p. 19, fig. 75. Riccardia
ultiJUUi Or. & Benn.f Nat. Arr. Brit. PL, p. 684, no. 1. Aneurapalmata

' i

488 Proeeed%ng% of the Royal Irish Aeademff.

Tar. major Nees, Kat. Eur. Leb., iii., p. 459. G. L. 9., Syn. Hepat., p.
498. Aneura lait/rtmt Lindb., Soc. Fl. Fenn., 1873. Bicetrdiaktifmi
lindb., Hepat. Hib., p. 513, 1875. Moore, Irish Hepat, p. 66S.
Aneura latifrotu JAndh., Pearson, Hepat. Brit. Isles, p. 454, plate 101

Districts L — ni. IV. V. VI. VII. VHI. .

Siah, — On moist toify banks, and decayed wood.

I, O'Sulliyan's Cascade, Killamey, 1873 : Lindberg 1875. Tpper
Glencar and Tore Waterfall : Scally 1890 ; and Sept. 1897 : Mci. 4
Lett. In the Dingle Peninsula it occurs as follows : — Anascanl, 1894 :
McA. ; and 1898 : Lett & McA. Lough Adoon, 1894 : McA. Conma*
nare Lakes, Bamanaghea Lough, Lough Nalachan on Brandcm, 1S9S:
LeU & McA. Caha Mtn., Co. Cork : McA. 1894.

III. 81ieye Bloom, Queen's Co. : McA. 1892 a.

IV. On decayed wood, summit of the waterfall at Powosconrt,
May 1897 : McA.

V. Bog at Ballykill, Howth : McA. 1897.

VI. Doon Bog, Clonbrock, Co. Galway : McA. 1896 h.
VIL Bog near Geashill, King's Co., 1890 : Russell.
VIII. Nephin, 1901 : Lett & McA.

4. Aneura sinnata Dicks., Dumort

Jungennania iinuata Dicks., PL Crypt. Brit., fasc. ii., p. 16. J-
multifida yar. ainuata Hook., Brit. Jung., tab. 45, fig. 2. Moore, Irish
Hepat., p. 667. Aneura einuata Dum., Gomm. Bot., p. 115. Jm^»
pinnatifida Garrington, Irish Crypt., 1863. Riccardia muUifida vtr.
mq/ar Lindberg, Hepat. Hib., 1875. Aneura ttnuata Dicks., Peanon,
Hepat. Brit. Isles, p. 456, plate 203.

Districts I. II. — IV. X. — Xn.

Hob. — On wet rocks, often within the spray of waterfalls.

I. Eagle's Nest, Cromaglown, and Glena, Killamey : Carringt^
1863 (under A, multifida var. major) ; Moore 1876. On Tore Cascade
YfitYi FruUania Eutehinsia: Carrington 1863 (under A. jn»nali/d4\
Fermoy, Co. Cork (T. Chandlee) : Carrington 1863.

II. Galtee Mountains, Co. Tipperary : Moore 1876.

IV. Luggielaw and Lough Bray : Moore 1876 (under JRiecerda
multifida var.) ; and McA. 1890. GlencuUen, 1887 : McA.

X. Killakeen, Co. Cavan, sparingly, 1893: McA. 1898 (under
Eieeardia).

XII. ToUymore Park, Co. Down, yery rare (Lett & Vaddeli; :
Lett 1890. Saintfield, Co. Down, Jan. 1903 : Waddell.

McArdlb — A List of Irish HepaticcB.

489

5. Anenra pingnis Linn., Dmnort

Jvngermania pinguis Liim., Sp. PL, p. 1136. Hook., Brit. Jung.,
tab. 46. Aneura pinguis Dumort., Oomm. Bot., p. 115. Rieeardia
pinguis Qray & Benn., Nat. Arr. Brit. PL, i., p. 684. Moore, Irish
Hepat., p. 668. Aneura pinguia Pearson, Hepat. Brit. Isles, p. 457,
plate 204.

Districts I. II. III. IV. V. VI. VII. Vin. — X. XI. XII.

Rab, — ^Damp shaded banks, among rocks in wet places, often
nibmerged in mountain streams, and among wet heather.

I. Cromaglown, Killamey, 1873 : Lindberg 1875. Waterville
md Glencar: Scully 1890. Brandon, June 1900, Connor HiU, and
[}oumanare Lakes, 1898 : Lett & McA. Maghanabo Olen, April 1897 :
P. W. M. & McA. Anascaul, 1894 : McA. Bog between Emalough
md Inch, 1899: Lett & McA. Prequent in the Dingle Peninsula :
tfcA. 1901.

II. Galtee Mountains, Co. Tipperary, June 1902: McA.

III. Stream bank, Slieve Bloom, Queen's Co.: McA. 1892 a. Gores-
bridge, Co. Carlow : McA. 1896 a.

IV. Lough Bray. 1876 (Moore) : MoA. 1890.

V. On wet sandy flats, Malahide : Moore 1876. Quarries near
button, and on a small bog at Ballykill, Howth, plentiful: McA.
893 a. Leixlip, Co. Kildare : McA. 1893 h. Side of the River Liffey
ear Lucan, Co. Dublin, June 1902 : McA. Anglesey Mountain and
^eath Qlen, Co. Louth : Lett 1890.

VI. Bog on Cam Seefin, Co. Clare : McA. 1895 a.

VII. Bogs at Welsh Island and Geashill railway station, King's
o. : McA. 1892 a.

VIII. Wood at Pontoon on Lough Conn, and on Nephin, May
901 : Lett & McA. Slopes of Devil's Mother, Sept. 1901 : Lett.

X. Derrytrama bog, Mintiaghs, Co. Armagh, 1888 : Lett.

XI. Errigal Waterfall, woods about Lough Eask, and Goat Island,
one 1903: McA.

XLK. On moist banks, Co. Antrim, 1838 : Moore. Palls Park, Co.
ntrim: Stewart 1888. Slieve Donard and Bencrom: Lett 1890.
ostreTor Mtn. (Waddell): Stewart 1888. Portstewart, Co. Derry:
tewart 1895.

var. dmtieulata Mich., Noes.

Lindberg in Acta Soc. Sci. Feun., z., p. 514.
In wet rocky places among Cylicocarpum Jfmgsoiii, Connor Hill,
0. Kerry, 1873 : Lindberg 1875.

490 Proceedings of the Royal Irish Academy.

G6nu»41. Metzgeria Raddi.
1. Ketzgeria pubesoens Schrank.

Junyermania pubescehs Schrank, Prim. Fl. Germ., i., p. 231, no.
860. Hook., Brit. Jung., p. 20, tab. 79. Metzgeria pvheseem Baddi,
in Att. Soc. Sci. Modena, xviii., p. 46. Moore, Irish Hepat., p. 666.
Pearson, Hepat. Brit. Isles, p. 460, plate 205.

Districts XII.

Sal. — On moist limestone rocks and banks in woods.

XII. Mountains near Belfast (Templeton): Taylor 1836. On
limestone rocks between Lame and Glenarm, and at Sallagh Braes, Co.
Antrim, 1837 (Moore) : Stewart 1888. Carr's Glen (Stewart), and
Sallagh Braes (Stewart, Lett, and Waddell) : Stewart 1888.

2. Ketzgeria fdrcata Linn., Kaddi.

Jungermaniaf areata Linn., Sp. PL, 1602, 1753. Hook., Brit. Jung.,
tab. 55, 56. Metzgeria glabra Eaddi, Jung. Etr., in Mem. Modena,
xviii., p. 43, tab. 7, fig. 1. Metzgeria fwreataJ)\xmoti.,'BjQ\, Jung.,
p. 26. Moore, Irish Hepat., p. 665. Pearson. Hepat. Brit. Isles,
p. 461, plate 206.

Districts I. II. ni. IV. V. VI. VII. VIII. IX. X. XI. XH.

Hah. — On the trunks of trees, on moist banks and rocks.

Tar. aruginosa Hook.

M./ureata y&t, fruticulosa Dicks., lindb. Monogr.

Mah, — On the trunks of trees.

Muckross, Killamey: Carrington 1863. Tore Waterfall, Sept.
1897 : McA. & Lett. Burnham Wood, and woods between Dingle and
Ventry, 1894: McA. Wood at Goresbridge, Co. Carlow: McA.
1896a. Cappard and Killeagh Abbey, Queen's Co., 1891 (Russell):
McA. 1892 a. On trunks of trees, Co. Wicklow, and frequent in Co.
Dublin: Moore 1878. Lough Allen and Slieveanierin Mtns., Co.
Leitrim : Stewart 1885. Famham Demesne and oak wood at Bally-
liaise, Co. Cavan: McA. 1898. Batt's Wood, Rathmelton, Co. Doneg^
July 1902 : Hunter. Eirkcassock and Gillhall, Co. Down : WaddelL

yar. proU/era McArdle.

McArdle, Proc. E. I. Acad., 3rd series, vol. iv., no. 1, p. 116, 1897.
IM. — On trees by a stream near the ground, and on a small bog
at BaUykill, both at Howth, March 1894: McA. 1897.

McArdle — A List of Irish Hepaticce.

491

3. Ketzgeria ooiyagata Lindberg.

Jungermania fureata (non Linn.), Weiss., PI. Crypt. Fl. Gott.,
p. 108, 1770, and others. Metzgeria eonjugata Lindberg, in Act. Soc.
F. & PI. Fenn., x., p. 495, no. 27, 1 875. Schiffner in Engler & Prantl,
Pflanzenfamilien, 91 and 92 Lief., p. 53, with figures, 1893. McArdle,
in Proc. R. I. Acad., 3rd ser., toI. iv., no. 5, with plate, 1898.
Pearson, Hepat. Brit. Isles, p. 463, plate 207.

Districts I. II. III. IV. V. VI. VII. VIII. — X. XI. XII.

Hob, — On shaded rocks and on the trunks of trees, and among the
larger Hepaticad and Mosses.

I. Glena, Tore Cascade and O'Sullivan's Cascade, among JTookeria
kttevirens, 1873 (Lindberg): Moore 1876. Glencar : Scully 1890.
Tore Waterfall, Sept. 1897 : McA. & Lett. Ross I., 1893 and 1899 :
McA. Frequent in the Dingle Peninsula: McA. 1901. DunboyWood,
Castletown Berehaven : McA. 1894.

II. Fertile in woods at Scariff, Galtee Mtns., Co. Tipperary, June
1902: McA.

III. On granite rocks, side of the River Barrow at Graigue, Co.
Carlow, and on trees near Goresbridge : McA. 1896 a.

IV. Altadore Glen, Co. Wicklow: McA. 1889. Near Ferns,
Co. Wexford, Dec. 1895: Greene. Killoughrim Oak Forest near
Enniscorthy, Co. Wexford, May 1899 : McA. 1903.

V. On FruUania Tamarisei among rocks near the Baily Lighthouse,
Howth : McA. 1893 a. Beauparc, Co. Meath, Sept. 1893 : McA.

VI. Clare Glen, Glenstal, Co. Limerick : Hart 1886. Cam Seefin,
Co. Clare: McA. 1895 a. Tycooly Wood, wood at Sheep-pool Bog,
uid very fine in old orchard, Clonbrock, Co. Gal way : McA. 1896 h,

Vn. Geashill, King's Co., 1891 (Russell): McA. 1892a.
Vm. Bangore, Sept. 1901 : Lett. Pontoon near Foxford, and on
S'ephin, May 1901 : Lett & McA.

X. Famham demesne, Ballyhaise Wood, shores of Lough Cultra
on Fndkmia), Co. Cavan, 1893 : McA. 1898.

XI. Rathmnllen Wood, July 1902 : Hunter. Wet rocks at Bun-
in Waterfall near Milford, Sept. 1902; Lough Eask, June 1903:
4cA.

Xn. Colin Glen near Belfast (Stewart), Sallagh Braes, Co. Antrim
Waddell) : Stewart 1888. ToUymore Park, and in Rostrevor Wood,
)o. Down (Waddell) : Lett 1890.

VAT. proUfora McArdle, I. Nat., vol. v., p. 238, 1896.
Tycooly Wood, Clonbrock, Co. Galway: McA. 1896 J.

492 Proceedings of the Royal Irish Aeademy.

4. KetEgeria hamata Lindberg.

Mettgerta hanuUa Lindberg, Soc. F. Fl. Fezm., 1874. Jwyemnia
fureata var. ehngata Hook., Brit. Jung. tab. 55. Metxgm^ ktmtU
Lindberg, Monogr. Metzg., p. 25, fig. 5. Mettgerta Uneurii non Aost.,
Lindberg, Hepat. Hib., p. 494, 1875. Mehyeria hamata Petwm,
Hepat. Brit. Isles, p. 464, plate 207.

Districts I. VIII. .

Hah. — On the ground among mountain heaths, on damp baiikSf
and by the sides of streams, often submerged.

I. Stream on Brandon, 1865: Moore; and 1897: Lett & McA.
Cromaglown, Killamey, 1873 : Lindberg 1875 ; and 1878 : McA. Tort
Waterfall, and MagiUicuddy's Reeks at 2500 ft. : Scully 1890. T«c
Waterfall, Sept. 1897 : McA. & Lett. On damp peat on the ascent to
Brandon near Gloghane, and on Mt. Eagle, July 1881 : F. W. M. t
McA. Bank of the Coumanare Lakes near Connor TTill, and it
Loughanscaul near Dingle, Sept. 1898 : Lett & McA.

VIII. Bangore, and on the slopes of Devil's Mother, Sept. 1901 :
Lett.

Order 2. MARCHANTIACELE.

Oenus 42. Karohantia March, fil.

Karohantia polymorpha Linn., Sp. PL, p. 1603, 1753.

Marchant fil., in Acta Gfal., 1718. Dill,, Muse., tab, 76-77, fig. 7.
Eng. Bot., tab. 100. Tayl., in Fl. Hib., pt. ii., p. 49. Moore, Irisb
Hepat., p. 601. Pearson, Hepat. Brit. Isles, p. 466, plate 208.

Districts I. IV. V. VI. — viiL — X. XI. xn,

JSab. — By the sides of streams, drain banks in bogs, old damp
walls. Flourishing on the surface mould of pot-plants in gardens.

I. Dunkerron, Co. Kerry: Taylor 1836. Frequent on moist rocks,
seldom fruiting : Carrington 1863. Common in the Dingle Peninsulu :
McA. 1901.

IV. Lough Bray: McA. 1890. Altadore Glen: McA. 1889.

V. Old damp wall near Drumcondra, Co. Dublin, 1878: KcA.
Anglesey Mtn., Co. Louth : Lett 1890.

VI. Co. Clare and Co. Limerick: Stewart 1890. Ballyrangbaxi :
McA, 1895 a. Drain on Sheep-pool Bog, Clonbrock, Co. Oidv^y.
abundant, male and female: McA. 1896 5. Abunduit on the bof
near Geashill railway station, King's Co., fertile : McA. 1892 •.

VIII. River bank, Westport, Sept. 1901 : McA.
X. Camlough Mtn., Co. Armagh, 1897 : Lett.

McArdle — A List of Irish Hepaticce. 493

XI. Goat Island near Lough Eask, fertile, June 1903 : McA.

XII. Frequent on moist shady banks near waterfalls, Co. Antrim,
1836: Moore. Common in the N.E. : Stewart 1888. Tollymore Park
and Black Stairs on SlieTe Donard, Co. Down : Lett 1890.

Oenu8 43. Conooephalus !N^eok.
Conocephalus oonicus Neck., Dumort.

Hepatiea vulgaris Mich., Not. F1. Gen., p. 3, 1729. Marehantia
eanieay Eng. Bot., tab. 504. Conoeephalus eonieus Dumort., Comm.,
p. 115. Steph., Sp. Hepat., p. 141, 1900. Fegatella eonica Corda,
in Opiz., Beitr., i., p. 649. Conoeephalus eonieus Moore, Irish Hepat.^
p. 601. Pearson. Hepat. Brit. Ides, p. 469, plate 209.

Districts I. II. III. IV. V. VI. VII. Vin. IX. X. XI. XII.

ffaib, — In damp shaded places sides of streams, abutments of
bridges, in cayems in mountainous places, on old walls, &c.

Genus 44. Beboulia Raddi.
Keboulia hemisphssrioa Eaddi.

Marehantia hemisphariea Linn., Sp. PL, p. 1604. Asterella hsmi-
sphceriea p. p., Beaur., in Encycl. Meth., suppl. i., p. 502. Moore,
Irish Hepat., p. 603. Rehoulia hemisphariea Eaddi, in Att. Soc. Sci.
Modena, ii., p. 357. Pearson, Hepat. Brit. Isles, p. 470, plate 210.

Districts I. V. VI. — VIII. XI. XII.

Mab, — On damp rocky places, walls of bridges and on damp sandy
ground.

I. Dunkerron, Co. Kerry : Taylor 1 836. Dingle Bay (Carrington) :
Moore 1876. In the crevices of moist rocks, Killamey : Wade Bar.
1804. Near Cork and Fermoy (I. Carroll) : Moore 1876.

V. On sandy ground at the North Bull, Dublin (McA.) : Moore
1876.

VI. Abundant on the walls at the bridge, Cong, Co. Galway : Moore
1876. In the crevices of moist rocks near Kilronan, Aran Islands:
McA. 1895 a. Doon Bog, Clonbrock, Co. Galway : McA. 1896 h.

VIII. Connemara: Wade Bar. 1804.

XI. Poisoned Glen : Hart 1886.

XII. Sallagh Braes, Co. Antrim : Moore. Shady wall, PameU'fr
Bridge, Tollymore Park, Co. Down, very rare: Lett 1890. Rathlin
Island (Stewart), basaltic rocks, Benevenagh, Co. Derry, 1836
(Moore) : Stewart 1888 (under Asterella).

494 Proceedings qfthe Boyal Lish Academy.

Genus 45. Preissia Corda.
PreiBsia commntata Lindenberg, Neea.

Marchantia hemuphartea Linn., Fl. Suec, no. 1052. MmtkMta
androgyna TajL, in Trans. Linn. Soc, xvii., p. 380, 1835. Frma
commutata Nees, Nat. Eur. Leb., iv., p. 117. Chomicearptm quairthu
Scop., Lindb., Muse. Scand., 1879. Freistia commutaia Mooie, Ins^
Hepat., p. 602. Pearson, Hepat. Brit. Isles, p. 473, plate 211.

Districts L V. Vin. XI. Xn.

Hah, — In fissures of damp rocks, and on damp ground in moun-
tainous places, more frequent in limestone districts, occasionaUy at
sea-level on damp sandy flats, among sand-hills.

I. Killamey: Moore. Eayine below tbe Ease's Nest, lS6o:
Oarrington. Dunkerron : Taylor. Kenmare ("W. Wilson) : Carrington
1863.

V. North Bull near Dublin, and in Co. Kildare : Moore. Bailey
bridge over Eyewater Biver, Leixlip, Co. Eildare : McA. 1893i.

YIII. Bocks above Ejlemore Castle and by the side of the lake at
Letterfrack, 1874 : Moore. Abundant near Cong, Co. Gbdway : K^>^
1876.

XI. In the crevices of rocks, Goat Island, Lough Eask, June 1^^' •
McA. MiUfield, Buncrana, and behind Batt's Wood, 1903 : Himtei.

XII. Sallagh Braes, Co. Antrim, 1837: Moore. Crega^ Gl»
and Carr's Olen : Stewart 1 888. Scarva demesne and Parkmore (Lett i,
Glendun, Co. Antrim (Brenan) : Stewart 1895.

Genus 46. Lunularia.
Lunularia oruoiata linn., Dumort.

Lunularia vulgaris Mich., Nov. Gen., p. 4, tab. 4, 1741. Mtr^sn^i*
cruoiata Linn., Sp. PL, p. 1137. Lunularia erueiaU Dumort, ComBu.
p. 116. Moore, Irish Hepat., p. 604. Pearson, Hepat Brit bk^,
p. 476, plate 212.

Districts I. IV. Y. — YII. X. XI. xn.

Haib, — On damp walls and banks, pathways, &c.| exoeedin^f
common in gardens. Very rare in fruit

I. Dunkerron, Co. Kerry, fertile iu August : Taylor 1836. Tore
Waterfall, Eillazney, 1873: Lindberg 1875. Plentiful on the^r&ll
and on the ground at the entrance to glen at Tore Waterfall, Sept
1897 : McA. & Lett. Near Kenmare : Carrington. Boea I., on ▼all-
tops, banks, and pathways, 1893: McA. 1900. Old walls abc>at

McArdle — A List of Irish Hepaticce. 495

Dingle : McA. Bumham "Wood, Ventry, 1898 ; bank betweett
Emalougli and Inch, 1899 ; old walls about dogbane and Fermoyle,
June 1900 : Lett & McA. Roadside, Connor Hill : McA.

IV. Altadore Glen, Co. Wicklow (Moore): McA. 1889. Wooden-
bridge, Co. "Wicklow, 1894 : McA. Damp bank, BaUybeg near Ferns,
Oct. 1896 (Greene): McA. 1903.

Y. In Glasneyin Botanic Gardens, once seen fertile : Moore
1876. Roadside near Finglas bridge, Co. Dublin, Aug. 1894 : Nellie
McArdle. Howth demesne : McA. 1897. Beauparc demesne, Co.
Meatb, 1893 : McA.

"711. Shady places, Geashill Rectory, King's Co., 1890 (Russell) :
McA. 1892 a.

X. Ditch bank, Famham demesne, Co. Cavan, 1893 : McA. 1898.

XI. Millfield, and behind Batt's Wood, Buncrana, 1902 : Hunter.

XII. Belfast Botanic Gardens, 1837 : Moore. Drumcro, Co.
Down ("Waddell), Aghaderg, Co. Down (Lett), moss-hole at Seymour
Hill, Co. Antrim, 1806 (Templeton) : Stewart 1895.

Genus 47. Dumortiera Nees.
Dumortiera irrig^a Wilson, Nees.

Marchantia irrigua "Wilson, in Hooker's Brit. Fl., vi., 106, no. 5.
ffygrophylla irrigua Taylor, in Fl. Hib., pt. ii., p. 54 ; Trans. Linn.
Soc. , xvii. , p. 303, tab. 15. Dumortiera irrigua Wilson, Nees, Nat. Eur.
Leb., iv., p. 159. Dumortiera hirmta Swartz, Steph., Sp. Hepat.,
p. 224. Dumortiera hirmta Sw., Tar. irrigua Tayl., Spruce, Hepat.
Amaz. et And., p. 566, 1885. Schiffn., in Eng. & Prantl, Pflanzenf.,
91 und 92 Lief., p. 36, 1893. Dumortiera irrigua Moore, Irish Hepat. ^
p. 602. Pearson, Hepat. Brit. Isles, p. 477, plate 213.

Districts I. IV. .

ffah, — In sheltered rocky places, about waterfalls, and in carems
in the mountain streams.

I. Blackwater Bridge near Dunkerron, 1820 (Taylor) ; Moore 1876;
Scully 1890. Tore Waterfall, KiUamey, 1829 (W. Wilson) : Moore
1876; Lindberg 1875. Plentiful within the spray of this waterfall,
Sept. 1897 : McA. & Lett. Maghanabo Glen near Fermoyle, Co.
Kerry, 1829 ^W. Wikon) : Moore 1876 ; also 1875 : McA. ; and May
1897 : F. W. M. & MoA. On wet rocks in the stream which flows
into Loughanscaul, Sept. 1897 : Lett & McA. BallinahassigGlen near
Cork: Power's F. & Fl. Cork, 1844. Dunscombe's Wood near Cork
(I. Carroll) : Moore 1876.

496 Proceedings of the Royal Irish Academy.

rV. Altadore Glen, Co. Wicklow (Lord Gongli) : Moore W7i
Collected in the same glen, 1872-74: Moore ; also 1887-9: McA.
Waterfall at Luggielaw sparingly ; Moore 1876.

GenuB 48. Targionia Micheli.
Targionia hypophylla Linn., Sp. PL, p. 1604.

Web. et Mohr, Crypt. Germ., p. 391 , tab. 1 2. Kng. Bot, tab. 2?:
Tayl., in Fl. Hib., pt. ii., p. 55. Targionia Miehelii Corda, in Op^
Beitr., i., p. 649. T. hypophyUa Moore, Irish Hepat, p. 605. Pesrsoa.
Hepat. Brit. Isles, p. 480, plate 214.

Distticts I. XII.

Rah. — On dry rocks, and on the ground among rocks.

I. Among rocks on bank near the chapel, Cahirdveen, April 1BT7:
McA. Near Muckross Abbey and Blackwater Bridge, Eenmare, Co.
Kerry: Scully 1890. On dry limestone rocks, Carrigaline De«rCerk
(I. Carroll) : Moore 1876.

XII. On Care Hill near Belfast (Templeton) : Moore 1876. Kt-
discovered there by Lett in June 1902. Veiy rare, only obserred •«
an exposed bank in the Little Deer Park, Glenarm, Co. Antrim, 18^^:
Moore in Ordn. Surv. Coll. ,

i
Genus 49. SphsBTOcarpuB Micheli. I

BphserocarpuB terrestrlB Mich. i

Spharoearpus terrestris Mich., Nov. PI. Gen., p. 4, tab. 3. Dil
Hist. Muse, p. 586, tab. 78, fig. 17. Smith, Eng. Bot., Ub. i^^
Targionia spharoearpus Dicks., PL Crypt. Brit., fasc. i., p- '
Spharoearpus lagsnarius Dumort., Comm. Bot., p. 78. Spharoesrft
terrsstris Mich., Moore, Irish fiepat., p. 669. Pearson, KepatBrl
Isles, p. 482, plate 215.

Districts XII.

Saib, — In clover fields and stubble land, and on day banks.

XII. On a wet clay bank at Colin Glen, near Belfast : David Ci
Dr. Moore writes — '' I have never seen any Irish specimens ^
plant, nor have I beard of its having been observed by any otk
person than Mr. Orr in Ireland."

Note. — Stubble fields and similar places have been neglected
-collectors, and tbere is no reason why the plant should not be itA
i^overed.

McArdlk — A List of Link Hepaticce, 497

Order 3. RICCIACEiE.

Genus 50. Biccia Micheli.

Biocia glauca Linn.

1. Eiccia glauca Linn., Sp. PL, p. 1606, 1763.

Eng. Bot., tab. 2546. Lindenberg, Monogr. Rice, p. 417, tab. 19.
Tayl., in Fl. Hib., pt. ii., p. 70. Moore, Irish Hepat., p. 606. Pear-
son, Hepat. Brit. Isles, p. 484, plate 216.

Districts I. IV. V. XI. XII,

S'ah. — On damp ground, and on wet clay banks.

I. Mud walls, Ross Bay, Killamey : Carrington 1863.

IV. Co. Wicklow: Moore 1878.

V. Co. Dublin : Moore 1878,

XI. Wet places beside the Mill River, Buncrana, July 1903, and
field near RathmuUen, Sept. 1903 : Hunter.

XII. Clay bank, Holy wood near Belfast, Jan. 1903: Hunter.
Colin Glen, and rocks at Bangor : Waddell in Guide to Belfast, 1902.

2. Biccia orystallina Linn.

Lichen palustris Dill., Hist. Muse., p. 535, tab. 78, fig. 12, 1741.
Riecia cry»tdllina Linn., Sp. PL, 1605, 1753. Pearson, Hepat. Brit.
Isles, p. 485, plate 217.

Districts XII.

Bah, — On damp clay in fields, and on wet banks.

XII. On moist banks near Glenarm and Cushendall, Co. Antrim,
1836: Moore.

3. Biocia sorooarpa Bischoff.

Rieeia soroearpa Bischoff, Hepat. Nov. Act. N. Car., xvii.,
p. 1053, tab. 71, fig. 11, 1835. Rieeia mtmma Linn., Schiffn., in Engl,
und Prantl Pfianzenf., 91 und 92 Lief., p. 15, 1893. Riccia toroearpa
Bisch., Moore, Irish Hepat., p. 606. Pearson, Hepat. Brit. Isles,
p. 487, plate 218.

Districts I. .

Mah. — Fissures and tops of old walls.

On an old wall near Dingle, 1873 (Lindberg & Moore) : Lindberg
1875. Wall top by the roadside leading from Dingle to Ventry, on
the Dingle side of the river near the TTnion Workhouse, Sept. 1898 :
Lett & McA.

498 Proceedings of the Royal Irish Academy.

4. Biooia glauoesoenB Carrington.

R. ylaueesesfu Carrington, in Can*. & Fears., Hepat. Brit Exsicc*^
fasc. 1, no. 66. Rieeia Michelii Baddi, Lindb., Muse. 8caDd./p. %
1879. Rieeia hi/urea Hoffm., Steph., Sp. Hepat., p. 30, 1900.
Rieeia ylaueeseens Carr., Pearson, Hepat. Brit. Isles, p. 493, pUte
222.

Districts XI. XII.

Mob. — On damp mud-covered rocks and stubble fields.

XI. In a stubble field near Lough Salt, Sept. 1902 : McA.

XII. On rocks in Glendun Biver, Co. Antrim, 1895: Brenan 4
Lett, Journal of Botany, vol. xxziii., p. 283, 1895.

Genus 51. Biooiella Braim.
Biooiella fluitanB Linn., Braun.

Rieeia fluitans Linn., Sp. PL, 1606. Lindenberg, Monogr. Bice.,
p. 443, tab. 24-25. Hook, et Taylor, Muscol. Brit, ed. iL, p. 21^
Dumort., Hepat. Europ., p. 171. RiecieUa fluitans Braun, Bot Zat,
p. 754, 1821. Moore, Irish Hepat., p. 606. Pearson, Hepat ^^
Isles, p. 495, plate 224.

Districts V. VI. IX. X. — Xn.

Hah. — Floating in stagnant pools and ditches.

V. Side of the Biver Boyne above Drogheda : Moore 1876. Dit^ife
at North Wall, Dublin : Wade Bar. 1804.

YI. Ditches near the Shannon, Co. Limerick : Moore.

IX. Among Lemna^ ditch by the Shannon below Cairick, Co.
Leitrim, June 1899 : Praeger.

X. Closet Biver and bog drains near Baughlan, Co. Armagh : I^
XII. Lough Neagh, where the canal joins the lough at Luifis^

Co. Down : Moore. Abundant in a drain at Meenan Bog, Lou^bri<.i-
land, Co. Down : Lett 1890; Stewart 1895. Several places in Kewry
Canal: Stewart 1888; Lett 1890. Lagan Canal at Eilmore, Co.
Down : Waddell in Guide to Belfast, 1902.

Genus 52. Biooiocarpus Corda.

Bicciooarpus natans Linn., Coida.

Rieeia natans Linn., Syst. Nat., ed. v., vol. xii., p. 2, 1755- Enc-
Bot., tab. 252. Rieeioearpus natans Corda, in Opiz., Nat^ p. ^I
Moore, Irish Hepat., p. 607. Pearson, Hepat Brit. Isles, p. 497, pUte
225.

McArdlr — A List of Irish Hepaticw. 499

Districts — V. YI. .

ffdb, — Floating on stagnant pools, drains, and ditches.

Y-. Drain near Newbridge, Co. Kildare, 1897 : Praeger. Ditch
near the Grand Canal at Ballyfennont near Inchioore, Co. Dublin,
August 1 893 : Colgan. Ditch by the Biy^ Bairow, three miles below
Athy, Co. Kildare, 1889: ScaUy. Abundant in a large pool about
half way between Drogheda and NaTan, near the railway, right-hand
side going from Navan to Drogheda, 1870: ICoore. Braganatown
Bog, Co. Louth, 1897 : Praeger.

YI. Ditch by the side of the Shannon, near Fortumna, Co. Gbl-
way : Moore. Ditches near Passy, Co. Limerick (Harvey) : Moore
1876,

Order 4. ANTH0CEB0TACE2E.

Genus 53. Anthooeros Micheli.

1. Anthooeros IsBvis Linn.

Anthoceros liBvis Linn., 8p. PL, p. 1606, 1753. Lindenb... Hepat.
Eur., p. 112. Nees, Europ. Leb., iv., p. 329. Moore, Lish Hepat.,
p. 670. Pearson, Hepat. Brit. Isles, p. 499, plate 226.

Districts I. .

Hiah, — On clay banks, fields, and pathways.

I. Wet clay bulk by the roadside leading from Dingle toYentry, July
1873 : Lindberg & Moore. On a pathway in Bumham Wood, Yentry,
Sept. 1898 (Lett & McA.) : McA. 1901.

2. Anthooeros punctatos Linn.

Anthoceros punetatus Linn., Sp. PL, p. 1606. Eng. Bot., tab. 1537.
Moore, Irish Hepat., p. 670. Pearson, Hepat. Brit. Isles, p. 500,
plate 227.

Districts I. II. — lY. Y. YIII. XL XII.

ffab, — ^In damp fallow fields, ditch banks, and by the sides of
streams.

I. Frequent in Co. Kerry : Moore. Moist banks near Eenmare :
Carrington. Brandon, 1829: W. Wilson. Brandon Head, Sept. 1898:
Lett & McA. Maghanabo Glen, Co. Kerry, 1875: McA.; 1881,
and April 1897 : F. W. M. & McA.

II. Co. Cork, frequent : Moore. Among paying-stones on the
ground at Queen's College, Cork, June 1899 : Prof. Hartog.

lY. Sugarloaf Mountain (E. P. Wright) : Moore 1876.

R. I. A. P&OO., YOL. XXTY., 8B0. B.] 2 B

600 Proceedings of the Royal Irish Academy.

Y. Kelly's Gtlen, Co. Dublin : Moore. Sandy flats and banks at
Malahide, Go. Dublin, 1880 : McA.

YIII. Slieyemore, Achill, Sept. 1901 : Lett.

XI. Field near Bathmullen, 1903 : Hunter.

XTT. Olendun, Co. Antrim : Moore. Belfast Botanic Gardens and
river bank in Olendun (Lett): Stewart 1895. Ditch bank shaded with
grass on the Stranmillis road, Belfast, 1803 (Templeton) ; and ligna-
pieste, Co. Derry, rare, 1836 (Moore) : Stewart 1888. Fallow Mi*,
Loughbrickland, Co. Down, 1897 : Lett. On a clay bank at Koly*
wood, Jan. 1903 : Hunter.

McArdlb — A List of Irish Hepaticm.

501

INDEX

PAOB

Acrobolbufl Wilaoni, 480

AdeUnthua decipiens, 440

ineura ftmbroaioidesy 487

latifionA, 487

multifida, 487

palmata, 486

pinguis, 489

oinuata, 488

Anthelia julftoea, 422

Anthoceros laBYifl, ..499

punctatus, 499

BaszaniA Peaztoni, 429

txiangularis, 428

tricrenata, 429

trilobata 427

Blasia puailla 484

Blepharostoma trichophyllum, . . 424

Blepharosia ciliazis, 424

Cephalosia bicuspidata, . . 432

catenulaU, 430

connirent, 434

currifolia, 435

denudata, 437

diyaricata, 437

elachista, 439

fluitana, 436

Franciaci, 436

hibemicay 434

Lammeraiaiia, 433

lettcantba» 439

limaJjBfolia, 432

pallida, 431

•pha^ni, 436

BteUulifera, 438

:««a ooialloidei, 478

ci«iiitlAtA, 479

obtuaa, 479

Tfailoacyphua poljantbos, , • 463

naamalocolea cnmiilolia, . . 462

PAOB

Conooephalui ooniousy .. 493

DiplopbyUum albicans, . . 449

Dicksoniy 460

obtiuilolium, 450

Dumortierairrigua, 496

Foflsombronia angulosa, . . 482

csBpitiformiBy 483

criatota, 482

Dumortieri, 483

pusilla, 481

FruUania dilatata, . . . . 406
fragilifolia, ..406

germana, 406

microphylla, 404

Tamariflci, 404

Harpanthus scutatus, 464

Herberta adonca, . . 428

Hygrobiella laxifolia, 440

Jubula HutcbinauB, . . 407

Jungermania alpeatria, . . 468

bantrienaia, 465

barbato, 471

bicrenata, 467

capitate, 466

cordifoUa, 459

crenulate, 461

exaecte, 469

ezaectsBlormia, 470

giacilia, 471

inoiaa, 469

iniOate, 463

lycopodioidea, .. 472

Lyoni, 470

minute, 472

oicadenaiiy 478

pumila, 469

riparia, 460

apbarocarpa, .. 460

turbinate, 464

502

Proceedings qf the Bot/al Irish Academy.

PAGE

Jungermania yentricoea,

. 467

Kantia arguta,

. 429

Trichomanis,

. 429

Lejeunea calcarea,

. 413

calyptrifolia,

. 416

diversiloba,

. 415

flava,

. 410

Holtii,

. 411

hamatifolia, . .

412

Mackaii,

. 408

microscopical

. 414

minutissimai

. 414

ovata,

. 412

patena,

. 409

Bossettiana,

. 413

aerpyllifolia,

. 408

ulicina,

. 416

I<epidozia cupressina, . .

. 425

Peanoni,

. 426

reptans,

. 426

setaoea,

. 426

. 427

Lophocolea bidentata,

. 460

cuspidata,

. 461

heterophylla.

. 461

spicata,

. 452

Lunularia craciata,

. 494

Marchantia polymorpha,

. 492

Manupella emarginata, . .

. 476

Funckii

. 478

sphacelata,

. 478

Mastigophora Woodaii, . .

. 423

Metzgeria conjugata,

. 491

f urcata,

. 490

hamata,

. 492

pubescenfl,

. 490

Mylia anomala,

. 455

Taylori,

. 454

Kardia compressa,

. 475

hyalina,

. 473

obovata,

. 474

scalaris,

. 476

Pallavicinia hibemica, . .

. 484

LyeUii

. 483

Pedinophyllum interruptum,

. 455

PeUia calycina,

. 485

epiphylla,

. 485

PAOI

Pellia Neeaiana, 486

Petalophyllum RalfBii, .. ..483

Plagiochila ambagioaa, . . . . 456

asplenioides, .. 456

ezigua, 4S^

punctata, 417

spinulosa, .. .. 4^

. tridenticulata, .. ..4^

Pleurozia cochlearifomiis, . . 4f I

Porella IfBTigata, 419

pinnata, 421

platypbylla, 420

livulanB, . . . . . . 411

Thuja, 421

PreiBsia commutata, .. ..494
Prionolobus Tumeri, . . . . 439
Radula aquilegia, . . . . 4IT

Carringtonii, 41S

complanata, .. .. ..419

Holtii, 4i:

voluta, 417

Beboulia hemisphsriea, . . . . 4$?

Biccia crystallina, . . . . 4S7

glauca, . . . . . . 49*

glauceacena, 49$

sorocarpa, 4dT

Bicciella fluitana, . . . . . . 4K

Ricciocarpus natans, ..4!^

Saccogyna yiticuloaa, . . . . 4$\'

Scalia Hookeri, 4S1

Scapania equiloba, .. 44i

aspera, .. .. .. 44i

compacta, .. .. ..441

curta, 44?

intermedia, .. .. ..447

irrigua, .. .. ..447

nemorOBa, .. .. ..444

nimbosa, . . . . . . 44a

omithopodioidea, ., ..445
reaupinata, .. .. ..41?

aubalpina, . . .. ..441

uliginoaa, 449

umbroaa, . . . . 449

undulata, 44€

Sphaerocazpua terreatzia, . . 4$t

Targionia hypophylla, . . 49c

Trichocolea tomantdU, . . 4«4

PROCEEDINGS

OF THB

ROYAL IRISH ACADEMY

VOLUME XXIV

SECTION C-ARCHJIOLOGY, LINGUISTIC, AND

LITERATURE

DUBLIN

PCBUSHED AT THE ACADEMY HOUSE, 19, DAWSON STBEET

80LD ALBO BT

HODOES, FIGGIS, k CO., Lmiibd, 104, OBAFTON STREET
sY 'VriLLIAMS k NOBGATE, LONDON, EDINBUBGH, AND OXFOBD

1902-1904

The Academy desire it to be understood that they are nd
answerable for any opinion^ represetitalion of facts^ or tram of
reasoning that may appear in any of the foUomng Papers, The
Authors of the several Essays are alone responsible far thetr
contents.

CONTENTS

SECTION C-ARCH^OLOGY, LINGUISTIC, AND
LITERATURE

Bjobt (Henby F.), M.A. : — pagb

' Notes on an Unpublished MS. Inquisition (a.d. 1268),
relating to the Dublin City Watercourse. From
the Muniments of the Earl of Meath, ... 89

BuBT (J. B.), M.A., LL.D., Litt.D. :—

* The Itinerary of Patrick in <]onnaught, according to

TlrechAn, 158

CovvBT (Oeobgb) : —

Some Monuments of the La Tine Period recently

discovered in Ireland. (Plates XVm.-XXn.), . 267

Faxjonkb (Gasab Litton), M.A. : —

The Irish Guards, 1661-1798 7

. Some Illustrations of the Commercial History of Dublin

in the Eighteenth Century. (Plates IX.-XII.), . 188
" The Counties of Ireland : An Historical Sketch of their

Origin, Constitution, and gradual Delimitation, . 169

I.AKB-P001<B (StANLBT), M.A., LiTT.D. : —

' An Arabic Inscription from Rhodesia. (Plate I.), . 47
The First Mohammadan Treaties with Christians, . 227

kluKBAT (Miss M. A.), F.S.A. Soot. :—

' Scarabs in the Dublin Museum. .... 81

Contents

O'Beillt (Joseph P.), C.E. — pagb

"Some further Notes on Ancient Horizontal Water-Mills,

Native and Foreign. (Plates II.-IV.), . . 65

'Notes on the Orientations and certain Architectural
Details of the Old Churches of Dalkey Town and
Dalkey Island. (Plates Xm.-XVII.), . . 195

Roberts (J. Maris), B.A.I. : —

On the Discovery of an Ancient Grave near Ardrahan,

Co. Galwa^^, 1

Westropp (Thomas Johnson), M.A. : —

" The Cists, Dolmens, and Pillars in the Eastern Half
of the County of Clare (Baronies of Bunratty).

(Plates V. and VI.), • 85

The Cists, Dolmens, and Pillars in the Eastern Half
of the County of Clare (Baronies of Tulla). (Plates
VII. andVni), 107

'* The Ancient Forts of Ireland." Being some further
Notes on a Paper of that name, especially as to
the Age of Motes in Ireland, .... 267

DATES OF PUBLICATION

Part 1. Pages 1 to 46. August, 1902.

„ 2. „ 47 „ 182. December, 1902.

„ 8. „ 188 „ 168. April, 1908.

„ 4. „ 169 „ 226. September, 1908.

„ 5. „ 227 „ 276. January, 1904.

ERRATA

SECTION C

Page 10, line 9, /or '^ witnesses" read '' witnessed".

„ 17, „ 21, for **and" read "or".

„ 19, „ 1, }br "1786" r*a<? "1686".

„ 20, „ 20, note 2, for "of" read "to"; /or "by" read "of".

„ 24, „ 18, far "were" read "was".
• „ 25, footnote, line 4, /or "Orlandais" read "irlandais".

„ „ „ „ 9,/or "s^ our" rdorf "s^jour".

28 ) ^y inadyertence these pages went to press uncorrected. A
' \ number of obvious errors occur in the Duke of Portland's

»» 29, J letter.

„ 30, line 3 from end, /or "general" read "several".

„ 85, line 8 from end, /or " 1870" read " 1879".

„ 86, paragraph 3, inaert "It" at commencement last line of
text ; for "ract" read "tract".

„ 89, line 13, /or "last" r^rf "east".

„ 91, Istlineof note, /or "Fich"rMrf"Fieh".

„ 95, line 1, /or **plants" rwi "planks".
„ ,, line 12, for "edges" read "ridges".

„ 110, line 29,/or'*broad"ri?a<^ "thick".

„ 118, note 3, last sentence, for "been" read "has been", and
add at end of sentence "for the lithograph".

„ 141, line 21, /or "Broadstreet"rifarf"Brad8treet".

„ 146y last line of footnote, /or "Commercial Buildings" read
"Grafton Street".

„ 136, first line of footnote, /or "Historical" read "Historic".

„ 176, line 30, /or "Greenville" rtf<irf"Geneville".
Pages 183-191, /or "Sydney" read "Sidney".
Page 188, lines 15 and 18, /or "0' Conors " rtfoi? "O'Connors".

„ 269, Une23,/or "1009" rdorf "1111".

PROCEEDINGS

THE ROYAL IRISH ACADEMY.

PAPERS READ BEFORE THE ACADEMY.

ON THE DISCOVERY OF AN ANCIENT GRAVE, NEAR
ARDRAHAN, COUNTY GALWAY.

J. HARRIS ROBERTS, B.A.I. (Dubldi).

[COHUUNICATES BT GEOBOE COFFEY, B.B.]

[Read Janvart U, 1901].

I HATS the honour to submit to the Members of the Royal Irish

Academy the following account of the discoyery of an ancient grave

in the neighbourhood of Ardrahan, Co. Galway. During the early

part of January, 1900, while Mr. St^hen Tarpey was excavating

gravel from a newly-opened pit on his farm, which Ues about 2^ statute

miles to the N.N.W. of Ardrahan, the workmen came upon a flag,

standing vertically in the ''face" of the pit. This feU out, disclosing

the grave (fig. 1). Inside the grave, which was formed of four flags

set on edge, with another on top acting as a cover, were a quantity

of bones and two urns. One of the latter, unfortunately, got broken

at the time, and, before the grave could be carefully examined, the

bones and urns were disturbed from their original positions. A

drawing of the urns, as shown (fig. 2), and a detailed report on the

boned by Dr. Brown of the Anthropometrical Laboratory, Trinity

CoUeg^f Publin, is appended. The internal dimensions of the grave

m.I.A. PBOO., VOL. XXIV., BIG. C] [1]

2 Froceedinga of the Royal Irush Academy.

were 4 feet in length, by 2 feet 3 inches in breadth, by 2 feet is
height. The plan and cross-section (figs. 1 and 3) will show the
arrangement of the flags. There was no flag in the bottx>m of the
grave, the contents resting directly on the grayel. This, by the war,
is limestone glacial detritus. The side flags were from 3 to 4 inches
in thickness, and were placed outside the end ones, which weie of
the same description. The eovering-stone was very roagh aod
irregular. The top of this coyer was situated 1 foot below ^
surface of the ground, and the grave was placed due north and south.
On inquiry I was informed that, when first found, the skulls were«t

*. *.• (\?^'«* I 111' 111 •• "^

c ravel:

FiQ. 1. — Sectioii of Ciflt.

the north end— one in each comer; the bones scattered along thf
grave, and the urns exactly at the centre at the east and w^
points, and resting on their bases.

The measurements of the complete urns are — ^\ inches in heigtt,
5^ inches across the rim, 2f inches across the base.

The remnant of the broken one measures — 5 inches acroM thciim-

There were no markings on the base. The ornamentation app«»
to be deeper and clearer on the broken one than on the one that i«
complete. On each there is a pattern inside the lip. When lo«wi»

{«)

Fio. ^.— Thi UuNg.

[!•]

4 Proceedings of the Royal Irish Academy,

they were quite wet and easily broken. The pottery of which titer
are composed is coarse and nnevenly burnt, quite black at the ceniit,
and brown outside and inside. Prior to the discovery of thiflgntt,
the workmen found some bones, but re-buried them. The pitissov

fC\' ,.'1' «.'.^"/r.^v' ' ;•

- 1. FF

.':■ ' . L'i .'^

f:w, ,___ .__ ^^ III;.

hmiliniE

Fio. 8.— Plan of Cut.

nearly half excarated, and it is quite possible that another dst vill
be unearthed.

No beads, or ornaments, or celts were found in the neighbomhood
of the grave, nor in the cist itself.

Eepobt bt Db. Chajlles Bbowitb.

AlTTHBOPOLOOICAL JjAStiBXSWJ-

Tbinitt Collbgb, BuBicr-

Having examined the remains submitted to me I have to rep^
hat they consist of —

(a) Bones which have been cremated or subjected to gre** ^^
These are in so fragmentary a state, and so distorted by the sctK*

BoBBBTs — Discovery of Ancient Gfrave, County Oalway. 6

of fire, that little or nothing can be said about them except that they
seem all to belong to one skeleton (there being no duplicate of parts),
which, it may be assumed from the size of the mastoid processes, was
that of an adult male.

{I) The greater part of a skeleton of an infant.
These bones show no traces of the action of fire.

{e) The skull and many of the other bones of a young person, all
much '' weathered," and extremely brittle from loss of animal matter.
The skull was in so fragile a condition, that in order to enable it to
stand the handling necessary for examination and measurement it had
to be painted over with a solution of gelatine.

The body evidently lay on its left side, as the bones of that side
(especially the bones of the skull) are much more affected by damp
than those of the right. The skull is that of a young person under
eighteen years of age and most probably a female. Much injured by
damp, the outer table of the left parietal bone and the left zygoma
being weathered away. It is of small size and symmetrical in shape.
Viewed in norma vertiealiSf the shape of the cranium is a broad blunt
oral, in norma lateralii the forehead is seen to be upright, with well-
marked frontal eminences, vertex high and prominent, occipital region
flattened off above superior curved line.

Mastoid processes very small, and glabella absent, all markings
slight. The cephalic index is on the border line between the mesati-
cephalic and brachycephalic classes. Frontal bones grooved. Face,
medium width. Left molar bone and zygoma weathered away. Nose,
evidently leptorhine. Orbits round, megaseme. Palate deep. Teeth
all present at time of death, small and sound. Third molars not
erupted.

Mandible, much weathered on left side, so no measurements are
obtainable. The angles are strongly marked, and the mental pro-
oesB prominent and square; slightly concave at symphysis, so that
ft has an almost forked appearance. Teeth were all present at time
of death, sound, and not much worn on crowns.

StUuree, — ^AU open (including the basilar suture), those of the
vault very complex ; three epactals in lambdoid suture.

There is little worthy of record to be observed in the other bones,
ribs, TertebrsB, clavicles, scapular radii, and ulnsB, fibulfle, pateUse,
bones of the hands, feet, and tibiae, except that by their stage of
derelopment they^aid in determining the age of the person to whom
they belonged, and that by the absence of any duplication of parts

Proceedings of (he Royal Irish Academy.

the presence of only the one body is ascertained. The impoitaot
skeletal bones are the femora, ossa innominata, and sacrum, and all
are those of a young person. The femora are much weathered, and
the epiphyses had not at the time of death joined the shafts of tlie
bones, but were among the bones recovered, and where those of ibe
right femur were placed in situ the whole bone thus set up measoied
415 mm.

The ossa innominata are not fully developed, the three primizj
divisions ilium, ischium, and pubes, all existing as separate bones.

Ute Sacrum. — The body of this bone is in good preservation, tnd
is rather broad and flat, lateral masses absent, the fourth and fifth
segments are united, but all the rest are detached.

From the condition of these bones and the skull it may be oon-
cluded that they belonged to a young person under seventeoi yaanof
age, and most probably a female.

On account of the weathered condition of the skull, and the yoath
of the subject, the cranial measurements are not very valuable, M
are given here in case they may prove of interest.

Calculating from the length of the right femur, the stature w
1482 mm., or about 4 feet 10 inches.

Cranial Measurements.

Glabello-oooipital length,
Mazimum breadth, . .
Baaio-bregmatio height,
Auriculo-Tertieal height,
Frontal longitudinal arc.
Parietal longitudinal arc,
Occipital longitudinal arc,
Bi-asterio width, . .

173
140
130
117
116
126
110
113

Horizontal drcumferenoe, .
Foramen magnum length, .
Foramen magnum breadth,
Baaio-alvedlar length, . .
Baaio-nasal length, . . .
Auriculo-nasal length, . .
Aurioulo-alyeolar length, .

Faee.

Face length, 104

Naaio-alyeolar length, .... 63

Facial breadth, 20

Nasal height, 44

Nasal breadth,
Orbital height.
Orbital width,

497
Si

9S
U

M
90

Indices.

Cephalic, 80*9

Altitudinal, 76-i

Auriculo-yertical, 67*6

Alveolar, ^^

Nasal, -«*•>

Orbital »l "^

Chaklbs R. Bwwsu

[ 7 ]

II.
THE IKISH GUABDS, 1661-1798.
By C. LITTON FALKINEB, M.A.
TRoad NoTRMBBB 80, 1901.]

Thb recent addition to the strength of the British Army of a B.egiment
of Irish Guards has been hailed with acclamation as an appropriate
compliment to the soldierly qualities of Irishmen, and as a graceful
recognition of the yalour displayed by Irish troops on the battle-fields
of South Africa. The innovation has also been criticised, on the other
handj as a somewhat tardy recognition of the claims of Ireland to
a share in the honour of furnishing those regiments which are most
closely associated with the personal service of the Sovereign, and
which have enjoyed for centuries a traditional precedence in the regi-
mental roll. It is not a little curious that a people, who, differing
among themselves in many things, are at one in their common pride
in those martial instincts which Irishmen have manifested wherever
snd whenever opportunity has served, should have so completely
forgotten an episode so interesting in the history of Irish arms as the
nuAing of the first regiment of Irish Guards. Yet it is a fact that
what has been greeted as a belated innovation is really only a revival
of a corpe which is coeval in antiquity with the institution of the
standing army, and which, under the title of ''His Majesty's Begiment
ot GnardB in Ireland," enjoyed a distinguished reputation for valour
and military efficiency at a most interesting period of Irish history.

The occasion, therefore, seems appropriate for an attempt to trace
the record of a regiment which anciently held a distinguished place at
the head of the military establishment of Ireland, and to recall the
history of the remarkable corps which constituted the flower of the
Iriah army from the Restoration to the Bevolution. And the
inquiry is not the less interesting because it is in this Restoration
Regiment of Irish Guards that we shall find the origin of one of the

8 Proceedings of the Royal Irish Academy^

most eminent of the distingmslied corps whicli subseqnently ccxnsti-
tuted the Irish Brigade abroad. For though disbanded after tk
Boyne, the regiment, taking service abroad, achieved under a sqcccI'
sion of brilliant officers^an honourable place in the military bistoiy of
eighteenth-century France. And preserving in exile that fealtj to
the principle of hereditary right which, combined with devotion to tb
Koman Catholic faith, had led its officers to adhere through evil dap
to the fallen fortunes of James the Second, it renewed, on the fall d
Louis XYI., its allegiance to the Sovereign of the Three Kingdoms,
and was re-enrolled for a brief period in the ranks of the Britisk
army.

The oblivion into which the origin of the regiment has fallen is,
however, not very surprising, and is explained in great part by Uie
circumstance that the compilers of Irish military history have g^Tea
but scanty attention to the records of Irish regiments at home. Fw
example, 0'Conor*s *' Military Memoirs of the Irish Nation,'' useliki
as an account of the exploits of the Irish Brigade abroad, i$
absolutely silent on the military establishment of Ireland at the
Restoration. D' Alton, again, in his ''Historical and G^ealogieil
Illustrations of King James's Army List," begins, as is natural, only
with TyrconnePs viceroyalty. And though O'Callaghan, in las
admirably minute and exhaustive ''History of the Irish Brigade is
the Service of France," does not omit all notice of the origin of the
distinguished regiments whose subsequent careers he traces in so miKb
detail, his references to their pre-Revolution story are brief and pares-
thetic. To this explanation of our ignorance of the earliest records
of the first regiment of Irish Guards it may be added, that it is odIt
in years comparatively recent that the materials for tradng the origia
of the regiment with any semblance of completeness have became avail-
able. No investigator in this field of our seventeenth-century history
can fail to acknowledge a large debt to our distinguished andbone&ted
academician, the late Sir John Gilbert, who, by his labours as editor
of the Ormonde Manuscripts and of the Becords of the Corporatioa oi
Dublin, has thrown open to the students of seventeenth-centurr
Ireland two splendid treasuries of historical, topographical, and anti-
quarian lore.

The process by which the regiments raised by various roymlift
officers became the parents of several of the most distinguished ot
existing regiments has its best known examples in the Grenadier
Guards and the Coldstream Guards, and need not be delin«afeed
here. And the circumstances which, immediately following on thf

Falkimer — The Irish Guards. 9

Bestoration, led to the institation of a standing army, and laid the
foundationB of the existing military system of the United Kingdom,
are familiar to erery student of our political and constitutional
history. But it may be well to glance at the beginning of the system
in Great Britain, since it was there that the model was proTided for
the military establishment which, on the appointment of the Duke
of Onnond to the Yiceroyalty, was at once instituted in Ireland.
Especially is this necessary to the elucidation of the origin of the Irish
Unaids, because the conception of a regiment directly associated with
the Crown, a regiment formed to be, in fact as well as name, *' His
Majesty's Guards,'' goes back to a period prior to the Bestoration.

Four years before the Bestoration, Charles II., hopeless of the
renewal of the ineffectual and half-hearted succour extended to him at
the beginning of his exile by the French Court, which under the
inspiration of Mazarin had become convinced of the permanence of the
Cromwellian reginU, imagined that he had found in Spain the assistance
necessary to regain his throne. In connexion with a project for the inva-
sion of England by a Spanish expedition, it was resolved to organise, for
service with the Spanish forces in the Low Countries, the considerable
soldiery which had accompanied their Sovereign abroad, and had
earned distinction in the armies commanded by Turenne. Accord-
ingly, several regiments, both British and Irish, were gathered to-
gether into a division, and placed under the Spanish commander in
Flanders. The English officers, by whom Charles was more imme-
diately surrounded, were formed into what was called a Boyal Begi-
ment of Guards under Lord Wentworth, and some regiments of Irish
were organised at the same time.^ The command of the largest of
these, a corps seven hundred strong, was assigned to the Marquis of
Onnond, and quartered near Bruges, and ultimately took part in the
unsuccessful operations at Dunkirk. The officers included many of
the Confederate Catholic officers who had fled from Ireland.'

* Clarendon's account of the matter id as follows :— '* The king resolved to raist'
one ngiment of Guards, the command thereof he gave to the lord Wentworth,
which was to do duty in the army as common men till his majesty should be in
mch a pooture that they might be brought about his person. The marquis of
Onnond had a regiment in order, to be commanded by his lieutenant-colonel, that
the Irish might be tempted to come over." — '* History of theRebellioni*' xr.,
p. 68.

* Sir F. Hamilton, in his '' History of the Grenadier Guards," mentions that
Charles I., during his stay at Oxford in 1642-3, had raised a regiment which was
known as " The King's Guards," and states that '* the Regiment of King's Guards,
as well as all the rest of the Royalist troops in England, ceased to exist as regiments

10 Proceedings of the Royal Irish Academy.

Wentworth's Kegimeot of Quaids survived the ill-saco»B o!
Charles the Second's negotiations for aid from Spain ; and remaining
abroad at the Eestoration as part of the garrison of Ihinkirk, it escaped
inclusion in that general disbandment of the army of the Cammon-
wealth, in September, 1660, which was almost the first act of the
restored monarchy. The young Sovereign, however, whose iHiok
conception of the kingly dignity was coloured by his familiarity iritli
continental courts, had no intention of remaining without a penontl
^ard ; and at the very moment which witnesses the dispersion of the
remnant of Cromwell's Ironsides, he entrusted Colonel John Eussell, a
brother of the Duke of Bedford, with a commission to raise a Begi-
ment of Foot Guards, twelve hundred strong, under the title of tiie
King's Eoyal Regiment of Guards. Lord Wentworth's earlier formed
regiment remained abroad until the sale of Dunkirk, when it came to
England, where it was maintained as a distinct corps during Went-
worth's life. But on the death of its colonel, three years later, en
the eve of the outbreak of the Dutch War, Wentworth's was merged
in Colonel Russell's regiment, to which the existing regiment d
Grenadier Guards proudly traces its origin. *

No one who has had occasion to consider the character of tiie
arrangements made upon the restoration for the machinery of tlte
constitution and the equipment of the public service, can have failed
to be struck by the closeness with which the institutions of every
sort set up in Great Britain were followed in the organisation of ^e

in 1646-7 ; and the English troops raised subsequently by Charles II., with vhidi
he endeavoured to recover the Crown of bis ancestors, were disbanded after the
battle of Worcester in 1651 ; so that though we trace among the offic«s of ^
Regiment of Guards which Charles II. raised in Flanders many Royalists who had
either served in the Eing*s Guards or in other corps during the Civil War^ both tm
the time of Charles I. and II., there is no connexion as a regiment between tbeea
two corps of Guards '* (vol. i., p. 8). It appears, however, from a letter pablisbed
in the '* Ormonde Papers '* (Hist. MSS. Comm., 14th Rep., vol. L, p. 97), that
Wentworth's regiment existed in some foim in 1649: — ** Thomas Wentworti
to Edward Broughton. Breda, June 24, 1649. You are to reeeive ssch
men as shall be delivered you on shipboard as part of a Regiment to {«tr) ths
King's Guards, and you to command them as Serjeant -Major to the said RegJaeBt.
and at your landing in Ireland you are to obey such orders and directions as jva
shall receive from the Marquis of Ormond, the Lieutenant-General of tiie kingdov
of Ireland.'' It is noticeable that this letter is addressed by the subsequent eokDsl
of Charles the Second's post-Restoration Guards, to an officer who subseqimdy
held a commission in that regiment. The letter is addressed, *' For Major Edvaii
Broughton, Major to the King's Guard of Foot."

1 Sir F. Hamilton's ** History of the Grenadier Guards." pp. 30-31.

Falkiner — The Irish Ouards. 11

Iriali GoTemment. The formal constitution of a standing army by
Charles II., and the formation of His Majesty's Begiment of Guards,
took place early in 1661. It does not appear how far, if at all, the
King's advisers then contemplated the proyision of a separate military
establishment for Ireland. It is probable that the question remained
in abeyance until after the selection of the first Restoration Viceroy,
an appointment which was delayed until the autumn of that year. When
the Duke of Ormond was appointed to the Yiceroyalty, he was careful
to imitate in all respects, as far as possible, the model provided in Eng-
land. The establishment for Ireland, both civil and military, followed
closely upon the lines laid down by Clarendon and the other advisers
of Charles II. Ormond was given a free hand in Ireland, *^ the
places, as well in the martial as civil list, being left freely to his
disposing." He at once proceeded to exercise his authority, by
providing for the civil and military needs of Ireland upon a scale of
great magnificence. And as a means, both of emphasising the dignity
of the Viceregal office, and of supplying an -efficient force for service
in emergency, one of his first steps was to procure a commission ta
raise a Regiment of Guards for service in Ireland. Accordingly, on
April 2drd, 1662, a commission was issued to the Viceroy.^

The Buke of Ormond' received his commission on April 2drd, 1662,
and he lost no time in acting on the authority thus given to him.
On the following day the regiment was formally constituted, and pro-
vision was made for the enrobnent of twelve companies of one hundred
men each. The Viceroy's second son. Lord Richard Butler, who was
immediately afterwards created Earl of Arran, was gazetted Colonel of
the regiment with the captaincy of a company ; and eleven other
officers were appointed to the remaining companies.' The establish-
ment of the regiment was calculated on a generous scale, no less a sum

* Th« following is the text of this CommiBsion : —

" Whereas we have already constituted and appointed James, Dnke of Ormond,
to be €k»vemor of our Kingdom of Ireland, and of all our armies there raised and to
to be raised : And whereas we hare thought fit to raise within this our kingdom
of Ireland, a regiment of 1200 foot to be our Regiment of Guards in our said
Kingdom of Ireland : We do give and grant to our said lieutenant and Chief
Ooremor full power, liberty and authority, by beat [of drums, proclamations, or
rjtherwiae, to raise the said number of men in England, and to conduct, lead and
transport them into Ireland, with power and authority to him to give and giant
romminioDB under his hand and seal to such persona as he shall think fit to be
ofllcers and commanders of the said regiment." — Ormonde MS., unprinted.

' Ormonde, MS., vol. i., 239.

» *• Sir William Petty's Political Anatomy."

12 Proceedings of the Royal Irish Academy.

than £24,518 8«. Si. per annum being allocated to its maintenanoe.
Its roll included, in addition to the Colonel, a Lieutenant-Colonel, a
Major and nine Captains of companies, twelve LieutenantSy twehre
Ensig:ns, forty sergeants, thirty-six corporals, a drum-major iritb
twenty-four drummers, a piper to the King's Company, and twebe
hundred soldiers. In addition to the fighting strength of the regi-
ment, there were attached a Chaplain, an Adjutant Quarter-Maeter,
a Surgeon and Surgeon's mate.^

It does not appear from any document from what dislzict the
rank and file of the regiment was recruited ; but it is evident that tt
the date of the commission to Ormond considerable progress bad been
already made in finding the men and arranging for tbeir equipmeoL
and the original list of officers included some who had served in tbc
regiment commanded by Ormond in Flanders. On April 14th, 166i,
the Vice-Treasurer received orders to pay to Lieutenant-Colonel Sir
William Flower, the sum of £1897 8«. 8 J., "towards the raising,
sending to the sea-side, and transporting into Ireland of the <^cers
and soldiers of the said regiment."' Two days later, a similsi
sum, ''being one month's pay of the fiegiment of Ouaids for
Ireland," was ordered to be paid to the same officer. On ApcQ
2lBt, orders were given for £663 14*. to be paid to John WaB,
''for 600 scarlet coats, bought of him for His Majesty's Regimest
of Guards for Ireland, and £755 12« to be paid to Henry Pk«80ott
for 661 red coats, and embroidering twenty-four drummer's oooti.
with sacks to pack them up in.'" This miiform is identical wi^
that prescribed for Colonel Russell's Regiment of Guards m Rnglind.
A little later Alderman Daniel Bellingham, afterwanls the fint
Lord Mayor of Dublin, received an order to furnish all the non-«om-
missioned officers and men with a "red cassock," a term not as yet
appropriated by the clergy, together with " cloth breeches^ twu
shirts, one pair of stockings, and one pair of shoes."*

No time was lost in transferring the newly raised regiment to its
destination. As early as May, the news-letters of the day chronicled
the embarkation of the Guards for Ireland.* "On the 9th instant*'
according to the Chester correspondent of Mercurius Publicos, ^' Six

* Ormonde, MS., vol. i.
« Carte Papers, 166, 3.

3 See Sir F. Hamilton's « History of the Orenadier Ghiaids."

* Orrery's StaU Lettert, p. 68.

^ Mercurius Puhlicus, May 9 and 28, 1662. See alM " M'Einnon*a ** Hu^orr
of the Coldstream Guards," i. 109, note.

Falkiver— The Irish Ouarch. 13

William Flower, who bad the conduct of HIb Majesty's Begiment of
(huada for Ireland^ tmder the command of the Earl of Arran, arrived
here with that regiment, in order to their transportation for Ireland,"
and on the 14th Hay, it was reported that " Sir William commenced
to ship twelve companies in eleven ships at Weston." We are further
informed that *' during the march from London with this regiment.
Sir William himself constantly marched with the men. Sir William
Flower, my Lord Gallan and other chief officers in the regiment were
entertained by the Mayor at Chester." They reached Dublin safely
before the end of May ; and on the 28th of that month, the same
journal announced that '' the King's Begiment of Foot, under the
command of the Earl of Arran, consisting of twelve companies, that
came this week from England, marched this day, completely armed
and clothed through the city, and are all quartered in and about it for
the Gfuards."

The conception of the regiment being that of a body-guard for the
person of the Lord Lieutenant as the representative of the King, it
WIS not contemplated that the corps should serve, in time of peace
at least, outside the capital. Accordingly, arrangements were at
once made for quartering the soldiers in Dublin, and for this purpose
communications passed between the Government and the City Corpora-
tions. Between the Court and the City the liveliest accord existed
throughout Ormond's Yiceroyalty, the Duke having, as one of his first
acts, secured a payment of £500 a year from the exchequer to the
Mayor in consideration of the loyalty of the city in the years
following the Bebellion of 1641, and of the civic poverty resultiog
from the Civil wars, and having exerted himself to the utmost at the
restoration for the protection and enlargement of the liberties of
Dublin. And it was to Ormond's intervention that the dignity of
lord Mayor shortly afterwards conferred on the head of the Corpora-
tion, the royal gift of a collar of SS. and cap of maintenance, and
other marks of royal favour, were directly due.^

The City Assembly was therefore prepared to comply with a loyal
alacrity with the direction of the Viceroy to provide quarters for the
Guards. On the 28th May the Lords Justices and the Council, by
direction from the Lord Lieutenant, ordered the sherifls of Dublin and
aeneschala of the Liberties " to provide lodging for the officers and
soldiers of His Majesty's Kegiment of Guards lately arrived out of

1 Speech of Sir W. Davys, the Beoorder, Dablin Corporation Becords, iy. p.,
579, and tee vol. i., p. 42.

14 Proceedings of the Royal Irish Academy.

England, in innB, wine-tayems, ale-housee, or victaalling hooaes.^ TIm
officers were likewise quartered on the city. On Jnne 1411i Oimoiid
wrote to the Mayor, and sheriffs requiring them ''forthwith to
appoint convenient quarters as near the Castle of Dublin aa nuT
he for our son Richard Earl of Arran, Colonel of His ICajesty'i
Eegiment of Guards and his servants'"; and shortly afterwardi
provision was made by the city, pursuant to his Excellency's wamnt
for the quartering of the commissioned officers of the Eing'i
Begiment in the city and suburbs. Thenceforward and down to the
Bevolution, Dublin appears to have continuously remained the head-
quarters of the Guards; and although the arrangements for tluii
lodging appear to have involved some burthen on the city, the best
relations seem, in general, to have been maintained between citixeDs sad
soldiery. The troops seem to have been quartered partly in the Castk,
partly through the city, especially at the city gate-houses,^ which, at
that time, were still utilised for residential purposes, as appears from
the complaint of one John Eastwood who had contracted to pay £4 per
annum to the city for St. Nicholas' Gate, but represented that '* the
said gate was taken up from him by the soldiers, by special orders froic
the Lord Lieutenant, to his very great damage." The provision of fire
and candlelight for the Guards were also constituted a charge upon tbt*
city, and assessments were annually made for this purpose on a warxant
from the Viceroy, this being, in the language of a resolution of 1665.
** required to be done by act of state and a business of public oonoei^-
ment to this city."* The amount of the assessment for this puipoee
was usually from £150 to £200 a year. The tax appears to have, in
general, been readily contributed, though in June, 1667, one John
Quelch, a freeman of the city and member of the Corporation, refused
'' in violation of his oath as freeman to pay his portion of the charge
amounting to half-a-crown " as unlawful and unwarrantable. ^

In addition to the occasional restiveness excited by the tax for Hieir
maintenance, the Guards appear to have provoked some unpopularity by
their demeanour towards the citizens. In August, 1 667, a petition was
presented to the Lord Lieutenant by the City Council "for a redress
against the several oppressions of the officers and soldiers on the
inhabitants of the city under the pretence of quartering." This,
however, was resented by the Colonel, Lord Arran, and the officers oi

1 Carte Papers, 37, 228.

* Corporation RecordB, iy,, p. 273. » /KA, p. 299.

* Dublin Corporation Eecords, iv., p. 347. ^/M., p. 4Sd«

Falkiner — The Irish Ouarcis. 16

tbe regiment, who, in a counter-petition, demanded an inquiry into the
matters complained of, averring their indignation at aapersions which
they stigmatised as ''a high reflection on the officers and soldiers of the
said Guards, either in conmiitting or suffering such oppressions to be
committed by those under their command/'^ But in general the rela-
tioDB between soldiery and civilians were harmonious, and Dublin was
proud of the regiment. In 1666' '' his grace the Buke of Ormond,
taking notice of the many buildings lately made on Oxmantown Green,
vhich have taken up so much room there that His Majesty's Horse and
Foot Guards and the City Militia have not conveniency to exercise as
formerly," and '' recommending the city to take present orders that
the grounds upon St. Stephen's Green, lately walled in, be forthwith
made fit for that purpose," the Assembly cheerfully ordered that the
ground should be levelled and made smooth with that object. This was
Accordingly done, and thenceforth St. Stephen's Green became the
parade-ground of the Guards. A review of the regiment on this ground
twenty years later is described in Clarendon's State Letters.*

A further memorial of the connexion of the Irish Guards with
Dublin is supplied in the records of two J)ublin parishes. The
regiment appears to have attended Divine service regularly every
Friday, sometimes in St. Michael's and sometimes in St. Audoen's,
and in 1671 Lord Arran contributed a sum of £150 towards the re-
building of the latter church. In requital of his liberality it was
ordered ** that the arms and supporters of the said Earl of Arran be
fairly presented and erected in the said church " ;* and further, that
every commissioned officer of the Royal regiment, from the said Earl
to the ensign, should henceforth enjoy all privileges and indemnities of
pmahioners in regard to marriage, christenings, and burials. The
I>arish of St. Michael was less fortunate, when two years later it
solicited a like contribution, notwithstanding that it was averred that
** for several years past the several companies of the Royal regiment
practised in this city have made use of the Church of St. Michael,
but in all that time nothing hath been contributed towards the repara-
tion of the said church or the seats thereof."

Mention has just been made of the City Militia, and some confusion
tnight easilj occur between the two bodies, which in the Assembly rolls

1 Dublin Corporation Becords, iv., p. 423.

*Ihid. p. 383, nth Aug. 1666.

> Clarendon's " State Letters," vol. i., 434, 8th June, 1686.

« Gilbert's ** History of Dublin," i., 281.

16 Proceedinga of the Royal Irinh Acackmy.

are Bometimes referred to indifferently as the GnardB of the citj. Tbe
two forces were, however, entirely distinct, and had no relation to each
other, save in so far as each was in its degree responsihle for the defence
of the city. A militia, 24,000 strong, was raised to supplement tb
regular army ; and in 1660 two foot regiments of city militia had heen
formed, one for service within the other without the city, the Major
for the time being acting -as Commander-in-Chief. The Mayor wb
likewise designated commander of a foot company through the good
offices of Sir Theophilus Jones, the Scout-Master-General of ihit anny,
a distinction which was so much appreciated by the city dignitary tbt
the city assembly voted a sum of £50 for a piece of plate to be pie-
sented to Lady Jones in recognition of her husband's exertions.' 8<nQe
friction seems occasionally to have been provoked between the (^
Guards and the King's regiment. The author of ''Ireland's Sad
Lamentation'" imputes to the latter a slackness little creditable to tiie
gallantry of the corps, alleging that the militia would not be suffeRd
to guard within the city, the King's Guard being appointed to defemi
the same, and were obliged to serve outside the walls, '' so that upoD
any attempt, our volunteer inhabitants might certainly have perished
before the King's soldiery who receive pay had entered into any
dangerous engagement." But this innuendo, with the rest of the pub-
lication in which it appeared, was declared by the city Assemblj to be
** a black and ugly libel."

Another force not to be confounded with his Majesty's Regiment of
Guards was the Lord Lieutenant's Guard of Halbertiers or Battle
Axes, which, during the reign of Charles II., from the opening of
Ormond's Yiceroyalty^ in 1661 down to 1665, was maintained as put
of the Military Establishment. This body which was known some-
times as the Company of Battle Axes, sometimes as the Guard of
Halbertiers, consisted of a captain, lieutenant, two sergeants, and sixtj
men, dressed in buff coats, and was modelled on the Yeomen of the
Guard.* The provision made at the Restoration for such a retinue to
attend the Viceroy was in accordance with the ancient tradition* c£
the Viceregal ofBce, for as early as the reign of Henry VIII. wh«^
the Earl of Surrey came over as Deputy, one hundred Yeomen of the
Guard were sent to Ireland with him to serve as his body-goari* I:

^Dublin Corporation Eecords, iy., p. 221.

« ** Ireland's Sad Lamentation," L681. Dublin Corporation Kecopis ▼. IV****-*

'Ormond Manuscripts, i., p. 406.

* " Sir W. Petty's PoUtical Anatomy."

* Preston's " Yeomen of the Guard/' p. 100.

Falkinbr— 2%« IrUh OtMrds. 17

would appear that in their uniform and accoutrements this Guard

closely followed its English prototype.^ On April 2, 1662, Colonel,

afterwards Sir Daniel, Treswell, who was appointed to its command,

receiTcd from Ormond a warrant for £275 4*. towards buying **64 buff

coats and 64 belts at £4 6«. for each coat and belt for our guard of

foot." The forces having been equipped in England came to Ireland

in that year, and ''for the more convenient performance of their

duty"' were ordered to be quartered as near to Dublin Castle as

possible. Treswell, their Colonel, who had come to Ireland in

1641 in command of a troop of horse, had ''faithfully served his

Majesty in honourable employment during the whole war in England

and Ireland," in the course of which he had commanded the Lord

Lieutenant's regiment of horse, and Ormond, loyal in prosperity

to his friends in adversity, not only rewarded his fidelity with the

comnumd of his Battle-axes,' but procured him, in 1665, the honour of

a baronetcy, and recommended him in the same year to the burgesses of

Downpatrick by whom he was returned to Parliament.*

In addition to the city guard the Lord Mayor, in emulation of the
Lord Lieutenant, seems also to have instituted a small body-guard of
balbertieTs ; but it is not surprising to learn that this force, six in
number, was " not found so useful as it was expected," and that it was
in consequence ordered that as many of them as the Lord Mayor and
sheriffB should think fit to be officers at mace should be so appointed,
and discharged from their place of bearing halberts.

•

That his Majesty's Begiment of Gruards was from the first intended

to hold the highest place in the regimental roll in Ireland there can be no
manner of doubt. When, during the Viceroyalty of Lord Clarendon, at
the opening of the reign of James II., several of the officers of the Guards
were displaced by Tyroonnel in pursuance of his programme to new-
model the Irish army on a Boman Catholic basis, Major Billingsley,
one of the displaced officers, in protesting against his removal, averred
that '^ to be a Major of the Royal Regiment of Guards is better and
more honourable than to be Lieutenant-Colonel of any other regiment."

' Carte Papen.

2 Order for quartering the Battle-axes, Dec. 8, 1662, Ormond MS., Dublin
'^orporution. Records, iv., p. 646.

' Hist. 1A8S. Com.» 6th Rep., 14th Beport, i. and ii.

4 Xbe following inicription appears upon a tomb in the chancel of the old church
t Fin^Iaa, near Dublin: — <*Heere under lyeth the body of Sir Daniel Treswell
night Bud. baionett who faithfully serred his Majesty in honourable employment
uring the ^whole war in England and Ireland and dyed the 24th day of May, 1670."

B^i.A- ymoc., VOL. xxnr., sac. c] [2]

] 8 Proceedings of the Royal Irish Acadentp.

The prestige of the regiment derived Selat at the outset from the Uti
that the commission for the raising of the regiment was given to the
Yiceroy. The Buke of Ormond was not alone the King's reps^aenta-
tive and the General-in-Chief of the army in Ireland, hut the first of
his Irish suhjects in rank, fame, and fortune. He had held the post
of Lieutenant-General or Commander-in-Chief of the aimy formed bj
Strafford as far hack aa 1640. His association with the regiment
would have heen sufficient of itself to stamp the corps with peculiar
distinction; and Ormond was careful to secure that its honour shonU
undergo no diminution in the persons of its officers, who were selected
largely from the ranks of the Irish nohiHty, and included several
who had followed his fortunes through the whole course of the civil
war and foreign exile.

Unable himself, with the multifarious duties of the Viceroyalty, to
assume the direct command, Ormond asserted in the most marked wax
his personal interest in the fortunes of the regiment by nominating
to the Colonelcy his second son Bichard, Earl of Arran, a nobLeman,
who, if less distinguished than his gifted brother. Lord Osaorj.
was yet a man of considerable ability, who, on more than ocv
occasion during Ormondes absence in England, filled the offioe of
Lord Deputy. Arran gave proofs of considerable military capacity
in command of his regiment, first in suppressing a formidable mutiny
of the soldiers of other regiments at Carrickfergus, in 1666, and later,
in 1673, by his distinguished conduct under the Duke of York, in the
sea-fight with the Butch in that year, in which, after the manner of
those days, the Guards took a part serving on board ship.^ For bis
services on this occasion, Arran was rewarded with an English
peerage. ** No man," says Carte, ** was more active, more eager, and
more intrepid in danger." During his tenure of the office of Deputy
in 1684, he exhibited great personal gallantry in dealing with a vt^
serious fire in Dublin Castle, by which a great part of the castlt
buildings was destroyed.' An address of congratulation was presentee
on this occasion by the citizens of Dublin, in which Arran's energy i*
eulogised in glowing terms : ** By your Excellency's presence of miiKl,
care, and conduct, in the midst of the devouring flames which
encompassed you, not only the remaining part of the buildings of rhe
Castle, but the great magazine of powder to which the fire had within
a few steps approached, was wonderfully preserved, and the ancicn-
records of this Kingdom, then also in the Castle, rescued from tb*js«

» Carte's *' Ormonde," ii., 544. « Dublin Corporatioa Reoonls, t.» p. * 1 2 .

Falkinbr — The Irish Guards. 19

flameB." On Lord Arran's premature deatli, early in 1 786, shortly after
his father had been recalled from the Irish Goyemment by James II.,
the direct association of the Ormond family with the Gnards was
maintained by the bestowal of the command of the regiment on Lord
Ofisoiy, son of the distinguished soldier-statesman of that name and
afterwards second Duke of Ormond, a selection which, as the new
Viceroy, Clarendon, reported to Sunderland, gave as lively a satisfaction
in Ireland as could be imagined.^

At the time of his original appointment, Lord Arran was too
junior to have acquired the military knowledge necessary in the
commander of the regiment in the field ; and for the Lieutenant-
Colonelcy Ormond selected, as we have seen. Sir William Flower, an
oiRcer who was well qualified by his experience to undertake the
effective control of the newly enrolled corps.' Flower, whose father
had come to Ireland towards the close of Queen Elizabeth's reign, and
had served in James the First's time as Governor of Waterford, had
been one of Ormond's officers in the troubled years that followed the
rebellion. As early as 1641, he had held a Captain's commission in
Ormond's own regiment of foot, which had its quarters in Christ-
church yard, and had formed part of the garrison of Dublin down to
1648 ; and he had risen to its command. He had suffered imprisonment
at the hands of the Parliamentary party on Ormond's departure from
Ireland in 1648. At the Restoration, he was at once raised to eminence
by his old patron, becoming a member of the Privy Council, with a
^i»i in the Irish Parliament as member for St. Canice, and being
appointed one of the trustees for satisfying the arrears of the '49
officers. He received considerable grants of land ; and his son
extending the family influence by a matrimonial alliance with the
daughter of Sir John Temple, the family became important enough to
willy in the person of Sir William Flower's grandson, the peerage
of Castle Durrow, a rank which, in the generation following, was
merged in the still existing dignity of the Viscounty of Ashbrook.*

The other officers appointed to the command of companies at the
institntion of the regiment were likewise persons of distinction. The
King's Company was given to Sir Nicholas Armorer, who had acted

' "dArendon Coirespondence,*' i., 229.

« Archdall's " Lodge's Peerage," vol. v., p. 283.

* Thczre Ib good reason to soBpect that during the eclipae of the rojelist fortunes
•*iower, like not a few of Ormond's Irish adherents, was among those who conformed
o the gowemment of Commonwealth, and to hare held a command in Fleetwood's
i"tnment. See the Leyhume-Popham Papers, Hist. M8S. Commissioners Report,
. 153.

[2*]

20 Proceedings of ike Royal Irish Academy.

as equerry to the Eong in exile, and was a dose friend of the Duke
of Ormond, by whose influence he was retnnied to Parliament for
Comity Wicklow, and appointed Governor of Gork.^ Sir John Stephens,
who, like Sir William Flower, had held a commission in Ormond'^
old regiment as far back as 1643, and who, after the Restoratioii,
represented Fethard in the Irish Parliament — who had married a
sister of Flower's, and held the office of Governor of Dublin Castle
— was appointed Major ; and the other officers included Lord CaUan,
afterwards the third Earl of Denbigh, Lord John Butler, Ormondes
youngest son, and Colonel Francis Willoughby, well known in the
ten years' warfare in Ireland, from 1641 to 1651. It is thus evident
that the note of pre-eminence and distinction which has ever bees
associated with the Guards in England was characteristic of the Irii^
regiment from the date of its institution.

A corps, whose sphere of service was restricted in time of peace to
the capital, and which even in war was only likely to be aetiveLj
employed in circumstances of emergency, was naturally deprived for
some years of many opportunities of distinguishing itself, and it is
not very easy to recall the record of the regiment in the first few
years of its existence. Its earliest active service appears to have been
in suppressing the mutiny at Carrickf ergus in 1 666 already noted,^ but
down to 1673, such mention of it as we find is chiefiy in connection
with ceremonial display. On the occasion of the Duke of Onno»i <
state entry into Dublin, in 1665, a pageant of unusual magnificen(v.
the regiment formed the guard of honour, from St. James's Gate t«'
the Castle, the King's Company being in close attendance on tb^-
Viceroy, and following immediately the Guard of Battle-axes. In
1672, they were ordered for service with the fleet on the outbreak vt

^ Cbobnondeley Papers, Hist. MSS. Com., 6th Bep.

' The following reference of the services by the Guards on this occasioB is taken
from McSkimmin*8 " History of Caixickf ergus," pp. 18, 19 : —

'* 1666, about the beginning of May, the garrison, consisting of about 200 isr^
mutinied for want of their pay, and, choosing corporal Dillon for their comim-rAeT.
seized the town and castle. On the 25th of the same month, the Earl of Am^.
son to the Duke of Ormond, arrived by sea in the Dartmouth frigate, widi focr
companies of Guards, and he assaulting the town by sea, and Sir WiUism Yiry^r-
by land, the mutineers were forced to retreat into the castle, with the k» '-c
Dillon their commander, and two others. The Earl also lost two soldiers, ^^^v.
day the Duke of Ormond arrived from Dublin with the Horse Gotrda, mxid tb^
mutineers surrendered at discretion. The corporation (of Carrickfeigitf) receiT^
thanks from the Government for their loytdty on this occasioOf and §&.*«' \
splendid entertainment to the Earl of Arran."

Falkinbii — The Irish Guards. 21

the Butch War, and two companies, of whicli Lord Arran's was one,
▼ere sent to Chester, and appear to have taken part in the action at
Solehay.*

The military annals of the Restoration still remain very scrappy
and imperfect ; and even the achieyements of the British Ghiards, have
been insufficiently recorded. Little or nothing is known of the career
of the Irish Guards from 1675 to 1685, when, as already mentioned,
the colonelcy passed to the young Lord Ossory on the death of his
uncie Lord Arran, although very full lists of its officers for several
years of this ohecure decade are still extant. The changes in the
ivgiment within this period do not seem to have heen many ; the most
important heing the appointment of Sir Charles Fielding — ^a member of
the ancient family of which the Earl of Denbigh is tihe head — to be
Lieutenant-Colonel on the death, in 1680, of Sir William Flower.
The Guards appear, however, to have been maintained in vigorou.
efficiency. On April 23, 1685, Major Billingsley reported to his
Colonel, that he ** drew out the Kegiment to solemnise the corona-
tion, which was performed after the usual way on state days."*
Loixl Clarendon, who superseded Ormond in the Irish Government in
1685, reported very favourably of their appearance in a letter to
James II. : — *' The other day," he wrote, " I saw your Majesty's
i^giment of Guards drawn out ; and though I am no soldier, yet I
may asaure your Majesty they exercise and perform all their duty
^ vreW as your Guards in England can do. If they had the honour
to be in your presence you would have no cause to be ashamed of
them.">

But the regiment was now about to become involved in those far-
reaching changes which shortly after the accession of James 11. became
po universal in every department of the public service, and were
(Ti" long to lead to such startling results. The King resolved on a
drastic reform of the personnel of the army, and Tyrconnel came
to Ireland to supciintend and carry out the changes which had been
nsolved upon. This is not the occasion on which to discuss the policy
of James the Second's dealing with his Irish forces prior to the events
which obliged him to rely upon their services in his unsuccessful
effort to retain his Crown. It must suffice here to observe, that under
Tyrconnel' 6 direction a sweeping reform was rapidly and even violently

* Sir F. Hamilton's '< History of the Grenadier Gnards," yol. i., p. 163.
' Onnonde MS.
** Clareodon Correspondence,*' i. 231.

22 Proceedings of the Moyal Irish Academy.

carried out. The process may be traced in the cortespondence of Lord
Clarendon, who, though unquestionably loyal to hie SoTereign, wa*
alarmed at the vehemence of the subordinate who was so shortly to
be his successor. Clarendon's letters, written during the period of hi»
Viceroyalty, shed a flood of clear light on events in Ireland in the
years immediately preceding the Revolution. Though of liberal
opinions on the Eoman Catholic question, he was, despite his close
family connexion with King James, far from endorsing every item in
the policy of his royal master, disliking the rapidity and violence
with which changes were introduced into the system of government
he was administering, and particularly resenting the interference of
Tyrconnel, who, as Lieutenant-General of the army in Ireland,
exercised plenary powers independently of the Viceroy. His letters,
descriptive of TyrconnePs proceedings, contain several references to
the Guards. * In letter after letter he represented to James and ta
his ministers his disapproval of proceedings which, apart from their
unfortunate effect in alienating a large section of the Irish population,
he considered injurious to the efficiency of the army in Ireland, and
especially to the Eegiment of Guards.

Pursuant, however, to the commands of the king who, as he told
Clarendon, was "resolved to employ his subjects of the Roman
Catholic religion," and " not to keep one man in his service who ever
served under the usurpers,"' Tyrconnel proceeded to put out of the
regiment such of the officers as were unlikely to lend themselves
to the new order of things, and at the same time to make large
changes in the personnel of the rank and file. The true reasons
for these alterations were not of course publicly avowed, the ostensible
reason being that, in the language of Tyrconnel, " the Scotch battalion,
which is newly come into England, has undone us; the King is so
pleased with it that he will have all his forces in the same posture.
We have here a great many old men, and of different statures : ' they
must be all turned out, for the King would have all his men young
and of one size " ; this, however, was only a pretext, for, according to
Clarendon, the new men were " full as little " as those who were
turned out.

On June 8th the Guards were reviewed in St. Stephen's Green by
Tyrconnel, who owned to Clarendon that ** it was a much better regi-
ment than he could have imagined, and that the men did their exercises

^ Clarendon State Papers, i. 433, $t teq, ' Ibid.f i., p. 431.

» Ibid., i., p. 468.

Falkiner — The Irish Ouards. 23

as well as any regiment in England " ; ^ bnt this did not prevent
Tjroonnel from proceeding -with his reforms. The new olBcers were
commissioned and presented to the regiment on parade. Sir Charles
Fielding, who had served with the regiment from its formation and
risen from ensign to be lieutenant-colonel, was superseded in his
command — the King, as Tyrconnel put it, '' being so well satisfied in
the long services of Sir Charles Fielding that he had removed him to
prefer him to a better post"; ' and Sir William Dorrington, a native of
England and the youngest major in the army, whose subsequent
career evinced considerable military ability, but who was a complete
stranger to his new command, was appointed in his place.* Other old
officers of long standing in the regiment, such as Major Billingsley
and Captain Margetson,^ a son of the Irish Primate, were likewise
suspended. The changes among the officers were followed by the
dismissal of 500 men, at least 350 of whom, according to Clarendon,
were " able and lusty men," and a credit to the regiment. The
hardship of their dismissal was aggravated by the fact that they had
just bought fresh uniforms by direction of their colonel, the young
Thike of Ormond, and were not reimbursed for their expenditure. To
fill the places of these men, Dorrington received orders to recruit in
such counties as he thought fit ; and accordingly despatched Arthur,'
ooe of his captains, to Connaught to raise men for the Guards — a pro-
ceeding much resented by Clarendon, who forbade Dorrington to
proceed in it.

So violent an exercise of authority inevitably excited alarm.
**A11 men,"* wrote Clarendon, "who have any consideration and
care of the King's service are extremely troubled at the method which
is taken of doing things. To turn out, in one day, 400 men of the
Regiment of Ghiaids, 800 of whom have no visible fault, and many of
them cheerfully went the last year first into the north and after-
wards into England, does put apprehensions into men's heads which
they would otherwise have no cause for, and putting in none but
natives in their rooms, who really to the eye, as to stature and

I Clarendon State Papers, i., p. 440.

« /Airf., i., p. 434.

' Ihid.^ ii., p. 45. Theie is no authority for D' Alton's statement, followed by
O'Callagban, that Dorrington was connected with the regiment from its formation.
His name does not appear in any of the early lists of officers.

« I&id,^ L, p. 435.

^ Ibid. » i., p. 678.

• Jhid.^ i., p. 476, July 4, 1686.

24 Proceedings of the Royul Irish Academy.

ability, makes worse figures than those that are put out, confinns
their jealous apprehensions."^ But though the composition of ihc
corps was largely altered, and the principal positions confi^ded to
officers of TyrconnePs way of thinking, there does not appear to
have been any general surrender of commissions by the old officers
who escaped immediate dismissal, and these appear to have remained
in the regiment down to the landing of William III. at Torbay,
when, with their Colonel, Lord Ossory, they embraced the cause of
the Prince of Orange.

From the sweeping changes inaugurated by Tyrconnel, it resulted
that, notwithstanding that the Colonel, Lord Ossory, who, in 1688,
succeeded to the Dukedom of Ormond, and had been left undisturbed
in his nominal command, went over to William III. as soon as he
landed at Torbay, the regiment took part with James II. in his
struggle for the Crown of the Three Kingdoms, though in numbers
considerably short of its proper strength. The colonelcy was then
given to Dorrington, ander whose command the Guards took part
in the siege of Deny, and subsequently were present at the Boyne
and Aughrim. In the latter battle Dorrington was taken prisoner, and
Barker, who had been appointed Lieutenant-Colonel, was killed ; and it
does not appear under what officers the last services of the Irish Guards
on Irish soil were rendered at the defence of Limerick. After the
capitulation of that city the Royal Regiment of Guards was the fore-
most of those which made choice of the cause of King James and exile,
and in that dramatic scene, so powerfully painted for us by Macaulay,
when the garrison of Limerick was ordered to pass in review before the
rival commanders, Gtinkell and Sarsfield, and those who wished to remain
in the Ireland of King William were directed to file off at a particular
spot, all but seven of the Guards, marching fourteen hundred strong,
went beyond the fatal point and embraced the tdtemative of exile.
Not all of these, however, adhered to their resolution, and only five
hundred appear to have been included in the thousands, who, in the
language of the historian, ** departed to leam in foreign camps that
discipline without which natural courage is of small avail, and to
retrieve, on distant fields of battle, the honour which had been lost
by a long series of defeats at home."*

Reference has been made above to the fact that the career of the
Irish Guards was not closed with the defeat of the cause with which

* Clarendon State Papers, i., p. 486, July 6.
' Macaulay* 8 Hittory of England^ chap. x"vii.

Falkiner — The Irish Guards. 26

their last yean in Jreland were identified. After 1690, indeed, they

disappeared from the roll of the regiments in the service of the

British Crown, and it is hardly surprising that William III. made no

attempt to revive a corps which had fought for his opponent. But

though exiled to France for ahove one hundred years, the identity of

the regiment was never completely lost. It still continued to be

recruited abroad fi*om the ** wild geese" who flocked in a continuous

stream from Ireland to the Continent through the course of the

eighteenth century. Under the leadership of Dorrington it served

with distinction at Loudon and Charleroy, and though broken up in

1698, after the Peace of Ryswick, when it ceased to retain its old

title, it was substantially re-embodied under its old chief, and

was known until his death, in 1718, as the Dorrington Regiment.

The regiment continued during this period, by desire of King

James II., to retain the uniform and colours it had worn in the British

Service.* Thenceforward it was distinguished by the names of its

successive Colonels, Counts Michael de Roth and Edward de Roth,

fiobert Dillon, Lord Roscommon, and Count Antoine Walsh de

^>errant, all of them representatives of old Irish families, and aU of

fhem soldiers of capacity. In the Marlborough wars, the regiment

served with the army of Flanders, and was present at Malplaquet

onder Count Michael de Roth ; it served with the Duke of Berwick

in Spain, and during the colonelcy of his son took part in tlie

battles of Dettingen and Fontenoy. Finally under Count Walsh

' See on this point, *< Mistorique du 87* Regiment d* Infanlerie de Ligne^ 1690-
Is9g.'* Par Capiiaine MaUagtUi, PariSf 1892 ^ from which the following extracts
are taken : —

**I1 semble que, d^« cette epoque (1698), lea regiments Orlandais et suisaes
etaient distingues par T habit rouge garance ; tandisque toute Tinfanterie f rancaise
portait rhabit gris-blano/' p. 16.

'* Notes aur Vuniformedu BegimetU de Dillon de 1690 a 1791.**—** Nous n'avona
pu trouver aucun renfleignement sur Tuniforme de Dillon pendant les quarante
premieres annees de son 86 our en France. Le premier ouvrage qui nous alt fumi
me donnee precise est la Carte abr^gee du militaire de la France (de Leman de
1a Jaise) qui, pour les ann6es 1730 et 1733 attribu6 k Dillon : habit rouge et pare-
nonU bleus," p. 75. The " habit rouge-garance " was worn continuously to 1791
>y all the Irish regiments in the French service. The facings yaried in colour,
lid in the case of the Irish Guards were of St. Patrick's blue. A representation
f the uniforms of the French army in 1772 shows the Guards or Roscommon
U'^^iment, as it was then called, to have worn a red coat or tunic with blue
irings, buff breeches, white Hessian boots, and a plumed helmet.

Tbe colours of the Regiment at this time showed a white cross on a ground of
t. Patrick's blue.

26 Proceedings of the Royal Irish Academy.

de Serrant the regiment maintained its old traditions down to the
Revolution, when it merged in the 92nd Regiment of the Army
of France. But its officers were still, for the most part, Irishmen,
and on the fall of the Bourbons, it was natural that the representatives
of a traditional loyalty to hereditary right should prefer the Fleur-de-
lys to the Tricolor. The successors of those who had refused to
concur in the English Revolution were too proud of their consistent
loyalty to be content to accept the French one. Almost without
exception its officers followed their Colonel, Count "Walsh, in his refusal
to serve under the banner of the Republic, and were among those who,
in 1794, accepted with alacrity the invitation conveyed to the Colonels
of the three surviving regiments of Dillon, Berwick, and "Walsh by
the Duke of Portland, to take service under the British Crown under
the title of the Irish Brigade.' It was intended that the regiment
should be placed upon the Irish Establishment, and be recruited
exclusively in Ireland for service abroad ; and its officers came over to
raise a fresh corps in Ireland. But the times were out of joint for
such an enterprise. The emigrant officers found Ireland in a turmoil
of agitation, which had much more in common with the France of the
Revolution than with that of the aneien r^yime, and their efforts were
almost entirely unsuccessful. The Rebellion of 1798 quickly follow-
ing, put a final end to whatever hopes might have previously been
entertained, by filling the English Government with misgivings as to
the use to which an Irish Catholic Brigade might possibly be turned
in spite of the unquestioned loyalty of its leaders. Recruits bein^
forthcoming in quite insufficient numbers, it was found necessary to
amalgamate the regiments forming the Brigade, with the result that no
place remained for many of the returned offi cers. "Weak and insufficient
in numbers the corps was sent to North America and the "West Indies,
but it was found impossible to maintain the Brigade as an independent
organization, and within a few years it had ceased to exist.

This last chapter in the history of the regiment is a sad one. Making
every allowance for the exacerbation of feeling at the time, the treat-
ment accorded to the returned officers was little creditable to Irishmen
of any shade of opinion ; whilst the conduct of the War Office in regard
to their pay and allowances was equally deserving of disapproval.
Wolfe Tone, in his Journal for 1796, describes how the officers
intending to go to Mass on Christmas Day in full uniform were
obliged to give up the idea for fear of being hustled by the populace

* See note added in the Press.

Fajlkinbr— 2%« Insh Guards. 27

of BuUin. On the other hand, the Duke of Fitz James, the descen-
dent of the great soldier Berwick, and the principal personage among
those to whom the invitation to join the British army had been
addressed, was insnlted by some observations from Lord Blaney in the
Irish House of Lords, and fought a duel with that nobleman in the
Phoenix Park in assertion of the honour of his eofifr^res,^ The un-
employed officers were treated with so little consideration by the
military authorities that some of them were reduced to a half -starving
condition, and had to wait several years for arrears of pay : while the
Colonels on the final disbandment of the Brigade were refused the
rank as half -pay officers for which they had stipulated when entering
the British service. Thus the final chapter in a story that had
extended over a space of above one hundred and thirty years was one
of misfortune and even humiliation. But none the less the record of
the Irish Guards, from their formation in 1662 to the final dispersal of
the last remnant of the regiment, is one in every respect creditable to
the martial traditions of Ireland; and rooted in the history of ita
country, whether as Jacobite or "Williamite, as loyalist or rebel, as
fighting for or against the Crown to which it owed its origin, its
career is one in which were exhibited at every stage the stainless
honour of Irish gentlemen, and the indomitable valour of the Irish
race.

NOTE ADDED IN THE PRESS.

Mr. Lecky, in his " BUtory of England in the Eighteenth Cen-
tury^^^ vol. vii., p. 254, gives some account of that final chapter in the
history of the Irish Brigade, to which O'Callaghan in his otherwise
I'xhaufltiTe narrative gives but scant attention. Reference is also made
to the episode in Mrs. M . A. O'Connell^s Last Colonel of the Irish Brigade,
Bat much the fullest authority for the later history of the Irish Guards
s to be found in a volume entitled: '* Une Famille Boyaliste, Irlandaise
i I^aneaiss^ et Le Prince Charlea-Edouard" privately printed at Nantes
Q 1901 by the Duo de la Tr6moille. In this work several documents
I'lating to the regiment under the Colonelcy of Antoine Count Walsh
le Serrant are reproduced. From it are extracted the documents f ollow-
Dg, vii:. : the letter of the Duke of Portland above referred to, and the

> Annual fiegiater, 1797.

28 Proceedings of the Royal Irish Academy,

Commission of George III. to the Comte de Serrant as a Colonel of
Infantry in the Irish Brigade : —

Lbtteb of the Duke of Pobtland to Count Walsh db Sbbsakt.

A Whitehall, cb 30 Sept. 1794.
M0N8IBUR,

Le Roi desirant remplir les intentions de la legifilature d'lilande, et de
douner k ses sujets catholiques de ce royaume, an prompte t^moinage de son. affectioa
et de sa confiance, s'est d^terminS k retablir le corps connfL cy deyant aout le non
de la brigade Irlandaise, et comme tous etiez colonel d'un des regiments doot elli*
etoit compos^ei Sa Majeste m'a donne Tordre de yous offiir dans ce nouTeaa coips
le mdme rang de colonel que tous teniez dans I'ancien.

L*intention de Sa Majeste est, que cette brigade soit maintenant compoeee dr
quatre regiments, le commandement de trois desquels, elle m'a ordonn^ d'offirir
aux colonels (ou k leurs represectans) qui out command^ lea trois corps qui cod*
posoient la brigade, lorsqu'elle 6toit au service de sa Majeste tr^s chretienne, et eclsi
du quatridme k Monsieur O^Connell, cy devant officier g^n^ral au aerrice dt
France, et certainement bien connu de yous et de tous les gentilahommea irlasdoi'
qui out servi dans ce corps.

II a aussi pier k Sa Majesty de determiner que tous les ofilciers, tant de T^tai-
major que les autres, except^ yous, Monsieur le comte et Monaieur le doc de
Fitz James, seront pris d'entre ceux de ses sujets qui sont n6s en Iilande, et qui
se seront distingu^s par leurs seryices, dans les memos grades dans la brigade, et
que si Ton manque d'officiers (comme il y a toute apparance) pour remplir \e<
grades inferieurs, on les choisisse dans les families des gentilshommes de la meB«
religion dout la demeure k toujours ^ en Irl&nde.

L^ intention de Sa Majesty est de plus, que cett^ brigade soit mise, dn moneiit
qu'ellc sera complette, sur I'^tat militaire de ce royaume, on de celui d'liUnde, ec
sorte que, d^s ce moment Ik, les officiers qui y tiendront des places, prendront xa?tg
avec les autr6s officiers des arm6es de Sa Majeste, et en cas que le ooips s>.':'
reforme, ils auront droit a la demi^re i)aye.

Sa Majesty recevra aussi la recommendation des colonels dans le cbotx d^^
officiers, et cela surtout, quand ces recommendations seront faites en faTeur de
ceux qui ont servi cy devant dans la brigade irlandoise. Mais elle ne pemrtsra
pas, qu'aucune consideration pecuni^re soit donnee pour obtenir aacua ras^
dans ce corps ; et en consequence, comme il n'aura 6te permis ik ancun offieiv «ie
quelque rang qu'il soit, de rien payer pour sa place, il doit comprendre claipsinent,
que sous aucun pretexte il ne lui sera permis de la vendre.

Sa Majeste m'a command^ aussi de vous informer qu'elle est determiaff a <-«
que ce corps soit sp6cialement affecte au aervicee des colonies de Sa Maj«^ daz>«
Itts Antilles, ou dans telle autre possession de Sa Majesty, hors de ce* dcus.
royaumes do la gnmde Bretagne et de Irlande, qu*il lui plaint de lea employer ; e-
que Sa Majeste d'attendra lii ce que tout officier de quelque rang qu'il soit, qui *
riionneur d'avoir un brevet dans ces corps, de tiendra comme indispensaUe^w^t
oblige de venir avec son regiment dans quelque partie de monde que se soit

Falkivbk— The Irish Guards. 2&

SaoB entiez dana de plus grand details sur se sujet, j'ajonterai seulement, k
I'oecaBion de yotre quality de colonel proprietaire d'un des regiiments de l*aucienn&
brigade irlandaiae, qu'il est tr^s essentiel que je tous rappelle, Monsieur le Comte,
qae la constitution de ce pajs-ci n'admet n'aucune propriety semblable, attendti
comme yous deyez yous le rappeler, que les fonds pour r^tablissement militaire ne
eontaccorde que pour I'ann^e, et que par consequent il ne peut avoir qu'une
existence annuelle.

Capendant, quoique place ne tous soit conflee par la legislature que pour un an,
ou doit en oonsid^rer la possession comme tous ^tant assur^e, durant yotre bonno
condoite, terme que je ne juis regarder de moindre dur^e que celui de votre yie.

Je yous ai maintenant expose toutes les circumstances qui m*ont paru n^cessaires
pour yous aider k d^rminer si yous devez accepter les offres gracieuses de Sa
Majeste ; je n*ai qu'ajouter, que, si apr^s mClre consideration, il yous parait plus
conyenable de ne pas yous en preyaloir, la bonte naturelle de Sa Majeste la
disposera k interpreter les motifs qui yous auront determine, de la mani^re la
plus fayorable pour yous ; et je juis mdme youx assurer, que dans le cas mdme oCi
reus accepteriez la proposition que je suis charge de yous faire, et que la guerre
finie, ou m^me pendant sa dur6e, yous ayez Tayis, de quitter le service de Sa
Majesty, et de reatrer k celui de Sa Majesty tr^s Chretienne, que vous trouyerez le
Roi dispose de m^e de yous accorder yotre congi, et de consid^rer cette mesure
srec sa bont^ accouturm^e.

Je ne scauroiB douter, que yous n'ayez la bont^ d*informer les officiers de la
brigade, qui out eu Thonneur de seryir sous yos ordres, des intentions du Hoi, k
leur egaid, selon la forme et les conditions que je yous ai specific cy-dessus ; et
que je yous yondrez bien aussi leur recommander, le peutdt possible, k quelque
sodrait oonvenable d'oil il pourront le plus commod^ment se rendre en Irlande, et
le mettre en ^tat de remplir les devoirs qui leur seront consignes de la part da Roi.
Je n'ai pas besoin de yous dire, que dans le cas, oCi vous yous decideriez k
accepter la proposition que Sa Majesty m'a autoris^ k yous faire, il n'y aura pas un
momi>iit k perdre pour yous rendre ici, & in de rdgler tout ce qui k rapport k la levde
des corps, le plus promptement possible.

II ne me reste qu^k yous prier assure, que je m'estime tr^s beureux d'ayoir ^t^
ttutoris^ k yous donner ce t^moignage, non Equivoque, de la bonne opinion et
I'estime de Sa Majesty.

J'ai rhonneur d'etre, Monsieur le Comte, yotre tr^s humble et tres obeissant
seryiteur.

Portland.

Palaia de st. James. 1", 1794. Brevet de colonel d'infanteiie (dans la brigade
irUmdaiae) pour Antoine Walsh, Comte de Serrant, au nom du Hoi Georges III.
sous la aignature de lord Portland.

George the Third, by the Grace of God, King of Great Britain, France, and
Ireland, Defender of the Faith, etc., to our trusty and well beloved Antony, Count
Walsh de Serrant, greeting: We reposing especial trust and confidence in your
loyaltj, coinage, and good conduct, do by their presents constitute and appoint
yoa to be Colonel of a Regiment of Foot, forming part of the corps known by the
name of the Irish Brigade, and likewise to be a Captain of a company in our said
regimenU You are therefore to take our said regiment as Colonel, and the said

30 Proceedings of the Royal Irish Academy.

•company as Captain into your care and charge, and duly to exereue w wD tbe
ofBcen aa soldiers thereof in arms, and to use your heat endeayoun to keep tbm
in good order and discipline ; and we do herehy command them to obey foo ti
their Colonel and Captain respectiyely ; and you are to oheerre and fcdlov tad
orders and directions from time to time as you shall reoeiTe from us, or any f&»
your superior officers, according to the rules and discipline of war, in punaaooe of
the trust we hereby repose in you.

Given at our Court at St. James's, the first day of October, 1794, in thediiitj-
fourth year of our reign,

By His Majesty's command,

PoKTUiKV.

Anthony Count Walsh bs Skkrant,
Colonel of a Regiment of Foot.

[The authorities on which this Paper is based are for the most pait
indicated in the foot-notes, or in the body of the text. The irriter
has also derived assistance from articles on the subject in the
Nineteenth Century^ for June, 1900, and in the Household Briftdf
Mayazine, for the same year, contributed respectively by the hXt
Fitzalan Manners, and by Lt.-Col. R. Holden, Secretary of th*'
United Service Institute. In addition the writer desires to express
his obligations to Major-General Sir Martin Dillon, K.C.B., aad tc
Mr. V. Hiissey Walsh, for much valuable information ; to Mr. F
Elrington Ball, M.R.I.A., for transcripts of documents in the Cart<
Papers at tbe Bodleian Library; to Dr. W. J. O'Donnavan, M.R.L1..
for references to general useful authorities ; and to the offidals of tht
Irish Eccord Office who have assisted his searches with their nsnal
courtesy and helpfulness. — C.L.F.]

1 " Une Famille Royaliste," Appendix, p. 95.

[ 31 ]

ni.

SCAEABS IN THE DUBLIN MUSEUM.

By miss M. a. MUREAY, F.S.A. Soot.

[comkitkicatei) bt-col. o. t. pluirksit, c.b.]

[Read April Uth, 1902.]

The Dublin Museum contains, among many other interesting Egyptian

antiquities, a fairly representative collection of " scarabs," those little

beetles made of stone or faience, which were held in high estimation

by the ancient dwellers on the Nile. The living scarabeeus beetle was

the symbol of the god Ehepra, the Creator, and was also emblematic

of the Resurrection ; its effigy is therefore appropriately deposited in

the tomb as the symbol of life hereafter and as placing the dead body

under the direct protection of its Maker. This, however, accounts

only for the scarabs found with the dead, and gives no clue to their

nse among the living. All scarabs, whether for the living or for the

<lead (with the exception of the so-called heart-scarabs which had a

special purpose), are pierced as if for threading, or for setting on a

swivel as the bezil of a ring, and are plainly intended for a more

definite use than mere ornament.

The underside of the scarab is flat, and this little oval space is
inscribed, the interest and value of the scarab depending entirely upon
the inscription. The reason for this use of the scarab has never been
explained^ nor, as I said before, has the real use of scarabs themselves
ever been satisfactorily demoDstrated. The generally-accepted theory
is that some were seals and some were charms, and this though not
dtogctlier satisfactory, serves as a convenient foundation for classifi-
L^tion.

The meaning of the signs in the inscriptions is one of the chief
lifficulties in the study of scarabs. Take, for instance, the very com-
Qon hieroglyph Neb, Lord, which appears continually on scarabs. It
s impossible to say whether it is inserted merely as being of a con-
enient shape to fill the curved ends of the oval, or as a semi-sacred
r<>rd, and therefore appropriate on a protective amulet. The latter
i'uaoa iroiild account for the constant use of other semi-sacred signs,

32 Proceedings of the Royal Irish Academy.

such as Nub, Gold, which convey no special meaning to us, andwliich
by their shape cannot hare been inserted as purely deooiatiye designs
to fill an otherwise empty space. Unfortunately the amulitic scanbs
have never been studied thoroughly and scientifically, and our know-
ledge on that subject is still very limited.
Scarabs may be divided into nine classes :

I. Kings' names.
II. Names of private pei'sons.
lit. Records of events,
rv. Titles, royal or priestly.

V. Names of gods.
VI. Sacred signs,
vn. Charms expressed in words,
vm. Sacred animals.
IX. Decorative designs.

I. Of royal names there are fourteen in this collection. Tbc
earliest is Nub-hetep (No. 1) of the XVIth dynasty. The dark-brown
of this scarab was not its original colour. It was once green, bat tt^
greens and blues of copper, with which scarabs are glazed, are fngitiTtr
under certain conditions, and the green changes to brown while tbe
blue fades to white. It is very tempting to place No. 52 in the Vth
dynasty, as the scarab of King An.* The name An is written with
a fish, but, as Professor Petrie pointed out to me, in this case the lotn^-
design is distinctly of the XlXth dynasty (compare the lotus in the
scarab of Eameses II., No. 11). This scarab, therefore, falls onder
class vm., and must be considered there. The scarabs of Menkhepef-
Ra (Thothmes III. of the XVIIIth dynasty) are the most numerous
of all royal names. There are several varieties in this collection. No. •
has the king's cartouche upheld by two kneeling figures, emblenutii
of the Upper and Lower Nile, symbolising the king's sovereignty ovt r
the Two Lands, i.e. North and South Egypt. No. 3 has the royal namtf
flanked on each side by a degenerate form of the crown of Lower
Egypt repeated four times. The crown of Lower Egypt, the Bed
Crown, appears to have had some peculiarly symbolic meaning, 9S it
is constantly found on scarabs. No. 4 shows the king as a sptkx*
beneath whom is the prostrate figure of an enemy. No. 5, a Teir

^ rWr " Petrie's Historical Scaral s."

Murray — Searabi in the Dublin Museum. 33

irom scarab of this king, with a rude representation on each side of
the crown of Lower Egypt. No. 6 is a square plaque engraved on
both sides ; obverse, the royal cartouche, flanked by serpents, wearing
respectively the crowns of Upper and Lower Egypt, symbolising, as in
the case of the two Niles, the sovereignty of the khig over the two
parts of Egypt. On the reverse, is the king as the sphinx, wearing the
double crown; behind him the serpent, emblem of power, and the
winged disk, emblem of protection ; beneath is the sign Neb, Lord.
No. 7 has merely the king's title, followed by the epithet ** Chosen of
Ra."

No. 8 is the throne-name of Thothmes lY., Menkheperu-Ra, finished
with the Neb sign below.

No. 9 is doubtful, though it may possibly be Neb-maat-Ba, the
throne-name of Amenhetep III. The throne-name was assumed by the
king when he actually succeeded to the crown. It is always com-
pounded with the name Ra, showing the king's descent from the sun-
god Ba. The throne-name is the one generally used on scarabs, though
the personal name is occasionally found.

No. 10 belongs to a very curious class of scarab which, as Professor
Petrie has shown, contain the names of two kings. In some scarabs —
unfortunately this collection has no specimen of the kind — ^the hiero-
glyphs are so arranged that one sign will do duty in both names. In
this scarab the names are Thothmes III. of the XYIIIth dynasty, and
Sety I. of the XlXth dynasty, two kings separated by a space of more
than a hundred years. Obverse, the throne-name of Sety I., Men-
xnaat-Ba, associated with the crown of Lower Egypt, a couchant lion,
imd the Neb sign. Eeverse, the throne-name of Thothmes III.,
Men-khex>er-Ea, the crown of Lower Egypt, the hieroglyphic titles of
the king of Upper and Lower Egypt, the sign Nefer (good luck or
happineea), and the Neb-sign. It is interesting to observe that in this
specimen the bee, the hieroglyph for the king of Lower Egypt, is much
larger than the hieroglyph for the parallel title of the king of Upper
Egypt, This, taken in conjunction with the constant occurrence of
the crown of Lower Egypt, would seem to show that the title has a
/ipeciallj symbolic significance. Another explanation is that these
scarabs were made in Lower Egypt. In all other places, except on
Hcarabsy the dominion of Upper Egypt takes precedence over Lower
Egypt, BO much so that it it is a generally received opinion that the
king of Upper Egypt conquered Lower Egypt and added the title to
the one be already possessed.

No. 1 1 is the throne-name of Bameses II.,^User-Maat-Ra, of the

S^I.A. 1PB0C., TOL. XXIT., 8SC. 0.] [3]

34 Proceedings of the Royal Irish Academy.

XlXth dynasty, surmounted by a design of lotus flowers and buds.
Next to Thothmes III., this king's scarabs are the most common.

No. 12, Ba-en-Ra, Merenptah, son and successor of Bameseg II.
This king is usually supposed to be the Pharaoh of the Exodus. "So. IS
is of an obscure king, Se-Amen, of whom little is known but theosme.

No. 14, the last of the royal scarabs in this collection, is that of
Shishak II., Kheper-sekhem-Ea, abbreviated to Kheper-Ra, and sur-
rounded by a conventional cord border. This^lShishak was the
descendant and successor of Shishak I. who defeated Rehoboam and
spoiled the temple of its golden shields. Nos. 15 and 16 are doubtful.
They may be royal, but I think that they belong to the next class.

II. Besides these there are only two private-name scarabs in this
collection.

No. 17, Mentu-sa.

No. 18, son of the sun, ? Nefer-Maat.

III. The extremely interesting series of scarabs, apparently struck
like medals to commemorate some great event, are unfortunately quite
unrepresented here. In Professor Petrie's collection there are several
specimens of Thothmes III. They are all of the same type — the king's
name in a cartouche, followed by the record of the event, e.y. bora in
Thebes, crowned in Thebes, and so on. The great scarabs of Amen-
hetep III., recording his hunting and lake-making exploits as well is
his marriage, are too well known to need description.

IV. This class of scarab presents many difficulties, and it is almost
impossible to say anything about them. They may have been sealf of
office, but some are probably amuHtic.

No. 19, of the Vlth dynasty, shows the bee of Lower Egypt
and a quadruped of uncertain character.

No. 20. The royal title Sa Ra (Son of the Sun), the unoccnpieil
space being filled with a lotus flower.

No. 21. Hieroglyphs giving the ordinary title of the sovBretiin,
King of Upper and Lower Egypt.

No. 22. The King as Lord of the Two Lands.

No. 23. The Living Horus, Lord of the Two Lands.

No. 24. The servant of Ra. This is read backwards.

y. Scarabs bearing the names of gods are apparently meK
charms, the wearer being placed in this manner under the epeoA
protection of the god.

Nos. 25-27. Amen-Ra.

1:1

DfclB. 1 ''^- 2 W 3 w 4

Go. S

Bn. 7

Bn. 6

W. 8

9 Bk. ID

w n

li"^

BI. 12

J3 Qjr- 1* Swrpwtme.te Bl 16

Bn. 17

w 18

SbuHk. IS

W 20 ^ a ft,. 22

W. 23 N^^ Bn. 26

W 24

W. »

Ga. 27

W. 29 ^ ^ Gn. a C3|.. 32

Bl. = Bine. Dk. bl. = Dark blue. Bn. = Brown. 6n. = Green. Gy. = Grey.

W - White.

36 Proceedings qf the Royal Irish Academy.

I am inclined to place in this class No. 28 witli the doubtful read-
ing **Amen-Ra, king of the North and South, Lord of the Two
Lands."

YI. AND YII. Sacred signs, and charms expressed in words are so
dosely connected that it is impossible to separate them with any
accuracy. They form the largest class of scarabs and are undoubtedly
amulets against evil. So little have scarabs been studied — ^Professor
Petrie's Historical Scarabs is the only book giving anything like a
classification of the subject, and he unfortunately confines himself
entirely to Royal Scarabs — ^that amulitic scarabs are still an unsolved
mystery.

No. 29. Khonsu as protection. Ehonsu appears from a popular
story to have been the chief protector against evil spirits.

Scarabs engraved with the Boat of Ea form a large division of the
amulitic class. They are placed by Professor Petrie in the XXIInd
dynasty.

No. 30. Worthy before the Boat of Ra.

No. 31. The Boat of Ra, [therefore] fear not.

Nos. 32 and 33. The legends on these are not decipherable.

No. 34. Gladdening [literally. Enlarging] of the heart, establish-
ing goodness, giving life.

No. 35. Life and Happiness. Or perhaps " Life and Luck " is a
better rendering.

Nos. 36-47 are untranslateable. No one has yet ventured to
suggest how the perfectly legible hieroglyphs of amulitic scarabs
should be read.

Nos. 48 and 49, though also untranslateable, show the worship of
the Sun-god Ra imder the form of an obelisk.

No. 60 gives the Crown of Lower Egypt, and two untranslateable
signs.

YIII. Sacred animals and figures are placed on scarabs with some
idea of protection.

The lion (No. 51), the fish (52), the lizard (53), and the croco-
diles (54 and 55), are animals who were supposed to possess powers
of enchantment. The double crocodiles figure largely in those
curious magical objects called Cippi of Horus, where the youthful
god is represented standing on two crocodiles.

Nos. 56 and 57 are figures of deities, associated with the ostrich
feather, the emblem of Truth and Righteousness.

1:1

w. »

Gn 35 W. 35 W 37

Schui. 34

V 38

39 -W. 40 Bb. « W. 42 ^ *•

(a. 44 W. 45

W 50

W. 46

Stone. 47

W. 48 W. 48

^^ W 52 Dk.M.» <» W

Stale. 63. W 66

Bl. = Blue. Bk. = Black. Bn » Brown. Dk. bn. = Dark Brown. Gn. = Green.
Gjr. = Grey. W. = White.

38 Proceedings of the Royal Irish Acadetny.

No8. 58 and 59. Hade representationB of the ape holding the sigD
Nefer, Luck.

No. 60. Two nondescript animals and a sphinx. This may possibly
be a charm invoking the king, as the sphinx bean the royal snake od
the forehead.

Very curious specimens of the Sacred-animal daas are the Yulture,
Beetle, and Snake scarabs, which are found in every variety of com-
bination, Nob. 61-69. Undoubtedly some special significance was
attached to the union of these three creatures together. The Beetle
is the emblem of Creation, the Vulture of Protection, and the Snake
of Power or of Death. The winged snake (61) is also symbolical
of Protection. Often the sacred animals are combined with sacred
signs, as in 64, where the Nefer and Neb signs appear ; 66, two sipg
of Life; 63, two Neb and Ankh (Life) signs; 69, the sign Hes,
Praise.

No. 68. Apparently a purely conventional design, but it still sho^
its origin, namely, four snakes and four crowns of Lower Egypt
symmetrically arranged.

1: 1

IX. The purely decorative designs are very commonly found, and
vary in beauty according to the period to which they belong. The
spiral and the lotus are the most usual forms of decoration.

No. 70. Spiral design, combined with the hieroglyphs Uaz and Ks.

No. 71. Concentric circles.

Of lotus designs there are only two in this collection, Nob. 1 1 and
52, in neither case appearing separately, but in combination.

[ 39 ]

IV.

NOTES ON AN UNPUBLISHED MS. INQUISITION (a.d.
1258), RELATING TO THE DUBLIN CITY WATER-
COURSE. FROM THE MUNIMENTS OF THE EARL
OF MEATH, By HENRY F. BERRY, m.a.

[Read Fbbbuaby 24, 1901.]

Lr the year 1244, Maurice FitzGerald, then Chief Justiciary of
Ireland, issued a writ directing an inquiry as to the hest and most
suitable place from which water might be diverted from its course, and
conveyed to the city of Dublin. The citizens, who appear to have badly
needed an additional supply, were prepared to pay the costs of the
necessary works, and special enquiries were to be made as to loss and
injury to property consequent on the formation of a watercourse, which
must necessarily run through the lands of divers persons. The under-
taking was duly carried out, and the ancient city watercourse, as we
Ftill know it, from its ** head " beyond Templeogue, where the river
Dodder is diverted, was constructed in pursuance of the Justiciary's
writ

Prior to this period, low lying portions of the city and suburbs
depended on the waters of the Poddle, which, flowing from Tymon
and the green hills of Tallaght, through Harold's Cross, lazily
meandered through the Liberties into the river Liffey. The more
ancient portion of the city, built on high ground, was supplied by
wells, and that the Castle itself had no other resort, is proved by an
entry in a Pipe RoU, 12 and 13 Henry III. In Easter Term, 1228-9,
the Sheriff of the Yale of Dublin made a payment of 2«. for a bucket
for the well of Dublin Castle.

It is certain that the authorities and the residents within the
precincts of the Castle were anxious to acquire a more abundant
supplj of water, as in the year 1245 (a year subsequent to the issue
of the above mandate), the King directed John FitzGefiErey, then

40 Proceedings of the Royal Irish Academy.

Justioiary, to have his Hall in Dublin finished, and water conveyed
thereto through a pipe from the city conduit, the work to be completed
by the ensuing summer.

In the course of the year 1254, water from the Dodder was flowing
into the conduit in High-street, which stood near the great gate of
the priory of the Holy Trinity, and the Liber Alhua of the corporation
of Dublin contains copies of water grants made in that year to
certain private citizens, and to some of the great ecclesiastical
foundations, (among them) to the said priory, and to the church of the
Holy Saviour, near the bridge of Dublin.

It is my privilege to bring before the Academy a hitherto un-
published,' and (I venture to think) xmknown document, which makes
us acquainted with some of the terms of an agreement, in connexion
with this water supply, made between the city authorities and the
Abbey of St. Thomas the Martyr, which stood in the western suburbs.
There are no contemporary documents known to be in existence
relating to these transactions, other than what have been above
indicated, so that this additional evidence, only four years later in
date than the period when the Dodder water was directly supplied to
the city, is of peculiar interest and importance. The next document
in point of date with which I am acquainted is some sixty jeeis
later.

Among the muniments preserved at Kilruddery is a pardmient
roll, containing the earliest grants {circ, 1177) connected with the
foundation of St. Thomas' Abbey by King Henry II. Attached to the
roll is an inquisition of 1258, taken in reference to the above agree-
ment, and this document Lord Meath has most kindly permitted me
to transcribe, with a view to submitting it to the Academy. These
archives of Thomas Court have been handed down in the Brabazon
family from the time of Sir William Brabazon, grantee in 1545 of it*
possessions.

The inquisition is as follows : —

Inquisitio facia a die Pasch^ in tres septimanas anno Te^ni dMsttni
'Regis Renrici xlhi^. coram domino galfrido de Forestel, tunc locum
justiciarii Hib^mtVc tenenU et aliis domim EegM et dojntni Bdwardi
fidelibwa p^r hreve domini Eegi« et dowini Edwardi de tr«nsgr«8ianil>««
factis domino abboti et conventui sancti Thom^ mor^nris jux^ Dublin

I It is not to be found in the Register of St. TkomM\ edited from the ongmal in
the Bodleian, by Sir John Gilbert. There \a no oopy in the RegiBter of the abWj*
preserved in the Library of this Academy, which I have examined.

BvRKY— Notes on an Unpublished i/S. Inquisition, 8fc. 41

p^ major^ffi et civea Dublin p^* stibscriptos, Haket filtum Eob^i^

Augnstmum, filtum Rogtfri, Will^/mum Pilets, Micho^Zem de Angulo,

Milonem Chever, Willtf/mmn Fichet, ^enrieum Galuy, ^cardum

Levayt, Alexandrum pistorem, Bobertum Dispensatorem, Thomam de

Athgo, Adam de Weston, £1 : Juvenem, Rog^mm Suineter, Eob^um

Tracy, Johannem le Peer, Milon^m le porter, WUlalmmn. MatH,

Joh0»nem de Stachkony, Yhilippum Macy, Andream Tyrell, et

Johannem, filfmn Bartho/bffM», Ghdiridum de Dondrom, Thomam

pr<»po«itum, Will^^iim f orestarium, Jordanem le Taylour, et Michaslem

de Stachkonny. Qui jwcati dicnnt sup^ oacramentum suum quod sic

conrenit pHuB int^r abbot^m sanc^ Thorns mor^tris jnz^a Dnblin et

majoretn et communitatem ejusd^m civitatis -per maDdatum domtni

Mauiiiii Slii Galfridi quod tantnm cap^ent de aqna sua currente de

Bother qtMmtum ciureret per medium mole cujiMdam rote plaustri et

nan plus sine assensu conventtM pro qtiinqt^ marc«« fine tacto de

qnibtM tres nuircas solverunt et pro una mor^a annul reddittM,

quemqiddem redditum nunqiiom recep^runt. Dicunt eciam qwod

pr^ic/» cives facere debent mumm lapideum supra aquam de Boder

ad custom suum proprium ctrca cap«d dtcte aque assumpte et nondum

fecerunt. Et pr^^ea dtc^MS abbas dtc^um murum sustinere deberet

pro pr^tcte marca annul redditiM pro hac aut^m c(mventtone omnes

c<^ntentibnes Inter dfc^tim ab&ot^m et elves dtc^ civitatis deb^ent

se-darl et padficart. Et dicunt quod jam ducterunt dtc^am aquam ad

daplum vel amplius et hoc ad dampnum moleDdinortim dtc^orum

ab^»> et conTenttM et molendinorttm do^nmi 'Regis qualibet septimana

ad multtiram unius molendini per unum diem unde estimatum dampnum

df'ctorum ab^att> et conventtM ad zu librae et dampnum domtni B^gi^

sex maicas. Et dicunt qiM>d malor etpr^tcti cives vendiderunt aquam

pr^dtc^am priori et conventui sancto TrinitatM Dublm, domum sancti

Joha^nls, sancd Salvatoris et sanc^i Francisci, set summam yendittonis

p^cunie nesciunt n^c recompensattbnem dampni inde proyenientt^.

lUm dicnitt qf^d pr^icti maior et elves injuriantur eisdon sup^r

lib^rtatibtM suis ledendis de captibne vadiorwm hominum suoru;/! pro

Alewrth, quod facere non debent. £t dicunt quod quedam insula

de Donour est de baronta dtcti abbotis et non portinet ad lib^atem

pr^f'c^e civitatM ubl vadia eorum sepe capta fuerunt contra lib^rtatem

pr^ic/ort<0t abbotMetconventtMp^prtfdic/os cives; dicunt eciam quod

ricus IntdT eccfenam soncto Katorine et forum equorum est de libera

.'lemosina portinens ad abbathiam sancti Thom^. Et ad istius

nq^toBitionis certificattbnem omnes jura^M suprosoriptl present!

nqujBtit^nt aigilla sua apposuerunt.

42 Proceedings of the Royal Irish Academy.

From the foregoing, it is clear that the citizens of Dublin and St.
Thomas' Abbey had had contests over the water supply derived from
the Foddle, prior to the construction of the watercourse from the
Dodder, and though, with the sanction of the Justiciary, a solemn
agreement had been entered into between the parties, for the
purpose of meeting such differences and difficulties as might arise
under the new arrangements, the citizens appear to have riokted
their part of the compact, and infringed on the undoubted ri^te
and privileges of Thomas Court. It is matter of history that,
until the suppression of the Abbey, this and other subjects of
controversy were frequent sources of litigation between the two
bodies.

In 1258, King Henry the Third and his son Edward, as Lord of
Ireland, issued a writ at suit of the Abbey, under which the forcing
inquisition was taken, and the findings of the inquisition may be brieiy
summed up as follows : —

The citizens were entitled to take from that portion of the Dodder
water appropriated to the Abbey a fixed supply, but in reality they
were drawing off double the stipulated quantity and even more.
They were entitled to take what would run **per medium ntdt
cujusdam rote plaustri," but as the word written " mole " (which
might possibly be read mete, and which may originally have had
a mark of contraction over it) is very indistinct, a perfectly satis-
factory conclusion as to the precise meaning of the phrase cannot be
arrived at.

At this early period, a more primitive mode of partitioning the
water than that afterwards constructed at the Tongue, may have been
used, and the clause in the inquisition may well have reference to this
point in the course. In the absence of the agreement, however, and
of a more specific description of the locality and surroundings of the
spot where the contrivance for limiting the supply was fixed^ thf
precise meaning of the expression must be matter of coniecture;
but it seems plain that a cartwheel of a circumference agreed
on (implied by the word, cujmdam) was to be the standard of
an outiet for regulating the quantity of water to be dravn
away.

A fine of five marks was to be paid for this accommodation, of
which three had been discharged, and in addition, a yearly rent of one
mark was fixed on, which, up to the date of the inquisition, had not
been paid. The jurors assessed the damage sustained by St. Thomas'
mUls and those of the King, consequent on the excessft^ with-

Bekky— iVb/es on an Unpublished MS. Inquisition ^ 8fc. 43

drawal of water, at £12 and six marks respectively, calculatmg at the
rate of the multure^ of one mill a day each week.

It may be well to explain here that the Dodder water, when
direrted, was conveyed in an open course or channel to the Tongue
(near Mount Argus), where by means of a stone pier, ending in an
acute angle, the water was partitioned, two thirds being conveyed to
the Liberty of Thomas Court and Donore, which supplied the mills
and tenantry of St. Thomas' Abbey. One third was brought, via
Dolphin's Bam to a large reservoir, which stood to the west of the
Abbey gate. From this cistern, as it was called, the water was
farther led to the conduit in High-street, whence it was conducted by
means of leaden pipes to the citizen's houses.

The jurors further found against the citizens on another count in
the Abbey's indictment, namely, that they had failed to construct
round the *'head" (as it was termed) a stone wall, which when built
(in consideration of the yearly rent before mentioned), the Abbey
▼as under terms to keep up. This ** head " was a dam or rampart of
stone, strong enough to resist floods, which was erected at a place
called Balrothery in the townland of Tallaght. When this was
damaged by very serious floods, the mayor and bailiffs were bound
to collect a number of the citizens and of those who had mills along
the water, with a view to its speedy repair.

Another of the findings was to the effect that water had been sold

by the city authorities to the following ecclesiastical foundations —

namely, the priory of the Holy Trinity, the House of St. John,

i>t. Saviour's and St. Francis*. In the Liber Albusoi the corporation

of Dublin are found entries of grants of water in 1254 to Holy Trinity

and to the church of the Holy Saviour near the bridge of Dublin.

St. John's was a poor house or hospital, outside the new gate, which

'•pencd to Thomas-street, and which was founded in 1 1 88. St. Francis'

must have been the house of Grey Friars, founded in 1235, which

stood in what is still called Francis -street. St. Saviour's lay in

(^xmantowiiy on the north side of the Liffey, occupying the site

of the present Four Courts, so that the water had of necessity to

k' broaght across the river ; for this purpose the bridge had to be

utilized, and the Friars bound themselves to carry out the works

without injuring it.

As numerous water grants to citizens for specified sums of money
re to he found in the Liber Albue, it is matter of conjecture why

* Toll or fee which a miller takes for grinding com.

44 Proceedings of the Royal Irish Academy.

the above special cases should have been proved against the city at
suit of the Abbey. In the case of Holy Trinity, though an agreement
is mentioned, no rent or payment is named, while the supply to
St. Saviour's is expressly stated to have been bestowed as perpetual
alms. St. Thomas' Abbey could hardly have objected to these grants,
unless it had some claim to a share in the profits, save on the ground
of the amount of water required to supply so many large establish-
ments, which might seriously affect its own interests.

The next finding deals with infringement of the Abbey's rights by
the city in taking pledges of their men for alewyth, in offering an
explanation of which I am much indebted to the researches of the
Deputy-Keeper of the Records, who is at present engaged in editing,
under the Master of the Rolls, the Justiciary and Plea Rolls of
Ireland, of the reign of King Edward I. Very little is known as to
the jurisdiction of the courts here at this early period, more especially
with regard to the practice of the inferior and petty courts, on the
origin and growth of which the publication in question must neces-
sarily throw much light.

The clause in our inquisition evidently refers to suits, prosecutions,
and fines in the baronial court of the lord abbot of Thomas Court, for
the liberty of Donore, which would be held at stated times by his
seneschal, and to similar proceedings in the rival court of the
mayor and citizens of Dublin. In those petty courts, as in the superior
ones, when any one had cause of complaint, he came in during a sitting,
stated his case and gave pledges for prosecution : in minor matters,
some article of more or less value would be deposited, and in a case of
greater magnitude, a friend or neighbour appeared as surety. These
were in no sense courts of record and the proceedings were carried on
without being committed to writing. On pledges being given, the
court was bound to summon the defendant to appear and answer at a
certain day, a summoner receiving instructions verbally from the court.
In the present instance, it seems plain that the city court had been
taking the pledges of the inhabitants within the jurisdiction of the
abbot of St. Thomas, the cause of action lying in the island of Donore.
I have no where else met with mention of any part of this district
being called an island, but it must have been some lowlying portion of
the large district and liberty of the name, insulated by the windings
of the Poddle.

Gilbert's Sistorio and Municipal Documents of Ireland contains
notices of appeals to the Justiciary of Ireland and to the King and
council in England from St. Thomas' Abbey against the city for draw-

B RRT — Notes on an Unpublwhed MS. Inqumtion^ 8fc. 45

ing to their court pleas of tenants of the former, which of right should
hare been pleaded in the harony court of the abbot.

Among the ** Laws and Usages of the City of Dublin," enrolled in
the Chain Book of the corporation, appears the following : —

Db Gbrueise.

Dautrepart, chescune ki aceresce paiera par an ij". pur ceruoise
qaele vend par an si ele neyt grace des bailiffs.

Dautrepart, si ele ne face si bone cerueise come ele fere deust ne
oe tient lassise come veisin et autre, ne si com est crie parmi la yile,
ele est en la mercy de xt deniers.

As this enactment deals with women brewers, it is to be supposed
that the bulk of the brewing in the neighbourhood was in small
quantities, and that women were principally engaged in it

In later times, the brewing trade was extensively carried on along
the line of the Poddle and the city watercourse, the water being of a
character peculiarly favourable for the purpose, and here the brewers,
especially about Donore, would have been subject to the jurisdiction
of the Abbey.

The saxon wyUy loiU, wytam were equivalent to the Latin mulcta,
fines, and the aUwyth of the inquisition was, doubtless, the ale mulct
or fine of 1 5 pence imposed on such as brewed bad ale or an article
not up to the standard of the assise of ale. Complaints were frequent as
to the assises of bread and ale not being strictly kept, and as to the
assay not being sufficiently frequent. These old-time ale brewers
would probably have had to give pledges in anticipation for payment
of this fine, which some of them were certain to incur, and the city
authorities were active in taking these, instead of allowing them to go
to their rightful tribunal, the court of the abbot, as baron. In con-
nexion with the subject of ale, it may be interesting to recall the fact
that one of the most ancient privileges conferred on the Abbey of
i>t. Thomas the Martyr, was the TolboU, a custom of the tribute of one
gallon and a-half of the best ale and mead to be rendered by every
brewer in Dublin out of each large brew. This had been granted to
Prince John, son of Henry the Second, and while lord of Ireland,
before his accession to the throne of England, he made a gift of the
imposition, in perpetual alms, to the abbey founded by his father.

The inquisition concludes with a finding that the street between
Ht. Catherine's church and the horse market was of free alms.

46 Proceedings of the Royal Irish Academy,

appertaining to St. Thomas' Abbey. The city, as it grew westwani.
began to encroach on the monastic precincts, for the Abbey, at it<
foundation, stood at a considerable distance from the walls, and soof
authoritative pronouncement on the extent and nature of the abbot's
jurisdiction in this direction miist have become necessary, (xilbeit's
Calendar of Anetmt Records of Dublin (vol. ii.), under the year 1571,
mentions houses built on the old horse market in St. Thomas-street,
which street Speed's map (1610) shows as then running east of
St. Catherine's, and it seems probable that this market hj nearer
St. Francis-street and the New Gate. The same calendar (toL i.,
p. 121) describes ground outside the New Gate, near the Franciscas^
convent, the grant of which reserves a place for holding pleas anniuUj
during the time of the fair. Another grant is entered, wherem &
curtilage in the city land where the fair was held, is mentioned as
lying outside the New Gate, in St. Francis-street.

1— S> a-c- ('f. O't^ .3 ^j

AN AKABIC INSCRIPTION FROM RHODESIA.
By STANLEY LANE-POOLE, M.A., Litt.D.

(Plate I.)

[Bead Mat 12, 1902.]

The Marble tablet represented on Plate I. is remarkable not only

as the first Arabic inscription so far discovered in Rhodesia, but as

a document relating to a very early settlement of Muslims in South

Africa, unrecorded in any Arabic history. Much has been published

during the past ten years on the ancient monuments of Rhodesia, and

the thirteen sites which formed the basis of Bent's Ruined Cities of

Moihonaland have now been multiplied by more recent exploration

till they are estimated^ at five hundred distinct groups of ruins, of

which boweyer scarcely half have been even partially surveyed, and

none has yet received thorough investigation by trained arch»ologists.

These interesting monuments, scattered over the immense stretch of

country between the Zambesi and Limpopo rivers, and bearing strong

points of resemblance to the remains of ancient buildings in Southern

Arabia, have naturally attracted much attention, and their origin is

one of the most curious problems that archeology has to solve. The

hypotbesis that they were the works of Sab»an miners of the period

when the South Arabian kingdoms were at the height of their power,

more than fifteen hundred years before the Christian era, and that

the numerous ancient gold-workings connected with these monuments

were really the source of the * gold of Ophir' which the * ships of

TarshiBh' brought for the adornment of Solomon's temple, as argued

by Professor A. H. Eeane* and others, has everything in its favour,

except epigraphic proof, and it may still be hoped that further ex-

1 HaU and Neal, AnemU Jtuint of Rhodetia^ 1902.

> The Gold of Ophir, whence brought and bff whom $ 1901.

K.I.A. PKOG., VOL. XXIV., hEC. c] [4]

48 Proceedings of the Royal Irish Academy.

ploration may reveal Himyaritic inscriptions. Nothing is more pro-
bable than that the sea-faring Arabs of the Yemen and Hadramawt
should have been in close commercial relations with the east coast of
Africa and have discovered the mineral resources of Bhodesia where
numerous gold-mines still testify to very ancient workings.

What the ancient Himyarites probably did in almost prehistoric
times, the mediaeval Arabs undoubtedly carried on. There is every
reason to believe that the predominant influence of Arab traders,
and in some parts even of Arab rulers, was continuously maintained
along the east coast of Africa as far south at least as Beira down
to the arrival of the Portuguese at the close of the fifteenth century.
When Yasco da Gama reached Sofala, the mediaeval port near Beira
(towards which the ancient sites and gold routes of Rhodesia evidently
converge), he found 'Moorish', i,e, Arab, traders employing natives
to work the gold mines^ and seized Arab dhows laden with gold dust
There is no doubt that thiis commerce had been going on for centoiies,
if not for some thousands of years. The references in the works of
Arabic geographers and travellers, scanty and vague as they are.
sufficiently prove that So&la was well-known as a port for the gold
trade. EL-Mas'udi, writing in the middle of the tenth century,
mentions SofSlla (which is itself an Arabic word, meaning 'low-
country') as the terminus of the voyages of the merchants of the
Persian Gulf, and adds that 'the country of Sofala and Wak-Wik
produces quantities of gold and other marvels." £1-Birunl, Ibn-Sa^id*
and Yakut refer to this trade, and el-Idrlsi says that in all the land
of Sofola gold is found in abundance, sometimes in nuggets of a {rmU^
pound's weight.*

Probably these commercial relations between Arabia and the ea^
coast of Africa had been uninterruptedly maintained from anciezit

^ J. de Barroe, Da Ana^ Dec. I, Hv. z, cap. 1.

' £d. Barbier de Meynard, i, 6, 7.

s Beinaud, Froffm. Ar,, 112 ; (?m^. ttAhoul/tda, Intr., HI.

* Jaubert, i, 66.

Lanb-Poolb— Om an Arabic Inaeripticn frwii Bhodesia. 49

times ; bat when the Muslim Araha first made settlements on the coast

is sot stated in any of the general Arabic histories. There was

evidently no definite invasion at the time of the great Mohammadan

expansion in the seventh century, or it would have been recorded.

The only authority we possess, and that at second hand, is a 'chronicle

of the kings of Quiloa' which was discoyered when Francisco de

Ahneida, the Portuguese yiceroy, took that island in 1505. An

abstract of this history — ^the original is apparently lost, — appears in

the celebrated Da Asia of Joao de Barros^ who seems to have had the

work at his disposal ; and a modem Arabic ms. from Zanzibar in the

British Museum (Or. 2666), entitled ij£=>jlsA ^ i^\ ^^\s£s\

contains a brief history of Xilwa (Quiloa) which has evidently been

compiled from some such earlier source as the Chronica dos Beys de

QnUoa cited by fiarros. According to this solitary authority there

were three independent settlements of MuaUms on the Zanzibar coast.

First a number of the schismatic sect of Zeydls — whose leader, Zeyd

ibn 'All, a descendant of the prophet, was executed for proclaiming

himself as the Mahdl in 740 by the Omayyad caliph Hisham', —

emigrated to the African coast, somewhat north of the modem

Zanzibar, to escape persecution. Barros calls them 'Emozaydij',

which, as Mr. Arnold suggests*, is probably a corruption of Umma

Zeydlya, ' the people of Zeyd'. These were followed in the first

half of the tenth century by a second (but orthodox) band of

fugitives, who left their homes near the Bahreyn on the Arabian

coast of the Persian Oulf in consequence of the oppression of the

amir of Lasah (probably el-Ha^), and settled at the same place as

the Zeydls, whom they drove into the interior. This second colony

founded the great port of Makdashu (Magadaxo) which became the

* Dec. I, liy. viii, cap. 4.

* Published by Mr. 8. Arthur Strong in the Jowmdl of iht R. Aiiatie Society ^
1896, 386-430.

' Et-Tab«ri, AnnaU$, ed. de Ooeje et alii, III, 1742 fF.

• •

« Arnold, Tht Preaching o/Itlam, 278, 279.

60 Proceedings of the Royal Lish Academy.

metropolis of the Arabs on that coast.^ The Morocco trayeller
Ibn-Batuta visited this city in 1332, and describes it as a vast town,
with numerous mosques, and under the rule of a Mohammadan saltan
called Sheykh Abu-Bekr ibn 'Omar. He mentions its trade Trith
Egypt, and says that Magadaxo was fifteen days' sail from Zeyla' on
the Bed Sea.' It was situated about half-way between Zanzibar and
Bab-el-Mandeb. The third settlement of Muslims came early in the
eleventh century from Shiraz in Persia. Sailing from Hurmiu in
the Persian Gulf, and avoiding orthodox Magadaxo— for the new-
comers belonged to the Shi a sect, — they proceeded further sonth \o
Kilwa (Quiloa), where they found a previous Muslim aettlem^it
and a mosque. Here they bmlt a fort, and ruled until the coming
of the Portuguese. This was the most important of all ^e Arab
settlements, for the kings of Quiloa extended their sway north over
Mombasa,' and south over So^la,* where they entered into relafaons
with the native ruler, whom the Portuguese called the Monomotapa
or Benomotapa, a name which Professor Keane explains aa Bantn for
4ord of the mines', but which the Portuguese understood as meaning
merely 'emperor'.*

Such is, in abstract, the little that we know about tiie Moham-
madan settlements on the east coast of Africa. Although the Quiloa
chronicle places the first arrival of Muslims not earlier than 7i0,
it is permissible to assume that other Muslims had preceded them,
since it is hardly probable that a band of persecuted fugitives would
have fled to an unexplored land, where the natives had the reputation
of cannibals, unless some others had shown them the way. "niat
there was some such early settlement, not only earlier than the date

1 Gp. Bigby : Report on Zanzibar Dom., 47, where the migratioa of Arabs of
the tribe of el-Harith from the Bahreyn to East Africa and the fonnilitinn of
Magadaxo ia placed about a.d. 924.

* Ed. De£r6mery and Sanguinetti, ii, 180 ff.
» Strong, J.B.A.8,y 1895, 430.

* BaiTOs, Dec. I, liv. viii, cap. 4 ; Wilson, Monomotapa, 109.

* BantM, Dec. I, liy. z, cap. 1 ; Keane, Opkir, 9.

Lanb-Poolb — On an Arabic Inscription ^/rom Rhodesia. 61

when the Zeydls established themselyes north of Zanzibar, but also
much farther south, is implied bj the inscription which has reached
us from Bhodesia. The text, chiselled on a small slab of white
marble, is quite legible: —

jJiS JJbj X)j ile. M ^

(The pointing in the original is reproduced above. There are no
points to ^^M^ori^ or to the i of JL«a or to ^J^ • ^^j^yoA omits
the first My and has superfluous dots under ^.)

In t}te name of God, the Compassionate^ the Merciful.
. I%ere is no god hut Ood. Mohammad is the apostle of God,
God bless and save him/ And this [m] the tomb
of SaUOm ibn Saldh [trAo] had forsaken
this world for the Last Abode^ and was [i.e. aftsr'\
the Rijra of the Prophet of the faithful five and ninety
years. The sentence ends. And God is aU-knowiny,
God.

There are several grammatical errors in the text. Jt^ jJb
fihould of course be ^ V^jb . It has been suggested that it may

52 Proceedings of the Royal Irieh Academy.

be^jQ \jjb , ' this is indeed the tomb'; but I baTe never met with
this classical fonn in an Arabic inscription, unless in Konn
quotations, y^^* ,13 , as third person feminine, does not agree with
its masc. subject ^ ]L . Possibly it is i^z^ii , ' thou hadst forssken*.
In any case it is a very peculiar phrase. j\j should be^,\jJl • 'Hie
phrase ^^jf^^\ ^ ijS^ c>^J» ®^'' ^^^ as it is expresaed,
can only mean that the Hijra happened ninety-five years before.
The formulas JU3\ J and ^J^ ^\ call for special notice. JU3\ J,
'the sentence ends', equivalent to *end of extract* or Jinii, maj
possibly imply that the inscription is copied from an earlier document^
aJ^ ^\ , ' God is All-knowing' or < God knows best' is a phrase,
like Jx\ i^\y that suggests some doubt on the part of the writer
as to the accuracy of the statement. As to the names of the
deceased, they may be read simply Salam ibn Sallh, 'Peace (or
security) son of Prayer' (for *jLrf»), which might possibly be namef
adopted by a native convert ; or the first name should have teshdld
and should be read Sallam, a not unusual Muslim name in the first
century of Islam, whilst the second may either be a mistake for
iLp , Silah, or possibly a form of the root u:.Jw> with the meaning of

LzJLa^, 'strong.'

We have therefore in this curious inscription an epitaph oa a
Muslim who is stated to have died in the year 95 of the Hijra,

^ Mr. A. G. EIIIb, of the Department of Oriental Hanuacripts in the Bttbik
Muaeum, informa me, however, that this phrase is characteristically Wat Abiaau
and that he has seen it in a draft inmsiiption in absurdly bad Arabic eonpoaed by
a West African natiye for a memorial stone to soldiers who fell in the recsBt
Ashanti expedition. He has also met with such names as Tawbid and Ta-«&
among West African negroes, which wonld bd parallels to * Peace' and * Pkafer*
in the present inscription. That this inscription is not West African, boverer,
is condusiyely proved, not only by its provenance, but by the pointing ol the /»
and kdfsy which are diflbiently pointed in the Haghrabi script.

Lane-Poole — On an Arabic Inscription from Rhodesia. 53

or A-D. 713-714. There is nothing whatever to suggest that it is a
forgery. Its history is perfectly straightforward. It was hrought
some eight or nine years ago from what appeared to he ' an ancient
temple' in Mataheleland — unfortunately all inquiries have failed
to trace the site — to Mr. P. Hanhury France, an agent of the
Union Steamship Company at Cape Town. Mr. France attached no
importance to it, and gave it as a curiosity to Dr. W. M. Russell,
a surgeon on that line of steamers, who afterwards practised at
Eimberley, and Dr. Eussell passed it on to Mr. G. S. Cary, of
Terenure, Co. Dublin, in whose possession it remains. No one in
South Africa could have forged it, nor is there any motive for
forgery. Moreover, forgers follow received types, and this in-
scription is peculiar in many ways. Nor do I believe that it was
imported. The inscription is too unusual in diction to have been
composed at any educated Mohammadan centre, but its peculiarities
and grammatical errors are natural in such an out-of-the-way place as
southern Rhodesia. I am told that there is no marble in Rhodesia,
bnt this remains to be proved. Arabs do not carry tombstones about
with them on their travels, nor can I imagine such an inscription
entering the mind of an Arab of Arabia or a Muslim of Egypt : the
language is too bizarre.

Assuming the inscription to have been engraved in Rhodesia
and set up over the tomb of this Sallam son of Salah, the question
remains, is it the original epitaph or merely a commemorative tablet
erected in later times ? The style of writing is no certain guide,
tance we possess no other specimens from the same region, and
^without dated examples epigraphic science cannot exist. The
Arabic character varies so greatly at different places in different
ages that it would be rash to draw conclusions from similar styles
of inscriptional nasthl elsewhere. Still, judging roughly by the
oharacter, it is impossible to believe that it goes back so far as the

54 Proceedinga of the Royal Irish Academy.

eighth century, and having r^ard to the peculiar fonnnlas ^the
sentence (cv record) ends', and ' God knows best', I indine to the
belief that there was some donbt as to the actual site of the tomb of
this Sall&m ibn Salah — possibly a local saint, — and that some Ifiter
Muslim pnt np the tablet, with all reserve, to commemorate the spot
identified by local tradition. Such tablets are not nnoommon in the
East oyer the graves of holy men, and to erect them is a pious set
which brings credit to the commemorator. There remains, howerer,
the possibility that the century of the date has been omitted, but tkii
I think improbable.

It will be seen that there are a good many problems connected
with this tablet which are not easUy solved. This much, howerer,
may be laid down. If not actually a contemporary tombstone of t
Muslim who died in southern Bhodesia in the beginning of the
eighth century, it shows at least that there was a local trsditiim
in regard to such a person strong enough to induce some one in
later times to set up a commemorative tablet recording bis nsiae
and date. As the solitary Arabic document from South Africa tbe
inscription is valuable ; but it is to be hoped that it will not loog
enjoy its unique eminence. A qualified archseological explontioB
of Bhodesia ought to bring to light other monuments of the Muslim
and possibly far earlier periods, and decide many questions in regaid
to the ancient and mediaeval history of South Africa which csn
never be settled until we have the evidence of trained explorers
and thorough excavation of the numerous sites which so fsr have
been scarcely more than looked at. Such an archseological surrey
should be undertaken without delay, and the results should be
collected in the Museum of the Bhodesia Scientific AssodatioB at
Bulawayo, whose members are fully alive to the importance of the
sub j Oct.

[ 55 ]

VI.

SOME FURTHER NOTES ON ANCIENT HORIZONTAL WATER-
MILLS, NATIVE AND FOREIGN. By JOSEPH P. O'REILLY,
C.E.

(PlATBS II., III., AJTD IV.)

[Bead Apbil U, 1902.]

In the paper on the '' Milesian Colonization of Ireland considered in
relation to Gold Mining,'* read before the Royal Irish Academy,
/anuary 22nd, 1900, 1 took occasion to cite from the work by Eugene
Tmtat on "The French Pyrenees" the names of the tools employed by
the gold washers at Pamiers in the Comte de Foix, and their probable
Celtic derivation, with a view to show the connexion that probably
existed between the tribes or peoples engaged in the working of the
precious metals in ancient times all over Europe, particularly in the
mountainous regions, and the consequent similarity not only in the
forms of the tools employed by them, but also in their names or desig-
nations. . Convinced that this path of inquiry is capable of leading to
very striking as well as useful results, as regards both Archaeology and
Philology, I have been expecting to meet with further opportunities
of pursuing it, and beg to submit the following remarks as to the
probable origin of certain ancient Irish water-mills. The subject
was suggested to me by the article which appeared in the Ulster
J<mmal of Arehtology^ vol. iv., 1856, p. 6, entitled "Ancient Water-
Mills," from which the following details are worth citing : —

**Thc accompanying drawing represents accurately an ancient
wooden water-wheel in the possession of Mr. James Bell, of Pros-
pect, near Ballymoney, county Antrim, excepting only that such
portions as are now imperfect, have been restored in the drawing
to eorreepond with the others.

" It was found a number of years ago, in the bog of Moycraig,
within one mile of Morsside, on a farm now 'occupied by William
Hamill, and which is comprised in the district called ' The Grange of
Drumtullogh.' The spot is low and flat, and no stream is at present
visible near it.

&.!.▲• FB0C.9 VOL. XXIV. y SBC. 0.] [5]

Ancient Irish Wa,ter-MtlL

ihrniheULsierJournaLi ofArch»j4tgy fmr,

O'Reilly — Ancient Water-mUhj Native and Foreign. 57

'^ T^e wheel here represented is a horisontal one, and is the most
perfect specimen yet found in Ireland. Portions of another of precisely
similar construction are now in the Belfast Museum which were found
in the county Down, near Eillinchey, beside an artificial island, or
water fastness, which is now occupied as a garden. The material of
the wheel now figured is of oak, and when found was quite soft and
spongy from long immersion in the bog; but on being dried, it
recoTered its hardness, and appeared perfectly sound. The water-
wheel consists of a nave and upright axle, both cut out of one solid
piece of wood, the entire length being 6 feet by 6 inches. Round the
nare are inserted nineteen buckets or ladles, curved in the manner
shown in the drawing, and which received the impulse of the stream of
water. Ten of these still remain perfect. At the upper end of the
axle is a deep groove 12 inches long, in which moves an oaken wedge,
used evidently for the purpose of raising or lowering a small millstone
which was placed above, or for what would be called now ' gristing
the mill.' The whole mechanism was supported by a stone pivot or
gudgeon secured by a wedge at the foot of the axle where it still
remains. This pivot, no doubt, revolved upon another stone hollowed
to fit it (a socket). A stone of this kind was in fact found near the
water-wheel at Eillinchey, and is preserved along With it in the Belfast
Museum, bearing evident marks of having been deeply perforated by
some pivot constantly revolving in it " (p. 7) : ' ' The buckets are ingeni-
ously ^tened into the nave by mortising, and are firmly secured by
an oaken pin driven in a sloping direction, from the outer circumference
ot the naye, in such a manner as to pass through the inner ends of
three, and at equal distances, each bucket in the wheel had three pins
passing through it, thus securing it completely to the two adjoining
ones and to the nave.

'* No tradition now remains among the people respecting the use of

water-mills of this construction in the country, but there is evidence

(which I give further on) to prove that they were common at lesst in

Ulster three centuries ago. However, down to that period and even

later, the use of the quern or hand-mill was quite general throughout

Ireland and its use is not yet given up in some of the western islands of

Scotland . So early as the thirteenth century legal means were adopted

in Scotland to compel the people to abandon the use of the hand-mill

for the larger water-mills then introduced. In 1284, in the reign of

Alexander III., it was enacted that ' ITo man i^ presume to grind

quheit, maishlock, or rye, with hand mylne, except he be compelled

he storm, or be lack of mOls, quhilk sould grind the samer ; and in this

[6 2]

58 Proceedings of the Royal Irish Academy.

case, gif a man grinds at bandmjlnes, he sail give the thretteiir
measure as multer; and gif anie man contraveins this out prohibi-
tion he sail tine his hand-myleres perpetuallie.' Yet in 1819,
M'Culloch ( Western hUs^ yol. ii., p. 30) states that the quern is foimd
in every house in St. Eilda, and the statistical account of Scotland,
published in 1845, mentions that in the parish of Sandsting in Shet-
land, there are ' querns or hand mills without number.'

''There seems to be reason, however, for believing that water-mills
were not unknown in Ireland at a very early period. Dr. O* Donovan,
in an article in the Dublin Fenny Journal, has quoted several paasa^
from the Brehon laws, which are of great antiquity, stating the
damages to which the miller and the millwright shall be respec-
tively liable in case of an accident occurring in a mill turned hf
water. He also gives references to many of the lives of Irish Saints,
in which water-mills are expressly mentioned as having been erected
by eeelesiatticSf proving that they were in use not long after the intro-
duction of Christianity.

*'Mr. Getty, in his account of Torry Island {Ulster Journal «/
Arehaoloffy, vol. i., pp. 143, 146), mentions the curious circnmstance
of a veiy ancient stone cross being fastened at its base into a miU-
stone ; and notes 'the tradition of the islanders, that all aadoit
buildings there have a millstone in their foundations.

'* In the notes to the translation of the Annals of Ulster (now in
course of publication in this Journal) at a.d. 587, it is stated from
the Breviary of Aberdeen, that Constantine, a King of Damnonia in
Britain, * having abdicated his throne, repaired to Ireland and
became miller to a monasteiy.' It is well known that a mill vas
sdmost always in connexion with religious houses of the Cistercian
Order.

<< In the Annals of Tiyhemach, one of the most trustworthy of our
old Irish chronicles, there is a curious passage at the year 561, where
mention is made of the slaughter of the sons of Blathmac, King of
Ireland, in the mill of Maclodran ; and a verse is quoted from an
ancient poem, in which the Bard fancifully addresses the mill thus :
' 0 mill ! what hast thou ground ? precious wheat ? Thou hast
ground not oats, but the sons of Cerbhall,' &c. (O'Connor, Eemni.
flibem. Scriptores, vol. ii, 198). The writers of the historical notes
to the Ordnance Survey of Londonderry gave quotations &xmi the
Book of Kells (MS. Trinity College), and the Registry of Clesmamoiss
(Clarendon MSS., Brit. Museum), in which grants of nulls U
monasteries in the eleventh century are mentioned; and various

O'Rbilly — Ancient Water-milh^ Native and Fweign, 69

passages maj be found scattered through our Irish Annals, in which

allusion is made to mills. Most writers who have mentioned the

subject seem to take it for granted that water-mills must have been

introduced into Ireland by Roman ecclesiastics, or at aU events from

wme country subject to Roman sway, especially as it is pretty well

ascertained that a mill of some kind was usually at each Roman

station in Britain ; and a decisiye evidence seemed to be afforded by

the similarity or rather identity of the Irish and Latin names for a

103JI, A little further examination of the question may perhaps

show that this is not so certain, at least so far as the North of

Ireland is concerned." The writer then examines the philological

argument first. He cites Cormac*s Glossary, and discusses the

derivation of the Irish term Muihan, The forms in which it appears

in the modem languages which are known to be directly descended

from the Latin, such as the Italian, Spanish, French, Walloon, &c.,

are mulino (It.) ; molino (Sp.) ; tnoulin (Fr.) ; molin (Walloon).

" But," he continues, '' if we examine further, we shall find the very

same root, little more changed than in the above examples in a

variety of other languages, which can claim an origin as independent

as the Latin, and are spoken by nations who were never influenced by

Homan away." He then gives a table of the equivalents for the word

' mill ' as used in the chief languages of Europe, exclusive of the

four already mentioned, twenty-eight in all. In each case the name

given is a slight modification of the word *mill.' This slightly

modified name is found in countries extending from the shores of the

Vediterranean to the far North, and from the coasts of Spain and

Ireland to the extremity of Russia. To complete the chain we

have only to note further, that in Persian tndl is ' to grind,' and

that in Sanskrit, the old lang:uage of India, malana signifies ' rubbing

or grinding.* The root is therefore common to all the extensive class

of languages known as the Indo-European family, as well as to

several outlying districts not included among them. There can be

little doubt, therefore, that it is one of extreme antiquity, and cannot

be claimed exclusively by the Latin any more than by the Celtic."

(p. 9.) — He then discusses the probability of the Romans having
introdnced the water-mill into the British Isles. He shows that the
hand-mill was no doubt in extensive use from a remote period, not
only in Italy, but all over Europe and the East. It must have been well
known to tiie Oauls and Britons in Caesar's tune, as he speaks of
their ** moUta etbarta^^* or ground breadstuff s.

(p. 10.) — ** The water-mill does not appear to have been a Roman

60 Proceedings of the Royal Irish Academy.

inTention. Strabo mentions that a mill of this kind was erected in
Pontus (in Asia) at the palace of King Mithridates (Strabo, zii., 3,
§ 30), which is the earliest of which we have any record. Indeed, ve
hare it on Roman authority that water-mills were not introduced m
Italy before the time of Julius Caesar (who died b.c. 44), and were Uwn
only used by a few indiyiduals (Vitruvius x., 5, 2). Pliny's slight
notice of them, which only occurs in one sentence of his entire great
work, shows that they were by no means common in his day (Plin.
lib. xviii., c. 10). He died a.b. 79. The earliest mentions of public
water-mills is about the year 398, under the Emperors Arcadius and
Honorius ; and the manner in which they are referred to in the lavs
of the period shows that they were then a novelty (Code Teod. 14, 15}.
Now it was at this very time that the Romans finally abandoned
Britain. It appears therefore that the Romans never used water-
mills to any great extent, nor have we any satisfactory proof tbtt
they established such mills at each of their military stations in
Britain. Many small millstones, indeed, belonging to the Bomsn
hand-TsnUlA, have been discovered on the sites of the Roman stations,
but so far as I am aware only a few doubtful cases have been bron^t
forward to prove the existence of water-mills at those places.

*^ For the foregoing reasons and from the consideration that there
never was a friendly intercourse maintained between Ireland and the
Roman province, it seems unlikely that water-mills were introduced
into this country from Roman Britain. We must therefore seek for
their origin in some other quarter, and, in my opinion, the weight of
probability rests on the North of Europe. Although the Danes and
Norwegians did not effect their conquest of Ireland for many oentozies
after the departure of the Romans from their British province, thej,
and the other maritime tribes in the neighbourhood of the Baltic hid
maintained an intercourse with these islands for an indefinite
period. The details of this intercourse is unknown to us farther
than what may be gathered from scanty allusions in old Irish Annal»
and Icelandic sagas. But there seems to be little doubt that daring
the obscure period alluded to these Gothic tribes had been gradoalfy
colonizing the east and north of Scotland, and of course bringing
within them whatever arts of civilization they possessed, which there
is reason to believe were greatiy superior to those existing in their
new colonies. A people who could send out fleets of well-equipped
vessels and armies of mailed warriors, sweeping the coasts of £niope»
and conquering wherever they appeared, must have possessed con*
siderable mechanical skill, and were not likely to be without wato-

O'BmhLY—Ancieni Waier-mtlh^ Native and Foreign. 61

millB for grinding their com. Their native hills abounded in
cascades suggesting the emplojrment of water-power, and their forests
furnished the materials for their miU-wheeh"

(p. 11.)— ''Now it so happens that the poetical account of the first

water-mill ever erected in Ireland (written by a bard who died

A.]>. 1024) and the popular tradition state the millwright who

oonstracted it was brought from Scotland. This was in the third

century, when, as the poet relates, the Monaroh Cormac, desirous of

flaring a beautiful bond-maid the labour of grinding com daily in a

quern, sent across the sea for a millwright who erected a mill on the

stream of Kith near Tara (Poem of Cuan O'Lochain, quoted in the

historical notes to the Ordnance Survey of Londonderry). We have

no description of this miU to assist us in forming a conception of

its form or constroction, but we may assume that it was of wood, and

of a simple form, probably not very different from the one which is

the subject of the present article. This traditional story, at all events,

points to the quarter from whence the invention was believed to have

come. Kow, if on examination, we should find that mills quite

similar to our specimen were in use, or are actually still in use, in a

number of districts in the British Islands and the islands adjoining,

known to have been peculiarly Scandinavian, and for centuries under

the government of the Northmen, it would be difOicult to avoid the

inference that these machines were introduced thither by them.

This I am enabled to show from various indepeAdent autliorities,

whose several notices of mills I now place together for comparison."

1 • In the Farde lihmda, — ' * The constmction of a water-mill in Faroe
is exceedingly simple. The building for the most part consists merely
oi woody the roof being supported by four posts or pillars ; but, to save
timber, these pillars are sometimes built of stone, mixed with mud ; it
is entirely open below, so that the water can have a free course through
it. On the groimd is placed a loose beam, having in the middle a
ineoe of iron, with a small hole in it, which, however, does not pasa
through the beam. This hole is made to receive the gudgeon of a
perpendicular axle, which proceeds up to the millstone, and this axle
supplies the place of a crown wheel and spindle. To the upper end
of tiie axle is fixed a round rod of iron, which passes through the
lower atone, and which supports the iron cross that bears the upper
millstone. At the lower end of the axle there are eight leaves or
boaida mortised into it, about 18 inches in length, a foot in breadth,
and from 1 to 1^ inch in thickness. These leaves, which perform the
part of a water-wheel, do not stand exactly in a perpendicular, but with

62 Proceedings of the Royal Irish Academy.

a somewhat oblique direction, so as to turn their flat sides towards the
water which falls upon them ; and the spout, which must give tiie
water a sudden fall, is placed with its lower end close to these leaTes.
From one end of the beam lying on the ground, which supports tk
axle and the upper millstone, a piece of wood rises in a perpendicalar
direction towa^ the millwork, where it rests on wedges, and by
pushing in or drawing out these wedges the upper stone can be
raised or lowered at pleasure. The millstone makes a hundred rero-
lutions in a minute ; but, as the stones in general are small, and bavie
no furrows in them, they grind slowly, and are not calculated for
the preparation of grits or barley. — ^Landt's * Faroe Islands,' 1810,
p. 293."

(p. 12.)— 2. In the Shetland Islands,—'' In skirting along tbe
harbour (* Rigseller Voe, in Shetland ') numerous slender rills were
observed ambling down the dales to pay their tribute to the Voe.
These occasionally served to supply some smaU mill, the presence of
which was signifled by a low shed of unhewn stones that stretched
across a diminutive streamlet, over which it was possible in many
places to stride. Compared with a water-mill of Scotland or England,
the grinding apparatus of Shetland seemed designed for a race d
pigmies. The milLstones are commonly formed of a micaceous gneiss.
being from 30 to 36 inches in diameter. Under the framework by
which they are supported is a sort of horizontal wheel of the same
diameter as the millstones, named ' Tirl* which consists of a stout
cylindrical post of wood, about 4 feet long, into which are mortised
twelve small float boards, placed in a slanting direction, or at an
oblique angle. It has a pivot at its under end which mns in s
hollowed iron plate fixed in a beam. A strong iron spindle, attached
to the upper end of the ' TtW,' passes through a hole in the under
millstone, and is firmly wedged in the upper one. A trough oondnctF
the water that faUs from the hill, upon the feathers of l^e ' Tiri,* at
an inclination of 40^ or 45^, which, giving motion to the upper mill-
stone, turns it slowly round. Such is a description of this exqnistie
piece of machinery, the invention of which is probably as old as the
time of Harold Harfagre."

3. In the Hehrides, — <' The mills at Lewis are probably the greatest
curiosity a stranger can meet with on the island. There is scarcelj a
stream along the coast, or any part of the island, on which a mill is
not to be seen. These mills are of very small size and of a very
simple construction. The water passes through their middle, whoe
the wheel, a solid piece of wood, generally 18 inches *n diameter,

O'Brilly — Ancient Water-milk^ Native and Foreign. 63

stands perpendicularly. A bar of iron nms through the centre of this
wheel. This bar of iron, or axle, rests on a point of steel, which is
fixed on a plank, the one end of which is fixed in the mill wall, the
other in the end of a piece of plank, which stands at right angles with
the plank on which the wheel rests. The upper end of the axle fits
into a cross-bar of iron, which is fitted into the upper millstone,
which is rested upon wooden beams or long stones. There is a pur-
chase upon the end of the said perpendicular beam or plank by which
the upper millstone can be raised or lowered (p. 13). There are nine
pieces of board, 8 inches broad and 1^ feet long, fixed in the wheel,
parallel and at equal distance from each other, upon which the water
is brought to bear ; which, together with a few sticks for roof and
some heather for hatch, constitutes a Lewis mill." — ** New Statistical
Account of Scotland," 1845.

H'GuUoch states that the quern was found in every house in
St. Kilda, and recommends the establishment of a water-mill to super-
sede it. He then gives a description of a water-mill almost identical
with those already described, and says : ''It would not be easy to
construct the horizontal mill on cheaper terms." — ^M*Culloch's ** West-
em Mes of Scotland," vol. ii., p. 80.

4. Isle of Man. — **Many of the rivers (or rather rivulets) not
having sufficient water to drive a mill the greatest part of the year,
necessity has put them on an invention of a cheap sort of mill, which,
as it costs very little, is no great loss, though it stands idle six months
in the year. The water-wheel, about 6 feet in diameter, lies hori-
zontal, consisting of a great many hollow ladles, against which the
water, brought down in a trough, strikes forcibly, and gives motion
to the upper stone, which, by a beam and iron is joined to the centre
of the water-wheel." — ** Gibson's Camden" (Isle of Man), vol. ii.,
p. 1448.

5. Uhier. — I conclude with a few remarks more, viz. : — **That from
the said long bogg (beside Newtownards, Co. Down), issue many rills
and streams, which make small brooks (some of them almost dry in
ye summer) that run to the sea on each side of ye upper lialf-barony,
and on them each townland almost had a little milln for grinding oats,
dryed in potts, or singed and leazed in ye straw, which was ye old
Irish custom, the mealle whereof, called ' greddane^^ was very coarse.
The mills are called ' Danish,' or ladle millies ; the axle-tree stood
upright and ye small stones or querns (such as are turned with hands)
on ye top thereof ; the water-wheel was fixed at ye lower end of ye
axle-tree, and did run horizontally among ye water, a small force

64 Proceedings of the Royal Irish Academy.

driving it. I have seen of them in ye Isle of Man, where the I>aiie»
domineered, as well as here in Ireland, and left their customs behind
them."— "Montgomery MSS.," p. 321.

(p. 14.) — '* Anyone by comparing the foregoing separate deecrip-
tions will at once perceive that the several mills mentioned are identical,
in principle and construction, with the one described in the present
paper, while differing in a few details, such as the number of backets
or paddles. It will also be noted that the districts in which they
are described as being commonly nsed form, when taken together, a
geographical chain, leading directly from the country of the North-
men through the old seats of their dominion in these countrieB, and
terminating on the eastern coast of our own province.

'' It wUl be seen likewise that the last of the extracts alhides
specially to the popular tradition, both in Ulster and in the Ide of
Man that these mills were Danish. The same passage, written about
the year 1698, shows also that in the county Down a short time
previously such mills were quite common. It is only remarkable that
more of these remains have not been discovered, but this has arisen no
doubt from the perishable nature of their materials." — Robsbt M^Adam.

To these citations may be added one from the '' Encydopadia
Brit." (9th ed.), vol. ix., p. 344, article " Flour Mills." The nature
of the water-miUs, which were formerly common in Great Britain and
Ireland, and which continued in use well into the present oentoij
(nineteenth), may be gathered from the following description of «Be
visited by Sir Walter Scott during his voyage in the Shetland Iskode,
&c., in 1814. ("Lockhart's Life"):— "In our return, pass the
upper end of the little lake of Cleik-him-in, which is divided by
a rude causeway from another small loch, communicating with it«
however, by a sluice for the purpose of driving a mill ; but rack a
null ! The wheel is horizontal, with the cogs turned diagonally to the
water ; the beam stands upright, and is inserted in a stone quern oi
the old-fashioned construction. This simple machine is enclosed ia «
hovel about the size of a pigstye, and there is the mill ! There aoe
about 600 such mills in Shetland^ each incapable of grinding saatL
than a sack at a time."

That mills, mechanically worked, were known and erected in
Ireland in the thirteenth century appears from the followiag entry ia
the Calendar of State Papers, Ireland 1171-1251, p. liv. (3941), Jvae

O'UniUsY-^Auctent JFater-miliSf Native and Foregin. 65

3id, 1248. Mandate from the JuBticiary of Ireland to assign to the
abbot and monks of St. Mary's, near Dublin, land or annual rent of
10 marks in compensation for the injury done to them by the erection
of the King's mills near the Castle of Dublin. That the hand-mills
were in common use at that time appears from the first entries in
that calendar, p. 1, entry 1, <' Barth. de Glanyille and others render
their account for 468 equippers (eskiperii), six hand-mills
(manumolendina) . ' '

Similar mentions occur in the entries 2, 6, 7, 8, 9, 10, 11, 12, 13,
14, 15, 16.

It would seem from these different citations that while the hand-
mill was in common usage all oyer Europe during ancient times, and
in the northern countries down to quite recently, water-mills, though
not so common, were also in use, particularly in these northern
countries, and that they presented generally, the peculiarity of con-
struction detailed in the before-mentioned descriptions, that is, were
horizontal wheels, with paddles of peculiar form, adapted to receive
the impulse of a smaU stream of water having a certain velocity, that
isy a certain sufficient head. Mr. M'Adam's conclusion that the
origin of this stylo of miU must be referred to the Danish or Norse
people seemed plausible enough, granting his assumptions, and taking
as ascertained, that no other such mills were known elsewhere in
Sorope or in the East, but Mr. M' Adam does not appear to have made
the necessary research in this respect, and hence it is desirable to
examine if there be any evidence for the existence in past or present
times of such mills in the countries of Europe and of the East which
have still remaining either monuments or records.

A pri^iy one might expect that the Chinese knew of water-mills,
as of many other mechanical appliances, long before any other nation
in the East, and in Chambers' Encyclopedia, under the heading, *' Wat^r-
V»er^^ p. 365, it is stated : '* Notably, amongst eastern nations, the
Chinese were conversant with water-motors from a very early period."

^' The first attempt to produce hydraulic machinery proper, as
the term is now understood, were made in the Greek schools at
Alexandria, which flourished under the Ptolemies, under whose
regime Ctesibius and Hiero invented the fountain of compression, the
siphon, and the force-pump about 120 b.c." That water-mills were
ioTented or introduced into Europe as early as that period would
appear from the following paragraph taken from Smith's '* Dictionary
of Oreek and Roman Antiquities," 2nd edition, 1859, under the

66 Proceedings of the Royal L^h Academy.

heading ^* Mola," mill and water-mill {vnola aquaria). The fiist water-
mill of which any record is preaeryed was connected with the palace of
Mithridates in Pontus (Straho, xii. 3, § 30). '< At Caheira (in Pontos)
was the palace of Mithridates (the Qreat, 120 B.C.), the water-mill,
the park for keeping the wild animals, the himting ground in the
neighbourhood, and the mines." That water-miUs were used at Rome
is manifest from the description of the them by Vitruvius (x. 6, edition
Schneider) : — '* A cogged wheel, attached to the axis of the water-
wheel, turned another which was attached to the axis of the upper
millstone, the corn to be ground fell between the stones out of a
hopper (infundibiliimi) which was fixed above them.*' (See Palkdio
de Ee rustica, 1, 42). Ausonius mentions their existence on the rirer
near Treves, and Yenantius Fortunatus, describing a castle built in
the sixth century on the banks of the Moselle, makes distinct mentioB
of a tail race, by which <* the tortuous stream is conducted in a strai^t
channel."

It is to be remarked that the water-wheel above described from
Vitruvius' work, was evidently a vertical wheel with harvumtal orti,
the cog-wheel placed on this, communicating its movement to tlie
vertical axis carrying the millstone, by means of another cog-wheel.
It was therefore not of the same description as the wheels forming the
subject of this paper ; rather indeed of the same style as those modem
forms so common in these countries in modem times, and which are
those generally mentioned in the various encyclopedias under the
heading Water-mill, It is very difficult to find any historical
description or mention of the horifumtal water-wheel in question in
these works, although the turbines, which they really are a rude f<ffn
of, are described, but as of quite modem invention. It can theref(B«
be only incidentally that such a description may be met with, asd
as such I have come across the following in the *' Lettres snr la
Gr^ce, PHellespont, et Constantinople (1811) premiere partie," hf
A. L. Castellan. Of this writer and artist the following short account
is given in the ** Biographic Universelle " : —

** ( Antoine Louis Castellan, peintre, graveur, and architecte fnn^;
n6 k Montpelier en 1772 : mort k Paris, 2 Avril, 1838. II se vwu
d'abord k la peinture, entra in 1788 dans I'atelier de Yalendennes ; il
acquit bientot pour le paysage une reputation m^ritee. Yoyagea dans
le Levant, visita Constantinople, la Grece, les ties, Tltalie, etlaSniaK;
recueUlant partout un grand nombre de documents, dedessins, etpnit-
ant dans ces riches contrees« un gout d'autant plus sur, qu'il ne se

O'Rbilly — Ancient Water-millSj Native and Foreign. 67

laissait pas aller k\m enthusiasme irreflechi. Fix6 k Paris des 1804,
il s'occupa de publier, divers ouyrages pleins d'int^ret oii se trouvent
consign^s les r^sultats de ses voyages et de ses obserTations. lis sent
ttccompagn^s de nombreuses ves de8sin6eB et gray6es par I'auteur ; tela
sont. ' Lettres sur la Mor6e et les lies de Gerigo, Hydra, et Zante/
1 vol. in 8~, Paris, 1808 : * Lettres sur Constantinople/ &c., in 8*°,
Paris, 1811.)"

At p. 87 of this last work he speaks as follows : —

** £n Gr^ce, on retrouve d cbaque pas la tradition des usages antiques,
et particulierement dans les arts m6caniques. 11 est bon d'obserrer
que la plupart des machines dont on se sert dans ce pays, sont d'une
8iniplicit6 qui, bien loin de marquer I'enfance de Part, semble au con-
traire ne pouvoir ^tre que le resultat de la reflexion, aid6e d'une longue
experience.

''Si I'on entend par mecanique. Tart d'augmenter les effets en simpli-
fiant les causes, on pourrait croire que les anciens I'entendaient mieux
qae nous, surtout si I'on en juge d'apr^s les entreprises gigantesques
qu'ils ont executees avec des machines qu'on pent appeler primitives ou
elementaires, et dont les notres ne sont que la complication."

(p. 90.) **Nous avons dit que le besoin seul 6tait le v6hicule des
uaciens dans I'invention des machines. £n effet, celle des moulins k
mondre le grain ne remonte qu'au si^cle d'Auguste. Avant cette 6po-
que, on B*6tait contente de moulins k bras, semblables k ceux qu'on
roit encore en Sicile (Voyage pittoresque de Sicile par M. Houel,
1782-87) (p. 91), etqui ne sont que de simples instruments de menage.
Ces moulins 6taient portatifs, occupaient les moins d'espace possible,
et devaient foumir k peu de f rais, assez de farine pour nourrir une
famille. On pouvait meme employer k ce travail jusques aux enfants,
et dans les maisons des riches, I'on en chargait les esdaves.

'* Mais lorsque le luxe s'introduisit k Rome, et que les besoins
aagmenterent en proportion des richesses de quelques particuliers,
tandis que le peuple s'appauvrissait d'autant, les grands dont, I'ambi-
tion etait de gouvemer, imagin^rent pour conquerir Topinion publique,
de donner des fetes magnifiques accompagn^es de distributions de vivres
et de pain."

(p. 92.)— '' C'cst alors que les moulins k bras devinrent insuffisants.
Oc f ut force d'avoir recours k des entrepreneurs pour foumir k ces im-
menses distributions. Ces hommes avidep, 6tant dans I'obligation de
payer un grand nombre d'eselaves, et qui meme employaientdesmoyens

68 Proceedings of the Royal IHsh Academy.

criminals pour s'en procurer (Th6odo8e fit en 389 uneloi poor repiimer
ces d^sordres qui duraient encore de son temps (Lebeau, ' Histoire da
Bas-Empire,' livre 24), cherch^rent k diminuer le nombre des bras en j
suppliant par les agents plus puissants et moins couteux, que derait
leur foumir la m6canique, et Ton inventa les moulins d eau.)

" L'6poque de cette d^courerte est fix6e d'une mani^re pr6cise par
I'epigramme suivante, faite k cette occasion ( Anthologie manuscrite de
la Bibliotb^que imp^riale et m6moires de TAcad^mie des Inscriptions
et belles lettres, vol. ii. p. 408. Edit" en 8***.) *Femmes, occupees k
moudre le bl6, cessez de fatiguer tos bras ; tous pouvez dormir sL Totre
■aise et laisser chanter les oiseaux dont la voix annonce le retour
d'aurore. C6r^s ordonne aux Naiades de faire ce qui faisaient to6
mains ; elles ob^issent, cUes s'^ancent jusqu'en haut d'une roue et
font toumer un essieu. L'essieu, par la moyen de rayons qui I'entour-
•ent, fait toumir avec violence, la pesanteur des meules creuses qu*il
entrafne. Nous voiU, reveuus d la vie beureuse de nos premiers p^res,
et d r6cueillir sans peine les fuits des travaux de C6r^s.'

*' n parait d'apr^s cette 6pigramme d'Antipater, que I'usage des
moulins d eau n'a commenq^ que du temps d'Auguste, et^Yitruve, son
contemporain, fait dans son dixi^me livre la description de ces moolinSy
qui peut meme servir de commentaire d I'epigramme Gr^que. Strabon
(lib. 12) remarque aussi une machine alors fort rare, et dont il parle
comme d'une singularity, k I'occasion de la viUe de Gabires et du palais
de Mithridate. 11 n*est pas douteux que les moulins qu'on voit encore
-dans I'Asie mineure et dans toute la Gr^ce ne soient des copi^ de
moulins antiques, et par cela il est int6re86ant de les faire connaltre.
D'ailleurs il est probable que ces m^mes machines nous ont 6t6 trans-
mises par la frequentation que nous avons cue avec ces pays. On
peut croire aussi que leur 6tablis8ement chez nous, ne remonte qu*au-
temps des Croisades, et qu'auparavant nous ne connaissions pas les
moulins 4 eau, les moulins d vent, et les puits 4 roue," &c.

(p. 94.) — **Les crois^s, au retour de leurs expeditions d'outre-mer,
introdusirent dans leur patrie ces machines et bien d'autres, qui se
sont perp6tu6eB et perfection6es en raison de nos besoins et de nos
lumi^res. H n*en est pas moins curieux de voir d'oii Ton est parti, ce
que nous devons aux peuples orientaux, et ce que nous avons ajout^ 4
leurs inventions.

" J'ai dejA donn6 le dessin d'un puits gr^c (Lettres sur la Moree, &c.,
2nd partie, p. 41) qui m'a paru remplir, a pen defrais, le m^me object
que des machines beaucoup plus compliqu6es. Je pourrai en dire
autant de deux sortes de moulins que j*ai de88in6 k Ltmipsaki; ils sont

O'RKiLhY— Ancient Water-mills^ Native and Foreign. 69

d'nne telle simplicity, que Tinspection seule dee dessinB doit suffire
poor en faire comprendre le m6cani8ine. (Plate II.)

" Ce mecanisme dans le premier mouKn (pi. 5) De consiete qu'oD
une roue horizontale devisee en rayons creus^s de mani^e k recevoir
et i opposer le plus de resistance possible k I'eau. L'axe en fer de
cette roue traverse la meule inf6rieure, se fixe, au mojen d'un tenon
en forme de hache, dont parle Yitruve, au centre de la meule superieure
qu'il met en mouvement. [Note. — Ge fer que les Latins appelent eub^
Mu est nomm6 Cothh par les Arabes ; ils ont aussi donne ce nom anx
poles du monde, et se figurent que les spheres des cieux toument sur
eax et i Tentour d'eux, comme sur des pivots, &c. (D'Herbelot au mot
Cothb)'\.

" Jusque Id il n'ja rien de fort ing^nieux dans ce m6oanisme; mais
cequi paiuit I'etre davantage, c*est d'avoir profite non seulement avec
discemement, du pen d'eau dont on pent disposer, mais encore d'en
avoir double Taction. Si Ton avait eu de I'eau eu abondance et dans
tous les temps, sa chute seule aurait sufi pour faire mouvoir une roue
rerticale, comme celle de nos moulins ; mais il s'agissait d'obvier 4 sa
rarit^ dans de certains temps, et de se d6bara88er saus peine de sa trop
grande affluence dans d'autres temps. A cet effet le canal est construit
de mani^re k ne contenir que la portion d*eau strictement n6ces8aire.
Le trop plain se diverse avant d'arriver a son extremit6, au moyen de
vannes indiqu6es dans mon dessin.

"A cette extremity du canal on a adapte une longue caisse en forme

de pyramide renvers6e; elle est fonn^e de planches 6paiBses, assemblies

et retenues par de fortes traverses. Les joints sont gamis d'6toupes

et goudronn^s, suivant I'expression de Yitruve, comme les navires.

Cette caisse embrasse la largeur du canal. Son autre bout, qui pen^tre

dans le mar du moulin, se r^trecit beaucoup, -et n'a qu'une fort petite

ouvertnre dirig6e vers les rayons de la roue horizontale. L'eau se

pp$cipite dans cette caisse, qui lui o£b:e un plan incline d'cl peu prds 46^,

et ne pouTant s'echapper que par cette etioite ouverture on pent juger

Je la violence avec laquelle elle sort. En effet, la cumulation du poids

de TeaUy augment^e par la hauteur de sa chute, jointe k son volume

qui Be trouTO doubl6 en quelque sorte par le retr6cissement des parois

du canal, doit donner une tr^s forte impulsion au jet qui s*6chappe de

cette sorte d'entonnoir, son action se portaut toute entiere sur les rayons

qtt'elle prend en flanc, et qui sont disposes horizontalement et creuse^

de maniere 4 le recevoir directement, ce jet, dis-je, doit faire toumer

la rone avec une rapidite sufflsante et pent Stre m6me plus grande que

celle produite par une complication de rouages."

70 Proceedings of the Royal Irish Academy.

(p. 97.) — " Nous ayons vu un autre moulin oii Ton a soiTi le meme
BTst^me de force motrice et le meme m^anisme & quelques modifica-
tions pr^s.

'' Au premier coup d'ceil elles paraissent pen avantageuses, et n'en
remplifisent pas moins Tobjet des conatructeurs. Le canal on aqneduc
J eet eleT6 & plusieurs pieda audessus du niveau de la roue : mais le
conduit dans laquelle Teau tombe, au lieu de repr^senter une pyramide
renyer86e et pos^e diagonalement, oSre un cone tronque et vertical.

"L*eau y entre par le retrecissement du sommet, et apres avoir
rempli I'ampleur de la base s'en echappe par un tujau perce homon-
talement et qui se trouve de niveau avec la roue. De plus, au lieu de
perdre, comme dans le precedent mouHn, la surabondance, ou la cnk
des eaux, on Ta rendue utile en divisant le courant de mani^re 4 faire
toumer les roues des deux moulins jumeaux et accol^e Tun & TautK.
A cet effet I'aqueduc s'61argit d, son extremity qui est divisee en deux
parties 6gales par une cloison en planches epaisses ; ce qui forme deux
canaux paralleles, gamis de leurs vannes; et qui aboutissent auxdeux
cones ou Teau se precipitant en meme temps, fait agir le doublt
mecanisme."

[Translation,]

In Greece is met with at every step the tradition of ancient usages
and particularly of those of the mechanical arts. It is worth obfierrinc
that the greater part of the machines made use of in this country an-
of a simplicity such, that far from thus marking the first efforts of art.
they seem on the contrary to be but the result of reflexion, assisted by
a long experience. If by mechanics is to be understood the art of
increasing certain effects while simplifying the causes, it might K'
believed that the ancients understood it better than we do, particalarly
if one is to judge after the gigantic enterprises which they have
executed, with the aid of machines which might be called primitive
or elementary, and of which ours are but the complicated form.

(p. 90.) — We have said that necessity alone was the guide of the
ancients on the invention of machinery. Thus, that of mOIs for
grinding com only goes back to the century of Augustus. Befort
this period people were content with hand-mills, similar to tboee tc>
be still seen in Sicily (see Picturesque Tour in Sicily, by M . Hout L
p. 91 (1782-87)), and which are but simple household instnuneDt^.
These mills were quite portable, occupied but little space, and were
capable of producing at but slight expense enough flour for the food
of a family. There might even be employed for this work the labour

O'Rbili.y — Ancient Water-milbj Native and Foreign. 71

of children, and in tho dwellings of the rich slaves were charged
therewith. But when luxury became introduced into Rome, and that
the requirements became greater in proportion to the riches of certain
individuals, whilst the people became poorer by so much, the great,
whose ambition it was to govern, set themselves to dominate public
opinion by giving magnificent festivals, with distributions of goods
and bread.

(p. 92.) — It was then that hand-mills became insufficient. People
were obliged to have recourse to undertakers in order to be able to
supply these immense distributions. These men, covetous of gain,
were under the necessity of paying a great number of slaves, and even
employed criminal means to procure them. (Thedosius in 389 passed
a law to repress these disorders, which persisted even to his time
(Lcbeau, History of the Low Empire, liv. 24), seeking the means
of diminishing the number of hands by the use of agents more
powerful and less costly, such as machinery could furnish, and thus
water-mills were invented.)

The period of this invention is fixed in a precise manner by the
following epigram, made on account of it (see MS. Anthologia of the
Imperial Library and Memoirs of the Academy of Inscriptions and
Belles Lettres, vol. ii., p. 408, edition in 8vo). ''Girls, occupied in
grinding the com, cease thus fatiguing your arms ; you may sleep as
you list, and leave the birds to sing, announcing thus the return of
morning. Ceres has commanded the Naiads to do the work you were
engaged on : obediently they leap up to the top of a wheel, and cause
to turn an axle. The axle, by means of the spokes which surround it,
causes it to turn rapidly the weight of the hollow millstones which it
draws with it. Thus are we brought back to the happy times of our
primitive fathers, and gather, without labour, the fruits of the works
of Ceres."

It would seem from this epigram of Antipater that the use of
water-miUs had not commenced until the time of Augustus, and
Vitruvius, his contemporary, gives in his tenth book a description of
these mills, which might even serve as a commentary on the Greek
epigram. Strabo (Book 12) remarks also on a machine at that time
very rare, and of which he speaks as a singularity, when describing
the town of Cabires and the palace of Mithridates. It is not doubtfiil
that the mills which are yet to be seen in Asia Minor and all through
Greece are copies of these ancient mills, and for that reason it is of
interest to describe them. Moreover, it is probable that these same
machines have been transmitted to us by reason of the intercourse

K.I.A. PROC., VOL. XIV., 810. c] [6]

72 Proceedings of the Royal Irish Acadetnt/.

we have had with these countries. It may even be believed that their
establishment with us does not go farther back than the time of the
GrusaderSy and that previously we had no knowledge of water-mills,
wind-mills, or wells with chains of buckets.

(p. 94). — The Crusaders, on their return from their foreign
expeditions, brought home with them these and many other machines,
which have been perpetuated and perfected by reason of our require-
ments and our learning. It is not the less curious to see from whence
they come, how much we owe to the Eastern peoples, and what we
have added to their inventions.

I have already given (in the Letters on the Morea, &c., 2nd part,
p. 41) a drawing of a well which seemed to me to fulfil cheaply the
same object as machines of much more complicated design. I may
say the same of two sorts of mills which I had occasion to sketch
at Lampsaki : they are of such great simplicity that the mere in-
spection of the drawing should be enough to make the mechanism
intelligible.

This mechanism in the first mill in question (pi. 5) merely consists
of a horizontal wheel divided into rays or spokes, hollowed out so as
receive and to oppose the greatest resistance to the water. The iron
axis of this wheel passes through the lower millstone, fixes itself
by means of a tenant in the form of a H, of which Yitruvius speaks, to
the centre of the upper millstone which it puts in movement. [Note. —
This iron, which the Latins called " suhsous" is caUed " Cothh*^ by
the Arabs : they have also given this name to the poles of the world,
and they imagine that the sphere of the heavens turns on them and
round them as if upon pivots (D*Herbelot, voce ** Cothb ").]

So far there is nothing particularly ingenious in the mechanism,
but that which really appears to be so is the care taken, not only to
take advantage of the small supply of water, with great skill, but
even to double, as it were, its action. If water had been abundant in
supply and at all periods, its fall alone would have been sufficient to
furnish the movement to a vertical wheel, such as that of our mills ;
but it was requisite to meet the difficulty of its insufficiency at certain
times, and also to get rid of its excess at other periods. For this
purpose the canal is so built as to contain or carry the quantity of
water strictly necessary. The excess overflows, before reaching the
end of the canal by means of a sluice-gate indicated in the drawing.

To this end of the canal has been adapted a long box, having the
form of a reversed pyramid; made of planks sufficiently thick, ad-
justed and held together by means of strong ties. The joints are

O'Reilly — Ancient Water-milla, Native and Foreign. 73

•caulked and pitched in the same manner, as Yitravius says, as a vessel's
seams. This box extends over the breadth of the canal. Its other
extremity, which passes through the wall of the mill, is much narrowed,
and has but a small opening at its lower extremity, directed towards
the spokes or paddles of the horizontal wheel. The water precipitates
itself into this box, which presents a plane inclined at about 45^, and
finding its only outlet, this narrow opening, rushes out with a violence
that can be easily conceived. In fact the accumulated weight of the
water increased in action by the height of the fall, added to its volume,
which becomes, as it were, doubled by the narrowing of the walls of
the canal, should give a very strong impulse to the jet as it escapes
from this sort of funnel, its action exercising itself entirely on the
spokes or buckets which it acts upon from the side, and which are
arranged horizontally, and hollowed out so as to receive normally this
jet, which must, therefore, cause the wheel to revolve with sufficient
rapidity, and, perhaps, with more than could be furnished by a compli-
cation of wheel-work.

(p. 97). — ^We had an opportunity of examining another null,
wherein the same system of motive-power and the same mechanism
have been employed with slight modifications.

At first sight they seem to offer few advantages, but yet fulfil the
intentions of the builders. The canal or watercourse is raised several
feet above the level of the wheel; but the conduit into which the
^vrater falls, instead of offering the form of a reversed pyramid and
l>eing disposed diagonally, presents a frustum of cone and is placed
vertically.

The water enters it by a narrow orifice at the summit, and filling
tbe column — broad at its base — escapes from it by a tube which
penetrates it horizontally, and which is placed at the level of the
^vrheel. Moreover, instead of allowing the surplus supply, as in the
previous example, to run to waste, it is utilized by dividing the current
«o as to make the water give motion to two wheels, thus twinned and
joined one to the other. With this object the water-course is enlarged
towards its extremity, and divided into two equal-sectioned channels
lyy a diaphragm, or division of thick planks, each fitted with a sluice,
and which are in connexion with the two cones into which the water
pr^ecipitates itself at the same time, thus putting in motion the double
mechanism.

From this very complete description given by Castellan it will be
^v4dent that the mill he speaks of is of precisely the same nature aa

[6 2]

74 Proceedings of the Royal JmA Academy.

tbat described by M'Adam, and shown by bim to be of oommon
occurreDce on several of the countries of northern Europe.

Moreover, it is to be noted that he speaks of those mills bem^
prevalent over all Greece, and to be found in Asia Minor ; lastly, they
are in Greece, at least in the example described by him, more carefullT
constructed and more skilfully disposed than in those northern
examples already described. Unfortunately, as to the origin of the
mechanism or the historical side of the question, there is nothing but
what may be interpreted from the epigram of Antipater; and although
the wording of this would allow of its being understood to refer to a
water-mill with ^tson^a/ buckets, such as described by Castellan, the
terms employed are not sufficiently clear, nor sufficiently technical, to
prevent them being equally applicable to a mill-wheel with kortMontd
axis and buckets^ such as are usually employed at present in these
countries. Two further points are also worth calling attention to:
Ist, that occurriDg in the citation from the Montgomery MSS., p. 321,
given by M'Adam, where it is said i ^^th$ water -wheel was fixed st
ye lower end of ye axle-tree , and did run horizontally among ye water, a
small foree driving ity This most important remark proT» that the
former users of this form of water-mill had ascertained by experi«ioe
the advantage it presents in being able to work when drowned^ that is
when completely covered by water. Now this is precisely one of the
advantages claimed for the most modem and most perfect fomu of
turbine, working as they do with remarkably small heads of water.
2ndly, in the citation from Lockhart's " Life of Sir Walter Scott,"
describing his tour in the Shetland Islands, he describes the oonditioiis
affecting the rude mill he meets with at Loch Cleik-him-in. '* It is
divided," he says, "by a rude causeway from another small lou|:b,
communicating with it, however, by a sluice for the purpose of driving
a mill ; but such a mill ! " Now it is evident that the upper lake was
taken advantage of to retain or pond a head of water, while the lower
lake, being allowed to find its own level, gave thus rise to a difference
of level or head of water utilized for driving the mill, the sluice
being established to allow the overflow of floods or high waters. It
is evident, therefore, that the ancient users of these mills knew how
to create the necessary head of water, simply by cutting off a part of
a lake or other surface of water by a dam or wall, taking care to
insert a sluice to meet the requirements of the necessary occasiona
overflow. Eut this simple device, so easily understood in the case of
a lough, is equally applicable in the case of small tidal estuaries or
inlets, such as occur with great frequency on the northern or Canta-

O'Reilly — Ancient Water-millSf Native aud Foreign. 76

brian coast of Spain, where I had occasion to sketch such a mill and
to witness the most simple, practical, and ingenioas use of the tidal
force, a problem often talked of in scientific journals in these countries,
but as a matter of fact never yet here realized, to my knowledge, up
to the present. Anyone acquainted with that Cantabrian coast, or
who even consults Uie Admiralty map of the Bay of Biscay, knows,
or can learn, that it is very bold, rocky, and accessible, for even small
shipping, in only a few points. It is penetrated by a number of small in-
lets narrow and rocky, which receive the many streams coming from the
foot ranges of the Pyrenees, and which, therefore, can be traversed by
a wall of no great length, and consequently of not excessive cost of
construction. Such conditions present themselves both to the east
and to the west of GomiUas, a fishing-village about 22 miles west of
Santander (and not even mentioned in any of the modem gazetteers).
At a distance of about 2 miles to the west of Gomillas there
occurs such a small inlet, relatively narrow, and presenting high
banks on either side ; across this, at a short distance from low-water
mark, a wall had been thrown, so as to retain the waters of a high tide,
and thus create a water power ; on the seaside was found established
one of these corn-mills (Plate III.), with horizontal wheel and radiat-
ing buckets or arms, quite similar to those described and represented
in general appearance by the accompanying sketch taken on the spot,
during the course of an excursion made to the locality in the summer
of the year 1857-58, or thereabouts. A sluice established in the dam
wftU allowed of the intake of the tide, and its being retained for the
purpose of working the little mill when necessary. These mills
correspond in the simplicity of their structural arrangements with
ipvhat has been described of the Shetland and other Scotch mills, so
completely that nothing further need be added than a reference to the
sketch. They are mostly intended for purely local and even personal
grants, and one would rarely see more com coming at a time to be
ground than can be carried on an ass's back. The great interest of
them lies in their adaptation to the tidal rise and fall, and the
suggestion they convey to other people, is sufficiently important to be
taken notice of. It is probable that further research in these countries
^would determine the existence of many other such mills, and perhaps
allow of some information being gathered, as to the period of their
introduction which^the ignorant peasantry would, in most cases, be
inclined to attribute to the " Moors." I am obliged to admit, how-
ever, that I have never yet come across mention of them in any of the
several works on Spain that I have read, with the exception of

76 Proceedifigs of the Royal Irish Academy.

Townsend's ** Travels,"* in the second yolmne of which the following
occurs : —

(pp. 69, 60). — ** Journey from the Asturias to the Eeeurial. — As we
approached the confines of the principality the scene changed greatly ;
for, instead of soft and swelline: hills, covered with grass or clothed
with woods, scarcely anything was to he seen but stupendous rocks of
limestone — some in long ridges, rising perpendicular to the height of
two or three hundred feet, others cragged and broken into a thousand
forms. In this route the way winds chiefly by the side of little rivers,
brooks, or torrents, till it has passed the summit of that vast chain of
mountains which separates the Asturias from Old Gastille; yet in
the midst of these stupendous mountains a few rich valleys intervene,
each with its little village, in size proportioned to the extent of
land susceptible of cultivation. In the ravines through which we
passed I observed that all the mills have horizontal water-wheels.
These grind the com very slowly, being fed by single grains; but
then to compensate for this defect, they place many near together,
and the same little stream, having conununicated motion to one wheel,
passes in succession to the rest. These are well suited to a oountiy
abounding with stone for building, where water runs with rapidity
down a steep descent, and where despatch is not required " : sinoe,
however, Comillas is hardly known to geography, its environs and
the coast along may be equaJly ignored by travellers.

From those different examples of this class of water-mill it maybe
fairly concluded that their origin can hardly be ascribed to the Banei,
finding them, as we do still, on the Gantabrian coast and in Greece. On
the other hand, nothing so far gives us any clue to their origin or intro-
duction into the Mediterranean countries. There is fair grounds for
presuming that they really come from the East or from Asia lCinor>
where all the arts were developed to so high a degree from the very
earliest times known to history, and that their remains or presence in
these countries and in the north of Spain and in these isles may be
regarded as offering testimony of the frequency and continuity of
the commercial relations between the southern trading peoples and
the '* Hyperboreans " or northern races inhabiting these islands, Hol-
land and Scandinavia. Ground is therefore furnished for research in
that direction, and, mayhap, it would be neither fruitless or unintenst-
ing. It may at once be pointed out that the Senehue-Mdr contains
some notices of mills which are worth quotation. In vol. i., p. 135,.

^ Three yolumeo. Published in London, 1791.

O'Rbilly — Ancient Water-milb^ Native and Foreign. 77

the eight parts of the mill aie referred to under the heading ^^DistresSy*^
" for the eight parts which constitute the mill, the ' spring,' the miU-
raoe, 'the land of the pond^' the stone, 'the shaft/ 'the support-
ing stone,' 'the shaft stone,' 'the paddle wheel,' 'the axis,' 'the
hopper (cup oomla),' so eaUed because ariginally the bond-maid was
bound to mind it." These different parts as mentioned fit in fairly
with those pertaining to the milh with horizontal wheel already
described.

At p. 141 this is commented on as follows : — "/br the ei^ht parte
which eomtitute the mill, i.e. about the eight parts which are necessary
to the mill, as we shall explain hereafter. ' The epring^' i.e. from
which water c^mes, i,e. the water which is drawn from the spring
rests on the land of the pond. ' 27ie millraee^* i.e. from the spring to
the pond, ' The land of the pond,^ i.e. they are the first requisite, i.e.
which is at the head of the ' en,' t.^. the water. ' The atone, i.e. the
second requisite, i.e. the upper stone. ^ The shaft,* (mol), i.e. the
thiid, i.e. this is its own proper name. The ' supporting stone,* i.e. the
fourth, i.e. the lower stone. The ^ shaft stone,* i.e. the fifth, t.^. the little
stone which is tmder the head of the shaft, and on which the shaft
tunis. ^ The paddle wheel' {^eireel*), i.e. the sixth, i.e. Qdar a eel*)
OTer its paddle the water flows. ' The axis,* i.e. the seyenth, the
burden <rfthe shaft is on it, i.e. the *ganuel.* ' The hopper* (cup)
i.e. the ei^th, because it drops the com out of itself into the upper
stone, i.e. the *tual,* i.e. the perforated iron. The ^ Comla,* i.e.
they are all in plaee ofe^ bond-maid to a person, i.e. the whole mill, t.#.
the mill common to them aU. Ihr the hond-maid was hound to mind it,
i.e. for she was bound to mind everything of these which a person
wished ; or everything that one has, which is worth a ' eumhai* is
entitled to a gate (' Comla*) to protect it, i.e. the whole mill, i.e. by a
gate (' Cewda '), the restitution of which should have a stay of one
day; because tiie bond-maid (' Cumhal*) is bound to protect it, and
one of its parts has a stay of one day, i.e. by a gate (' Comla '), the
value of which is four pennies for every man in the place. If both are
not supplied, it is foil honour price ; if less, it is half honour price."
There are several entries in vol. iii. relative to " miUe ** of which the
following is interesting, p. 391 : — " And when a man has the site of a
kiln, or of a mill of rightful land, or when he shall purchase, such it
makes a native freeman of him."

Also similar entries in vol. iv., but not bearing on the question of
construction or form.

In the preface to volume i. of this work (1864) the origin and

78 Proceedings of the Royal Irish Academy.

antiquity of it is elucidated and commented on, and it may be gathered
therefrom that certain parts of it were of high antiquity, *' sack as
the first judgment respecting distress," by Sean, son of Aighe, who is
supposed to have flourished about 100 b.c. It is reasonable to anume
that the matters forming the subjects of the laws were themselYes still
more ancient, and that the mills referred to therein were of this dass,
and, consequently, it may be inferred that water-mills of the style to
which the ''eight parts" already mentioned may be considered u
belonging were already ancient in Ireland 100 b.c. This inference
would certainly be in harmony with the details contained in Messrs.
Bennett & Elton's " History of Commilling," vol. ii. (1899), "Water-
mills and Wind-mills."

They examine and discuss the references to be found inthedaoical
authors with reference to water-mills, including the epigram of Anti-
pater of Thessalonica (to whom they assigned a date of 85 b.c., while
Smith, in his '' Dictionary of Greek and Roman Biography," places
him between 10 b.c. and 38 a.d. ; and Hafer in the '* Nouvelk
Biographie g6n6rale " indicates merely that he liyed under Augastns
and Tiberius, t.^.^between 63 b.c. and 37 a.d. They state (p. 5), ''The
extended researches of modem times have, perhaps, brought us into
closer view of actual facts ; the sum and substance of these, as wiU be
shown, establishing water-milling as of Greek origin, at a period only
shortly preceding the birth of Christ." At p. 6 they describe the Greek
mill, horizontal type, and cite the epigram of Antipater as the earlieet
known allusion tiiereto. They give two translationB thereof one by
Tennant, in verse, as follows : —

« Te maids, who toiled so faithful at the mill.
Now cease your work and from those toils be still.
Sleep now till dawn, and let the birds with glee,
Sing to the ruddy mom, on bush and tree.
For what your hands performed so long and tme,
Geres has charged the water-nymphs to do."

'' The epigram is to be taken as recording the invention of water-
milling about 85 B.C."

They also examine and discuss the mention made by Strabo as to
the water-mill having existed at Cabira in the time of Mithnilat^ of
Fontus, and called by him " hudraletes " without farther descriptMo*
and which word is agreed on by commentators to indicate a water-milL
[They add as a note, " In some Latin translations of Strabo this woid
is stated to be omitted, as in that of Jan8en"( Amsterdam, 1652, iL,196V^

O'Reilly — Ancient JFater-milkf Native and Foreign, 79

Strabo does not distmctly claim it as a novelty, nor yet as a Cappa-
docian invention. Page 8, tbey examine the mention made in Pliny's
Natural Hist, xviii. 23, and point out how doubtful is the true
meaning of the passage. Page 9, they say, ^* In classic times no
evidence occurs indicative of the nature of the Greek water-mills ; and
in modem ages its existence has been almost entirely overlooked.
Still, there are at hand abundant means not only of proving its existence
as above, but of judging, from other sources, of its form and construc-
tion." Page 10, ''Its use spread throughout Europe, till about the
4?ighth century it was generally superseded by the larger and more power-
ful Koman water-mill, and at the present day in Europe and Asia either
the mill itself is found in use, or its prehistoric relics testify to its
former existence. In Greece it survived till late in the Middle Ages.
The sixteenth century French naturalist and traveller, Belon, saw at
Mount Athos, in Greece, mills driven by streams no thicker than a
man's arm, the wheel small, and 'made in a different manner to
ours,' but, nevertheless, capable of turning millstones as large as
might be desired. ' In this mountain-mill overlooking the JSgean
Sea, with its water-wheel differing from that of the French mills, we
may, doubtless, recognise the still perpetuated, primitive little
machine which evoked the wonder and inspired the ode of Antipater
of Thessalonica.' "

As on Mount Athos, so in the Holy Land, on Mount Lebanon, and
Mount Carmel, the same little mill was seen in 1668 byB'Arvieux, the
politician and traveller. ** The mills on Lebanon and Carmel bear a
great resemblance to those found in many parts of Italy. They are
exceedingly simple and cost little. The millstone and wheel are
fastened on the same axis. The wheel (if it can be so called) consists
of eight hollow boards, shaped like a shovel, placed across the axis."

'' Italy also is thus seen using the mill as in the time of Pliny."

(p. 11.) — In France the mill is described by Paul Henzer in
1588 (Itin. Gall. 56, 262): '*0n the €hu*onne they have a curiously
made mill, in which the wheel is much smaller than in ours, and has
a shaft inserted in the centre of the floats or vanes, which revolve
with great rapidity." " The wheel is not set perpendicularly upon the
water, but moves horizontally in it. The millstones are much larger
than ours, and are composed of so many pieces, skillfully joined
together, that one stone is estimated to be worth a thousand crowns."

At p. 12, they give details as to the ** Norse milV^ In Korthem
and Western Europe, and in Asia, the primitive mill, with its hori-
zontal water-wheel of Greek type, has been in general use from

80 Proceedings of the Rcyal Irish Academy.

prehistoric times and in some places surviyes still. The mill had early
become established in Britain. There seems to be no remaining
evidence of its nse in England; but there can be no donbt that
when the mill was equally common in Scotland, the Isle of Man, and
Ireland, it was also ordinary in this part of the kingdom.

(p. 13.) — In Ireland, over half a century ago, the discovery of
relics of water-mills of remarkable form led to a systematic exploration,
which eventually established the fact that the Norse horizontal mill
had been extensively in use there from, at all events, the seventii to
the eleventh century. The Irish laws ascribed to King Cormac, of the
third century, as well as certain traditions of the same king, seem to
refer to these mills; but the actual date, both of enactments and
legends, is so extremely doubtful as acarcely to warrant their aooept-
ance as evidence of any Irish mills existing in the third century. The
Kilkenny Archseological Society, under whose auspices the investiga-
tions were carried out, found the black oaken remains of these ancient
machines in the dried-up channels of old streams, covered eometimeB
with turf, and sometimes with beds of clay from 6 to 10 feet thick,
the clay having evidently been purposely placed upon the mill, in some
time of rapine, to conceal them from marauders.

At Ballymartin was found, at a depth of 6 feet below the sutfaee,
a framework of black oak, placed across traces of an ancient water-
course. It measured 1 1 feet by 6 feet, and supported a flooring of
boards 2 inches thick, and some of them 3 feet wide — all having been
evidently dressed with the adze. At Bramblestown, near Crowran, in
deepening a river-course, a similar platform framework, together with
fragments of millstones, was exposed. One of the latter had been
2 feet 5 inches in diameter, and the marks of the position of the lynd
on the upper stone was still apparent. (Kilk. Archaeolog. See., i. 164.)

(p. 14.) — Discoveries at Bantry and Mallow revealed the aamie
general features ; but here also were found water-troughs of black
oak, about 12 feet in length, which had evidently been need for the
purpose of conveying water to the mills to create a fall. The oaken
shaft or spindle was also found. In the neighbourhood of these dis-
coveries is one of the raths or enclosed prehistoric camps. Two pair
of millstones here were neatly finished and well faced, the upper
being 2 feet in diameter, and Ij^-inch thick at the eye at the centre.
The stream was very small, and had a fall of only 5 feet. At Sbanna-
cashel, Co. Cork, Uie mill seemed to have been burned down ; but
on the floor were a pair of millstones, a wooden shovel, and th«
shaft of a wheel. The upper stone was 8 inches thick and 2 fert

O'Reilly — Ancient Waier-mtlls, Native and Foreign. 81

in diameter, the nnderstone being only 3 inches thick at the centre.
The shaft, about 6 feet in length, was rather of an ornamental
character, and contained at the lower end a series of mortises for the
reception of the water-vanes. The mill was found buried deeply
under turf. In 1838 three relics of the floats or ranes of a Norse
mill-wheel were discovered at Banagher, Co. Berry, all (except one
of yew) being of black oak. They were of scoop-like shape, the
dished-end serving to catch the force of the stream more effectively
than a flat board. Each float was 14 inches in leogth, perforated
and fitted with a projecting ledge at the narrow end for fixing into
the shaft.

(p. 15.)^0ne of the most complete relics was described, in 1856,
by Mr. If 'Adam {Ulster Journal of Arehaology, 1856, p. 6).

(p. 16.)— In Scotland the mill has long been known under the
designation '' Norse mill." In the Islands of Golonsay and Oronsay
the small meal-mills built across streams, and now driven by ordinary
vertical wheels, were anciently worked by horizontal Norse-wheels,
and known under the term '' Muilean-dt^h" or black mills. One of
them has been pointed out at the farm of Ballerdomin-mor. The
sides of the small stream had been built up with dry stone- walling for a
length of 8 or 10 feet to a height of about 4 feet, with a circular
recess, in which the horizontal wheel turned. The stream was
bridged by four or five long undressed stones, upon which the mill
hoist had been constructed, and in one of these remained a portion of
the hole through which the spindle from the wheel beneath had passed
to the millstone. Other ruins are mentioned at Machrines, Bulnahard,
XTragang, and Aidskinnish. No information could be obtained as to
tbese latter water-mills having been utilized for grinding meal during
living memory, though it was said they had been to some extent
utilized for bruising malt for making smuggled spirits. Throughout
the district the Norse-mill, either in use or in ruins, abounds.
Wherever a small stream runs rapidly down to the sea may be found
a series of these little mills at no great distance from one another
(Soc. Scot. Antiq., 1888, 292). In some instances a double miUrace
running under the structure with a wheel and pair of stones at each
end of it, the mill, of course, possessing no cog-gearing, and a separate
wheel being thus always required for each pair of stones.

(p. 19.) — ^In Lewis and the Shetlands the Norse-miU is stiU com-
mon, being continued, says Professor Mitchell, rather from choice than
from necessity or ignorance. The same authority has given an excel-
lent description of these curious survivals of bygone times. The mills

82 Proceedings of the Royal Irish Academy.

are small and entail no great expenditure either of building or working.
They are convenient and easy to operate, and, though grinding but
slowly, are amply able to meet the small wants of the country-side.
<< Many of the people who build these mills know as well as any of
us the general superiority of an overshot water-mill and the unfitnesB
of the wheel they use to do anything more than the small amount of
work which they require of it, and not a few of them thoroughly under-
stand the waste of power in the mill ; but, to use the words of one
of the crofters, * If I get all the power I need from the bum, as it
flows past, where is the foolishness in leaving the rest unused?'" —
** Past and Present," Mitchell, 1876, 39.

At p. 20 is given an illustration of a Shetland mill (exterior
view), as shown in Mitchell's sketches, the little hurst of timber,
v^oofed with thatch or turf, is of merely sufficient size to contain the
^11. There is no resident miller. The door usually stands open to
all comers, precisely as in the ancient laws of Bohemia, together with
the Church, the Court, and the Hall, is stated always to do.

(p. 22.) — The same type of mill is described in a paper read bj
Mr. James Jardine to the Hawick Archseological Society, to have
abounded in that district, a list of no fewer than fifty-one being
enumerated within a radius of about eight miles. The usual diameter
of the stones was from 2j^ to 3 feet, and the upper was usually concave
on the lower side. When the controversy as to the identity of tbe
early Hibernian mills was in progress, Mr. K. Chambers, who bad
then recently visited Norway, recognised the type as that of the
horizontal mills of that country, and published the fact in " ^ TUa- m
Norway, ^^ in his popular Journals The Norwegian Korse-null is still to
be found in ordinary use, '* housed in structures as rude as may have
been that seen by Antipater nearly 2000 years ago."

(p. 23.)— Mr. E. C. Hart (Eobinson & Son, Eochdale) remarks of
the Norwegian mills : — ** In western Norway we find many of these
little mills, in all sorts of places. The spindle is made out of a pine
tree, with verticAl teeth, there being paddle-blades at one end, and
stones at the other."

(p. 24.) — In Eoumania (valley of the Danube) they have been
seen recently at work by Mr. Wilson Marriage, of Colchester, who
entertains a high opinion of their value for the kind of work required
from them. Mr. Marriage, in a contribution to Millifiy, accompanied
by a photograph, says : '* The Norwegian mill bears a s;bikiog
resemblance to the mills one sees in the Carpathians ; and I should
think that the mills of Norway and Roumania are almost identical in

O'liEihhY— Ancient Waler-milh^ Native and Foreign. 83

the method of working. A wooden upright has a home-made tnrbine
at the foot, and drives a single pair of stones. The mill is started by
shifting the wooden flume conveying the water of the mountain
stream on to the wheel. I saw several of these mills at work with
no attendant. The owner brings a supply of grain, fills the hopper,
sets the mill going, locks the door, and does not need to return for a
day or two. It is a far cry from Norway to the Carpathians, yet wo
see here two mills which might have been constructed by the same
workman, so similar are they in almost every detail, from the
foundation of rough stones to the ' log cabin ' mode of building.''

(p. 26.) — In the ** Great West " of China the same horizontal
mills were seen in frequent use, within the last two or three years,
1895 to 1899, by the traveller, Mrs. Bishop (Isabella L. Bird), p.b.g.s.,
who, in a recent communication to us, states that she saw them in
large nnmbers, especially on the great Ching-tu plain, where no doubt
they have been in continuous use from very primitive times.

Starting in the present paper with the occurrence of this form of
horizontal water-mill in Ireland, during ancient times, and following
the descriptions of the different authors cited, it has been found to
exist or to have existed and been in use in Scotland, the Shetland
and Orkney Islands, in Sweden and Norway, in France, in Spain (on
the northern coast), in Italy, in Roumania (in the Carpathians), in
the Korea, the Holy Land, and Asia Minor, and, finally, it is last seen
in western China, by Mrs. Bishop, a most intelligent and observing
traveller. The writers last cited, Messrs. Bennet and Elton, seem to
rely much on the epigram of Antipater, as placing the invention of
tliis form of water-mill in Greece, and seem disposed to date the
invention at about 85 b.c. But taking into consideration the wide
extent of Europe and Asia, over which its former and recent presence
has been determined, and the still more important fact that it has been
frequently met with by Mrs. Bishop on Uie western plains of China
in her very recent travels in that country, that these plains adjoin the
great central plateau of Asia, from which it is generally admitted the
earliest emigrations of a civilized race proceeded towards the west, it
is not exceeding the limits of the probable in presuming that the
^vrater-mill in question formed part of the industrial appliances
developed by the people who inhabited the central plateau and its
dependencies, and that it passed with the successive hordes of
emigrants in their slow march towards the west, and must therefore
l>e considered of very high antiquity, much more ancient than even

84 Proceedings of the Royal Irish Academy.

the introduction of the arts into Greece. To substantiate this
presumed antiquity, it would be necessary to bring forward farther
examples from the three eastern countries, and thus proTe the
connexion of the different links of the chain. Meanwhile every new
example, if carefully described, helps in the elucidation of this veiy
interesting problem, and must be considered as having a certain
historical value. As bearing on the question, and as tending to
illustrate one of the details, there is submitted herewith a stone in
which has been mechanically worked a conical cavity, and which has
all the characters of a socket-stone of one of these water-mills. It is
the property of the Be v. F. A. O'BeiUy, cc, at present administzator
of the parish of Eillala, County Mayo, and he has very kindly
entrusted it to the author of the paper, giving at the same time the
following details : — *' The stone vnis found about eight yean ago at
Ardmore, in the Mullet, about three miles from Belmullet, at the
bottom of a cut-away bog, that is to say on the graveL" It has been
in Father O'Beilly's possession for the last four years. I had the
occasion of seeing it in 1898, and was much puzzled to account for the
fineness of the lines of wear at that time. Sir John Evans also sair it,
and pronounced it to be a socket of a water-mill, and subsequent con-
sideration of the whole question leads me to the same condusion. I
have therefore prepared photos of it, and made a section to accompany
the present paper. (Flate IT.) The material of the stone is pure
white crystalline quartz, and the shape shows that it is a water-won
pebble or cobble, such as might be found on a sea-shore or in the bed
of a mountain torrent.

[ 85 ]

VII.

THE CISTS, DOLMENS, AND PILLARS, IN THE EASTERN
HALF OF THE COUNTY OF CLARE. Br THOMAS
JOHNSON WESTROPP, M.A.

[PlaTKS V. AMD VI,]

[Read Ap&il 14, 1902.]

So important are methodical field surveys to antiquaiies, and so
few persons have even attempted to treat exhaustively the early
remains of any of the counties of Ireland, that I venture again to
trespass on the time of the Academy by further notes on the dolmens
oi the county Clare. The present paper is an attempt to give the
position and description^of all the oldest sepulchral remains, cists
and pillars of rough stone, in the baronies of Upper and Lower Runratty.
I do not anticipate that it will prove exhaustive ; it is still probable
that other cists may be found in the northern portion of upper
Bunratty. There, among a network of fields with rocks and boulders
(very like dolmens when seen at a little distance), some actual cists
may even yet be found. It is an undulating country difficult to
examine without passing through every valley, if not through nearly
erery field, and on this account this paper only clauns to contain de-
scriptions (or notes where the monuments have perished undescribed)
of the eleven dolmens in Upper Bunratty, and seven in Lower Run-
nitty, as marked on the maps of the Ordnance Survey of 1843, together
with ten which I have been able to add to the list of the first barony
during a series of researches from 1870, but more especially since 1892.
I hope to continue these notes to include the other dolmens of Clare.
This {Mtper is a continuation, or rather an expansion, of one — ^' The
Bistribntion of Cromlechs in the County of Clare" — read before the
Academy in May, 1897, and (as any detailed descriptions of the Bun-
ratty dolmens in Mr. Wm. Copeland Borlase's book, <' The Dolmens
of Ireland," were from my notes), I must here ask forgiveness for
any repetitions needful for the completion of this paper.

86 Proceedings of the Royal Irinh Academy,

Sections I akd 2 — The Baronlks ojf Buneattt.

The^ dolmens in the district of Bunratty have been strangely
neglected by previous antiquaries. None have been noted in the
Ordnance Survey letters ; in either of Miss M. Stokes' lists of Irish
dolmens, or in Canon Dwyer's " Diocese of Killaloe." Mr. Jam»
Frost only notes Croaghane and its destroyed neighbour.^ Mr. Borlase
gives my notes and illustrations of the remains at Caheraphuca, Caber-
loghan (with the cists wrongly located as in Moymore in the baronv
of Upper Tulla), Rylane, Knappoge, and Ballinphunta. Plans an*
given of the three last.

The baronies of Bunratty extend from the borders of Ghdway to
the Shannon. The northern extremity is a grassy and hiUy district
abounding in small lakes, and the frequent occurrence of the place-
names of "Derry" and **Durra" confirm the allusions in earh
history, showing that it was for the most part wild and wooded.
South of this lie large tracts of bog, then a region of crags or grassy
districts (in part well cultivated) with many forts of earth and stone.
Along the south lie the rich corcasses of the Shannon.

is conventionally divided from the barony of Tulla, but for
archaeological purposes the whole district may be counted as one froxL
the Fergus to the hills of Slieve Bemagh and Slieve Aughty.

The tribal arrangements even as recorded in the earliest l^end
and history are probably too late to have any bearing on the dolmeos.'
We may briefly note that Lower Bunratty nearly corresponds to the
ancient Tradree (Tradraighe), and is dearly marked by the tvf
streams — the Rine or Gissagh* flowing into Latoon creek to the north,
and the Owennagamagh or Eaite to the east. This ract, as notec

» " History and Topography of Clare" (ed. 1898), p. 11.

3 The age of the smaller cists is very doubtful. Some may be poat-ChiistiaE.
The '* Tripartite Life of St. Patrick " (ed. W. Stokes, p. 123) mentions a " Gkiix'*
Grave" 120 feet long, dating 100 years before the Saint's mission. Acootdins ti:>
the Leabhar na hXJidhre {fievue Celtique, vol. xiii., 1892, p. 64) Fothaidh Airgtbeacii .
King of Erin, killed by Caeilte in a.d. 286, was buried in a cairn in ** a chert c-
stone." The alleged erection of the Clochogle dolmen, near Ballina, in the sixts
century, is not supported by the original narrative, and in any case could only i»?i?
a secondary burial. See Mr. H . T. Knox in Journal of Royal Society of Antiqoira^
of Ireland, vol. xxvii. (1893), p. 430, and vol. xxviii., p. 284.

3 The nameU ** Misagh " in MSS., R.I.A., 24 D 10, a poem " on theFranoKsks
Monastery (Quin) on the Misagh." The stream is " the Gissagh at Quia'* tr
Hugh Norton's account of Clare, 1696, MSS. T.C.D., I. I, 2, p. 236. He< '
the Sixmilebridge river " the Kney." The ** Goasogh " and ** Goaseogh"
in early seventeenth-century maps.

Wbstropp — CisiSy Doltnensy and Pillara ofEa%t Clare. 87

before, was the first portion of county Clare oyerrun by the Dalcassians
under Lnghad Meann, King of Monster, before a.d. 370, and so formed
the mensal land of the kings of Thomond, the O'Briens as named in
later times. Probably on this acconnt it was seized by the Norsemen
in the tenth century, and by the Normans in the thirteenth. It re-
mained the special appanage of the Earls of Thomond till 1712, and
was then sold in fee-farm to yarious English families. No dolmen
remains, probably from Tradree having been so long under cultiyation.
The northern portion of this barony (with portions of the upper
barony, and of the barony of Tulla) forms a rather bleak plain, '* the
beautiful cold Magh Adhair."^ It was a legendary settlement of a clan
of the Huamorian Firbolgs ; and its name contracted from a district to a
townland (Tuanamoyri'in 1584 and 1685), then to two fields ''Moyri,"
or " Moyross Parks " in 1839, and to a single field *< Moyars Park " at
the present time.

The oldest allusion to any dolmen in Clare is to that of Enot^ka-
lappa, or "hill's bed," byThomas Dyneley in 1680 ;' there is, as already
noted, no detailed description of any in eastern Clare till 1897.

The dolmens are here described topographically, and, where pos-
sible, in groups; the townland names are always giyen, and are
followed by the sheets of the Ordnance Survey maps, which, with the
sections of each sheet on that of 25 inches to the mile, are given in
brackets.

We may classify the remains in the baronies of Bunratty as — (1)

simple dstS'Kilvoydan, Toonagh (three), Caherloghan (six). Bally*

hickey, Monanoe (site), Ballymacloon, Enocknalappa, Drumullan

^northern), Ballysheen (site) ; (2) cists, with two or more chambers —

Caheraphuca, Ballymaconna (?), Bylane (western), Ballinphunta ; (3)

dolmen with enclosure — Enappoge; (4) long dolmens — Ballyogan

(Jtwo^ one removed); (5) circle — Clooney (eastern); (6) enclosures of

blocks — ^Dooneen, Clooney (western) ; (7) doubtful and destroyed —

liylane (eastern), Clooney (blocks), Drumullan (southern) (site),

Kilcoraan (perhaps multiple chambered), Brickhill (site), Lacht(?)

i>ite} ; (8) pillars —Enocknafearbreaga (five), Magh Adhair. None of

the stones to my knowledge exhibit any sign of dressing, and only one

1 •« Circuit of Irebmd," a.d. 941, by Cotmacan Eigeu, p. 43.
s Xiuuumojre in •* Castle List/' 1584, MSS. T.G.D., Tuanamoree, 1655,
I'etty'* Surrey (Vallancey'a copy Tuanamoret).

9 Royal Hist, and Arcb«ol. Assoc. Ireland, vol. ix. (1867), p. 176, repeated by
yg^^ f*rt»t in **Histoi7 and Topography of County Clare," p. 543.

^•I.A. FHOC, YOL. VIII. (xxiv J aBC. c] [7]

CAHCRAPHUCA - («"'<>•*)

KILVOYDAN

BALLMACONNA

RYLANE

TOONAGH

BALLYHICKCY

KNAPPOGC

KNOCKNALAPPA

KILCORNAN

^ tt

tCAkt

ft art

5«.^.-gi -f-

BALLINPHUNTA

Flans of Dolmena in the BaroniM of Bunratty, Coanl7 Of-

Wbstropp — Ckts^ Dolmensj and Pillars ofEa%t Clare. 89

(fttKilcoman) of scribing, wliereas liammer-dresdng occuxs in seyeral
•of those made of limestone slabs in the districts of Bunen and
Inchiqnin.

(1). Cahbbaphuca, Inchicronan Parish (0. S. Sheet 26, I^o. 2). — ^A.
Teryperfect double-chambered dst made of fiye side stonesand two ooyeis
lies dose to the north west of the main road from Ennis to Gh)rt, not
far to the south of the Tillage of Cruaheen. It is in perfect preserva-
tion, and is a most interesting specimen ; but, like eyeiy other dolmen
(except the closely similar Ballinphunta), known to me in this county,
ithas been opened. There seem to be slight traces of the earth mound
in which it was once embedded. It is of the usual type, getting
aarrower and lower towards the last. Its axis (as is usual) lies E.S.E.
and W.N.W. ; and the main chamber, which is exactly 8 feet long
inside, tapers from 5 feet 5 inches to 4 feet 4 inches. The sides are

Fig. 13. — Caheraphuca.

4 feet high to the west and 2 feet 7 inches to the east, the lesser

and lower chamber being only 18 inches high at its eastern end

and 4 feet long. The block closing the western end measures 5 feet

3 inches by 4 feet 2 inches by 12 inches. The north block of the

larger chamber, 7 feet 8 inches by 13 inches thick ; the southern, 8 feet

long, 4 feet to 2 feet 7 inches high, by 13 inches thick. The blocks are

<yf the coarse and irregular gritstone of the district and exhibit no

^reeaing. A fine and picturesque hawthorn springs at the north side

near the junction of the two chambers, and is, I fear, slowly overtum-

in|^ the structure. The dolmen has been illustrated by Mr. Borlase.'

Of the defaced Caher, which gives its name to the townland, I have

failed to find any legend. It is possible that it was deserted at an

^early date, and the people (before the townland name was fixed)

believed that it was haunted by the phuca — ^that mischievous goblin

pony or goat which, even in this age of unbelief, is still a reality to

Bome of the peasantry. This connexion of the phuca with forts and

1 «<Dolmeii5 of Ireland," vol. i., p. 82. See plan, p. 88, fig. 1, tupra, and
PUtoV.,fig. 1.

[7»]

90 Proceedings of the Royal Irish Acadetny,

rocks is sufficiently marked in other places ; as at Carrigaphuca, ia
Cork, Glochaunaphuca in Kerry, the Dun of Glopoke in the Qaee&'s
County, and the dolmen of Poulaphuca in the Burren, and other
examples might be cited.

(2). KiLTOTDAX, Inchicronan Parish (0. 8. Sheet 26, Xo. 14).— A
defaced dolmen, which occupies a beautiful position on the snnunit of
an abrupt grassy hill, with a fine view of two lakes and the iried peel-
tower called O'Brien's Castle. To the north, at the foot of the hill and
near Durra House, are the very ancient graveyard of Kilyoydan,^ irith
a large basin stone and a dry holy well, and the remains of a ring fort

The dolmen is to all appearance an example of the very cuiioiu
(though scarcely credible) type which occurs near Louisburg, in Majo,
as figured by Mr. G. H. Kinahan, of which, as Mr. Borlase points out,
examples occur in Portugal. These structures are formed of slopiog
blocks overlapping till they nearly meet at the top, and then roofed
with small slabs. The section of the Kilvoydnn cist closely lesemUes
this type, as may be -seen in the illustration (p. 88, fig. 2, #if/yr«); bat
after careful consideration it seems more probable that the slipping
inward of the sides tilted the massive cover over the north side, against
which it now leans. The cell is 12 feet long, with single blocks to the
north and south. These measure, respectively, 9 feet 6 inches by 18
inches thick and 5 feet by 20 inches thick, being at present odIt
about 3 feet high. The cover measures 11 feet 8 inches long, aboat
6 feet wide, and 22 inches thick. All the slabs are of massive gntBtono
or conglomerate.'

(3) . Ballymacootta, Kilraghtis Parish (0. 8. Sheet 26, No. 1 0).— Not
far to the west of Kilvoydan three other monuments lie near Kilzag^
church, and with the first named may bo called the Kilraghtis groap.
The district is diversified and interesting, formed by a group of bw
rounded hiUs with a curious fortress-like outcrop of stratiOed rock at
Dromgloon. The church was called '' Kilrathusa " in the Papal tu*
ation of 1302, and is at present a plain building of the later fifteentb
century.'

The cist lies in the remains of a cairn on the slope of the hill to

1 Not the KiWoydan near Corofin described by Dr. HaoNamaia in Joanal
U.S.A.I., Yol. xzx. (1900}.

> See section and plan, p. 88, figs. 2 and 3, stipra.

' The fields in which these dolmens stand (we were told) were a few yaan a^
coyered by an assembly of myriadii of rats. The great meeting was hdd for ti^vvnl
days, and then broke up, and marched eastward in squadrons which must ha^^e
dispersed as they went, fur they could only be traced for a couple of miks.

Wrstropp — CUta^ Dolmens^ and Pillars of East Clare. 91

the north of the church. It seems to have two chambei-s, but is much
buried in the cairn. The principal cell is formed of two massive
blocks rising 3 feet above the stones and 7 feet long, being about
3 feet apart and tapering eastward. The cover is nearly level, and is
a strong slab of gritstone, over a foot in thickness and about 7 feet
square. Both the north side and the east end are deeply buried in
fimall stones, which nearly fill the interior.^

A large slab 6 feet 6 inches long protrudes from the cairn about
9 feet from the end of the cist, and may be the top of a second chamber ;
it does not seem to rest on side blocks. The dolmen is locally calleil
the " Lobba." The names of Dermot and Grania seem to have passed
out of local memory at this place.

(4). Balltogan,* Kilraghtis Parish (O.S. Sheet 26, ^o. 10).— On the
eastern slope of the same hill, in the adjoining townland of Ballyogan,
lie the foundations of a largo and massive stone ring-wall called
Gahereiny ; it has the remains of a souterrain, but is most completely
defaced. A fort-like knoll of rock juts up not far away, rising from
a cultivated field.

(a) The bohereen to the west of the caher did not exist when the
O. 8. survey of 1840 took place ; but the older labourers remember its
construction, and state that this led to the removal of a ''Giant's
Gi-ave '' very similar to and not far from the existing monument. It
was a '' long grave," as we learned from two independent descriptions,
each side consisted of five or six blocks, across the middle of which
rested a large slab 5 feet or 6 feet each way. There had once been '' two
otlier covers," but few remembered them at the time of the demolition.
Only for the positive statements of the men and the corroborative
recollections of the late Mr. Pierce O'Brien of Durra (who gave
me much kind aid when studying the dolmens of his neighboui'-
hood), 1 should have suspected some confusion with the existing
^'giant's grave" which it so closely resembled. I was shown t}ie
approximate site a couple of fields from the latter and to the nortli-
west. No antiquities were unearthed in the removal.

(Jb) The existing " long grave " lies nearly covered with grass

1 See plan, p. 88, fig. 4, 9upra.

' Ballyogan, in 1640-1668, was the residence of Maoilin M'Brody (MacBruodin)
aind his wife, Margaret Molony, whose son, the well-kno^ni monastic historian,
JiDthony '* Bruodinus,*' was bom there. It is called Gortnefunchin in the 1675
Sook of Sanrey at Edenvale. " Ashgrove *' in the townland is probably a mistrans-
l&tion of this name. We may note that the adjoining townland Beamaf uoshin is
^Iso called Ballyfinshan in the 1703 estate-maps of the Earl of Tbomond.

92 Proceedhuj9 of Vie Royal IrvA Academy.

and ftones in a tilled field to the north of the second hoheieen. Il»
axis lies K.K.E. and 8.8. W., and it consisted of fonr rows of stonesw
The complete portion is 24 feet long, and tapers eastward from 7 feet
4 inches to 5 feet 4 inches over all. The end slab remains to the
west. Bows of stones stood about 3 feet from and parallel to each md»
of the central enclosore. Only two remain to the north and three to
the south. It is eyen possible that the most western of the latter
belonged to the inner row. The longest slab remaining is 7 feet
6 inches long. If the minute sketch-plan on the Ordnance Surrey
map of 1840 may be trusted, two long blocks projected from the
southern side, and (apparently) a coyer remained in ittu. It is still
called, as on the maps, " the Giant's Graye/' but no fuller legead
suryiycs.

This monument is one of a yery interesting and wide-spread type,
though of less frequent occurrence than the cist type so common
among the dolmens of Ireland, Portugal, and Spain.' The long
grayes are closely akin to (in some case identical with) the all6ee
couyertes of France, and to passages in the tumuli in Scandinayia
and other parts of Europe, as far south as Sardinia. For comparison

with the Ballyogan graye, we
select a few others — Lachtneill,
Cork, 12 feet long, 3 feet wide
inside; Slieye-owen, Cork, 20
feet long, 3 feet to 1 foot wide,
with parallel rows of slabs at
each side ; Burren, Cayan, two
grayes, respectiyely, 29 feet by
5 feet to 4 feet, and 17 feet S
Pig. 14.— Ballyogwi. inches by 4 feet 9 inches to S

feet 2 inches ; Carry glass, Tyrone, 40 feet by 4 feet to 2 feet 3 inches,
with parallel rows of slabs at each side ; Coolbuck, Fermanagh, 33 feet
by 4 feet to 3 feet ; Proleck, Louth, 22 feet 6 inches long, 6 feet to
2 feet 6 inches wide; and Moiglisha, near Arklow, in Wicklow. With
these we may compare the ** Hun's Beds " (Hiinebed), in the district
of Drenthe, in Holland, which consist of long chambers, outside ot
which are parallel rows. Mr. Ferguson says these differ from those
of France and Ireland by haying closed ends ; but this is not the case,

i '< Dolmens of Ireland/' toI. i., pp. 29, 31, 202, 214, 321, 817, 414 ; vol. ii.
p. 639. Ferguson ** Bude Stone Monuments of all Countries, their age and uses **
(Ed. 1872), pp. 320, 366.

Westropp— C&fe, Dolmefis^ and Pilhrs of East Clare. 93

as we hare learned, even from tlie defaced " long grave " of Balljogan.
Similar structures have been also found under long barrows, as at
Kerlescant, which is 52 feet long and 5 feet wide, with rows of stones
parallel to its sides. The nearest example to Ballyogan is probably
the much more perfect long grave of Formoylemore on the brow of
that steep hill, up which runs the road from Broadf ord to Limerick.

I was much interested to see ''microlithic" representatives of
these "long graves" in certain churchyards in Kerry, where slabs
set on edge and covered with other slabs enclose the coffin. They
rise above the level of the churchyard, and are covered by miniature
cairns.

(6). RvL-iNB, Clooney Parish (0.8. Sheet 96, No. 16).— The two
dolmens at this place are not named on the map of 1840, but may be
found marked by two small squares to the south-west of a rath, and
near a second fort directly above the " E " of the parish name. They
were first noted by Mr. Arthur Gethin Creagh, of Fiaghmore, to
whom I am also indebted for particulars about the destroyed burial-
place and the hearths at Coolosluasta Lake.

(a) The more southern monument is called the '* Labba." It is a
cist about 23 feet long, and 1 1 feet 6 inches wide, and had at least
three, if not four, chambers. It has been illustrated and described
from my notes by Mr. Borlase. It probably consisted of a parallel-
sided enclosure, with an outer line of slabs loxaul it. The eastern
end is nearly perfect.^

(J) The second " Giant's Grave," as it is called, is greatly
defaced. It is of larger blocks than the "Lobba." Its southi
side Ues east and west by compass, and it tapered eastward. It lies
in a circular patch of stones and mounds, much overgrown, and
evidently the base of a cairn or tumulus. Some of the blocks are
5 to 6 feet long and 4 feet high. Near it is the almost levelled ring
of a small rath ; while a much more perfect and larger earth fort
occupies the summit of a low green hill to the north-west. It is girt
with a deep fosse, which frequently contains water.

B'ot far away to the north-east, in a field at the Donoghue's house,
and about 500 yards from Maghera cross roads, was found (on
Pebmary 4th, 1897)' an interesting early burial-place. The field
was dug ixp for the first time in human memory. Scarcely 2 feet

I •'Dohnent of Ireland," vol. i., p. 82 ; see plan, p. 88, fig. 6, »upr0.
« Journal R.8.A.I., vol. xxvii. (1897), p. 178.

94 Proceedings of the Royal Irish Academy.

below the surface lay a passage 10 feet long and 2 feet 6 incheB wide,
of rather small dry masonry roofed with thick sandstone flags. The
passage ran N.N.W. and S.S.E. At its southern end was a small
circular cell, domed in the usual way by courses overlapping till the
space could be closed with a Blab. Another slab was laid outside the
flank of the dome, as if to ease the pressure of the earth at that point
Fragments of a human skuU, including a portion of the lower jair
with six teeth, lay at the north-west end of the grave ; and a lew
yertcbrae in the last stage of decay lay along the passage. Around the
remains lay charcoal and burned earth, while other traces of a fleroe ^le
were apparent on the side walls. Nothing was found in the round
cell, nor were there any traces of either metal or pottery. On the top
of the covering slabs lay portions of the skull of a horse and " bones
of a goat or pig." Similarly, in an early cairn-burial near St. Cemin
de TArche, near Brive, in France, tbe bones of a horse lay outside a
cist covered by a cairn. In another cairn lay the skeleton of a woman,
the upper part of which was partly cremated. An oval hearth of
sandstone blocks, including a portion of a quern, lay 1 foot 2 inches
south of the Clare grave, and 4 feet higher than its floor. It was about
2 feet 6 inches under the sui*f ace of the field. Mr. Creagfa at once
wrote to me ; but when I next was able to visit Clare, in the following
April, the whole structure had been removed and stacked against the
fence of the field, and com was sprouting in the slight depresDon
that marked its site.

Some years before this, discoveries of some interest were made at
the little bog-pool of Coolosluasta Lough.^ The peat had evidently
grown in and greatly diminished the lake, and the Carrahan drainage
works considerably lowered its waters. Mr. Creagh found several
planks under 4 feet of good peat, which had been covered by about
6 feet of water. One plank was worn along the edges as if by a rope.
It was 5 feet long, and had a round hole cut in it. The other boards
were in fragments, and so soft that the hand sank into them. All

> Journal R.S.A.I., vol. zxv. (1895), p. 179. Fich and CoolaliiOBty appear ia
Petty*8 '< Book of Distribution" (1655), p. 2. The latter is noted by Canon Dvycr
fromKing's'^ChurchHistoryof Ireland/ Supp1.Tol.,p. 1047. *<0*Slual8tifraBCSn
Osluaisti'* ; and others, ** these were they who stole the horses, the mules, and the
aases of the Cardinal who came from Rome to instruct in the time of Domhnall Mhr
O'Brien, King of Munster; and it was on that account the Cowarba (t.#. thesaeoeMor)
of Peter sold the rent and right of Erin to the Saxon.** If so thia little natnhj ifot
is one of the most historic sites in the world, but the story needs for better aBtib»-
rity to support it.

Wbstropp— Cm^, Dolmens j and Pillars of East Clare. 95

around and under tlie plants was a mass of deer's bones, recalling the
popular interpretation of the townland name Fiaghmore, '^the big
deer." Directly beneath were found three human skulls, "two
females and one male." All the teeth were sound but much worn.
The lower jaws were, it is said, still in the sockets, and the arm-bones
with the shoulder-blades. Among these bones lay two long oak poles,
one neatly shaped with a sharp implement. 2^o metal was found,
but it may have simk in the deep peat below.

A fragment of a very neat gritstone quern is pi^eserved ; it was about
20 inches it diameter, with a ring 8 inches wide and a central hole,
with three concentric and rounded edges, each an inch wide. A straight
band crossed these, and girdled a small handle-hole, which did not pass
through the slab. I could not ascertain the locality of this " find."

Around the margin of the lake were found some twenty hearths of
gritstones; they measured about 6 feet across, with wonderfully fresh
charcoal and remams of pigs and goats, the long bones were broken
for the marrow. Great stems and roots of bog-deal lay everywhere ;
all had fallen towards the east ; the tops, and in some cases the roots,
had been burned, or in a few cases cut.

(6)» Cloombt, Clooney Parish (0. S. Sheet 34, No. 7).— In the pic-
tnresque demesne of Mr. Joseph Hall, and at no great distance from
the fifteenth-century castle and church, are two remarkable remains.

(0) The first is of a type not very common
in Ireland ; it occupies the summit of a low

'k ;^"

natural mound near a stream, and is much ^ ^Ba^ ^^ \^

oTcrthrown. Enough, fortunately, remains i^^\jF /,

to show the plan. The late Mr. Borlase* /pHj .^ ^, Q

(who had not seen it) considered it a " boat- \/0i ' yii '

shaped enclosure" ; but, as I pointed out to ^ ^"

him when sending the plan, the structure

is a nearly straight-sided but not rectangular | [

oblong indosure, with the angles cut off. It VC^ ^ -'t}W^ ^

is formed of two rows of slabs, equidistant •5J*?*%!5^

side actually points E.S.E. and W.N.W., and -^ *-'- - V*

is 10 feet long; the northern is 12 feet, and

the remaining sides about 20 feet each. A ^^8- 15.— Clooney.

small entrance, with two side-blocks, opened eastward ; and a slab stood

in line with its northern jamb projecting at right angles from the inner

' *< Dolmens of Ireland,** vol. i., p. 82.

96 Proceedings of the Royal Irish Academy.

face of the ** western " side. Thirteen slabs of the inner and seven of
the outer row are in position, many others remain out of pkce. I repeat
the pkn from the earlier paper.^ The enclosure and knoll are thickly
planted, and there is no trace of a cist or any other structure in the
^larth.

(b) A second monument lies nearly eastward from the last in a
groye of fine beech-trees. It was a circle, once probably double, and
is much defaced. The western portion is, however, partly preserved*
Beginning at the south, and going westward, there are four outer
blocks, measuring respectively 12 feet 6 inches long by 8 feet by
25 inches, 9 feet 3 inches by 6 feet 6 inches by 24 inches to 30 inches
thick, 4 feet 2 inches by 3 feet by 1 1 inches, and 3 feet by 2 feet
4 inches by 12 inches. Inside there are two fallen blocks, one 5 feet
6 inches long. A single fallen and nearly buried block lies to the
north, and another, 5 feet by 4 feet by 21 inches, to the south. At
some distance outside the grove another set block suggests a third
ring.

(c) Three blocks are set in line, N.N.E. and 8.8.W., in another
grove, and possibly represent the remains of a third monument.

(7). KvocxirAFEABBREAOA, Clooney Parish (O.S. Sheet 34, Xo. 3). —
Near Classagh House, but in the townland of Knockunonra, is a low
cultivated hill, called Knocknafearbreaga. On it stands a line of five
pillars, lying N.N.E, and S.S.W. Noting Uiese from the north they
measure respectively — the first, 4 feet 10 inches high by 3 le^
8 inches by 1 foot 2 inches ; the second, 4 feet 5 inches by 1 1 inches
by 17 inches; the third, 5 feet 2 inches by 3 feet 10 inches by
10 inches ; the foui*th is broken, the stump being only 1 foot 6 indies
high ; and the fifth leans towards the north-west, and is 6 feet 2 inebea
by 2 feet 3 inches by 10 inches. There are said to have been two
others in human memory.'

Such alignments of pillars are not unknown in Ireland, SootLaad,
France, and elsewhere. Examples occur in Caithness, Wiltshire, and
Brittany, varying from a single line to the great group of Camae.
The Rev. S. Baring Gould, in his interesting account of those on l>ait*
moor, advances the ingenious theory that they are tribal monuments,
the stones varying from under 3 to over 6 feet ; and, on this suppoaitioii,
varying accoi*ding to the number or individual strength of each family.*

^ Proc. R.I. A., Tol. iv., eer. iii., p. 646.

» See Plate VI., fig. 1.

» **Book of Dartmoor" (1900), pp. 00, &o.

Wkstropp— (7w^^ DolmenSf and Pillars of East Clare, 97

In Ireland, we may note, for comparison, the alignment near Lough
Gur, in Limerick, and that in the townland of Reenaree and parish of
Kilnamartrj, Cork. The rows in the latter case are 4 feet apart, and
the highest gallan is 5 feet 6 inches high.*

Local tradition states that the Clare pillars were seven robhers who
** kept about the place in old ancient times." Now there was a saint
(possibly Mochnlla) at Tolla who was building the church there ; and
he was so busy, '* he had no time to cook his food." The holy man for-
tunately possessed '' a blessed bull," and used to send it with bags to
Ennis Abbey, where the monks used to pack a supply of cooked
prorisions for the church-builder. Hearing of this, the robbers way-
laid the faithful animal, and proceeded to ravage the bags. The bull
thereupon roared so loudly that he was heard at TuUa ; and the saint
stopped building, '< and he prayed and cursed at the one that was
biuting his bull." Kapid was the retribution — in the twinkling of an
eye, the seven robbers were ** struck and turned into fearbreags," or
sham-men, on the spot where they stood.

The legend is late in form, but has ancient equivalents in all ages
and counties ; we need only note some Irish and Scottish forms. We
find at lona, a '' Portanfhirbhreig," so called from a tall rock, supposed
to resemble a man's figure.' Another late Scottish legend, *' Fionn's
Enchantment,"* tells how Fionn and his lads had been hunting in a
snow-storm, and, while waiting for their ** bird-stew" to be cooked, a
haro ran into the house and kicked up the ashes. All the men ran
out after her, and followed her to a hut, which proved to be the abode
of a giant magician, named " Yellow Face," who lived on enchanted
boi»rs and human flesh. The giant called in the intruders to help him,
and, on their crossing tbe threshold, struck them with his rod of magic,
and ** they became pillars of stone ; and he set them on the north side
of the door to stop the sleety wind " (like the shelter slabs at the doors
of dochauns in Corcaguiney). The curing of Fionn, whose legs had
been burned off, and the disenchantment of the petrified youths, do
not bear directly on the subject of Fearbreags. St. Brendan is said (in
an Irish legend) to have saved a young man from murderers by
changing him into a pillar, and a pillar into his form. The villains

s ** Coik Historical and Archaeological Societ}'*8 Journal,'* toI. iy., ser. ii.,
289.

» ^daronan'a " Life of 8. Columba ** (od. Dr. Reeves), p. 429.
* Jierfte Critique, vol. i. (1870), p. 19G.

•98 Proceedings of the Royal Lnsh Academy.

stabbed the pillar and- cut off its head, and carried it away to another
place, "and still that stone remains."

At Kilross, in S1igo» two men endeavoured to steal a magician's
cow ; but the owner pursued, and, striking them with a wand, turned
them into stones, which are still shown. It is possible that the " cow-
stone" and ** thief -stone," near Gallcrus, in Kerry, commemorate
some such legend. In the ** Book of P'eenagh," when the Druids of
Fergna **do corrguineacht " against St. Caillin, Aedli Dubh, son of
Fergna, orders his soldiers to attack them. '* No," said Caillin, *' we
will not exercise human power upon them ; but it is my will (if it be
the will of my God of Heayen and Earth) that the Druids may bo
changed into stones forthwith." Theitjupon the Druids were imme-
diately turned into foiins of stone. Fergna, in his wrath at his son's
defection, and because '* his Druids were transformed into the shape
of stone columns," brings on himself the vengeance of Heaven, and
dies. The whole subject of Fearbreags, as John Windele has long
since noted, is much in want of elucidation. It crops up in Monaghan,
Cork, Clare, Tipperary, and, as we have seen, in lona and elsewhere
in Scotland. The name is applied to a stone circle near Kimalta
(Keeper Hill), in Tipperary, and to a cairn at Kilcolman, in Cork.*

Natural rocks, called Fearbreags, occur at Fanygalvan, in the
Burren, in the hills near Broadford, and at other places in Clare.

Mr. Borlase notes its connexion with the name and legend of the wolf
(breag) and were-wolf, and that wolf -names are connected with cairns
and tumuli in Germany (as, ^.y., the wolf hiigel) and Bohemia, aa well
as in Ireland.'

It is, however, possible that the pillars at Classagh, like those not
far from the Fearbrcga Ilock of Fanygalvan, form some long-forgotten
tribal boundary rather than sepulchral monuments, though » in the
historic period, they seem to have coincided only with townland
borders. It will be remembered how Cuchullain, when mortally
wounded in battle, went to drink at a lake. *' Now a great mearing
went westward from the lake, and his eye lit on it ; and he went to a
pillar-stone which is in the plain, and he put his breast-giidle round

* *• Brendaniana, or St. Brendan the Voyager in Story and Legend,'* Rev. De«i<
0*Donoghue (ed. 1893), pp. 16, 17. See alao ** Battle of Moy Leana,** p. 31 il:
<* Traces of the Elder Faiths of Ireland," Col. Wood-Martin, vol. ii., p. 214 ; ar..2
*' The Book of Feenagh" (ed. W. Hennessy and D. H. Kelly, 1875), p. 117.

' ** Dolmens of Ireland,'* vol iii., pp. 912-915. See also some ralnabU notes
on the name in Dr. Joyce's '' Irish Names of Places." Seties ii., pp. 411» 412.

Wbstropp — Ci8t8^ DolmetiSj and Pillars of East Clare. 9&

it that he might not die seated or lying down,"* *« Cormac's Glossary "^
also has a sentence: *' They are not neighhonrs till their properties are
meared with boundaries of pillar-stones." A manuscript in the Library
of Trinity College, Dublin, says : * * Land is secured by the joint memorirf
of two territories — i.e.^ the ogam on the gallan," Unfortunately the
Clasaagh pillars are uninscribed.*

(8). Maqh Adha-ib, Clooney Parish (0. 8. Sheet 34, No. 12).— Near
this fully-described' place of inauguration, at the opposite side of the
stream horn the moat, cairn, and basin-stone, and in the townland of
Corbally, is a rude limestone pillar. It measures 6 feet 3 inches high,
by 3 feet to 2 feet 6 inches wide, and 10 inches thick. It stands in
line with the mote and its sloping descent and the cairn. Between
it and the stream is a shattered block set firmly in the ground, and
possibly the base of another pillar. Another stone lies near Drumbaun
fort, two fields to the west of Moyars Park.

(9). TooNAOH, Clooney Parish (0. 8. 8heet 34, No. 8).— Not far to
the north-east of Magh Adhair, near the same riyulet, is a group of
defaced cists. They are not marked even on the new map, but were
shown me by the Kev. J. B. Greer, of Tulla, They lie on the 1899
map at the apex of a practically equilateral triangle, resting on the
main road from Clooney to Tulla, between the bench-marks 121 and
120*4, and south of the road. The romains of two, if not of three^
de&ced cists lie in a furzy hollow near a small brook.

(a) Of the western cist only the sides remain ; the northern is still
standing, a coarse gritstone block, 7 feet 6 inches long and 4 feet
8 inches by 20 inches ; beside it lies a slab (the &llen south side
lO feet by 4 feet 3 inches by 20 inches); the axis of the standing
slab lies S.S.E. and N.N.W.

{t) Another thick block, 5 feet long and 16 inches thick, set north
and south, lies at a short distance to the north-east, and may be the
urest end of a cist.

(e) On higher ground, to the north of the last, aro the remains of
a little cist. The south side lies north-west and south-east, and is
about 6 feet long, 32 inches high, and 12 inches thick. The north
side has been much broken, and lies from 4 feet 4 inches to 4 feet

s Ktvus dUiquif vol. iii., p. 182. From the ** Book of Leinster," p. 178.
s «*ConDao'8 GloMary*' (ed. W. Stokes), p. 84, MSS., T.C.D., H. 3, 18,
230. For '« meer stones " see Notes on the Landnamabok (by Bey. £. T. Ellwood) ,

pi 69.

» B.8.A.I., vol. xMi. (1891, p. 463 n.), with illustrations. Proc. R.I.A., toI.

iir., aer. iiL,p. 66.

100 Proceedings of the Royal Irish Academy.

3 inches away ; only 4 feet 6 inclies remain — the rest is broken to the
ground. The west end is 4 feet 4 inches long and 3 feet high ; tlie
cist tapers eastward to 3 feet 9 inches internally, and the broken cover
lies near it.^ A short distance from the east end is a large round esini
of mossy stones overgrown with bushes. In this, near its noitheni
^ge, a large block is set east and west.

A line of slabs 3 feet to 6 feet long is set in the ground southward
from the northern cist to a small pillar 22 inches square, 9 feet to the
west of which lie a fallen slab and traces of an old-looking curved
earthwork.

(10). Gaherloohak, Glooney Parish (0. 8. Sheet 35, No. 8 and
2^0. 12). — ^This townland adjoins the barony of TuUa; and its menu*
4nents group naturally with the once numerous cists of Milltown and
Moymore. Our plan of adhering to the lines of the nup neoe«i-
tates their separation. The group of four small cists were (by a
mistake of my own as to the townland bounds) given as in Moymore
by Mr. Borlase. They were shown to me by the Rev. J. B. Oreer, and
lie near a farmhouse between the Moymore bridges, being marked on
the new survey, whose officers were careful to insert any unmarked
monuments pointed out to them ; but unfortunately (so far as I know)
only Dr. George U. Macnamara and I took any trouble with the
marking of pre-historic remains of Clare on the new mapa.

(a) The first or southern cist hardly rises over the field ; its cover
only measures 4 feet by 2 feet 3 inches by 12 inches ; it rests on three
•other blocks, one being 3 feet square. There is no trace of a moimd
•or cairn about it.

(h) Two blocks of similar character lie side by side ; but it ia not
•certain that they formed a cist.

{o) In the north wall of the field is set a block 4 feet by 2 feet
8 inches by 12 inches, and is said to have been part of a " Lobba," of
which the other portions were used for the wall.

{d and e) Beyond this are two small cists entire : the nortkern
xesembles a demi^olmen ; its cover measures 6 feet east and west, and
6 feet 8 inches north and south, being about 12 inches thick. Hie
standing-block, supporting it to the east, measures 33 inches by
14 inches, and rises only a foot above the ground. The neighbouring
cist has an irregular top slab 4 feet 4 inches to 7 feet 3 inches east
and west, and 4 feet 2 inches north and south, and is 16 inches thick.
The f^ides are nearly buried.

^ See plan, p. S8, figs. 6 and 7, 9uprm»

Westropp — Ci9t8f Dolmens^ and Pillurs of Ea%t Clare. 101

(/) There are remains of a cairn, with a small and defaced cist of
four Uocks, in a patch of bushes, near (but outside) the west wall of
the field. The cist was about 4 feet square.

It may be noted that, both in size and in the lack of definite
orientation, these tiny cists differ greatly from the usual type in
Clare, whether of the huge dolmens of Foulnabrone, Ballyganner, or
Fanygalvan, or the little cists at Poulaphuca, Farknabinnia, or Toonagh,
which taper eastward, and are identical in every respect save size*

(y) The remains of a larger dolmen are found at the opposite
(southern) edge of the townland, beside the road to Magh Adhoir. In
liuman memory, it was " a great box of stones''; but unfortunately
a farmer overthrew it when clearing the field. He removed the sides,
but found the top too heavy ; and (unable or afraid to blast it, though
popular belief does not extend its protection to dolmens or cahers as it
does to earthen forts) he set it up on edge, where it remuins
propped by lesser stones, and measuring 8 feet by 5 feet by 12 inches
to 15 inches thick.

(11). Ballyhickey, Clooney Parish (0. 8. 34, No. 15). — This small
cist of coarse gritstones is quite perfect, and is unusual in having parallel
sides and level cover. The axis lies E.N.E. The north side is of one
block 6 feet 8 inches long and 1 6 inches thick. The south side has
two, parallel to which, and about 3 feet away, is another and thinner
slab. The west end i^ 7 feet 2 inches long and 8 inches to 9 inches
thick. The cist is 8 feet 1 inch long, and the interior 7 feet 4 inches '
east and west by 4 feet 2 inches north and south. The cover is some-
what pear-shaped ; and, broken into two, it does not overlap the west
endJ It lies in a plantation to the side of Hazel wood House, and is
shown correctly in a little sketch on the map of 1840.

(12). DooKSSK, Doora
Parish (0. S. Sheet 34,
No. 6). — ^There are only
two dolmens in Doora
parish, much of which
(as its name implies) is
swampy. The Dooneen
monument is called a
" Giant's Grave," and
is not recognized as a
*• Lobbn " ; it lies in a field to the north of the rood from MoyreiRk

Fig. 16.— Dooneen.

See plan, p. 8S, fig. 8, 9Hpra.

102 Proceedings of the Royal Irish Academy.

Cross, past Maryfield House. It is a defaced enclosure of deep-set
slabs from 9 feet to 6 feet long, and measures internally 22 feet
8 inches east and west, and 13 feet north and south. The two largest
slabs at the eastern end make a slight angle. Of the other side hlocb
only one remains to the west, one to the north, and two to the aouth.
It has been considerably defaced since I first saw it. Ferguson figures
a somewhat similar Continental example at Eginlar.

(13). MovAMOS, Doora Parish (0. S. Sheet 34, No. 10).— A small cist
stood on a grarel hill not far from a large stone-faced earth fort. I
failed to find it on my earlier visits, owing to an altered bohereen and
field-bounds. When aided by the new map of 1894, I found the place
the monument had been removed and its very site deeply dug out for
gravel.

(14). Balltmacloox, Quin Parish (0. S. Sheet 42, No. 4).— This
parish — one of the most historic parts of Clare — only possesses tiro
dolmens. One stands near a small lake in a field with outcrops of
rock in Ballymadoon. It is entirely overthrown, but was of rude
and massive slabs of gritstone. One measures 7 feet, east and west,
by 5 feet 3 inches. It is shown on the map of 1840 as a massive
block, resting on three lesser stones. Near it is a dolmen-like slab
resting on a large boulder, but probably natural, and certainly beaiing
no marks of human workmanship.

(15). KiTAPPOGB,* Quin Parish (O. S. Sheet 42, No. 8).— This cist lies
sou^ of the road from Knappoge Castle to Kilkishen, and has suffered
from the hands of the " improver," having been partly removed and
blasted, the blocks still bearing marks of the crowbar. Its western
end defied the vandals, and still rests on t^ie end and two side blocks.
It is hard in its present condition to speak with confidence as to its
original design. The cist appears to have stood in an oval earthm
mound with a kerbing of blocks round the base, llie cover meumes
8 feet north and south, 4 feet 7 inches east and west, and is about
10 inches thick. The west end is 5 feet by 1 foot thick, and lises
18 inches over the mound. The north is 6 feet long by 12 inches;
some ten blocks of the ring seem to lie in position. Mj diawring
and a rough sketch-plan are figured by Mr. Borlase.'

1 Or Knoppoge.

3 «< Dolmens of Ireland," toI. i., p. 84. See also plan, p. 88» fig. 9,

Westropp— (7««^«, Dolmens^ and Pilfara of East Clare, 103

BuioiATTY Lower.

It is noteworthy that in the ancient Tradree^ from Latoon and the
Fergus to the Shannon and Owen na Geama no dolmens occur. This
prohahly arises from the district having heen (as pointed out) in an
unusual state of cultivation and clearance as the special appanage of
the native princes. It is only hetwocn the Lakes of Eossroe and
Mountcashel that any monuments remain , for the dolmens at Croagh-
ane really helong to the group on the Slieve Bemagh Hills.

(16). Kkocknalappa, Kilmurry na Gall Parish (0. S. Sheet 43,
No. 11). — ^This is named in Dineley's sketch of Rossroe, in 1680, aa
•* Knockalappa, anglice the Hill's Bed,"' but is not drawn. It rests on
a low green hill, at the foot of which stands the massive ivied tower
of Bossroe Castle, and the large lake called after it.

The cist is of large blocks, less shapely than most other ** Lobbas "
in this county. It is at present 1 1 feet long, and tapers eastward from
4 feet 7 inches to 3 feet internally. A single block, 6 feet 3 inches
long, and 2 feet 8 inches thick, remains ; a second was, I hear, blown up ;
and other stones bear marks of crowbars. For some reason not stated,
the destruction is stopped ; and there is no present intention of removal.
The west end is 5 feet 7 inches long, and from 17 inches to 24 inches
thick. The cover is irregulai-, about 8 feet 6 inches by 8 feet, and of
varying thickness.'

Close beside the ''Lobba" was found a gold fibula, described to
me as about 3 inches across and as thick as a cedar pencil, with,
however, slightly expanded ends, but without cups ; my informant
roughly sketched it for me. I could not learn to whom it was
sold.

(17). Drummdllan, Kilmurry na GallParish (0. S. Sheet 43, No. 9).
— There were two cists in this townland. One lies north of the road
to Fenloe. It is so rude and defaced that, only for the orientation and
tapering of its side blocks, it would be haixl to believe it a dolmen at
all. Indeed, despite this and the 1840 map, I am scarcely satisfied
that the blocks are not a split rock. Each slab is about 10 feet long ;
the northern is much broken. Two lesser blocks lie at the opposite

1 Tratraigho of the Fiibolg Eace, Mac Firbis : see " Irish Nennius,'* p. 266.
9 Journal R.S.A.I. (R.H.A.A.I.), vol. ix. (1867), p. 176.
3 See plan, p. 8S, fig. II, and Plate YI., fig. 2.

K.. 1. A. FROC, VOL. XXIV., SEC. c] [8]

104 Proceedings of the Royal Irish Academy.

side of the fence in whicli it is embedded. Another dolmen lay netr
Enockacunag Lake, and has been entirely removed since 1839. It
is shown as a small, square cist on the 1840 map.

(18) KiLCOBHAN, Eilmurry na Gall Parish (0. S. Sheet 43, No. 5,
No. 6). — ^A greatly defaced but massive and interesting monument lies
within a ring of lakes, and has a pleasing view of the long-wooded
ridge of CuUaun, crowned with its turret. The grave is covered with
earth, stones, and tangled masses of hazel scrub. It was hard to
understand its nature until a plan was made ; when it became evident
as a group of some three or four compartments, an airangement not
imexampled in other Irish monuments, but, I think, otherwise un-
known in Clare. To the north-west angle is a block 5 feet long, lying
north and south ; near its northern end are blocks at right an^e^ to
the last and .6 feet long. To the east of this lies a cover of inegular
shape, 4 feet 8 inches by 4 feet, deeply marked by the nanow lines
of a large cross. The graving must be of no little age, as tlie edges
are worn and the grooves mossed. South of this are two paiallel
blocks, 4 feet 6 inches long and 9 inches apart; and west of these, in
line with them and its west end — also in line with the north-w^
block — is a large irregular stone about 5 feet long. Iti s called a
"Giant's Grave."*

(19). Balltsheekbeo, Kilfinaghty Parish (O.S. Sheet 52, No.4).—
Major Walton, of Ballysheen, informs me that it was a small box-
like cist. It was destroyed about 1852. It is shown by two small
marks on some maps of the 1840 Survey.

(20). Brickhill, Kilfintinan Parish (O.S. Sheet 62, No. 2).— This
lay to the west of the dolmen of Ballinphunta and Croaghane Choich.
At least one large block remained in 1839 ; but I only found a low
green mound of earth and stones on the site, which may or may not
conceal the slab.

At no great distance is a place called Lacht, where I found no
remains of a cist. However, ** Lobba," and not " Lacht," is tiie
received local name for a dolmen.

(21). Ballinphtnta oe Ceoaghake, Kilfintinan Parish (O.S. Sheet
52. No. 2). — One of the most perfect cists in the county stands in the
tilled field south of the defaced church of Croaghane, in full sight of

^ See plan, p. 88, fig. 10, tnpra.

Wesi'ropf — CistSf Dolmens^ and Pillars of East Clare. 105

passengeiB on the railway which runs along an embankment directly
east of the remains. It is now much buried in field rubbish, and
oTergrown with brambles and ash-plants. However, as this may add
to its chance of survival in its endangered position, we may the less
regret the fact. It was (so far as I could find when making its plan
in 1887) unopened. In naming this to local antiquaries we suggest
great caution and consideration in any action to be taken.

Having given to Mr. Borlase (and here repeating) the plans and
elevations, I need only note that this cist is double, lying east and
west. The west chamber has a large end-slab, with two stones to

^5^iiiiww,^^|]S* ^^i^rfLiaa

Fig. 17.

«ach side, and three low stones parallel to the end, which, with two
others near the east end, show that this dolmen had a kerbing of
smaller blocks around it.

The cover measures 6 feet 7 inches by 5 feet 6 inches by 10 to
13 inches thick. The eastern chamber is far lower, so that its slab is
partly overlapped by the western cover. It measui'es 4 feet 8 inches
!bj 4 feet 2 inches. It was nearly buried even when I first remember
it in about 1881. »

This, to our present knowledge, completes the survey of the
dolmens in the baronies of Bunratty. The problems raised must be
reserved for solution till the subject is more advanced. Much as the
g^round has been cleared by Mr. Borlase's great work, much more re-
yna^ina to be done. We want field-work, spade-work, and folklore at
present. Later on we may proceed to clearer light than is afforded by
tlie theoretical portion of our only general survey. Till a map can
bo prepared of each Irish district (not necessarily a county), little
progress can be made ; and |as one short step towards this map, this
j>aper is offered to the Academy. We are at present bewildered in

» See plan, p. 88, fig. 12, $upra. ** Dolmens of Ireland," vol. i., p. 86.

[8*]

106 Proceedings of the Royal Irish Academy.

mist and darkness ; but every district completed is at least a step
towards the light.^

[Note. — The group in each parish is marked off by a broader
space.]

1 Of the dolmens named in this paper there are photographs of the following
in the photographic collection of the Royal Society of Antiquaries of Ireland:—
Caheraphuca, Bally maconna, Eilvoydan, Clooney Circle, Ballymacloon, Knappogue.
Knockualappa, and Bullinphunta.

[ 107 I

VIII.

THE CISTS, DOLMENS, AND PILLARS IN THE EASTERN
HALF OF THE COUNTY OF CLARE. By THOMAS
JOHNSON WE8TR0PP, M.A.

[Plates VII. and VIII.]

[Bead Junb 23bd, 1902.]

SEcnoKS 3 AND 4 — ^The Babokies of Tulla.

The baronies of Upper and Lower Tulla form, witb those of Biinratty,
the well-marked eastern half of Clare. It seems desirable to take them
next in order to the district of which I gave a survey of the early
monuments of rough stone in a paper read before this Academy in last
April. The barony of Upper Tulla corresponds to part of the ancient
Hy Caisin and Hy Ronghaile, with Tuath Echtghe (or Feakle parish)
and Cinel Donghaile (the O'Grady's country). Lower Tulla covers
Hymbloid, Hy Turlough, and some obscure little states along the edge
of the hills, iis in the Bunratty districts, it is probable that none of
these arrangements extend far enough into the past to affect the
dolmens or cists. Topographically the divisions are equally defective ;
eastern Clare falls naturally into the plain land and the hills of Slieve
Echtghe and Slieve Bemagh. In this paper wo are obliged to refer
strictly to the conventional divisions and the maps of the Ordnance
Survey, which show the sites, but do not always mark the dolmens.

The earlier writers have, as usual, passed by the prehistoric menu*
mcnts without any detailed description — indeed, rarely with even casual
mention. In 1839 O'Donovanand O'Curry described, not very clearly,
two monuments at Miltown and two at Ballycroum. Beyond this they
barely allude to the *' Broken Giants' Graves" of Drummin and Bally-
kelly. The maps of 1840 omit four at Miltown, two each at Fomerla,
Kiltanon, and Dromandoora, and one each at Tyredagh, Ballycroum,
Maryfort, Derrjmore, Elmhill, and Cloonyconry. Mr. M. Brogan de-
scribed the upper dolmen of Dromandoora to this Academy in 1865.
Miss Stokes, in both her lists of dolmens, omits all the monuments
(even those marked on the maps) to the east of the Fergus. Canon

1 08 Proceedings of the Royal Irish Academy.

Dwyer and Mr. Prost, in their histories and topographies of the district^
are equally silent.

At last, in 1897, Mr. W. Gopeland Borlase published several descrip-
tions and a fairly complete list in '^ The Dolmens of Ireland." He
gives his own notes on ones at Miltown, Newgrove, Cloonyconry, and
Formoyle ; Mr. M. Brogan's notes on Dromandoora ; O'Curry's not€s
on Miltown (2) and Ballycroum (2), and my notes on Tyredagh (2),
Miltown, Maryfoi*t, Eosedara, Elmhill, and Ardnataggle. The only
other contributions are two shorj; descriptions of Tyredagh Upper and
the lower dolmen of Gorbehagh in 1897, and a paper on the three
monuments at Ballycroum in 1 900 — both in the Proceedings of this
Academy.

Though these descriptions need only be noted briefly in the present
paper, it is necessary to repeat the plans for comparative study.

The Babony of Ttlla Uppek.

In the following survey I give the name of the townland and
parish, the sheet of the Ordnance Survey of 25 inches to the mile, and
the description (if the structure is hitherto undescribed) at some length.

TULLA.

(22). TraEDAGH Upper, TuUa Parish (0, S. Sheet 27, No. 13).—
This townland, the Tir Aodha of 1390,* contains a very remarkaWe
monument not marked on the maps of 1 840. It lies in a pleasant little
recess or shallow valley, hemmed in to the north-east by low but
picturesque cliffs. A little stream flows past and has undermined its
west end ; one block has fallen down the bank. It has been fnlly
described, and the plan and view of it published from my notes by
Mr. Borlase. We may briefly note that it is of five compartmentff,
with an extension to the north-east ; that it tapers slightly eastwani ;
that the covers have fallen ; that it measures at present 27 feet

^*'Tire«heeda" in the list of lands in the termon of TulU Churdi, 1397,
copied (it appears) from the ancient *< Black Book of St. MoehoQa*' into the
Inquisition taken at Ennis in 1611. The termon comprised: — Tnlla, EiUeeiy
Lisoffin, Cloonteen, Dromlig (Knockdromleague), Mojmore, Fomeria, SfltaaoR,
Tiresheeda, Dromcaha, or Kildonalballagh (Ardbooly, according to a Molonj tad
Westropp deed of November 1720], Ballyore, Cregancryan, Drom>ghmartia>
Bonavorey, Furhee, Loghan, Cutteen or Caheroutteen, and (appai«ntlj) Kae.
0' Curry, without citing any andent writing, says the name is *<properij Tir
Biada." It is pronounced ** Tir'eeda.*'

iiwft^wa^

Dolmens in the Burony of Upper Tulla.

110 Proceedings of the Royal IrUh Academy,

east and west, and that it tapers eastward from 7 feet 6 inches to
6 feet 6 inches in 20 feet. The axis lies north-east and south-west*

(23). Tybedaoh Lowee, Tulla Parish fO. 8. Sheet 27, No. 13).—
Not far away from the last in the adjoining townland, separated from
Tyredagh Upper by the road from Carrahan to Tulla, is a large dolmen
lying in one of the back yards of Tyredagh House. Borlase has
published very roughly a view and sketch-plan from my notes. The
dolmen is from 16 feet to 19 feet long, and tapers eastward from
5 feet 10 inches to 3 feet 11 inches internally, the sides sloping in the
same direction. A fine horse-chestnut tree growing in the endosnre
has helped to destroy the western end. Three of the side blocks
remain respectively to the north and south, with two others projecting
from them, and two end blocks to the west and one to the east. Three
large slabs of the broken and rather thin cover remain.^

In the adjoining field are the reputed remains of another dolmen,
being two rows of rather small blocks lying east and west, and neaily
parallel. No cover remains ; and I am more than doubtful as to the
Inature of the structure.

> In a plantation to the cast, beyond the yards and garden, is a
small earthen ring-mound, far too small to be a rath. One can only
recall the tomb of Dathi at Hathcroghan and the passage in Keating's
** Three Bitter Shafts of Death "« (1620), where it is steted that the
pagans were laid facing the east, and a small rath raised round
with a leacht or cairn, or an earthen rath without a monument. No
Istones remain in the Tyredagh ring.

In the field still farther eastward is a pillar slab, 9 feet high, made
of a very thin flag of limestone, tapering upwards from 31 inches to
24 inches wide* and 6 inches broad.' A gi-aveyard lies near it to the
north, and north of the latter, at the opposite side of the road and in
line with Tyredagh Castle, lies a large sandstone boulder with abuUann
or basin ground into it.

The dolmens in the two townlands have been described from my

j ^ Flan, p. 109, fig. 1, supra,

* Plan, p. 109, fig. 2, mpra,

»See note in «*The Battle of Gablira" (edited by N. O'Kearney) in the
publications of the Ossianic Society ; also Kilkenny and South-East of Irdaad
Bociety (R. S. A. I.) Journal, 1854, for ** Tulachs" or burial-mounds.
I * Misprinted bfeet in Proceedings of the Royal Irish Academy, ut in/r^.

fi<< Dolmens of Ireland," i., p. 88, sketches and plans; the sketch of tht
dolmen of Tyredagh Lower being very rude and defective. Pioc B. I. A.,
Ser. iii., vol, iv., Plate ix., p. 646.

Wesi'kopp — Cists^ Dolmens^ and Pillars of East Clare. Ill

notes by Mr. Borlase, and that in Tyredagh Upper is noted in these
Prooeed^gs."

(24). KiLTANON (0. 8. Sheet 27, No. 13), Tulla Parish.— The
remains are greatly defaced, and are not mai'ked on the maps. Four
stones stand in line, side by side, lying east and west on a low mound
among hawthorns to the south-east of the outer bridge on the front
ayenne ; it is probably the remains of a small dolmen. The blocks
measure 4 feet by 80 inches to 24 inches by 7 inches ; the second and
third are each 4 feet 4 inches wide ; other dimensions similar. Another
similar slab lies to the west on the slope of the mound, which is
18 feet long.

South of Kiltanon House, in a grove of trees, lies a sandstone block
resting on a slab, 3 feet by 4 feet. It has a bullaun, 13 inches across,
ground in it. This has been illustrated elsewhere in these pages.^
Three other slabs lie in the grove. The side of an undoubted dolmen
stands in a disused burial-place for children, at no great distance from
the last remains. It tapers upwards from 9 feet to 5 feet 9 inches long,
and is 3 feet 8 inches high and 16 inches thick, of good sandstone, and
belonged, the old people say, to a ** box " cist.

(25). MiLTOWN or Balltvollen, Tulla Parish (0. S. Sheet 35,
No. 1). — ^Miltown is separated from Kiltanon by the Affock river
running through the range of caverns called the Toomines, so bom-
bastically described by John Lloyd in 1 778 ;' but which form an interest-
ing and picturesque natural gallery, lit by openings hung with ferns
and ivy,and with a gravelly strand beside the stream. The English name
of the townland is a mistranslation, as the place was 0*Moylan*s town,
and is called baile ui TTlaoilni, from the family of O'Moylan or
Mullens, in the Macnamara's rental, 1390. It is named Ballyworryn
(Ballyvollyn) in the Inquisition post mortem of Donat Macnamara
lieagh in 1591, and Ballymoilin in the Edenvale ** Survey of Clare"
about 1675.

One of the finest groups of cists in Munster existed in human
memory among the craggy fields of Miltown. Unfortunately,
the majority have been "improved off the face of the earth," and
only one remains perfect. I can faintly remember two others that
have since been destroyed, but which stood nearly perfect some twenty-
five years ago. The others were all in existence so late as 1839,
though some are not marked on the maps of the Ordnance Survey
of 1840.

J Pioc, R.I.A., Ser. iii., vol. iv., Plate ix., p. 647.
« *' An Impartial Tour in Clare.'*

112 Proceedings of the Royal IrUh Academy.

(a) Not far north of the picturesque and ivied peel-tower of
Miltown lies among the crags a large, rugged block of limestone,
supported on lesser stones. It is apparently artificial, and formed a
cist of difEerent and ruder type than the others in the townland.

{h) In the Kennedy's farms, to the west of the side road, stood in
my recollection a fine dolmen. It has been described to mc b j a
farmer, who remembered it as '^ the tallest labba in the place; a
great table of a rock on top of four other stones as high as yoor
shoulder " (say over 5 feet high). '' It was open below ; the top slab
was very thick, and it stood on the highest point of the field some
distance from the fence." My sister, Mrs. O'Callaghan, says that it
closely resembled the perfect one on Mr. Sheehan's farm, and wu of
four thick blocks, with a large coyer, being over 5 feet high, and
partly covered with thick ivy. These descriptions are independent of
each other, and, as will be seen, closely correspond. My own recollec-
tions of this monument are very vague. It is called ''Dermotand
Grania's Bed " on the map. Unfortunately, I cannot find that any
sketch or measurement is preserved ; and it is only shown as a cist on
the 0. S. map, and lies E.N.E. and W.S.W. It was removed about
twenty years ago.

(c) In what is now the same field are the mutilated remains of
two smaller dolmens. In the more northern (so far as I oonld
examine it through a thick overgrowth of sloe-bushes and brambles),
the top slab is from 31 inches to 36 inches thick, and haa cniahed
down the south side which lies under it — ^if , indeed, these are sot tiie
remnants of the destroyed monument last described ; though I was
told that this was not the case.

{d) In the ditch south of the same field is a defaced little ci&t«
12 feet long and 4 feet 9 inches wide. The sides and ends ronain;
and a cover-slab, now partly buried, lies on the north- west slab. The
south side lies E.N.E. and W.S.W,^

(#) Another cist, facing north-west and south-east, lay south-west
from the last gi'oup towards the ^ewgrove Bridge.' It has been long
removed ; and I could not get any very clear description, or the date of
its destruction; but it seems to have been a ruined ''box" of slabs.

(/) To the east of the side road was another dolmen, which I
sketched about 1 883. Unfortunately, the many opportunities of vimt-
ing Miltown led to my postponing my intended survey of its remains

1 Plan, p. 109, fig. 8, $upra,

' It is marked over the I of the townland name on the 1840 map.

Westropp— Ci«/«, Dolmens^ and Pillars of East Clare. 1 13^

till too late, for this dolmen wqb blown up by the tenant of the farm
in July, 1890. So far as I remember it and my sketch shows, the cist
was a small box of three slabs and a cover, and sloped towards the
south ; it lay east and west.

(jg) South from the last, in the angle made by the two roads
through Miltown, lay three monuments marked on the map of 1839.

The first is called "Giant's Grave" on the Survey of 1840, and
has been noted by Eugene 0' Curry in the Ordnance Survey Letters.*
It lay to the norUi-east of the perfect one on the Sheehan's farm, and
was in "the form of an ordinary grave (coifin), measuring 19 feet
6 inches in length, 4 feet 5 inches in breadth at the foot, and 6 feet
4 inches in breadth at the head, the thickness of the stones all round
being included in the measurement. This was enclosed by a number
of large stones placed at a few feet distance, and following the form
of the grave." He continues : " The grave of Sliabh Georr, near
Glen CuUen, in the county Dublin, is of the same form with this, as
are some moro in the eastern parishes of Clare. These long coffin-
like graves can hardly be supposed to belong to the same period of
time as the square chest-like and sometimes irregularly formed monu-
ments, to be met with in several parts of the Barony of Burren, &c."

The long and enclosed type also occurs at Faunaroosca and Iskan-
cnllin,' in the Burren. I do not know of any extant example in " the
eastern parishes of county Clare." Not a trace of this monument is
now to be found, nor do I remember any in 1877. The site lies near
a nearly levelled rath.

(A) Another undescribed dolmen lay near the last ; it has been
removed.

(t) Another monument, further to the west, has shared the same
fate ; it is shown as a cist, called " Dermot and Grania's Bed," lying
in the field near the junction of the by-road and the TuUa road, and
to the east of the former.

(j ) I was told that certain blocks on a little mound near the road
are tiie sides of a labba. If so, they are possibly not in ntUy and may
faaye belonged to the neighbouring dolmen.

{k) The sole intact survivor of this once fine group stands near the
laat ones to the south of the main road from Newgrove to Tulla, on the
Slieeban's farm, near the old lead and silver mine. It is thus noticed

1 H SS. B.I.A., H B 24, p. 255.

* See Journal of Royal Society of Anti^uariefl of Ireland, vol. xxxi. (1901),
pp. 277-285.

114 Proceedings of the Royal Irish Academy.

by O'Ciirry in the Ordnance Survey Letters : " There are seven giants'
graves in Miltown (Baile Ui ' Mhaoilin, O'MuUen's town), one of
them near Green's House, ^ in perfect preservation, of an irregular
square form, serving as a pig-sty, with a rick of turf built over it at
present." Then after describing the giant's grave already given, be
adds : '' The other graves in this townland deserve no particular
description." Students will regret that this feeling towards pre-
historic remains dominates the Letters on C?lare, and deprives us of
much valuable information, then easily accessible and recoverable.

In this case the house and yards have disappeared, leaving no trace,
while the cromlech remains. It is a low, clumsy cist, of five massive
gritstone blocks, and is capped with a deep heap of earth, covered with
a rich growth of shamrocks and flowers. The cell measures 9 feet
6 inches by 5 feet 6 inches to 5 feet, tapering eastward. The west,
north, south, and east blocks are, respectively, 7 feet by 2 feet by
1^ feet; 9 feet 6 inches by 2 feet 9 inches by 1 foot 9 incbes; 5\ feet
by 2 feet 9 inches by 18 inches, and 5 feet by 1 foot 3 inches by
18 inches. Its axis lies slightly to the south of last. It has been
described and illustrated by Mr. Eorlase from his own notes.*

There are some large slabs in the fence of a field south of the road,
some distance to the west of the last.

(26). Newgrove or Balltslattebt, Tulla Parish (0. S. 8beet 34,
No. 4). — This townland was assigned by the Mac Shanes to the
O^Slatterys (from whom it takes its Irish name) in 1493, by a deed
published in the Tramaetions of this Academy.' It adjoins Kiltanon
and Tyredagh, lying across the Alfock river from Miltown. The
** giant's grave," as it was called in 1839, lies in the demesne to the
west of the avenue ; it is now called '* Lobba 'yiermudh." It is
nearly pei*fect, and is a large cist about 5 feet by 9 feet internally,
with a bullaun groimd in a block at the east end, and a surrounding
fence of slabs set on end, like the demolished '^ giant's grave" in
Miltown and others. Three are still standing to the north and two to
the south. The cover measures 9 feet by 9 feet 6 inches by 18 inches
thick. The basin stone found with this and other Irish dolmens is a
very interesting feature. Similar basins occur with dolmens outside

1 Colonel 0*CalIagban, the owner, tells me that Green's holding is nov the
Sheehan's farms. The leases are now at Maryfort. This clearly identifies the
monument described by 0* Curry.

2 ** Dolmens of Ireland," vol. i., p. 91, plan and view. Also plan, p. 109, fig. 3,
aupra,

3 Transactions R. I. A., vol. xv. (Antiquities), p. 62.

Westropp — Ciats^ Dolmens^ and Pillars of East Clare. 115

our islands, as at Mont d'Algeda, in Portugal, and at dolmens in
Syria and Moab. Basins are found near overturned dolmens near the
site of Dan. Other basins are found in Palestine,^ possibly marking
the sites of sacred pillars, stone circles, and dolmens, destroyed by
Israelitish reformers when they ** removed the high places and brake
the pillars." The Newgrove monument has been described and
figured by Mr. Borlase, and also by me in these Proceedings} In the
plantation near the river stands a pillar marking the mearing with
Tyredagh.

(27). FoMKBLA (0. S. Sheet, No. 34).— Two little cists lay, I am
told, near the stream not far from the remains of Fomerla Castle. I
could not find tbem ; but believe that, at any rate, one exists.

(28). MomoBE, Tulla Parish (0. S. Sheet 35, No. 5).— A small
cist of four stones and a cover, called " Dermot and Crania's Bed,"
lay near the north-east angle of the townland, but has been removed'
since 1839.

Another cist is intact, but buried in field rubbish, and with a
hawthorn growing on it; it was also called *'Dennot and Crania's
Bed" on the Survey of 1840. It is 10 feet long and 6 feet 8 inches
wide externally, and 5 feet 4 inches internally, at the east end. The
north block (so far as I coidd measure) is 6 feet 4 inches long, and'
8 inches thick ; the south the same length, and 14 inches thick. Several
cover blocks lie on the top, the most western being 4 feet north and
south, and 3 feet east and west, the most eastern 6 feet 4 inches by
6 feet across.' Across the stream, near Moymore Bridge, the group
of small cists, already described,* lie in Gaherloghan.

(29). Mabyfobt* or Lismbhan, Tulla Parish (0, S. Sheet 35,
No. 3), — The townland frequently appears in notices of the
Jiacnamara lands and castles, but it bears the name of the

1 Conder's " Heth and Moab," pp. 230—249. Cups and circles nppear on rude
stone monuments even in Fiji.

' « Dolmens of Ireland," vol. i., pp. 89, 90, plan and illustration, Proc.
R.I. A., scr. ill., vol. iv., p. 547, Plate ix., illustration. Also plan, p. 109, fig. 5,

•Plan, p. 109, fig. 6, supra.

* Plan, p. 109, mpra,

^ The costlo is said to have hecn built by Mahon Mac Shane Macnamara early
in the fifteenth century (S. H. O'Grady's Catalogue of Iiish MSS. in the British.
Museum, p. 73). The modem name was given to the house about 1760 by Balph
Westropp, of LiBmehan, in honour of his wife Mary Johnson, and, folloMing (it
is said] the bad example of Robert Westropp, who renamed Fertane as " Fort-
Anne," after his wife Anna, some fifty years earlier.

116 l^roceedings of the Royal IrUh Academy.

O'Meehans, who were possibly its earlier owners. Liop inio6acain
appears in the Macnamara rental oi circa 1390.^ The fragments of
the destroyed castle, built about 1440, lie on what seems to have been
an earlier entrenchment. Two other lisses are on the rising ground to
the south ; a third on the beautifully wooded hill behind Marjfort
House. The slight remains of a double earth-fort lie in the marsh neir
the castle ; and on the higher field, east of the castle, where the tnoe
.of an old road crosses the avenue, is a small cist. It stands on •
grassy knoll in the remains of a mound. The ends are gone, the
south side has fallen, and only one block, the north side, ^nd the
cover, still leaning upon it, can be measured. They are two lime-
stone blocks ; the cover is 5 feet 6 inches east and west, 6 feet 3 inches
north and south, by 9 inches ; the side measures 4 feet 8 inches by

3 feet 6 inches by 3 inches, and lies E.N.E. and W.S.W. It htt
been illustrated by Mr. Eorlase^ from my sketch; but the back-
ground has got rather altered in the engraver's hands.

(30). EossLABA, TuUa Parish (0. S. Sheet 27, No. 16).— This,
when 1 sketched and planned it, was a fine and fairly perfect dolmen,
on a low, grassy hill, overlooking the Castle Lake and Maryfort Lake,
overhung by the wooded hill already named, and with an open view
to the hills behind Feakle. It was not far from the " rude ribs of
the ancient castle '' of Fertanemore or Rosslara. It fell naturally in
the spring of 1898. Its blocks lie untouched and almost overgrown
in the bushes beside the fence and hedge along the top of the hill.
The cover measures 9 feet to 7 feet 6 inches by 6 feet 6 inches by
14 inches to 16 inches. It was supported by two stones to the south
and three to the north, forming a cist 12 feet 6 inches long, and

4 feet 8 inches to 4 feet wide, and about 4 feet high. The ends were
removed, which, with the lowering of the field (so frequently notioe-
able at dolmens), led to the settlement and collapse of the monument
It is figured and described from my notes by Mr. Borlase,' but, as
usual, the views have suffered in re-sketching.

This completes, so far as I am aware, the dolmens in the parish
of Tulla. Lewis's '* Topographical Dictionary," however, mentions

* Tliore is & chaiter of Tiuge i Meadchain, son of Conor, to the Macnamatas,
dated 15l7 (Catalogue of Irish MSS. in the British Museum, S. H. O'Gnidy,
p. 155 ; Egerton Charters, No. 97).

' ** Dolmens of Ireland," toI. i., p. p. 94, view. Plan, p. 109, fig. 9, fiyr*.

^Ibid., vol. i. p. 93, view, plan and elevations. Alao plan, p. 109, fig. II,
aupra.

Westropp — C/«/«, Dolm€H8, and Pillura of East Clare. 117

another lying on the hill of Tulla, of which I can find no trace, and
which is not on the maps, though this, of course, disproves nothing.

KiLNOE.

(31.) CxiOGnEfi or Deruyuoue, Xiluoc Polish (0. S. Sheet 35,
No. 5).^ This townland was called Glogher before 1651, and
deriyeB its name from the many great limestone blocks and out-
crops of crag rising in its fields. It is, however, best known as
Derrymore, as it forms that very picturesque demesne lying along the
winding shores of Lough Breeda. That the place was inhabited by
more than wandering hunters in early times is shown by various
defaced forts and two crannoges on tbe lake.

The dolmen is one of the most curious, and I think the most
mafisiTe, examples among the monuments of the county. My attention
was first called to it by Mrs. Gore, of Derrymore, it being apparently
a mere natural block. It lies in a grove of trees south of the first turn
on the older (northern) avenue. At the opposite side of the avenue
we notice a great block similar to the cover of the dolmen. This is
embedded in a bank of drift about 6 feet above the level of the foot of
tbe bank ; and the gravel underneath has been partly hollowed out.
This fact and the appearance of similar blocks (lying near it on low
ridges of drift and gravel) lead us to suppose that the dolmen cover
also rested on such a mound, under which the grave-makers ran a
tunnel, inserting upright blocks underneath so as to form a cist ; this
i>va8 donble- walled to provide against the pressure of the cover. Even-
tually— whether soonufter the intennent or at some other time during
the long subsequent centuries — the cover "kicked out" the outer
slabs and crushed and splintered the weaker inner ones. The broken
and fallen blocks, however, still bore up the mass sufficiently to allow
one to explore and make a plan of the structure. The cover is a
massive irregular slab of rough limestone, 12 feet 7 inches east and
^v^est, 10 feet 6 inches north and south, and from 3 feet and 3 feet 6 inches
nt tbe ends to 4 feet 6 inches in the middle. The west face is 6 feet
lon^y the south-west 8 feet 7 inches, the eastern 7 feet 6 inches, the
remaining edges 3 feet and 5 feet. Underneath, as may be seen by the
plan* was a cist 9 feet 5 inches by 7 feet 4 inches, with double lines of
BlAtpB to the sides. The east end slab is 7 feet 4 inches long, the
Tiortbem inncrside is 8 feet ; the southern is broken ; they are 9 inches
to 10 inches thick. The outer north slabs are more massive, 7 feet
G inches by 7 feet 5 inches by 12 inches to 18 inches, and 5 feet

118

Proceedings of the Boi/al Irish Academy.

4 inches by 7 feet 9 inches. The axis of the cist lies north-west and
south-east. The dolmen is not marked on the maps.^

Feaklr.

Feakle parish lies for the most part among the hills round Lough
Graney. It is a place almost devoid of history and of ancient build-
ings of the historical period, having been, it appears, densely wooded.
However, a few cahers and earth fort«, together with six dolmens,
show that it was inhabited in a few spots. A curious plate of gold was
found near the village of Feakle about 80 years ago, and is de8cril>ed
in the sixth volume of the Academy's Trans-
actions in the following words : ** A few years
ago Mr. Samuel Johns, a working silversmith
in Limerick, bought a very curious piece of
thin gold of this shape " [a figure with con-
cave sides and straight ends is shown], '^ and
of an exceeding fine quality, supposed to be a
shield or breastplate. He got it from a
labourer, who found it in the parish of Feakle
and county Clare; it weighed upwards of 12
ounces. He gave £3 8*. 3rf. per ounce for it,
and sold it afterwards to Mr. William D.
Moore, goldsmith, in Dublin, for £4 an ounce-
The dolmens, except the wrecked one at
Corracloonbeg, have been fully described, but
must be briefly noted here.

(32.) CoRBEHAOH or Druhanboora, Feakle
Parish (0. S. Sheet 19, No. 3).— The upper
Labba and the rock- markings have been de-
scribed by Mr. Brogan' in these Proceedings,
The Labba on the ridge consisted, in 1866, of ^^^-^*"'"ff- Corb<^b:xgi.
an enclosure tapering eastward. It measured intemallv 5 feet tv
3 feet 6 inches by 15 feet 6 inches. A cover restetl on the western
end, and another lay near the N.E. corner. The ends were nearly intact.
It had, besides the end blocks, threel ong ones to the east ; and to the
west the middle was removed. The lower " lobba " lies in a tilled fitUl,

1 Plan, p. 109, fig. 7, supra; Plate vi., 2.

«Trnna. R.LA., vol. vi.. p. 32.

'Proc. R.I. A., vol. X. (ISfil-lSGC), p. 440, Plate xxix., views of upper Ool a.
and carvings; plan of the first. The view of the carving been re-sktrttl-
inaccurately.

Wbstrofp— C7sa/«, Dolmens^ and Pillars of East Clare. 119

and is a low massiye cist of seven blocks, measuring internally 8 feet
2 inehes long and 6 feet 5 inches to 3 feet 3 inches, and beyond the
end slab to 1 feet 8 inches wide, tapering boldly eastward, with a
cover block 1 1 feet long, 8 feet 2 inches to 2 feet 2 inches wide,
and 16 inches thick. ^ Not far away, the outline of a footstep and a
curiously combined ornament, formed of spirals, loops, and curved
ends, are incised in the natural rocks. I have described the remains
and given a sketch in a previous volume of the Proceedings.*

(33). Balltcboux, Feakle Parish (0. S. Sheet 19, No. 16).— This
important group stands in the basin of a mountain bog, and consists of
the long dolmen Altoir TJltach ; the cist Tobergrania,' once considered
to be a holy well and a long cist found by me, near the first. The first
two are marked on the 0. S. maps of 1840, and described, though
vaguely and incorrectly, by O'Curry. This account was published by
Mr. Borlase, who did not visit them. They are noted at some length
in the Proceedings of this Academy for 1900.^ I repeat the plans.

(84). CoRBACLooNBBo, Feakle Parish (0. S. Sheet, 20, No. 14).— A
steep ridge (crowned with a small table of rock, closely resembling a
dolmen, but natural) rises 629 feet above the sea, and commands a broad
view to the Shannon, the Fergus, and Lough Derg, and across the lake-
studded plain to Slieve Bemagh. Towards the north, we look up the
valley to Lough Graney and the woods of Caher, and, down the slope,
but on the brow of a blufP, we see the defaced dolmen.

A few shattered blocks and a row of three slabs (each 3 feet long,
and a fourth, 5 feet long, set deeply in the ground, and running east
and west for 14 feet) alone remain. Two loose blocks, 6 feet 6 inches
by 2 feet, and 3 feet 3 inches by 1 foot 3 inches by 5 inches, lie near.
It is shown as a line of scattered blocks on the survey of 1840.

TOMORAHET.

(35). Cloqhlba, Tomgraney Parish (0. S Sheet 28, No. 12).—
South of the road, on the bounds of the rectory grounds, stands a pillar
called Cloghlea. The slab is 6 feet 7 inches high, and about 4 feet

> Plan, p 109, fig. 4, iupra.

s Proc. B.I.A., ser. iii., vol. iv., p. 546. (Inaccurate). They seem to have been
independently discovered and a query inserted in the Journal B.S.A.I., vol. xxi.,
p. 86, in 1890, by Be v. J. Halpio, of Scariff. They are now marked on the
Ordnanoe Survey maps.

> There is actually a holy well, Tobergrania, in Drummaneen, near Crusheen.

« "Dolmens of Ireland," vol. i., p. 96; Proc. B.I.A., ser. iii., vol. v., p. 86.
lUofltration and three plans. See also phms, p. 109, figs. 10, 11, 12, tupra.

JUX.A. FttOC., VOL. XXIV., SEC. c] [9]

120 Proceedings of the Eoynl Irish Academy.

2 inches wide ; but it has split into two layers^ eack 10 inches to
7 inches thick. It is thickly ivied, and may have been adopted or
erected by the monks of the neighbouiing yery ancient monastvy,
to mark the bounds of their lands.

MOTKOS.

(36). CAPPAeHASA^w MouiTEAnr or Gappaghbavk,* Moynoe Parish
(0. 8. Sheet 21, No. 5).— This dolmen lies near the boonds of due
and (Mway, in a secluded nook in the hills above Loug^ Berg. It is
shown in the maps of 1839 as a large irregular block, resting on at
least three stones, and with other outlying blocks. It lies on a grassy
patch, on the slope of a heathy hill, covered with bog. The site
commands an extensive and beautiful view over Lough Berg, with its
islets and wooded shores and promontories. The hills of Ogonnelloe,
the woods of Raheen and Caher, and the point of Aughinish, dose in
the view ; and through the gaps we see a further reach of the lake
towaxds Killaloe, and the great rounded mountain of Tliountinna
(where Fintan slept securely under the waters of the Belugp, as told
in our older legends) and tbe heights of Slieve Bomagh over the valky
of Killokennedy. Behind us, the Bow river and the streams of Sheeaun,
Olencullin, and Bamaminnaun, fall down the long valleys from the
heatheiy uplands towards Lough Atoiick, the highest point in
Oappaghbane being 1126 feet above the sea.

The structure is called '' Bermot and Grania's bed" on the 1640
map. I found no one in that lonely valley to tell me any legend or
name connected with it on either day when I visited and planned tite
monument.

It is a small cist of three great sandstone blocks, tapering eastward,
and measuring internally 4 feet 2 inches to 3 feet 3 inches wide, and
6 feet 9 inches long. The blocks are about 2 feet thick, and 2 feet
6 inches high ; it has been opened, and the west stones lie outside it.
The walls are double, and have externally three blocks to the north.
The most western is a pillar, 5 feet by 25 inches by 12 inches to
14 inches of conglomerate, with pink and scarlet pebbles fanbedded
in it. There is one block, 4 feet 2 inches long, and 8 inches thi^ to
the west, and three blocks not parullel to the inner side, and
respectively 4 feet 7 inches, 2 feet, and 3 feet 10 inches long, S feet
high, and, except the middle flag, I foot thick. The interspace is filled
with turf, and makes sides 31 inches to 36 inches thick. The cover

I Locally ** Cappabane," pronounced Gappa-bahn.

1 BALLYKCLLY

^^'^

2 FORMOYLCMORE

I CLOOYlYCONRYMORC

I ELMHILt « CLOONYCONRYHORE

O
• K1U.0K£NN&DV

iQt?rto+

i^SS-

; t)^-*^*^'/

» KK0CK8HANV0

• ARDNA7A6CLE

t..,.r..,.r

a ORUMMIN

\W1

Ik^tfti^wiff

Dolmens in the Barony of Lower T uUa.

[9*]

122 Proceedings of the Royal Irish Academy.

block is 7 feet 3 inches east and west, and about 5 feet 6 inches to
6 feet north and south. It is capped with a mass of turf and heather,
and has broken across and hangs into the cist ; but is still otherwise
perfect and in position.^

I noticed a very similar pillar and several large loose blocks lymg
about in the fourth field south of the road to the east of Cappaghbane
school-house. It may be a wrecked dolmen, though suitable blocks
abound up these valleys. I have seen no others that are not manifestly
untouched by man.

Babont of Tttlla Low£b.

This barony corresponds to the old states of Tuath Ua gCon^iaile
(Ogonnelloe),Ui Thoirdhealbhaith (south of Killaloe), Tuath Ui bFloiim
(Kilseily and Clonlea), and Olenomra. It consists mainly of the great
slate and sandstone hills of Slieve Bemagh, rising to heights of 174S
feet, 1729 feet, 1458 feet, 1353 feet, &c., above the sea.

The dolmens lie, with one exception, in the circuit of the hills; aod
several of them — such as Enockshanvo, BaUykelly, Formoyle, domj-
conry, Killokennedy Drummin, and Lackareagh— occupy prominent
positions.

There seems to be little prehistoric legend connected with the hiUs,
save the late one of the death of the poisoned Ard High Crimthann at
Glennagross, about a.d. 377 ; but the connection of Aibhell, the Great
Banshee of the Dalcassians, with Craglea, is of very antique
complexion. At least two examples of Fearbreagas occur : one near
Knockaphunta, and one near Killokennedy, neither being vety far
from dolmens.

We may divide the dolmens, as usual, into simple cists of five slabs
— Elmhill, Knockshanvo, Yiolet Hill ; simple cists of several a)ab&
— Drummin, Cloonyconrymore (two), Cloghoolia, Lackareagh (?) ;
complex cists of more than one compartment — Aidnataggle, Lcmg-
graves, Formoylemore, BaUykelly, Killokennedy; doubtful and sites —
Bealkelly-Purdon, Ardskeagh, Cloonyconrymore (upper dolmen).

KlLLUKAN.

(37). Elmhill, Killuran Parish (0. S. Sheet 86, No. 9).— The
remains of this cist stand on a grassy ridge not far from Doon Lake,
and command a striking view of the highest part of Slieve Benugh.

> Plan, p. 109, fig. 13, 9upra\ Plate til, fig. 1.

Westropp— Cm^, Dolmens^ and Pillars of East Clare. 123

It is not marked on the 1840 map, and was first noted by Mrs.
O'Callaghan, of Maryfort, to whose interest and constant help I am
much indebted for the completion of this paper. The monument
was a cist of four blocks and a coyer. The northern side measures
7 feet by three feet, by 1 foot 4 inches, lying E.S.E. and W.N.W.
The eastern block has fallen, and is 3 feet by 3 feet by 1 foot

3 inches ; the other 6 feet by 3 feet to 4 feet 2 inches by 1 foot

4 inches. The cover is tilted up, and measures 5 feet 3 inches by
4 feet 3 inches by 1 foot. In the adjoining field to the east is the
disused graveyard of ** Lackbrack." Whether this is an old name for
the dolmen (of which it would be most descriptive) I cannot now
learn. Mr. Borlase has described it from my notes.'

Ogonnellor.

(38). BEALEELLT-FuBDoir or Bkhebnagh, Ogonnelloe Parish (0. S.
Sheet 37, No. 1). — This dolmen had been overthrown before 1839.
Some blocks still remain to mark the site high up the hills over Lough
Derg.

EnSEILT.

(39). Drummin, Kilseily Parish (0. S. Sheet 44, No. 2).— This

dolmen is shown on the map of 1840 as four radiating blocks on the

western slope of the high rounded hill called Laghtaiagat,' perhaps

from the monument. O'Curry calls it a ** broken giant's-grave." It

IB reaUy a cist, of which the top slab has been removed. The west

end is a block of conglomerate, 4 feet 10 inches long by 3 feet by

i 1 inches. The dbrthem side consists apparently of 2 slabs (one of

slate, one of conglomerate), and is 10 feet long. The south had three

blocks (the western of slate, the eastern of conglomerate), and is

9 feet 3 inches long. The cist measures internally about 4 feet

4 inches wide, and 8 feet long. The axis lies £.N.£ and W.S.W.

It is nearly buried in small stones, and stands in a cidtivated field,

^with a wide view to the north of the hills beyond the river Oraney,

and lies over 600 feet above the sea.'

There are some blocks in line E.N.E. and W.S.W. among furze in
O'Sliea's acres, south of the last ; but I am not certain whether they

1 *• Dohnens of Ireland," vol. i., p. 96. See alao plan, p. 121, fig. 6, tupra.
^ It is 968 feet high. Broadford ii only 100 feet ahove the sea, and lies at it
oot to the louth-west.

> Flan, p. 121, fig. 11, iupra.

124 ProeeediUffB of the Royal IrM Academy.

were portion of a dolmen, though they do not eeem to belong to a liiie
of fence.

(40). YioiJET Hill, KilseUy Pariah (0. 8. Sheet 44, No. 2).— Itk
not marked on the 1840 map, and lies above the beautifally sitiiatod
house, looking out across southern and central Clare, with the wooded
shores and lakes of Boon and Gullane, and the bluff hills of Enocksise
behind Kilseilj Church. The cist is nearly buried in furse and a cain
of small stones, and consists of a sandstone slab, 5 feet 9 inches bj

5 feet 7 inches, resting on two smaller blocks, about 8 feet apart and
3 feet long. Whether the cist continues farther eastward I am unable
to find ; but a slate slab, 3 feet 3 inches long, and 4 indies thick,
is set in the ground, 3 feet from the west end. Another sandflAone
block, only 31 inches long and 9 inches thick, lies east and west 2 feet
east of the cover ; and another block, 4 feet by 2,f eet 8 inches, lies 3 feet

6 inches farther east, in line with a side slab south of the dat, which
is loaning outward, near the centre of the south side, and is 3 feet

7 inches by 2 feet. Two other large blocks lie down the slope.
I was told of another dilapidated caiin, called a ** giant's grars,"
by Mr. James Going, in 1893, but could not find it on my later
visit, the hill-top being much covered with furze.'

(41). Ardskeaou or Buoadfoed, KilseUy Parish — (O.S. Sheet 44,
No. 6).— This place is the Ard Sgiath of the 1390 rental of the
Macnamaras. In 1839, three large blocks of stone, lying east aad
west, occupied a little rounded knoll near the old road from Bioadfbid
to Kilbane, at a place called Knockaunnafinnoge, not far from a hollow,
called Poulamuckagh, in which stands a large boulder. The dolmen
was probably removed when the field was cultivated ; and we bare
seen com cut and bound on its site. It is on the lower slope of tike
hill, not on high ground as stated by Mr. Borlase.

(42). Balltkellt, Kilseily Parish (0.8. Sheet 44, No. 6).— lliis
occupies a noble station on a shoulder of the high hill of Xnw^aiee
south of the entrance of the valley at Broadf ord, and about 550 feet
above the sea. It commands a view out to Slieve Aug^ty and Canaa^
with the Shannon, the Fergus estuary, and a crowd of lakes. Beneatt
it lies the picturesque lake of Boon, with its wooded ahores and
cnmnoges; beyond it lie bogs pink with heath in the season. The
massive tower of Tierovannan, the white houses of (VOallaghaa's
Mills, and beyond them tho wooded demesnes of Kilkishen, Kilgoray,
Derrymore, and Fortanne, with Maryfort on its woody hilL Mr.

■ Plan, p. 121, fig. 4, «njm*«.

Wbstropp— Cb/^ Dolmens, and Pillars of East Cktre. 125

Borkwe figures and describes the dolmen.^ Ho lays stress on its having
been called '' Old Grania " by a woman ; but such names as " Granny's
beds " are not uncommon. 0' Curry mentions it, with Drummin, as a
'' broken giant's-graTO." It is a long dolmen, consisting of a row of
seren blocks,' extending for nearly 25 feet towards the E.N.E. They
are of irregular height ; but get lower eastward. The most western
is 6 feet 6 inches high, andhas a slab set parallel to it in the interior.
Three of the coTcrs still rest against the northern side ; and scarcely
any of the southern blocks are undisturbed.' The first and highest
slab in the west end and the other blocks (save the second and fourth,
which are of grey conglomerate) aie of green slate.

(43). Kkocbbhahvo, Kilseily Parish (0. 8. Sheet 44, No. 18).—
This picturesque and perfect cist lies far up the hill-side, aboye the
Dromsillagh river in a rushy field, sheeted with scabious and bilberry,
while long ferns grow in its chamber. The ridge of Knockaphunta,
pmple with heather, rises boldly not far away to the north-west ; and
to the east, beyond a picturesque ** screen " of fir-trees, we get a fine
view of Slieve Eimalta— the Keeper Hill— «nd its attendant ridges, and
the broken edge of Tipperary and Limerick to the serrated peaks of
the Galtees ; southward, we overlook the plateau and the valley of
the Drumsillagh stream, down to Trough.

The monument is shown on the 1840 map as a large block
supported at its western end. Tradition says it was used for the mass
doling the stress of the penal laws ; and a hollow near Knockaphunta
is also said to have been a place of secret worship.

The cist is of massive sand-stone slabs; they measure respectively:
the northern, nearly 10 feet long by 16 inches to 18 inches thick,
and 3 feet 8 inches high inside, the peat rising over a foot outside it ;
the southern, 7 feet 2 inches long and 13 inches thick, being the same
height ; it leans slightly outwards ; the east, 3 feet 4 inches long ; it is
slightly displaced ; and a slab 32 inches wide has been removed from
the south-east angle ; the west, 4 feet by 9 inches, and has fallen out.
The cist tapers from 3 feet 4 inches to 3 feet ; the cover does not slope,
and is a finely-shaped sandstone slab, from 4 feet 6 inches at the
west to 3 feet 3 inches at the east ; 7 feet 6 inches long, and 17 inches
to 1 8 inches thick ; it has curious coiTugations and a small round hole,
pexhaps an '' elf mill,'* such as occurs in the cover of more than one

» *• Dohneni of Ireland," vol. i., p. 97. l^ew and plan.

• Ths •« granite " mentioned by Mr. Borlase is, I think, a grey conglomerate
oeeuning eUewhere on the bills.

* See plan, p. 121, fig. 1, wpru ; also Plate vni., fig. 1.

126 Proceedings of the Royal IrUh Acadetny.

of the Swedish dolmens, and even in some Irish ones. The axis of the
cist lies towards the E.N.E.^

SIlLLOKBirXEDT.

(44), KiLLOKBNNEDT,« Killokenncdy Parish (O.S. Sheet 44, No. 3}.
— The townland and parish deriTe their names from the church of the
O'Eennedys, a once powerful clan of the Hymbloid, expelled by the
O'Briens and Macnamaras after the battle of Dysert O'Dea, in 1318.
The church was probably founded by Cronan (perhaps of Tomgraney},
to whom its well was dedicated. Passing the steep bohereen past \U
iyied and broken walls, we find the " giant's grave," high up the steep
hillside to the north. The dolmen stands on a little drift moond,
projecting from the slope, and is nearly 900 feet above the sea, and
650 above the road from Broadford to Kilbane. The hills rvshfi
behind it up to Gragnamurragh, which is 1729 feet high, and Glen-
nagalliagh mountain, 1746 feet high, the highest point in county
Glare. A slip has taken place above, and partly buried the monument,
which is a "long grave." The outlook is very fine, far over Glcno-
mera to the Galtees, and Knockfiema ; flanked by the rounded hills
of Glennagalliagh and Gloonyconry ; southward, the view includes
the pleasing (if not strikingly picturesque) valley to Hordleston.

The complete monument is 17 feet long, or 27 feet if we include
a compartment at its western end, of which the two northern slabs
(6 feet 8 inches and 3 feet 6 inches long) remain. These are greatly
distorted by the slipping of the earth. The west end is a stroog
block, measuring (so far as we are able to reach among the other stones
in which it is embedded) 5 feet 10 inches long, 20 inches thick, sod
at least 4 feet high. The nmin structure had three side slabs to the
south, about 5 feet 5 inches, 5 feet 10 inches, and 4 feet 6 inches long;
the eastern is prostrate ; on it lie two other slabs, evidently the eastern
end slab and the cover. There are four cover blocks, each from 6 fei t
to 6 feet 6 inches long, and partly buried on the north. A nairov
block, 5 feet 8 inches long, slopes out from the south-west angle, tod
a great block of slate rock (evidently natural) projects diagonally to
the south-east, and a little stream runs down the face. It is haid to
give further measurements of a structure whose slabs lie piled two and
three deep ; but from the ends it seems to have tapered eastward from
7 feet '2 inches to 3 feet 3 inches.

» Plan, p. 121, fig. 10, 8upra ; Plato viii., tig. 2.
2 Killoj^ennedid in the Papal Taxation, 1302.

Westropp— Ciifo, Dolmem^ and Pillars of East Clare. 127

Four other dolmens lay along the southern side of the valley in this
parish.

(45). FoBMOTLEMOKE, Killokonnedy Parish (0.8. Sheet 44, No. 9).
— ^This is the Formaol of the 1390 rental of the Macnamaras. A fine
long dolmen stands (as the name implies) on a conspicuous bare ridge,
to the cast of the road from Limerick to Broadf ord, before it dips boldly
into the valley near the perfect earth fort of lisnagry. It has been
planned by Mr. Borlase^ in his usual careful manner. It was embedded
in a modem house, of which (like the dolmens of Slievenaglasha and
Commons in this country) it formed a part. The buildings are nearly
demolished; but the dolmen is almost perfect. It consists of four
blocks of sandstone and slate to the north, and five longer blocks to
the south ; outside and parallel to these are others — one to the north and
two to the south, showing that, like the grave at Ballyogan, it had
side rows. Two large cover slabs rest upon it, and others lie about.
Its axis lie E.8.E. and W.N.W. It measures 17 feet long, and tapers
eastward from 3 feet to 2 feet 4 inches.

(46). G1.00VTCONBTHORE, Killokennedy Parish (O.S. Sheet 44,
No. 7).— This townland, the Cluan TJi Chonaire of the 1390 rental,
possessed a group of three dolmens. The two existing ones are on a
rounded shoulder; the larger on a choice situation, crowning a hummock
of cultivated land. It is in sight of the dolmens of Lackareagh,
Killokennedy, Ardskeagh, and, I think, Formoyle, and looks out of the
mouth of the valley over Broadford. The larger dolmen is of coarse
and shapeless sandstone blocks, and has been planned and figured
by Mr. Borlase.' It has tiie west end, two blocks to the north and one
to the south. The cover measures 7 feet 6 inches by 4 feet 3 inches.
Seven of the side blocks remain.'

The second cist lies lower than the first, which is in sight of it. It is
not marked on themapsor noted by Mr. Borlase. Ithas a smaU chamber,
8 feet 3 inches by 3 feet 8 inches to 3 feet 4 inches internally, and
tapering eastward. There are two slabs to the north, one 6 feet by 2 feet
by 2 feet thick to the south, and one at the west end ; the top has been
removed.^ The third lay high up the second ridge of the hill. I saw
no trace of it ; and the new maps mark its site under a fence, by

> '' Dolmens of Ireland,'* voL i., pp. 08, 99. Plan. See aleo p. 121, t%. 2,

* Jhid,^ pp. 99, 100. View and plan.
•Plan, p. 121, fig. 3, tupra; Plate vii., fig. 2.

* Plan, p. 121, tig. 6, ntpra.

128 PtvceedingB of the Boyal Link Academy.

which it was destroyed or covered. It is shown on the 1S40 flutps
as a large slab supported by smaller ones, and was eyidenily a
cist.

O'BRiEir's Bbidoe.

O'Brien's Bridge is a straggling late parish, and cuts into the
Glenomera and Broadford yalleys at their junction to the nortii-east
Kcar this point stood a dolmen, and another one lay at the fuitlier
and lowest ridges of the hills to the south-east, three miles away.

(47). Lackabsagh, O'Brien's Bridge Parish (O. 8. Sheet 44, Ko. 8).
— This cist is shown on the map of 1840 as a laige rDgnlardah,
evidently the north side, against which a long and more irregular dab,
of about the same size, is leaning, with its nearer end on the grovod.
It is there called " Dermot and Grania's Bed." When Mr. Borliss
visited the site, there were only the remains of a cairn ; and it is
only shown as a ''site" on the map made in 1893 (published in
1898). It stood near the summit of the hill above Glennagallii^^
Valley, nearly 1 1 80 feet above the sea. The stream and gorge of
AiUenagommaun runs down the flank of the hill to the north.

(48). Abonatagglb,' O'Brien's Bridge Parish (0.8. Sheet M,
No. 4). — This dolmen lies on the northern slope of a low ridge, B0t
very far from Bridgetown and O'Brien's Bridge, near the so-caDed
" Cromwell's-road." It commands a fine view of the great blnffi of
purple and brown which we have been exploring. It is embedded in
deep heather and bracken; and its interior is a veritable garden of
delicate ferns and sorrel. Mr. Borlase,' in publishing a version of my
notes, considered that there had been a row of enclosing slabs roand
the tomb. I do not think this was the case, as none of the siqh
posed '' peristyle " is standing, and no loose blocks (even) lie to the
north. It is shown on the 1840 map as a large block, supported on
two others. It is really a complex cist, with two if not three compart-
ments ; the most perfect resembling the little sketch on the map, and
is about 8 feet 6 inches. At the west end are two lai^ slabs, 6 lect
6 inches long, the more eastern forming the end of the cist. The
first compartment of this is formed of irregular gritstone Uocks; tiie
cover is 7 feet 8 inches long, 5 feet to 6 feet 6 inches wide, and
11 feet to 20 inches thick. Beyond this the cist eontiniies in a very

^ Locally '* 01ounagoll6ch.*'

' Ardataggle of the Ordnance Surrey maps.

s (• Dolmens of Ireland," vol. i., p. 131. See plan, p. Itl, fig. 9, m^tw.

Westropp— Ci«f«, Dolmet^ and JPillan qfEwt Clare. 12&

deiaeed conditioii for 1 1 foot, and, porhaps, had throe compartments^
The azia lies E.N.E. and WAW. It is now locally known as the
"lobha," bat is oaUed "Dermot and Grania's Bod" on the 184(^
map.

(49). Clo©hoolia, Clonlea Parish (0.8. Sheet 62, No. 4).— Far
from all other dolmens, in a yallcy in the plateau of Slievc Bomagh,
stands the wreck of a small cist. It lies near Oatfield, and to the
north of the road running from Sixmilebridge, through the woods of
Mount lorers, and past Trough to the pretty village of Clonlara and
the falls of the Shannon at Doonass. It is in a meadow, whence, to
the east, wc see the great dome of the EeeperHill, and to the south, oyer
a low bo^y hollow and ridge, the blue lake of Cobneen, where, in 1 3 1 3,
Lochlain Macnamara, chief of that powerful clan, was beheaded by his
foes during De Clare's wars. " Spray-showering, wind-swept Loch
Colmin of the easy landing-places and green shores, . . • Loch of
Colmin that has a cruel story."' The cist is called <'Dermot and
Orania*s Bed " on the 1840 map, and was possibly perfect ntthat date,
for it is shown as an irregular oblong enclosure. It is nearly buried
in the field ; the west end, two southern blocks, and, perhaps, part of
the north slab, remain. The first is 4 feet 6 inches by 6 inches, and
only rises 2 feet above the field ; the other two are 2 feet 4 inches by
6 inches, and 6 feet 10 inches by 10 inches to 7 inches ; the east end
level with the ground. It sloped and tapered eastward, and, if the
block north of the east end be in siiu^ was 9 feet 4 inches by from

2 feet 6 inches to 12 inches wide. A tiny holly-bush springs between
the side blocks. Two other slabs lie to the south, one 6 feet east and
west, and 4 feet 4 inches north and south ; and probably tlie cover lies

3 feet from the east end ; the other is 3 feet 9 inches long, and
12 inches thick, and may have been a north block: all are of fine
brown sandstone.*

This concludes these notes on the dolmens of Eastern Clare. We
ba:v6 described all those marked on the Ordnance Survey maps or
known to us. The oairns, which are usually dilapidated, and some-
tiaies of doubtful age, are, however, omitted.

The vast majority of the dolmens are considered by the peasantry
to be " Dermot and Orania'sBeds" ; but the legend, save from modem
books, 1^, I believe, extinct. In some cases they are recognised as

1 <' Wan of Turlougb, 1318 ; see alao Joomal Royal Society of Antiqitariea of
Inlnd, vol zzi. (1891), p. 887.
* Plaa, p. 121, fig. 7, 9upm.

130 Proceedings of the Royal Irish Academy.

graves, occaBionally as ''giant's graves,"^ but no legends of their
occupants seem to exist. In other cases they are supposed to be
Christian altars (as Altoir TJltach and EnockshanTo) or wells (as
Tobcrgrania). I only met one legend, and that not from a local person,
but from a servant, that the " Druids ** used to offer black cocks upon
the Maryfort cist. I have also heard on local authority that (over fifty
years ago) a black cock "without a white feather" was actually
offered on the giant's grave at Carnelly in the same county. This was
intended to bring about the fulfilment of the sacrificer's dearest wish ;
but was also believed to have brought misfortune in its train. Whether
the dolmen was an accidental rather than an essential adjunct of these
unholy rites is not clear; probably the ** Druidical " pseudo-archsBology
of the earlier nineteenth century filtered into the minds of some of the
peasantry, superseding their own rational tradition that the dolmens
were sepulchral by that of the belief that they were sacrificial altars
of the pagans.

In later days (about 1879) great excitement and anger was caused
in a place about a mile from the Maryfort dolmen, by four quarters of
a beast having been found '' offered" at the four comers of a certain
field. The comparatively recent date of the latter event preyents my
saying more on this very obscure but curious subject, though I am
acquainted with the names and circumstances ; but these two cases
show that it is not impossible that {minus the " Druids *') tiie Maryfort
story may have, at least, some probability. The most general impression
seems to be that they were graves. All seem to have been opened
before living memory, except, perhaps, Ballinphunta. Only one find^
that of a gold fibula, is recorded (Knocknalappa). Owing to the lapse
of time since they were explored, all memory of finds of bones (as at
various dolmens in the Burren) or pottery is lost.

To sunmiarise for the four baronies surveyed in these Papers, there
are simple cists of four blocks and a cover — Ballyhickey, Ballyma-
conna, Ball3rmacloon, Caherloghan (4 ? 5), Kilvoydan, Monanoe,
Toonagh (2 ? 3), Ballysheen, DromuUan (?). Tobergrania, Kiltanon,
Maryfort, Miltown (at least 6), Moymore, Fomerla (1 ?2) Elmhill,
Violet Hill, Knockshanvo, Lackareaghmore. (In all 26 or 29.)

Simple cists of more than four side-blocks and a cover — ^Knappoge
(with enclosure), Knocknalappa, Altoir Ultach, Ballycroum, Drumin-
doora (2), Cappaghbane, Gorracloonbeg, Kiltanon, Newgiove (with

^ Dermot and Qrania are Christian « saints from Feakle,'* Gnnia bung a maa
in the legend at Ballycroum. See Proc. R.I.A., vol. iv., set. iiL, p. 9K

Wbstropp — cuts J Dolmens J and Pillars of Hast Clare. 131

enclosure), Bosalara, Tyrcdagh Lower, Drummin, Ardskeagh (f ),
Cloghoolia. (In all 13.)

Cists of several chambers — Rjlane, Ballinphunta, Gaheraphuca>
Kilcoman, Tyredagb Upper, Ardnataggle. (In all 6.)

Long Qraves — Ballybogan (2\ Miltown, Balljkelly, Formoylemore,
Killokennedy. (In all 6.)

Enclosures of blocks and circles — Clooney (2) Dooneen. (In all 3.)

MassiTe top block and double- walled cist — ^Derrymore.

Pillars — Cranagher (5), Magh Adbair, Newgrore, Tyredagb,
Tomgraney. (In all 9.)

Making the total of some 66 dolmens, 9 pillars, 3 enclosures.'

ILLUSTKATIONS OF DOLMENS.

Pl4ts V. 1. Caherapbuca — Cnuheen, from south.

2. Knoclmafearbreaga — pillars, from south.

Pi, ATI YI. 1. Knocknalappa — Sixmilebridge, Bossroe Castlu and Lake in
distance.
2. Derrymore-Tiilla.

PukTB YII. 1. Cappaghbane — ScarifF, from north-east ; Lough Derg uud
hills of Thountinna, Ogoimelloe and Glennagalliagh in the
distance.

2. Cloonjconrymore — Broadford, from south; Knocksiso, Doon
Lake, and Broadford in distance.

Plate YIII. 1. Ballykelly — Broadford, from south-west ; Doon and Kilgoroy
Lakes in distance.
2. Knockshanvo — Broadford, from west ; Keeper HiU in dibtance.

1 The following photographs of dolmens in the baronies of Tulla aro in the
collection of the Boyal Society of Antiquaries of Ireland : — Tyredagh, Miltown,
XewgTOTe, Maryfort, Bosslara, Altoir-Ultach, Tobergrania, Corracloon, Elmhill,
Cloooyconry (2), Pormoyle, Killokennedy, Knockshanvo, Cloghoolia, Violet Hill,
X>rummin, Ballykelly, and Cappaghbane.

My thanks are due to my sister Mrs. O'Callaghan, Col. 0*Callaghan Westropp,
ReT. J' B. Greer, Mr. A. G. Creagh, and the late Mr. Pierce 0*Brien, for assistance
ia finding and examining the dolmens of Eastern Clare.

132

Proceidmgs of the Ib^ Iruh Acadimg.

INDEX

TO THE ABOVE TWO PAPERS, TAKING TBB NUMBERS
OF THE LOCALITIES.

l7%eJ!ffurM refer to the «Ar<tofU.]

Ardnatoggle, 48.
Ardskeagh, 41.

BaUinphunta, 21.
Bally croum, 33.
Ballyhickey, eee Hazelwood.
Ballykelly, 42.
Bully madoon, 14.
Ballymoconna, 3.
Ballymullen, eee Miltovii.
Bally ogan, 4.
Ballyaheenbeg, 19.
Ballyslattery, eee Newgrove.
Bealkelly-Purdon, 38.
Brickhill, 20.

Caherapbuca, 1.
Gaherloghan, 10.
•Cappaghbane, 36.
Olassagh, »ee Knooknafearbreaga.
•Clogher, tee Derrymore.
Cloghlea, tee Tomgraney.
•Cloghoolia, 49.
Clooney, 6.
<31oonycoiirymore, 46.
Corbeliagh, tee Drumandooiiu
•CoiTocloonbeg, 84.
•CitMighane, tee Balliiiphunta.

Derrymore, 31.
Booneen, 12.
DrumandooTu, 32.
Brummin, 39.
Drummullan, 17.

Elmliil], 37.

Fomerla, 27.
FormoyleiDore, 4o.

Hazelvood, 11.

Eilooman, 18.
KiUokeniiedy, 44.
KiltanoD, 24.
Kilvoydan, 2.
Knappoge, 15.
Knocknafearbreaga, 7.
Knocknali4>pa, 16.
KnockahanTO, 43.

Laekareagh, 47.
Lismekan, aee Maryfort.

Mac^ Adhair, 8.
Mcryfoit, 29.
Miltown, 2d.
Monanoe, 13.
Moymore, 28.

Newgrove, 26.

Rofislara, 30.
Rylane, 5.

Tooigzaneyy 3o.
Toonagh, 9.
Tyredagb, 22, 23.

Violet Hill, 40.

Proc. B, I. A., Vol. XXrV., Sec C

Plate I.

,/'

?A&fjJt9fi>-^j';-#"^]

-xi.

^••'•^■

•.;-v

f ♦

AUi

Arabic Inscription from Rhodbsia.
Dated a.k. 95 (a.d. 713-4).

Proc. R. I. A., Vol. XXrV., Sec. C.

Plate IL

is
a:

►3 «

si
S "S

il
II

Proc. R. I. A., Vol. XXrV., Sec. C.

Plate m.

y. p. (TReiliy

Wateh-mill with*Hoiuzontal Wmebl.
;Tvo miles W. of Comillas, Province of Santnnder, N. Spain, 1859-60.

Proc. R. I. A., Vol. XXIV., Sec. C.

Plate rV.

SOCKET-STONB OF AN IrISH HORIZONTAL WaTEB-WHEEL.

Proc. R. I. Acad., Vol. XXIV., Sec. C.

Plate V.

1. Cahrbapuuca, Ckvshebn.

2. Knocknapbaebubaoa, Cloonby.

Proc. R. I. Acad., Vol. XXIV., Sec. C.

Plate VI.

1. KnoCKXALAFPA, SlXMILEBllIDQE.

2. DSUUYMOUB, TULLA.

Proo. R. I. Acad., VoL XXIV., Sec. C.

Plate VII.

1. Cappaohbane, Scariff.

2. Cloonyconbymorb, Broadford.

Proc. R.I.Acad., Vol. XXIV., Sec. C.

Plate VIII.

1. Balltkbllt, Buoadforo.

2. KNOCKBHAMVOy BbuADFOED.

2t3x^ ISo^.Zo

\* /

[133 1

IX.

SOME ILLXISTKATIONS OF THE COMMEECIAL HISTOEY
OF DUBLIN IN THE EIGHTEENTH CENTFBT,

(PtATW IX.-XIL)

Bt C. LITTON FALKINEE, M.A.

[Bead Juki 9, and Jura 23, 1902.]

Ths Yolumes placed at the disposal of the Council by Colonel Welch
came into bis possession as ezecntor of the late Charles Haliday,
by whose zeal as a collector of Irish books and manuscripts the
Academy has so largely profited. They consist of two volumes of
minutes, each bearing closely on the mgin and early history of two
important Dublin institutions, viz. the Port and Docks Board and
the Chamber of Commerce. Of the two volumes, the first in date, if
not in point of interest, is an old folio bound in calf, and labelled
** Ballast Office, 1708 to 1712." It contains the minutes of the pro-
ceedings of the Committee of Directors for the Ballast Office during
the first four years of the existence of that body. These minutes add
considerably to our knowledge of the development of the Port of
Dublin. A portion of them has been published in an abbreviated
form in Mr. William Gibbon's notes to those ''Observations Explanatory
of Sir Bernard de Gomme's map, made a.d. 1673," which are printed
aa an appendix to Mr. HaHday's " Scandinavian Kingdom of Dublin,"
and which form perhaps the fullest account yet attempted of the history
of the Fort of Dublin. Through the courtesy of Mr. Proud, the
secretary, the writer has been permitted to examine the records of the
Fort and Docks Board, the successors of the Ballast Committee, and
has ascertained that the earliest volume of minutes in the possession of
that body is the Committee Book of the Ballast Committee, commencing
March 3rd, 1721. The volume acquired by the Academy is thus
aereral years earlier than the oldest official record, and as elucidating
the condition of the harbour of Dublin at the very commencement of
the eighteenth century, it is of considerable value to all who are
interested in the history of the development of our city.

The second and perhaps the more important of these volumes is a
folio manuscript book, bound in green boards, and labelled *' Merchants'

m.2.A. PHOC, TOL. XXIT., SBC. C.J [10]

134 ProceeiUuga of the Moyal IrUh Aeademf/.

Bough Book. ' ' It contains the minates of a body called the Committee
of Merchants, apparently a sort of Council of the Guild of Merchants,
which appears to have taken charge of the commercial interests of
Dublin during a considerable portion of the eighteenth century. The
entries in this volume cover a period of fifteen years, viz. from 10th
February, 1768, to 10th February, 1783 ; and the importance of thiF
record in relation to the history of our capital may be measured by
the fact that it opens with a statement of the circumstances in which tiie
fine building, long known as the Boyal Exchange, and now familiar to
us as the City Hall, originated, and doses with a *' Plan for institating
a Chamber of Commerce in this city," which was the direct origin of
the flourishing mercantile association so well known to us now under
that name. Incidentally the volume covers a number of topics of
interest touching on the development of Dublin, as, for instance, the
building of the present Custom House — ^a project vehemently opposed
by the merchants of the day, on the ground that it tended to shift the
commercial centre of gravity in Dublin from Essex-bridge sad
Dame-street, the neighbourhood of the old Custom Hoaeey to the
inconvenient and then scarcely accessible slobland of the K<xth
Lotto.

As in the case of the Ballast Committee's minutes, so in this^fte
writer has been enabled to consult the minutes of the modem body to
whose chronicles the book relates, and has ascertained that flwii^
the minutes of the Chamber of Commerce are extant for ten jeszv
immediately succeeding its institution in 1783, no document sur-
vives to indicate in what manner the Chamber came into existence.
The Bough Minute Book is therefore valuable as containing as
authentic statement of the circumstances in which one of the
most important of our Dublin corporate bodies came to be fonned.
Advantage has been taken of this acquisition of volumes besiing
so directly on two important Dublin institutions which date fron
the eighteenth century, not only to give a brief de8cripti<m ol
the nature of their contents, but to offer some account of the
origin of those weU-known corporations, the Port and Docks Bosid,
formerly known as the Ballast Board, and the Dublin Chamber ol
Commerce. The history of both institutions throws considenhk
light on the commercial development of Dublin ; and a valuable
sidelight is thrown on the same topic by the story of the Ooael
Qalley Society, which is also included in the present paper in con-
nexion with one of the Society's (}old Medals, lately added to
the Academ3r's collection.

Falkiner— iHi«^n/^/aw« of Commercial History of Dublin. 13S

I. — Osienr of th£ Ballast Offxck akd Port akd Docks BoAkd.

Projects for the improvement of the harbour of Dublin and the

better regulation of the shipping of the port appear to have been

frequent in the last quarter of the seventeenth century. The fear

lest the audacity of the Dutch and the defenceless condition of the

envirous should expose the capital to attack had led, in 1673, to

Sir Bernard de Gomme's well-known ** Survey of the city of Dublin

and part of the harbour below Bingsend ; '' and although this survey

was undertaken from purely military considerations, it naturally drew

the attention of mercantile people to the deficiencies of the port from a

commercial standpoint. The control of the port was vested at this

period in the Corporation of Dublin, to whom it had belonged from

the time of King John, when a royal charter had endowed the

citizens^ with so much of the river and estuary of the Liffey as ran

within the city franchises.' The Corporation does not appear to have

paid close attention to that part of its responsibilities which concerned

the harbour ; but in the year following De Gomme's visit their

attention was called to the matter by the visit of Andrew Yarranton,

an expert on harbour improvement.' Yarranton, '' acquainting the

Lord Mayor with his thoughts as to the making a very good harbour

at Bings^," was '' importuned to bestow some time in a survey and

diacovery thereof," and devoted three weeks to this task. But though

the survey was made, no steps were taken by the Corporation, and the

first step towards providing a proper machinery for the control of the

port was left to private enterprise. In 1676 one Thomas Howard

petitioned the Irish Privy Council for a patent for the provision

of a Ballast Office in all the ports of Ireland. Howard's proposal

stirred the city fathers to activity. Protesting against the petition,

00 far as it related to Dublin, as an encroachment on their civic rights,

they appointed a committee to consider the erection of a Ballast OfficOi

** the profits whereof is intended for the King's Hospital," and prayed

the Lord Lieutenant that no patent should pass to Howard. The

protest of the Corporation was effective, and Howard, though he had

1 GQbert'e Hittorical and Municipal Documents of Ireland, 1172-1320.

* The Mayor of Dublin anciently exercised, as Admiral of tHe Port of Dublin,
« jurisdietion which appears to have extended from Skerries to Arklow, and the
city was entitled to the customs of all merchandise within those limits (Halidsy's
Beandioamn Kingdom of Dublin, pages 139 and 246).

> Haliday's Bcandinavian Kingdom of Dublin, p. 242.

136 Proceedings of the Royal Irish Academy.

obtained a patent in England for the erection of a Ballast Office in
Ireland, was nnsaccesafal in lus application. Accordingly bis next
move was to petition tbe dtj, in association with bis brotber, for a
lease of tbe port of Dublin at £50 a-year, in return for wbicb be
nndortook to surrender bis English patent. A lease for thiity-oae
years was granted ; but as tbe Howards took no step to perfect it,
it was three years later declared void, and petition was made by ibe
Corporation for a patent to the city for a Ballast Office. Tbe activify
displayed on this as on the preyious occasion was due to tbe exerticiis
of a priyate individual who had taken up Howard's project.

In the year 1697 one Captain Dayison bad made a proposal to
tbe city to erect on or near the bar of Dublin a ligbtboase^ forty
feet above water, wbicb should be enclosed with a small fort of tbiitf
guns capable of defending tbe harbour, and at tbe same time be
proposed a Ballast Office, " by which ships should be supplied with
ballast from such places only as should tend to tbe bettering tibe
harbour." In 1700, having obtained tbe approval of the Dablin
merchants and captains of ships trading there, and being encouraged
by the Irish Gk)vemment, Davison proceeded to London, and
petitioned William III. for authority to proceed with tbe work, and
for a grant of the lighthouse and Ballast Office. His petition was
referred to the* Irish Lords Justices, who reported that tbe design
was useful and " absolutely necessary for tbe preserving tbe trade of
tbe place '' ; but stated that tbe '* city desired that tbe grant thereof
might be made to them." Tbe Lords Justices accordingly re-
commended that '' lest it should be thought a business of damoor to
grant such a thing away from a whole city, the grant should be made
to Davison as the instrument of the Corporation."

The matter was then referred to the Committee of tbe Frivy
Council for the affairs of Ireland, *' to investigate the daim of the
several parties pretending to a right in tbe carrying on of this wwk,"
several other persons having meantime sought a patent. The
Committee found the claims of Davison infinitely superior to those of
all private rivals ; but tbe city of Dublin alleging several anciokt
charters by which they had titie to the ground from whence tbe said
ballast was proposed to be taken, '' and having in the sitting of tbe
last Parliament obtained a bill to be sent over for the establishment of
a Ballast Office," they recommended the claims of tbe citiirens to Her

^ Memoiial about the Light House at Dublin. Brit. Mumuid, Add. MS.
21136 folio 82. Printed in Dublin Corporation Keoords, vi. p. 609.

Falkimbr — Ittmtratiom of Commercial Historic of Dublin. 137

Hajesty^s fayour in preference to those of any priTate persons. They
at the same time expressed an opinion that, if the authority were
given to the city of Dablin, Captain Davison should be employed on
the work.

Ko action appears to have been taken upon this report, and in
1702 Davison renewed his application,^ which was again opposed by
the Dublin civic authorities as highly prejudicial to the city, and the
project seems to have remained in abeyance for some years. In 1707|
however, a petition under the city seal was ordered to be addressed
to His Boyal Highness Prince George of Denmark, Queen Anne's
Consort, then Lord High Admiral of Great Britain and Ireland,
for erecting a Ballast Office. This petition set forth that ** the port
and river of Dublin are almost choked up, and are very unsafe
by the irregular taking in and throwing out of ballast," and
besought favourable consideration for a fresh bill which had been
sent over for erecting a Ballast Office, the petitioners being advised
that without legislation no duty for the support of such office
when erected could be imposed on shipping. The petition further
averred that ^* nothing can contribute more to the safety of the lives
of seafaring men who resort hither than the mending of one of the
most dangerous ports in Her Majesty's dominions " ; and in order to
obviate the possibility of a grant to any private individual rather than
to the city, it expressed the willingness of the assembly that all
profits arising from the Ballast Office ^' should be applied towards the
maintenance of the poor boys in the Blue Coat Hospital in this city,
whereby they are instructed in navigation to qualify them for Her
Majesty's sea service."' In a letter from the Lord Mayor to Prince
Gborge, in furtherance of the city claim, it was also stated that the
port was so unsafe that there was scarce depth of water left for a
small vessel to ride, where some years before a man of war could
aaf ely anchor.'

These applications were not favourably entertained by the
Admiralty, Pxince George of Denmark being of opinion that the erecting
of a Ballast Office by Act of Parliament was a direct infringe-
ment of the lights of his office of Lord High Admiral. He therefore
expressed his intention of opposing the bill.* But His Boyal Highness
*• having a paiticular regard to the cleansing of the port of Dublin,"

> Dublin Corporation Beoords, toI. yi. p. 272. ' lb. p. 374-6. > lb. p. 616.
« Letter of Josiah Borchett, Secretary to the Admiralty. Dublin Corporation
Beeorda, vol. vL p. 618.

138 Proceedings of the Royal Irish Academy.

was content '' if the Lord Mayor would make proper application to
him and to him only/' to grant a lease of a Ballast Office to the dtj of
Dublin for a term of years, provided that the surplns of the port does
should be applied to the benefit of the Blue Coat School in tiie
manner already mentioned. The objections then raised by the
Admiralty were combated in a very vigorous letter addressed to
Lord Sunderland, the Secretary of State, in which it was pointed out
that the sand and soil whence the ballast was to be taken was the
inheritance of the city of Dublin, which by several charters had the
jurisdiction of the Admiralty granted to it, notwithstanding which the
city would be willing to waive all such rights and take a lease from the
Lord High Admiral, were it not that powers under an Act of Pailis-
ment were absolutely necessary, as a means of obviating the difficulty
raised by Prince George, to enforce payment of harbour dues.^ And
in token of the readiness of the city to admit the claims of tbe
Admiralty, an offer was made on the part of the Corporation to
add to the bill a clause saving the Admiralty jurisdiction, by pro-
viding in the following quaint terms for the city's '' yielding and
paying therefor and thereout to His Royal Highness Prince George
of Denmark, Lord High Admiral of Great Britain, and to his
successors. Lord High Admirals of the same, one hundred yaids
of best Holland duck, that shall be made or manufactured within
the realm of Ireland, at tbe Admiralty Office of London on
every first day of January for ever hereafter." The solution thns
propcped was accepted by the Admiralty, and the heads of the boll
liaving been approved in England, there was passed through the
Irish Parliament in 1707 the statute of the 6th Anne, chaptorSO,
entitled, <' An Act for Cleansing the Port, Harbour, and Biver of
Dublin, and for erecting a Ballast Office in the said City."

The minute-book acquired by the Academy contains the recoid
of the steps first taken to put this Act in motion, and must fbnn the
materials for the first chapter in any history of the Ballast Office, or
of its successor, the Port and Docks Board (see Appendix I.).

II.— Obioin of thx Dublin Chambbr of Commbbob.

No record exists of the circumstances under which our DuUin
Chamber of Commerce was founded, and inquiries recently institaM
regarding its origin show that, save in so far as they are contained in

' Dublin Coi-poralion Kecoi-ds, yoI. vi. p. 621.

Falkinbr — lllmtratiam of Commercial Hhtory of Dublin. 139

the Rough Minute-Book of the Committee of Merchants recently
acquired hj the Academy, those circumstances cannot now he traced.
For although the Chamher of Commerce still possesses among its
records the first minute-hook of the Chamher, that volume throws no
light upon the mode in which the Chamher of Commerce was first
constituted. It begins with an entry dated March 18, 1783, which
records the calling of a meeting for March 22 ensuing to elect a
President, two Vice-Presidents, and a Treasurer, and to determine on
the duties of a Secretary. And the next entry duly announces the
election of those officers and the appointment of one William Shannon
88 Secretary at an annual salary of £30. But of the circumstances
leading up to these proceedings no trace remains. The mintite-hook of
the Committee of Merchants not only unexpectedly supplies the lost
details, hut incidentally gives us a very interesting chapter in the
history of the mercantile development of Dublin.

In the account given by Sir John GKlbert in his History of Dublin
of the origin of the Royal Exchange (now the City Hall), mention is
made of an association of merchants formed to resist the exactions of
one Thomas Allen, who, having been appointed in the year 1763 to
the office of Tasterof Wines, end^vouredto enforce for his own advantage
a fee of two shillings per tun on all wines and other liquors imported into
Ireland. The struggle against this arbitrary tax did not, according
to the authority quoted by Gilbert, last long ; *' and turning their
thoughts to the best mode of applying the redundant subscriptions
raised to conduct the opposition," the members unanimously adopted
the idea of building a commodious building for the meeting of
merchants and traders. A situation having been fixed upon, the
purchase-money, £13,000, was obtained from Parliament by the
zeal and activity of Dr. Lucas, then one of the city representatives.
The building so erected was the Royal Exchange, of which the foun-
diition stone was laid in 1769, which was opened ten years later.*
It is to the proceedings of the Committee of Merchants, by whom the
building of tiie Exchange was promoted and conducted, that this Rough
Minute-Book relates ; and the record shows that the committee not only
performed for many years many of the functions now discharged by the
Chamber of Commerce, but was tbe actual parent of that institution.

The minute-book opens with the record of a resolution ^' that the
ground for building an Exchange be conveyed to the Corporation of
the Guild of Merchants, and the planning of the building and carry-

> Gilbert*! History of Dublin, iL 66.

140 Proceedings of (he Royal IrUh Academy.

ing into the ezecution of the Exchange conducted by a committee
t>f certain citizens therein named, together with fifteen wholfimle
merchants, freemen of the Guild of Merchants to be chosen bj the
wholesale freemen of the Guild of Merchants from among theoudTes."
The earlier entries in the book are concerned with the steps taken to
raise funds for the erection of the Exchange, the monej Toted by
Parliament being absorbed by the cost of the site. These funds irere
for the most part obtained by means of lotteries. On Feb. 23, 1768,
it was resolved '^ that a scheme be grafted on the State Lottery now
depending in England in order to raise a further sum towards the
expense of erecting an Exchange on the reserred ground on Cork Hill,
and that an advertisement for that purpose be published in due time
in all the Dublin papers, except the Gazette." The minute-book is
crowded with entries, between the dates 1768 and 1778, relatingto
the progress of the building, including a resolution of 24th Feb., 1769,
for the payment of the bills ''for the expenses of entertaining theLoid
Lieutenant on the occasion of his laying the foundation stone," not-
withstanding the Conmiittee are of opinion they are exceedingly
extravagant. The bills amounted to £298 13«. Hd.

But the Conmiittee of Merchants was concerned with topics more
serious than these. They busied themselves from the first in sack
matters as the procuring an amendment in the Irish Bankruptcy Iawb,
in movements for the direct importation of spirits from the British
plantations without first landing them in Great Britain, and other
questions directly affecting the conunercial interests of Ireland. That
they also took a lively interest in the mercantile development of their
own city is evident from the space devoted in their records to sneh
topics as the building pf the new Custom House, and a proposal f<ff
greeting Law Courts in College-green. Both of these projects wei«
npposed by the merchants on the ground that they tended to diiit
the commerce of Dublin from its old centre in the neighbourhood of
Essex-quay ; the latter scheme was especially obnoxious as tending
'' to the erection of a bridge east of Essex Bridge '' ; and the fonner
was formally condenmed as '* extremely injurious to the int^ests of
thousands of individuals, and highly prejudicial to the commeroe of
this city in general." ' It is interesting to note that the erection of tht

* Oa 30Ui Dec., 1773, it was nMolred:— *«That the remoral of theCiiitoa
House below Temple Lane slip will tend to draw the inhabitants of the dty farther
down the river, and so furnish a pretext for building a bridge to tlie east of S0(s
Bridge, which would be still more injurious to private property, totxmde, and to «ri-
gadon than even the removal of the Custom House.'*— (A<r»rf/rwwJfiiiaO hti)

FALKWER-~Illustration$ of Commercial History of Dublin. 141

lonner Cnatom HouBe had two generationB earlier led to Bimilar com*
plaints. But the objections of the merchants were, of course, unavail-
ing* The Commissioners of Beyenue pointed out that the increase of
building had been of late so rapid that the town which was formerly
terminated to the east of Essex Bridge was now divided by that
stmctore into equal parts, east and west, that the eastern portion had
no communication across the river save by ferries, and that as the city
must naturally continue to develop in an easterly direction, they would
•be highly blamable in preventing such a communication in the future.
The merchants, however, did not surrender without a struggle ; they
interviewed the Viceroy, petitioned Parliament, and invoked the aid of
ihe merchants of London ; and they voted gold snuff-boxes to two
London merchants who had interested themselves in promoting opposi-
tion among the traders of the English capital. The result of their
efforts was to retard the erection of the new Custom House for about-
ten years. But in 1781 the Commissioners of Bevenue were at length
empowered to build the Custom House on the site so much objected
to, and although at a public meeting summoned by the merchants
under the presidency of the Lord Mayor, a further petition was ordered
to be presented to the Viceroy by the members for the city, Mr*
Clements and Sir Samuel Broadstreet, the protest was unavailing.
The Custom House was built where it still stands, Carlisle (now
O'Connell) Bridge became an immediate necessity, and the develop-
ment of the city to the east and south-east at once proceeded apace.

It was probably a sense of the deficient authority of the Merchants'
Committee, as revealed by the failure of their opposition to the Custom
House scheme, which led to the institution of the more formal organi-
zution of a Chamber of Commerce. The change may also have been
hastened by on investigation into the .conduct of the lotteries held by
the Committee, which appears to have provoked some scandal, though
no proofs of fraud were established. It is certain, at all events, that
little more than a year later the Committee was convened to meet at
the Boyal Exchange on Eebruary 10, 1783, for the special purpose of
taking into consideration the '< Plan for instituting a Chamber of
Commerce in this city," a copy of which is printed as an Appendix to
this Paper (see Appendix II.). Resolutions affirming the plan were
at once adopted, and the Committee of Merchants, after a useful and
interesting existence of exactly fifteen years, merged in the Chamber
of Commerce of Dublin.

Although it is not the province of this Paper to further pursue the
history of the Chamber of Commerce, it appears desirable, inasmuch

142 Proceedings of the- Boyal Irish Academy.

M that liifltory has never been written, to note the steps which were
taken to proyide the new association with a formal conBtitatioB pur-
snant to the resolution jnst chronicled. One month after the final
meeting of the Committee of Merchants a ballot was held for the elec-
tion of a Council of forty-one members.' One hundred and flfty-tiiree
persons appear to have Toted, and Mr. Travers Hartley, l(mg the most
active member of the old Committee, who had been for many years a
representatiye of Dublin in the College-green Parliament as a follower
of Qrattan, was returned at the head of the list. At a farther maetiiig
held on March 22, for the election of officers, Mr. Hartley was eleeted
President of the Chumber^ — ^a position which he appears to have held
continuously down to 1788. In that year rules were drawn up
for the annual election of officers of the Chamber, but no electioa
under these rules is recorded in the minute-book, which is a Uank
-fxom March 29, 1788, to 1805, except for a single entry in 1791,
Whether or not the Chamber met during this long interval does not
certainly appear ; but from the fact that the first minute-book in the
possession of the Chamber of Commerce is indexed as '' Old Chamber,''
and that what is referred to as the '^ second " Chamber began to sit
in 1805, it may be assumed that the Chamber as originally started
failed to meet for several years, and was, in fact, during a period of
seventeen years a less efficient guardian of mercantile interests tiian
the old Committee of Merchants wliich it had replaced. The minute*
book ends with the year 1807. No records exist of any meetings from
that year until 1820, when the Chamber appears to have been reoon-
stituted ; and it is doubtful for how many years its proceedings were
suspended. From the latter date the manuscript records have been
preserved in perfect sequence, and are in the custody of the present
Secretary of the Chamber, Mr. Perry. The printed reports «f the
Chamber date from 1821.

III. — The Ouzel Galley Sogobtt.

At the end of the seventeenth century, in the closing years of the
reign o! William III., a vessel known as the '' Ouzel/' in the owner-'
ship of a Dublin merchant, and engaged, it is believed, in the Smyrna
trade, sailed from Kingsend for the Levant. Prior to her departure aha
had been insured against risks, with Dublin underwriters, in theusnal
way. In the ordinary course her absence would have been a lengthened

^ Minute8 of Chamber of Commerce.

Fai^iner — Illustrations of Commetvial History o/Dubliti. 143

one ; but when, after a lapse of some years, nothing had been heard of
her, she was assumed to have been lost at sea with all hands. The
owners duly claimed their insurance-money, which was paid by the
underwriters ; the ship was deemed to hare made her last voyage, and
the commercial transactions in respect of her to have been finally
closed. But it fell out that not very long afterwards, to the astonish-
ment of all concerned, the '* Ouzel Galley " cast anchor in the port of
Dublin. The captain had a strange tale to tell. Proceeding in her
eastern course down the Mediterranean, the '' Ouzel " had fallen a
victim to the Algerine corsairs, who in those days, and, indeed, for
long after, were still the scourge of the mercantile marine, and being
a large and well-found ship, she had been appropriated by her captors
to their own uses. But by some fortunate chance the crew of the
*• Ouzel '' were enabled to turn the tables on their conquerors, to
repossess themselyes of their ship and its oarn^o, and to return in
safety to the port from whence they had sailed.

So far all was for the best. But the return of the ** Ouzel,"
unfortunately, proved the occasion of a knotty legal difficulty involy^
ing troublesome litigation, which in one form or another lasted for
seyeral years. The '* Ouzel " brought home in her hold, not alone the
peaceful merchandize which it wus her mission to carry, but the
piratical spoils of her sometime Algerine masters. This loot wus of a
value far exceeding that of the legitimate cargo, and immensely in
excess of the amount for which the ship had been insured, and for
which the owners had been compensated. A question at once arose
as to the ownership of the plunder. Was the booty the property of the
original owners under whose auspices it had been gained? Or did it pass
to the underwriters in virtue of their completion of the contract of
indemnity ? The point was a nice one, which apparently had not then
been settled, and the gentlemen of the Law Coui'ts exerted their
ingenuity in the endeavour to determine the destination of so rich a
prize, No records of this litigation are now traceable; but it is
reputed to have engaged the Courts for years without any result being
reached ; and the case was ultimately referred to the arbitration of a
oommittee of merchants, through whom a compromise was effected,
and the litigation terminated.

To celebrate this triumph of the elastic principle of arbitration
over the unaccommodating and dilatory procedure of the Courts, the
merchants of Dublin resolved to found a society which should have for
ita object the settlement of all commercial disputes without having
recourse to the winding mazes of the law ; and they gave to their

144 Proceedings of the Royal Imh Academy,

ABSociation the name of the vessel which had been the means of
bringing it into being. Accordingly, about the year 1705, the Oaid
Galley Society was founded.

The books of the Proceedings of the Society for the first hslf-
century of its existence have long been irrecoTerably lost, and only
the more recent minute-books are now extant. But its rules and
regulations, with a list of members, were printed in 1859, as col-
lected from the books of Proceedings which were then aTailaUie.
These rules and regulations include the Report of a Committee ol tiis
Society appointed in 1799 " to inquire into and prepare a dedantion
of the rules, orders, and customs of the Galley." We are thus
enabled to understand the precise objects of the Society and the mode
in which it was organized. From this it appears that it was the dutj
of all members of the Galley to sit as arbitrators [in the settlement of
such disputes as might be referred to them, '* provided all the aiUtn-
tors chosen are members of the Galley." Parties were prohibited froo
making any personal applications to members respecting any matter in
dispute, and all proceedings were regulated under the guidance of ai
officer known as the Registrar, to whom a sum of money, arranged
according to a fixed scale, was payable by the parties seeldng arbitxa*
tion, "to insure the payment of the (Galley Fees," which irere
appropriated, after payment of the costs of the award, to a charitable
fund. Within the limits of the Society parties were entitled to the
choice of their arbitrators, but with the arbitrators when chosen lay
the appointment of an umpire.

Such were the purposes for which the Society was fonnallj
constituted ; but it had, or grew to have, other functions at once
beneyolent and convivial, which appear in time to have engroBsed a
large share of the attention of its members. From the year 1770 tlie
subscription appears to have been a guinea ; but on November 11,
1801, "it appearing by the bursar's accounts that the subscription of
one guinea per annum is insufficient to pay the annual dinners," it
Was raised to a guinea and a-half . Two years later, no doubt for the
same reason, it was raised to £2 5a. 6<^. ; and the frequent occntroKe
of the word '' dinner " in its rules may, perhaps, be held to aoooant
for the mourning accents with which surviving members still speak of
this ancient Society. Most of the business of the Society was trans-
acted at or after dinner, except at the November meeting, which vas
held immediately before dinner. Certain it is, at all events, that
while continuing to perform its more serious functions, the Onsd
Galley Society became highly popular among the merchants of PaUii

Falkinkr — lUuUraiions of Commercial Sistoi-y of Dublin. 145

as a oonTiyial association. Its roll being limited to forty members,
admission to it was bighly prized. The list of its members for a
period of a hundred and forty years, contains, it is no exaggeration
to say, representatiyes of all that is most honourable in mercantile
Dublin, and attests the high character the Society continuously
enjoyed. The names of La Touche, Guinness, Hone, Pim, Jameson,
Hatley, Colyill, and others equally familiar constantly recur.

The esteem in which the Society was borne, and the hold it had on
the affections of its members, was strengthened by the quaint and
characteristic customs which its constitution ordained and its rules
enforced. It was organized, in deference to its marine origin, on a
nautical basis. The affairs of the Ouzel Galley were administered by
a Council, of which the officers were : — ''The captain, two lieutenants,
Blaster, bursar, boatswain, gunner, carpenter, master's mate, coxswain,
boatswain's mate, and carpenter's mate " ; and a peremptory regulation
enacted that at the meetings of the Galley, of which three were held
annually, ** the captain, or in his absence the senior officer on board,
has supreme command, and any disobedience to him is mutiny." The
introduction of officers and new members was conducted "according
to the ancient and immemorial usage of the Galley," part of the
ceremony being, it is understood, the draining, at a single draaght, of
a bumper of claret from the glass cup, a beautiful example of Irish
glass-work, a photograph of which is reproduced with this Paper*
Guests could only be introduced on the inyitation of the '* captain,
officers, and crew of the Ouzel Galley." At each meeting members
were bound, on pain of a fine, to wear a gold medal^ pendant from
an orange ribbon. Finally, the members were "piped to dinner"
T?ith a boatswain's whistie ; and the minutes for 1754 record that a
ailyer whisUe, probably that of which a representation appears below,

I ordered to be provided by the carpenter for the boatswain's use.*

> The records of the Society for Feb., 13, 1772, contain the following :—

" Ozdeied, that the medal be made of gold. That on one side of the medal the
* Ousel Galley ' be lepresented, and the motto * Steady.' That on the reverse be
i«preeented the figure of * Equity,' with the motto ' cuique suum.' "

These medals appear to have been struck at different periods. That acquired
by the Academy £■ believed to be from the design of Parks, a Dublin architect.

» The captain's oath, in 1764, was as follows:— «• I, ul, J, do swear that I
will be faithful to our Sovereign Lord King George the Second ; and this galley,
entrusted to my command, I wUl, to the best of my power, defend against all
piimtes either by sea and land ; the rules and orders established on board I will see
observed to the utmost of my power, and justice admiiistered to the crew, and aU
who put any freight on board. I will continue to be a good fellow, and, as long
as I can, hearty and merry."

146 Proceedings of the Rotjal Irish Acadetn^f.

That at these convivial meetings the charitable objects associated
with them were by no means ignored ap{)eaiB from the regalation that
the bnrsar should keep two accounts ; one for the Subscription Fofid^
and the other for the Charitabie Fund ; and that after such dinner if
was customary to vote away in charity the earnings of the Galley.
And it is certain that the Society enjoyed throughout its existoioe a
high reputation for practical benevolence.

The meetings of the Ouzel Galley Society were held throughout
the nineteenth century at the Commercial Buildings, and many ^
recall these gatherings which each November were held in the open
square behind the Chamber of Commerce. In the latter part of the
eighteenth century, and for many years subsequently, the dinners
appear to have been held at Atwell's Tavern in Dame-street^

From the foregoing account, it is easy to understand that a society
of this kind must, in time, have outgrown the drcumstances in which
it originated. Though as a benevolent association it continued to
serve a useful purpose, its functions as an institution for promoting
arbitration gradually fell into desuetude, as legal procedure adapted
itself more closely to the needs of the mercantile community. From
a printed account of awards made in each year from 1799 to 1869, it
appears that 364 awards, many of them dealing with matters of gTC«t
magnitude, were made within that period. But of these nearly tvo-
thirds were made in the first quarter of the nineteenth century. I&
1888, accordingly, the Ouzel Galley was voluntarily wound-up and
dissolved by an order of the Court of Chancery, which provided for
the distribution of its funds, to the amount of £3300, among chaiitaUe
institutions connected with the city in which the Society had so loi^
flourished.

Many citizens of Dublin must be familiar with the large painting
of a full-rigged ship which hangs over the door of the news-room in
the Chamber of Commerce, with the legend, " Hie Ouzel Galley,"

^ The meeting-places of the Society, a« recorded in their TniuMtioBB, tfafotr
interesting light on the taverns or eating-houses of BahHn and its enrirans, ia tb«
second half of the eighteenth century. In 1748 the Galley met in Che FImvux
Tavern, Werhurgh«street ; in 1761, at the Ship Tavern, Ch^Mlixod ; in 1775, si
the Rose and Bottle, Bame-strset; in 1770, at the Eagle Inn, Sualaoe-itiv^t >
in 1776, at Power's, Booterstown ; in 1796, at Harrington's, (haflosi-strset; sd^
in 1800, at Atwell's Commercial Tavern, Dame-street. In tiie early pait of tl»
nineteenth century the favourite resorts were Leech's Boyal HoCd, Kildare-ilivM ;
Horrrison's, in Nassau-street ; the Hilton, in Saokville-street ; and Jade's Held,
Commercial Buildings.

Palkinbr— IllustraiioM of Commercial History of Dublin. 147

beneath it. It seems right that in this notice of the Society the pedigree
of this painting should be preserved so far as it can be collected from
the records of the Society.

The painting appears to have been presented to the Society as far
back as 1752 by Alderman John Macorrell, the then captain of the
Galley. Wliether or not it was a merely fancy picture, or an
authentic representation of the actual ship from which the Societ}
took its name, cannot be stated, for nothing further is known of the
date of the picture or of the artist In the minutes of the meeting
of the Galley held at Chapc>lizod m August, 1753, a receipt is in-
serted, in which one John Morris acknowledges the receipt of *' a
large painted piece representing the Ouzel Galley, which is put u!p in
the great room in my house," and admits the picture to be the pro-
perty of the Galley. Morris was probably the owner of the inn or
tavern in which the Society was then in the habit of meeting.

Nineteen years later, 16th July, 1772, the minutes record tlie
appointment of a committee *' to inquire after and recover the picture
of the Galley presented to the Society by Alderman Macarrell," but
the result of the inquiry is not given in any subsequent minute. It
may be presumed, however, that the picture was recovered, and is
identical with that which still hangs in the Chamber of Commerce,
and is thus referred to in the entry for drd June, 1870 : — '* That the
offer of the Chamber of Commerce to place the old painting of the
Galley in a more conspicuous place be accepted."

[No aooount of the Ouzel Galley Society has ever appeared in
print, save a brief notice in Whitelaw & Walsh's " History of
DnUin," vol. li., p. 914. The account given above of the origin of
the Society, and its history prior to 1753, is not sustained by any
documentary anthority, but is derived from oral tradition preserved
among its members. The writer has to express his .cordial obliga-
tions to Mr. R. F. S. Colvill, of Coolock — whose father, the late
James Chaigneau Colvill, was the senior officer of the Society at
the date of its dissolution — for much information and assistance.
To Mr. Colvill, also, as the custodian of the glass cup and silver
whittle, and the possessor of one of the medals shown in the illustra-
UonBj the Academy is indebted for permission to photograph these
interesting relics.^-C, L. F.]

148 Pr&eeedingB of the Royal Irish Academy,

APPENDIX,

BxDre XHB Fntsr Ehtet iv the MnrrxB-BooK of ihx
Ballisi Opugs Committsb.

^* The Committee appointed to consider the proper methodi for
aetUing the Ballast Office, &c., aie come to the following lesolalionp,
which they hnmhlj offer to yonr Loidahips and the Ajnemblj as
f ollowB : —

Impiimis. — ^That it is necessary there should be a standing com-
mittee of 3 or 5, who shall be called Qoyemors and IHrectcffB,
and haye the management of the business, and report their
proceedings to the Lord Mayor and Assembly qnaiterly at their
meetings, and oftener if needful — ^the Committee to be altered
every Assembly if thought fit.

2ndly. — That there be a proper officer appointed called the Master
and Treasurer of the Ballast Office, who shall duly attend the
said office in person, and observe such directions as he shall
receive from the Committee of Directors ; and that he have a
good Clerk for receiving and paying, etc., for whom he will be
answerable, but if recommended by the Assembly then aecniity
to be given by him to the Directors,

3rdly. — That there be a good and sober d^k, called Accountant
and Registrar, to officiate ; also a Secretary^ to the Diredon,
and to attend the General Assembly with .the Eegistry, and
other books of the proceedings of the office when requu^od.

4thly. — That there may be an officer knowing in Shipping, bj
the name of Chief Gauger and Supervisor, to gauge the Slips,
and inspect into the working of the Lighters and Gabbardi,
and make return to the office of what baUast is put Qnboazd,etc.

5thly. — That there may be a sober and careful person appointed
to be messenger and office keeper, who may be frequently
employed to assist in other matters in the daytime.

Falkinbr — Itlmiratiom of Commercial Eiatory of Dublin. 149

ethly. — That there ought to be an office immediately appointed
in a proper place, as near to the Custom House as can con-
Teniently, and where boats and Ringsend cars may come
without distiirbance to the street.

That at Temple-bar there is a proper place if none more conve-
nient be found ; it has large rooms and warehouses, at £15 unn.

7thly. — That there ought to be a convenient boat with 2 boat-
men, to attend the Ganger and Supervisor, and other services.

8thly. — That there being no Gubbards in this port of the kinds
of the Lighters used at London, for the raising the ballast
with expedition and ease — it is necessary that 2 Lighters be
immediately built, one of twenty the other of thirty tun, with-
out decks or bends. When these Lighters are set to work by
the men belonging to the office, they will show what quantity
of ballast can be raised in a certain time, and what the cost
will be to put each tun on board ; and this will be the guide
either to build more of these Lighters, or come to agreements
with mastesB of Gabbards. The Governors and Directors, after
they have met 2 or 3 weeks, will be able (it is belieyed) to
inform your Lordship and the Assembly what are the more
proper steps to be taken in relation to the raising the ballast.
This Committee are humbly of opinion that no salaries can
well be settled till 3 months after the 1st of May next.

11th Dec, 1707.

'' Robert Cheatham.
Wm. Quaile.
Tho. Kirkwood.
Tho. Thome.
John Nevill.
Ed. Surdevill.
Thos. Wilkinson.

John Pearson
Matthew Pearson.
Humphry Jervis.
John Rogerson.
"Wm. Fownes.
John Eccles.
John Godley.
Nath. Whitwell.*'

K.l.A. PKOC., VOL. XXIV., SEC. c]

[11]

150 Proceedings of the Royal Tinsh Amdeniy,

11.

Bbsolutiohs of Dubuv Mkbchaivts, and of thx CoxxnTEi or
Mbkchants, kelatite to the Establishmknt of a Chambek
of comhebce.

BoTAL Exchange, Dubldt,

nh Febrtury, 1783.

Present — Travers Hartley, Esq., in the Chair. Messrs. William
Colville, James Horan, John Binns, Denis Thomas O'Brien, David
Dick, Alexander Armstrong, George Lang, Hemy Lyons, John
Cowan, Samuel Dick, Bobert Magee, Arthur Bryan, Paul Patrick,
James Anderson, George Luneli, Edward Forhes, Edward Patrick,
William Bruce Dunn, Daniel Marston, Joshua Pirn, Frederick Geale,
George Sutton, Leland Crosthwaite, Thomas Mitchell, Bobert Black.

A paper having been introduced, containing '' Propositiona for the
EstHblishment of a Chamber of Commerce in the City " —

Besolvkd — That the said Paper be referred to the Committee of
Merchants, and their opinion requested thereon.

The meeting adjourned to Tuesday evening next at seven o'clock,
when the answer of the Committee of Merchants will be received.

BOYAL EXCHAKGK,

February 10, 1783.

Present —

Mr. Colvill.
Mr. Hartley. Mr. O'Brien.

Alderman Sutton. Mr. Cosgrave.
Mr. Carothers.

At a meeting of the Committee of Merchants regularly convened by
summons for the special purpose of taking into consideration a plan oi
instituting a Chamber of Commerce in this city, Mr. John Patrick aiui
Mr. Joshua Pirn presented to the Committee the plan hereunto annexed,
which being received, read, and considered, the following reaolntioos
were entered into : —

^^ That we highly approve of the said plan as forming a broad and
firm foundation on which may be expected to arise a superstractore
of eminent usefulness in the commercial department.

That from this measure the trading interest is likely to derive
great additional importance and respect, and the public in general the
advantages consequent thereto.

Falkinrb — Illustrations of Commercial History of Dublin. 151

That on the great change expected shortly to take place in the
commercial system of Oreat Britain and Ireland, and prohably in that
of some other countries, it is highly necessary and peculiarly seasonable
by a scheme of this nature to collect the experience and abilities of
every intelligent trader in the various lines of commerce and manufac-
tures that their united knowledge may be happily directed to the
general good.

That this Committee do therefore most heartily recommend to
their fellow-citizens the carrying said plan into effect as speedily as
possible, and they will think themselves happy in resigning their ap-
pointment as the Committee of Merchants when on the liberal and
extensive plan now proposed a Council of the Chamber of Commerce
shall be elected.

Plan for Institutiko a Chamber of Commerce nr this Citt.

''The present important situation of this country, its lately renewed
constitution, its fond hopes of rising commerce, and consequently in-
creasing opulence, the variety of commercial i-egulations necessarily
incident to this change of circumstances, and j)articularly requisite
from the late revolution in the political system ; every consideration
appearing to demand a general union among traders and a constant
unwearied attention to their common interests ; from a view whereof,
to promote these laudable objects in this particular district, and to
hold forth an example for imitation and co-operation to the rest of the
kingdom, it is proposed to institute forthwith a Chamber of Commerce
for the city of Dublin.

That any merchant or trader resident within the said city or its
dependencies shall be eligible as a member of this Chamber on his
paying one guinea to Mr. John Patrick or Mr. Joshua Pim, who have
kindly undertaken to act as Treasurers until a person shall be elected
to that office ; such subscribers to continue members as long as they
shall respectively comply with the rules which shall be adopted by the
said Chamber for its good government ; and for the continuation of a
fund to answer the purposes of its institution.

That when the subscribers shall amount to one hundred the said
temporary treasurers shall call a meeting by public advertisement,
at which said first meeting of the Chamber, or at an adjournment of
said meeting, the members present shall choose by ballot a certain
number of persons who shall be called the Council of the Chamber of
Commerce^ to continue in office until the 1st of May, 1784 ;. and that

ll»

162 Pt-oceediiigs of the Eoyal Irish Academy.

on annual ballot for such Council shall be held on OTeiy fint day of
May, not being Sunday, and when Sunday, on the 2nd day of Kay.

That it shall be the business of said Council to attend to the in-
terests of commerce, and for that purpose to hold frequent meetings, to
oonfer when necessary with persons in high stations or others, to haTe
a watchful attention to the proceedings of Parliament respecting trade
in both kingdoms ; to inspect into the methods of transacting bosineBs
in Dublin, and to continue and recommend improvements therein when
such shall be thought expedient.

That the said Council for the time being shall choose by ballot fram
among themselves a President, two Vice-Presidents, and a Treasurer,
and shall appoint a Secretary with a fixed salary suitable to his services.

That it be understood that the members of the Chamber of Con-
merce shall be peculiarly entitled to the protection of the institation
on every proper occasion."

[ 153 J

THE ITINERARY OF PATRICK IN CONNAUGHT,
ACCORDING TO tIrECHAN.

Br J. B. BURY, M.A., LL.D., Litt.D.,
Regius Professor of Modem History in the TJniversitj of Cambridge.

[Read Januaut 26, 1903.]

If we attempt to trace on a map the itinerary which Tfrechdn marks
out for St. Patrick through the kingdom of Connaught, we are met by
seyeial difficulties, but by none perhaps more awkward at first sight
than that which arises at the very outset in regard to the point where
Patrick crossed the Shannon. In the present paper I propose to
ahow that this difficulty is only apparent, being due to an erroneous
identification which has been accepted without question, and to deter-
mine as nearly as possible the alleged route of Patrick from Granard
to Rath Crochan. Further, I shall hare occasion to point out a funda-
mental confusion which pervades Tirechdn's memoir.

To avoid misconception, it may be well to state explicitiy that I
am concerned here merely with the interpretation of that document ;
not directiy with what Patrick did, but with what Tirech&n says
he did.

§ 1. At the end of Book i., our text of X^rechdn thus marks the
progress of Patrick from the caeumen Oraneret to the Shannon : —

Tenit in campiun Rein (31 15 RoUb ed.) ;'

yenitque P. ad alueum Sinone ad locum in quo mortuus fuit auriga illius
Boidmalus et sepultus ibi in quo dicitur Call Boidmail usque in hunc diem (311 9}.

That is : Patrick proceeded through Mag Rein, and reached a place
<m the Shannon, which, in the writer's time, was Cail Boidmail. Mag
Bein included the southern part of County Leitrim ; and the name is

> While I supply the reference to the Rolls edition, I gire the text of the
passages which I quote from the proof-sheets of Dr. Gwynn's edition of the Codex
Annaeiianus which is shortly to appear.

154 Proceedings of the Royal L-uh Academy.

still preserred in Lake Binn and the river Rinn. We have no means
of identifying the exact rite of Gail Boidmail ; but its whereaboats
seems to be indicated cleu*l7 enongh. It was ad alueum Simtm^
The qnestion arises : why did Tfrech&n, whose langonge is always
plain and unadorned^ use this phrase, which recnrs 312,, and 313^,
instead of saying simply ad Sinanam ? There can, I think, be only
one answer. Requiring a Latin word to express the swellings or
lakes of the Shannon, Tirechim adopted aluetu as the best equiTal»it
he could find.^ Otherwise alueus in these passages is perfectly
unmeaning, i^ow the Shannon-swelling which Patrick would reach,
adyandng westward from Granard through the plain of Binn, is that
which is known as lakes Bofin and Boderg. The inference is that
Gail Boidmail was somewhere on the eastern bank of these lakes.

§ 2. Digremon on Mag SUeht, — An interesting question presents
itself here, bearing on the criticism of Tfrech£n's text. In later
biographies, which depend largely on Muirchu and Tirechto, we find
a notice that Patrick virited Mag Slecht, where Grom Gmach was
worshipped, and cast down the idol. iN^ow, this incident is not recorded
in the documents contained in the Godex Annachanus ; and therefore it
might seem reasonable to infer that it was a story of later origin^than the
events, whether legendary or historical, recorded by Tfredi£n and
Muirchu. On general grounds I do not feel that such an inference
would be quite safe ; but there are certain particular oonsiderationBin
this case which must make us heritate. The later biographies, to
which I referred, are those which it is usual to designate, following
Golgan's nomenclature, as the Yita Tertia, Yita Quarta, and Tita
Tripartita. Now, in the Yita Tertia, the story of Mag Sledit (c 46)
is inserted immediately after the incidents connected with Goirpre and
Gonall, SODS of Niall (c. 43 and 44), and immediately before the tale of
thedarkness which the magicians drawdown upon Mag'Ai, when Patrick
entered Gonnaught. In the Yita Quarta, the virit to Mag Slecht
(c. 53) occurs in exactly the same porition (between Goirpre and
Gonall (c. 51, 52), and the darkness on Mag 'Ai (c. 54)). As these two
Lives are quite independent of each other, this is hig^y significant,
for it shows that both depended here on a common source in which
these incidents were related in thU order. Now the story of the two
sons of Niall, and the legend of the magic darkness, are derived from
Tfrechdn ; so that the condusLon which naturally presents itself is

* The SMOciation of alueus,* river-bed,' witho/wM, * paunch,* explains the om of
the former word by Tirech&n.

Burt — The Itinerary of Patrick in Connaught. 165

that the common source of Y4 and Y, here was an intermediate docu-
ment in which the compiler inserted at this point the story of Mag
Slecht.

Turning to the Vita Tripartita, in which large portions of
T{rech£n's memoir haTe heen reproduced, we find that the visit to Mag
Slecht immediately precedes the crossing of the Shannon (pp. 90-92,
ed. Rolls), and follows the visit to Granard. This confirms our con-
clusion. The coincidence in these three documents points to an older
document, in which the episode of Mag Slecht immediately preceded
the crossing of the Shannon.

Now, it is easy to see why a compiler who was following the
memoir of Tfrech&n might have heen tempted to introduce from
another source the Mag Slecht incident just at this point. The
following words in Tfrechdn's text ohviously might supply the
fnotive :

mittens autem Patricias methbraini ad fossam Slecht barbarum Patricii propin-
quiun qni dicebat mirabilia in dec iiera (31 In).

The mention of Rath Slecht here might have readily induced a
compiler, who was at a loss where to insert the story of Crom Cruach,
to choose this place as appropriate. If so, the author of the Tripartite
Life, Part ii. (or his source), has gone further ; and in the process of
inserting the story, has altered a point in Tirechdn's narrative.
Having recounted the overthrow of the idol, the Tripartite proceeds to
reproduce as follows the passage which I have just quoted from
Tirechin:—

Forothaigsiiun [dano] eolais isininutsin .1. Domnach Maige S16cht, ecus
forftocaib and Habran Barbaras Patricii, oognatasque ei et propheta.

Apart from the notice of the foundation of Domnach Maige
Slecht, which is not mentioned by Tlrechdn, there is an important
discrepancy between the two passages. In Tfrechan's memoir,
Patrick, from some place in Mag Rein, tmds his relative to Rath
Slecht ; in the Tripartite he leaves his relative in Rath Slecht. This
difference could of course be accounted for, as due to an alteration
entailed by the insertion of the Mag Slecht story.

^ Hov is tbis to be reconciled with tnabran in tbe Tripartite ? Must we not
suppose that m is in both cases an error for ni, the name being Niabrain ? Of.
L. B. fo. 15 a, b. For nieih, op. Ann. Ult. ^.d. 693, and Rhys, Welsh People,
p. 51. A similar nustake occurs in the Biburg MS. of the Vita Tertia (Colgan,
p. 26, 0. Ixuii) : Mothfer for Niothfer. Colgan's note to tbe passage shows that
this Is not a misprint.

156 Proceedings of the Royal IrUh Academy.

But I confess that I do not feel quite satisfied withthiBexplaaatMii
•f the notable coincidence between Vita 3, Vita 4, and the Tripaztite.
It is the only possible explanation if we assume that the text ol
Tlrech&n is right, as it stands in the Liber Armachanus. But theie
are grounds for questioning this assumption. In the first place, there
is some corruption, whether large or small, in the passage quoted
aboTe {mitiena autem . . , in Deo uera) ; for there is no finite TeA for
the subject Patridm. The scribe of the Armagh Ms. noted the
difficulty of the passage by his symbol s in the margin. In the
second place, it seems strange that Tirechan should not have
mentioned explicitly that the purpose of the mission of Patrick's
follower was to take charge of a church which had been founded at
Rath Slecht. This is evidently implied; but we expect it to be
stated. Combining these two considerations, we cannot avoid the
conclusion that there is a lacuna here. For the sense, it is neceflsaiy
only to asume a short lacuna ; the sentence might have been completed
by a few words referring to Domnach Maige Slecht But the
suspicion forces itself upon us that the lacuna may have been of larger
compass, and that the original text of T(rechdn may have coatained
a notice of the visit to the Field of Adorations.

§3. Having brought Patrick to the bank of an aluem Sinane in
Book i., Tfrechdn thus resumes his journey in Book ii. : —

TTenit ergo Patricias sanotus per alueum fluminlB Sinnae per uadum
ftmum in campum 'Ai (3123i) ;

et uenienint per alueum fluminis Sinnae quae dicitur Bandea ad toowliui
Oradi (313|) ;

uenierunt ad campum Glais et in illo poeoit celolam magnam quae a
'cellula magna (ib. lo) ;

deinde uenit ad Auicum et Bitteum et ad magos qui fiienmt de
Oorcuchoiiluain Bono et Ith fratres. Alter auscepit Patricium et aanctos eins cob
gaudio et immolauit sibi domum suam. Et eziit ad ImbHuch Horaon . . . Bt
pOBuit ibi Assicum &c. (ib. u).

Fathcius uero uenit de f onte Alofind ad Dumecham nepotum Ailello et f usdndt is
illo loco aeclessiam quae sic uocatur Senella Cella Dumiche usque huno diena (314i»)w

Patrick is thus said to have proceeded to Mag 'Ai by crossing the
Shannon at Vadum Duarum Auium^ the Latin equivalent of the Irish
Snimh-da-6n. The plain known as Mag 'Ai comprises a large part of
County Roscommon, stretching from the town of Boacommoa
northward beyond Elphin. It is in the north part of this plain that
we find Patrick when he has crossed the river ; first of all, be does
certain things in Mag Glais, a district whose name still surviTea (as

Burt — The Itinerary of Patrick in Cannaught. 157

we shall see) dose to Boderg; and then he goes on to Elphin,
evidently croaeing the Baune (Badgna) hills, which divide Mag Glais
from Mag'Ai. Accordingly, if we had no other knowledge, we
should, without much diffidence, conclude that the Yadum Duarum
Auinm was in the neighbourhood of Boderg and Bofin.

But when we consult modem authorities on Irish topography, we
find the Snimh-da-en placed, without any hesitation, far from the
scene of the events described by Tirechdn. It is shown by
O'Donovan tliat it was an old name for a part of the Shannon close to
Clonmacnois. It is mentioned as a landmark in a description of the
boundaries of the Hy Many, in a context which shows that it was
south of Athlonc ^ ; and the situation near Clonmacnois is implied in
the story of the Aided Diarmadaj published in Mr. O'Grady's Silva
Gadelica,' and in a passage in the AgaUamh na Sendrach,^

It may be said without the least reserve that this situation is quite
irreconcilable with the narrative of Tfrechdn. It would mean that
this writer supposed Patrick to have travelled southward from Mag
Bein (in Leitrim) to a point south of L. Ree in order to cross the river,
and returned northward again along the western bank just for the
purpose of reaching Mag Glais. It is as if one who wanted to reach
Battersea from Hampstead were to go round by way of Henley. Mag
Bein lay on one side of the Shannon ; Mag Glais opposite to it on the
other. If Tfrech&n had intended to bring Patrick round by this
circuit of seventy miles, it is inconceivable that he should not have
said something to explain it or indicated more precisely the route ;
nor is there any imaginable cause why such a route should have been
chosen, if it were not for the purpose of preaching and founding
churches in the districts through which it lay. Not a hint is given
of any such activity in the territory through which Patrick would
have passed, and the Yadum Duarum Auium is introduced as if it were
the direct and natural passage from Mag Rein to the northern part of
Hag 'Ai.

In the case of another chronicler, we might suspect that, through
ignonuice of topography, he had mixed up his information and failed to
perceive the incongruity of his story. But, as I have pointed out at

1 0'Dooovan, Hy Many, p. 5 ; compare the map.

'Text, pp. 72-3; translation, p. 76: <*two birds that Nar son of Gonall
<3«maeh'8 son Finncha killed there on Eistine the Amajson's shoulder, whenoe it
is named Sn&mh-dfi-^.*

*Jh. Text, p. 184 ; trnnsl. p. 147.

158 Proceedings of the Royal Irish Academy.

length elsewhere,^ Tfrechdn was personally acquainted with the
geography of Connaught and Meath ; and we must feel the utmost
hesitation in imputing to him the apparent ahsurdity.

These considerations seem to me so weighty as to he fatal to the
notion that Tfrechdn supposed Patrick to have crossed the river in the
neighhourhood of Clonmacnois. But the argument hecomes simply
irresistible when we turn to the details which the memoir supplies as
to the crossing. The crossing was at a river-swelling {alueusJiumiHii) ;
and this condition is not fulfilled by O'Donovan's Snamh-dd-6n. In
order to meet the difficulty, it might be proposed to take the words
psr alueum fluminU per uadum duarum auium in an unnatural way, so
as to mean that Patrick, having travelled along the left shore of L. Bee
{alu,flum,)y proceeded down the river to the Sn£mh-dd-6n, and there
crossed. But if we could entertain such a forced explanation, it
would be only to encounter a new difficulty on the other bank.
Having crossed over by the uadum, Patrick and his companions came
to another river-swelling : —

£t uenierunt per aliieum fluminis Sinnae quae dicitur Bandea ad tamulum
Gradi (SlSs).

Thus Patrick, having already crossed the Shannon by the
has again to cross the alueus ' Bandea/ in order to reach Duma Graid.'
Unfortunately Duma Graid no longer bears that name ; and we cannot
make use of it to determine the situations of other places. But it
was clearly in Connaught, on the western side of the Shannon, on the
same side as Mag Glais ; for Patrick proceeds from it into Mag Olais
without again crossing the river.

If any doubts be still felt as to the justice of my negative criticiBm
on the view that Patrick (according to Tirechan) crossed by O'Dono-
van's Sndmh-dd4n, they must yield to the positive fact that there is
another place on the Sliannon which satisfies fully the conditions iA
the problem. The essential condition is that having crossed by a
river-swelling, Patrick should then come to another river-swelling

^ English HiBtorical Heriew, April, 1902.

' The only way out of this conclusion would be to assume that here per abmmn
does not mean ' across ' but *• along the banks of *—per in these passages being and
in different senses with alueum and with uadum. In that case the alueut Bmtdet
might be sought anywhere (except in L. Ree, which, on this theory, would be a
different alueus) f since, ex hypothesis circuitous routes not traced by the writer are
admissible. But such possible attempts at exegesis will not satisfy a reasoiiable
critic.

Bury— TAe Itinerary of Patrick in Connaught. 159

which he should also cross without returning to the left bank. This
seems imperative ; but it will also be admitted as a desirable, if not
indispensable, condition that the required river-passage should lie in
the direct, or more or less direct, route from Mag Bein to Mag Olais.
Kow, these conditions are exactly fulfilled by the riTer-swellings
which are known as Bofin and Kilglass. We saw above, following
T£rech&n'B route, that the natural place to locate Cail Boidmail was on
the eastern bank of Bofin or Boderg. We may say, more generally,
that any one passing from Mag Bein to Mag Glais, in the north part
of the barony of North Ballintobcr, would be sure to cross the river
somewhere between Boosky and Drum sua.

But if the second condition is satisfied, the first condition is also
strikingly fulfilled. If he crossed L. Bofin, Patrick would have found
himself on the river-girt promontory (which forms part of the modem
parish of Kilglass), with L. Bofin on one side, and L. Kilglass on the
other. In order to reach Mag Olais, whither he was proceeding, he
would have to cross L. Kilglass, unless he took a long d6tour round
the south extremity of this river-lake. It is manifest that this topo-
graphy conforms precisely to the requirements of the narration of
Tirech^. Having crossed a first alueus, L. Bofin, the saint then goes
on to cross a second, L. Kilglass, by which means he is able to proceed
into Mag Glais. It follows that Bandea was the old name of the
branch sweUing which is now known as the lake of Kilglass.^

Having crossed Bandea, Patrick went to Duma Oraid. Topographers
expect to find this appellation in the form Doogary, a place-name

1 The Tripartite Life, Part ii., which depends here on Tirech&n, gives an
additional piece of topographical information, derived from an unknown source.
The pauage is translated by Dr. Stokes as follows (p. 93) : ** There Fatrick
found iheferta* (bar ?, bank P), namely, the earth was raised up under Patrick in
the ford ; and the learned still find that ridge. And he went into the harbour at
onoe, and there died Buad-moel, Patrick's charioteer, and was buried in that place.
Cell Buadni^il is ito name, and it belongs to Patrick.*' [Cell B6adm&il should be
corrected, after the text of Tirech&n, to Cail B6adm&il.] While this abbreriates
Tirecb&n's account, it adds the token that at the passage there was a ridge of raised
earth in the ri?er-bed. This notice is far more likely to have a basis of fact than
to be a pure inrention. There is no reason to suppose that the name Cail B6adm&il
had disappeared between Tireoh&n's date and the composition of the Vita Tripartita ;
and we need not hare many scruples in accepting the statement that near Cail
BCiadm&il there was a bank in the river which, according to the people of the
neighbouriiood, rose out of the bed as the saint was crossing. It Mould be interest-
ing to know whether there are traces of this bank in L. Bofin. There is at all
erents an island.

160 ProceedingH of the Royal Truth Academy.

which occurs in different parts of Ireland.^ But in the neighbourhood
where we seek Tirechan's Duma Oraid, near Mag Glais, there is none
to be found.' Thence Patrick and his companions went to Mag Glais.
The name of the plain of Olas has survived unchanged since Tfrechin's
day, though with a far narrower signification. Moyglass is now a
small townland adjoining L. Tap in the parish of Kilmore.' The
ancient Moyglass included the modem parish of Eilmore, of which it
is now only a small portion. This follows from the fact that the
Patrician church which gave its name to the parish of Kilmore was in
Moyglass, combined with the geographical consideration which suggests
the probability that there was a name to designate the whole district
between the Baune hills and the Shannon. It is possible that the
territory thus named extended considerably beyond the parish of
Kilmore, south-westward, into the barony of Boscommon. This may
be inferred from the existence (a mile or so west from the south
extremity of L. Eilglass) of another townland, Moyglass, which looks
as if it too preserved the denomination of the original Mag Glais ; and
likewise from the name of the * Church of Glas,' from which L. Bandea
came to be called L. Eilglass.

In this district,Patrick founded a large cellula called CeUula Magna^
that is, in Irish, Cell Mor. This foundation has been preserved, and
the original cell was, we may assume, not very far from the modem
church, about two miles north of the bridge which spans the mouth of
L. Eilglass.

The circumstance that Cellula Magna in campo Glak is situated
close to that part of the Shannon which, in other respects, conforms to
the conditions which are implied in Tfrechdn's narrative, strongly
corroborates my conclusion that this writer makes Patrick cross the
Shannon at L. Bofin. We must now return to the original difficulty.
While, as has been shown, the details of Tirechdn's story make it clear
that the crossing was at L. Bofin, Tfrechdn designates the place dt

^ There is a L. Doogary in Leitrim ; another in Armagh ; there are Doogarys
in Cavan, Fermanagh, Tyrone, Monaghan, Down, Kerry. The nearest places to
Hag Glais are Doogarymore in the barony of Ballintober South, near L. Bee
(Ordnance Map of Roscommon, Sheet 40), and Doogary in barony of Boyle (»^.,
•heet 4) ; but neither is possible.

' Tet it seems possible that the name surviyes in a corrupted form in the islet
which is known as Dookery's island at the mouth of L. Kilglass. If so, we
might infer that Duma Graid was opposite this island in the townland of Eushport.

' Rightly identiHed by Mr. Hennessy in a note to his translation of the Vlte
Tripartita in M. F. Cusack's Trias Thaumaturga (p. 427, n, 8).

Burt — The Itineraft/ of Patrick in Vannaught. 161

crossing as Yadum Duorum [«tV] Auium, or Snamh-dd-6ii, whieh was
a wholly different place.

In tiie case of a more commonplace name, one might, with some
reason, leap to the conclusion that there were two places so called on the
Shannon — one at L. Bofin, and one near the later monastery of Clon-
macnois. But the ' Swimming-place of the Two Birds * hardly lends
itself to such a facile explanation, which we should have no difficulty
in accepting if the name were, for example, the ' Swimming-piace of
the Ox ' ; and it seems to me that we can hardly escape the conclusion
that T(rechin did not intend to associate Patrick's crossing-place with
the name Snamh-di-6n, and that an error has crept into his text. The
thought naturally occurs that the Yadum might have heen known hy
the name of the two cows, the red cow and the white cow, Boderg and
Bofin, which gave its names to the riyer-swelling. If Tfrechdn wrote
Vadum duarum vaeearum (to translate Sndmh-d£-b6), and if vaeearum
fell out accidentally (through homoeoteleuton), it is easy to conceive that
duarum might have heen corrected to duarum auium by a scribe to whom
the name of the Sndroh-di-en was familiar, but who had no aecurate
knowledge of the geography of the Shannon.^

§ 4. From Moyglass, the saint proceeds, in the pages of Tirechan, to
the territory of the * Corcu-chonliiain ' ; ' and one of the chiefs of this
tribe (one of two brothers, named Ith and Hono, described as magi)
welcomed Patrick, 'et immolauit sibi domum suam et exiit ad
Imbliuch Homon.' It seems probable that Homon is an error for
Hbnon (genitire of Brno) ; and this is the view suggested by the
Tripartite Life (p. 94), where ' Imlech Onand ' is the dwelling of
Ono, 'de quo TJi Onach.' If this correction is legitimate, one

* As there was a Druim-d&-6n near the Sn&mh-d&-£n on the river-reach
belowAthlone, so it is possible that, if there was a Sn&nih-d&-b6, there may have
been a ridge of corresponding name. The modem Drum-sna is north of the riyer-
swelling ; but it may at least he suggested that the ridge from whieh the place
flleriTes its name was called fnim the ancient Yadum— the ridge of the Snimh
{d£ hd). At all eyents Drum-sna must he short for a fuller name in which the
particular tndmh was designated.

* Hie name (suggesting tUreus etminum) is pnxzling ; but the Cotvu Oehland (so
Vit. Trip. 94) are meanL Their territory is descrihed in Vita Trip, (ib,) as < on
tlu« aide of the land of the Hy Ailella, and to the north of Sliay Baune.' See
O'Bonoyan, < Annals of Four Masters,* a.d. 1256, p. 458, note ; and * Topogra-
pbieal Poems of John O'Duhhagain, &c., notes, p. xl, on the Corca Sheachlann or
Oorea Achlann, one of the three tuathas which formed a deanery in the diocese
o£ Slphin.

162 Proceediuga of the Royal Irish Academy.

ambiguity .will be removed from the narrative, which, at this point, is
not quite clear. It is not made evident which of the two brothen
received Patrick ; but if Imhliueh Honon is the true readiag, it viU
follow that^it was ffono (as the Tripartite Life assumes).

Another ambiguity lies in exiiL The sentence reads as i{ Bm
were the subject ; but if this means, as it would naturally mean, that
Hono went thither alone without Patrick, a difficulty arises as to the
reference of ibi which occurs just after. The text is : —

et immolauit aibi domuin suam et exiit ad imbliuch^ homon et dixit ili
Patriciu8 Semen tuum erit benedictam et de tao aemine erant sacerdotet donuoi tt
principes digni in mea elimoyaina et tua heredxtate et posuit ibi aMiciimaibediiraii
filiam fratris aasici et ci plant matrem bethel episcopi.

The awkwardness is increased when, reading on, we find that the
place from which Patrick started when he had thus setupAssiciu
and Betheus was a place which he was not said to have reached—
foM Ahfind(Z\Ax^).

Now, it seems certain that ihi means Alofind, for there can be
little doubt that Assicus was stationed there.' The inference might
seem to be that one sentence at least has fallen out, in which the
coming of Patrick to Alofind was mentioned. But if we turn U) the
Tripartite Life, we find a solution which may enable us to dis][X'Qse
with the assumption of a lacuna. There we find Alofind identified
with Imbliuch Honon (94so). ^"^ interpretation implies that
Patrick, received by Hono somewhere in the territory of Gonn-
chonliiain, went with him to Imbliuch Honon — Uiat is, Elphin—and
founded a church there.

It must be allowed that the text of Tfrechan, just as it stands,
admits of this interpretation. A different punctuation from that
adopted in the Rolls text will make it clear : —

Alter suscepit Patricium et aanctos eius oum gaudio et immolauit sibi domus
auam. Et exUt ad Imbliuch Homon et dudt ill! Patridua : '* Semea tuumcnt"
hereditate.*' Et posuit ibi Assicum, &c.

The subject of exiit is Patricius, who proceeds to the Imbliach of
Hono ; and it is to be observed that, with this inteipretation, the pn>-
bability that Honon^ to which the following Hit would refer, is a troe
correction for Homony approaches certainty.

' Compare Tirech&n, SlSze.

Bury — The Itinerary of Patrick in Connaught. - 163

Tet it remaiDB strange that the close proximity of Alofind to
Imbliuch Honon is not more clearly brought out, and also that the
foundation of the Ecdesia there is not formally mentioned.

§ 5. In any case, Alofind gives a fixed point in the territory of the
Corcu-chonluain ; and for the purpose of marking the itinerary, it is
enough to determine that, from Eilmore, Patrick, having crossed
Sliav Baune, proceeded south-westward to Elphin. From there he
passed on to Dumeoha in the country of the Hy Ailello, and founded
the church called Senella Cella. The ' land of Ailill ' has survived in
the name of a portion of western Sligo ; but the barony called Tir-
enriU corresponds only to a part of the original territory, and the
present passage has topographical importance in proving the southern
extension of the territory of that tribe.^ For it is clearly right to
seek the Senella Gella to which Fatrick passed from Alofind, in the
district of Shaakill, which is dose to Elphin. Thus the territories of
the Hy Ailello and the Corcu-chonl6ain would have adjoined close to
Alofind.

But the church which Patrick founded here at Dumecha cannot
have borne the name of ' Old Church ' when it was newly founded by
him. Tlrech&n speaks as if it were so called from its very foundation ;
but it must have been in contrast to some newer establishment that
the eeU of Dumecha was distinguished as old. Wo are here in
presence of the same kind of problem that is puzzling Boman
archaeologists in regard to the name of that early church of which
the plan has recently been discovered in the Forum. But S. Maria
Antaca is more baffling than the Cella Senella of Dumecha. The
cine seems to lie in the close vicinity to Alofind. If we suppose that
the church was situated on ecclesiastical ground near the cemetery at
the Shankill crossroads, a mile from Elphin, then the natural con-
jecture would be that the foundation of Dumecha was earlier than
that at Alofind, and that, when the newer church was planted, the
earlier came to be distinguished from it as the ' Old Cell.'

The obvious objection to this conjecture is that it contradicts the
narrative of T(rechdn, who represents the foundation at Dumecha
as subsequent to that at Imbliuch Honon, or Alofind. This objection,
however, is not fatal. In fact, we come here into close quarters
with a problem of great importance regarding Tirech&n's itinerary.
He tells us himself that Patrick peruenit per Sinonam^ that is, visited

0*Donoyan, Leabhar na g-Ceurt, p. 101.

164 Proceedings of the Roffal Irish Academy.

Connauglit three times (d29ia) ; and the question suggests itself
whether Tfrechftn, in collecting his information from various souroes,
has not gathered up and compressed into the one visit to Connaught,
which he descrihes, incidents which really helonged to other visits.
At the end of this paper I will adduce a larger argument to prove that
T£rech&n was guilty of such a confusion ; hut here I may point out
that some of the incidents mentioned hy Tfrech&n imply a previouB
visit. Thus, we have a statement that, on crossing the Shannon,
Patrick ordained Ailhe, eui indieaiuit altar$ mirMle lapideum in mtnUe
nepoium AiMh (31 85). The natural implication is that Patrick had,
on a previous occasion, visited Sliab hua nAilello, and seen the altar.
Again, it is important to observe that, when Patrick comes to GcRrea
Ochland, he is described as coming not only to the chie& Hono and
1th, but to Assicus and Betheus, his disciples (8l3i,). Thus Assicns
and Betheus were already stationed in the district ; and the inf eiaiiee
may be that Patrick had visited it before, and planted a small
Christian oommunity somewhere. If so, the conjecture that the
Senella Cella had been founded on the occasion of the previous visit
seems plausible.

§ 6. At Senella Cella, Patrick was visited by Mathona, the sister
of Benignus; and here we encounter a diificult passage, whidi requires
elucidation : —

St uenit apnd ae filia felix in perigrinationem nomine Mathona aorar Beaigni
aoecestoria Patricii quae tenuit pallium apud Patricium et Bodaaom. Monaeba
fuit illis. £t exiit per montem filiorum AileUo et plantauit aecleanam liboam
hiTamnuch. Et hononta fuerat adeo et hominibus, et ipea fecit amiotiaai ad
reliquiafl aancti Bodani, et aucceaaores iUine epulalMuitur ad inuioem.

It is obvious that the words et exiit . . . hiTamniuek interrupt the con-
text most awkwardly, and that the sentence should run : Memmehafmt
iUiSf et honorata fuerttt adeo et hominihm. Et ipea^ ^. Moreover, the
idea forces itseU upon us that the subject of exiit vmlpUmUmU in the
inserted clause must be Patricius, not Mathona. And, turning to the
Tripartite Life, we find that, in the text in the Bolls ed. (p. 98),
though it agrees with the Armagh text, the clause in question i»
referred to Patrick.'

^ So» too, in Colgan*8 Latin translation (Tr. Th., p. 13d); where, hovever, the
sentence et ipta fecit, ftc, becomes et ipte feeit, ftc, and is postpoaed to the
following notice of the ordination of bishop Caiiell, to whom ipm ia made to
refer. — In regard to the clause /oMat^u ineclait sair AiTamnMJk (p. 98u, ed. BolU),

Bury — The Itinetwy qf Patrick in Cmnaught. 166

ThiB oriticiBm is borne out, and the problem defined, when we
diBOoyer that almost the same dause recurs in f . 15 r^ a (328]) :

et eziit tzans montem fiHorum Ailello et fundauit aeclessiam ibi, id Tamnach
et Echienach et Cell Angle et Cell Senchuae.

Now, the fact that this crossing of Sliab mace nAilello and founding
of the ohuroh in Tamnaoh (in Tirerrill) is mentioned in almost the same
words in two different contexts, at two different stages of the itinerary
which T(rechdn has marked out for Patrick, is highly significant. It
seems dear, in the first place, that the foundation of Tamnach was
not the work of an excursion of the saint from Shankill on this occa-
sion, but belongs to the context of other work in the region of
Tirerrill. And in the second place, it seems probable that the founda-
tion of the churches in Tirerrill (as described in the passage just
quoted) belonged to a previous yisit to Connaught. For it is natural
to suppose that it was from Tamnach that Mathona came in peregrina-
iionem to see Patrick, at Shankill ; and if so, the Tamnach community
was already established.

Moreover, the mention of the m»ns filiorum AiMlo corroborateB
these inferences. This chain of hills can hardly be any other than
the Bralieve mountains which divide Tirerrill from Leitrini. There-
fore if Patrick crossed these hills to reach the districts of Tamnach
and Senchua (Tawnagh and Shancoe), he must have come from the
Leitrim side. This confirms the conclusion that the work in those
districts belonged to a different visit.

A corollary of considerable importance may be drawn. We cannot
easily explain this particular confusion unless Tlrechdn had a written
source before him, in which the crossing of the Mens filiorum Ailello was
distinctly recorded in connexion with the foundations in Tirerrill. If
his material had been merely oral, he would have been less likely *to
fall into the topographical inconsistency which helps to reveal his
methods to us. But, having a written authentic statement before
him : exiit trans mmtemfliorum Aihtto^he simply wrote it out without
(Titicism.

In both the passages where he repeats this statement (814u and
d2S|), there are signs of patchwork. We have seen how the Mathona
passage is dislocated ; but there is also an awkwardness, though of a

the editor translates ** founded the church $ati in Tamnaob," and observes, in a
note, that Colgan gives imigmm §e«Mam^ *'as if for Mtr, his texU had tHr^
* noble.* '* But surely Tirech&n's lihtram shows that the word is ffl«r, ' free.*

B. I. A. PBOO., VOL. ZZnr., BBC. 0.] [12]

166 Proceedings of the Royal Irish Academy.

slighter kind, in tibe Tirerrill passage. The singular aeehaiam,
introducing fonr churohes, is corioas; and Dr. Stokes, feeing this|
was led to suggest aeeUstias quatnor. But aseletiiam seems to me not
to be a textual error, but to let out a secret of compilation. As
I hare said, Tfrech&n was here using a written source ; he used it
both for the Matiiona passage and forthe Tirerrill passage. Why did
he tdunk of luing it for the Mathona passage ? The obrious conjecture
is that it mentioned Mathona in connexion with Tamnach. This con-
jecture at once supplies the explanation of the angular Meeftnmmm,
In the source, the wordB/wuUmit aeclestiam liheram ihi^ id mi Tttmmmk
(or hi2hmnueh\ were followed by a notice of Maihoiia's association
with that community, after which the foundations of the otker
churches (Echenach, &c.) were enumerated. But Tfrechin had
worked the notice of Mathona into his account of ShankiU aad
Rodanus; and, consequently, he dropped it out whrai he came to speak
of the communities (^ Tirerrill. Bntindoing so, he left the «Mi«itMiii
(which in his source applied only to Tamnach), although he added, im
dependence on the same verb, the names of three other foundations.

S 7. From Elphin and Shankill, Patrick went on to Rath Cfochan,
seren or eight miles to the soutk*west (SHw) ; and there I must leave
Mm.

The two things which I have endeavoured to do im the fore^
going pages are (1) to identify the place at which, according ts
Tfrech&i's memoir, Patrick crcwsed the Shannon ; and (2) to show
that — assuming the author's statements as to Patrick's doings in Con-
naught to be more or less authentic — ^we are forced to infer that, in
ptitting together his material, he has worked into the frame oi
one visit events which must have belonged to different visits ; because
he has unwittingly left certain implications which betray this oneon-
seious contamination.

But the suggestion tbat ^e events of more than one expedition to
Onmaught have been confused and conbuied in the narrative of
Tfrechin admits of a clearer and more trenchant demonstration,
which touches the whole plan of his memoir. The motive id the
circular tour which the writer deseribeft is represented to have been
the meeting of Patrick with the sons of Amolngaid* It was arranged
by Endae, one of these brethren, and Patrick that they should travel
together to Endae's country in north-western Connaught, to establish
the Christian faith in those regions. But the route followed by
Patridi is quite inconsistent with this motive. In the first place, he
spends a long time in missionary or eccledastioal work in Meath

Bury— JA^ Itifierary of Patrick in Connaught. 167

before he enters Connaugbt, And when he crosses the Shannon, he
makes a long tonr in Boscommon and Maye before he cornea te
Tirawley. The goal of his Jonmey is entirely lost from view;
Tirawley is ahnost the last part of Connanght he visits. It is maaU
f estly absurd to suppose that Endae and his followers undertook te
aecompany the i^slle on this long round of missionary activity. Nor
is there, in the itinerary itself, the slightest indication that they did
80. Endae and his arrangement to travel with Patrick are completely
forgotten in Tfrechdn's story, until suddenly — after the lapse of
months, or years — ^he reappears with his son Conall, as Patrick's
companion, when the saint at length crosses the Moy and enters
Tirawley.

At this point, indeed, the suture in Tfrechdn's compilation is
visible. The route can be traced from stage to stage through Bos-
common and Mayo to Mount Egli in Murrisk. After his fast on the
mountain, Patrick proceeds to the region of Corcu-themne, which
seems to have been near tiie Partry mountains and L. Mask. Then we
find him in ri^ionihua maice H$reae in Diehuil et Aurehuil (d24M), and
in the White Plain in regionihu nepotum maini. This was probably
in southern Roscommon. Then there is an extraordinary leap :^

Per Mnadam uero uenit «t ecoe audierunt magi flliorum Anobigid quod sanctua
iiir ueniiset, etc. (325m}.

The break here in the itinerary is manifest, and exhibits very clearly
the method of Tfrechin. The narrative between SlOi, and 826n—
between the starting for Tirawley and the coming to Tirawley — is
wholly or mainly concerned with the incidents of another journey,
or other journeys, than that which was taken expressly for the
purpose of converting the tribe of Amolngaid.

In one passage Tlrech^ himself betrays a consciousness of the
incongruity. He states— inconsistentiy with the context and the
situation — the object of the expedition of Patrick and Endae to have
been Mount Eg^ (310,), whereas the tenor of his own account
implies that it was Tirawley. This is the only attempt he makes
to conciliate the actual itinerary with the avowed motive of the
joum^.

This investigation confirms the suspicion which I hazarded in a
former paper on Tfrechdn, that, while the notices of the particular
incidents which he records depend on sources written or oral, and may
is many cases be credible, yet the actual route which he traces and

168 . Proceedings of the Boyal Irish Academy. .

the chronological order whicli he assames may be due to his own
combinations. I must add that further study and . moie . minate
analysis of Tfrechdn's text have led me to conclude tiiat he had
more written material at his disposal than I was before incHnedto
suppose.^

I See Tirech&n's Memoir of St. Patrick in fing. Hist. Review, April l^^
pp. 236, tqq.

Proc. E. I. Acad., Vol. XXIV., Sec. C.

rittto IX.

Glass Lovixc-cup of Ouzel Galley Socibtit.

Proc. B. I. Acad., Vol. XXIV., Sec. C.

Plate X*

Gla68 LoviMO-cvi' OF Oi'ZEL Gallbt Society.

Proc. R. I. Acad., Vol. XXIV., Sec. C.

riate XL

Boatswain's Whistle and Bgyeuse of Medals of Ouzel Galley Societv.

Troc. B. I. Acad., Vol. XXIV., Sec. C.

Plate XII.

BoAi8WAU('s Whistle and Obviiksb of Msuals of Oczel Oaubt Societt.

[ 169 ]

XI.

THE COUNTIES OF IRELAND: AN HISTORICAL SKETCH
OF THEIR ORIGIN, CONSTITUTION, AND GRADUAL
DELIMITATION.

By C. LITTON FALKINER, M.A.

Read Noybmbbr 29, 1902.

Not the least of the many merits of that most luminous of nineteenth-
centuiy historians, the late John Richard Green, is his insistence on
the importance of the relation in which geography stands to history.
By geography Mr. Green meant not so much physical as political
geography. The dominating influence upon the development of any
given race or people of the main physical characteristics of the land in
which their lot is cast has long been understood by historians ; and
the effects produced on the history of the world — in modem times, by
the insular position of Great Britain, or, in the world of the ancients,
by the peninsular position of Greece — are among the commonplaces
of historical criticism. What is not so much a commonplace is the
extent of the influence exerted upon the domestic history of any
community by the accidents of its early local history, and the degree
in which archaic conditions of tribal division may survive in the
modem organisation. For these divisions often continue for long
centuries after their origin has passed into the partial oblivion of
unexplained tradition, to mould the shape and form of a more advanced
civilization.

The application of this principle to the case of Ireland is direct
and obvious. For the local history of Ireland is, as has been acutely
observed, in a special degree, the backbone and foundation of its
general history. Owing to what may be described as the inorganic
character of the social structure in the Ireland of the Middle Ages, to
the absence of a strong central government or settled constitution,
capable of giving to the country and the people the impress of its own
Qidformity, it is almost exclusively to clan or sept history, and to the
history of the particular areas with which the septs were associated,

s. X. A. psoo., TOL. xzrr., bkc. c] [13]

170 Proceedings of the Royal Irish Academy.

that we must chiefly look if we would seek to realise the body politic
of the Ireland of a not very remote past. If this statement should
appear at all exaggerated, let it suffice to note two simple but striking
illustrations. As late as the reign of Henry YIII., in a memorandum on
the State of Ireland, which is among the most instructive documents in
the Tudor State Papers, the names of the ^' Irish regions," and not Hie
territorial divisions to which we are accustomed, are the units employed
by the writer to describe by far the greater portion of the country.'
And in the Elizabethan Map of Ireland, drawn by Dean Kowel, in
the third quarter of the sixteenth century, division by territories, or
'' chief eries," and not that by counties, is the method adopted;* for
down to the reign of Philip and Mary, as Sir John Davies observes in
the lucid paragraphs devoted to the history of the shiring of Ireland in
his well-known work : — *' The provinces of Connaught and Ulster, and
a good part of Leinster, were not reduced, to shire ground. And
though Munster were anciently divided into counties, the people were
so degenerate as no justice durst execute his conmiission among them."^
It is the main object of this Paper to indicate the process by which
these large districts were gradually brought within the ambit of
English administration, and by which the counties of Ireland, as we
now know them, came to be formed.

" The civil distribution of Ireland," to quote Bishop Beeves*s
most valuable Paper on * The Townland Distribution of Ireland,' ** in
the descending scale, is into Provinces, Counties, Baronies, Parishes,
and Townlands."^ But this highly convenient division of the sur&ce
of Ireland, as the Bishop goes on to say, is characterised neither by
unity of design nor by chronological order in its development. '^ The

i^'Who list make sunnifle to the King for the reformatzon of his land of
Ireland, it is necessary to show him the estate of all the nohle folk of the same, as
weU of the Song's suhjects and English rebels, as of the Irish enemies. And first
of all to make His Grace understand that there may be more than 60 countries,
called regions in Ireland, inhahited with the King's Irish enemies ; some re^on as
big as a shire, some more, some less, unto a little ; some as big as half a shire and
some a little less ; where reigneth more than 60 chief captains . . . that liveth
only by the sword and obeyeth to no other temporal persons, but only to himadf
that is strong . . also there is no folk daily subject to the King's laws but half the
county of Uriel, half the county of Meath, half the county of Dublin, and half the
county of Kildare." <*The State of Ireland and Plan for its Reformation.*'
*' State Papers Henry VIII.," vol. ii., Part iii., p. 1.:

s Copy of an ancient map in the British Museum by Laurence Nowel, Dean of
Lichfield, oh. 1576. Printed by Ordnance Survey.

* *' Discovery of the True causes why Ireland was never entirely subdued," &c.

« « Proceedings of the Royal Irish Academy," vol. vii., p. 473.

Falkinbk — I%e Oountien of Ireland. 171

provinces, subject to one suppression and some interchange of adjacent
territories, represent a very ancient native partition which in the
twelfth century was adopted for ecclesiastical purposes. The counties
and baronies, though principally baaed on groupings of native lordships,
are of Ang^o-Norman origin, and range, in the date of their creation,
from the reign of King John to that of James I. The parochial
division is entirely borrowed from the Church, under which it was
matured probably about the middle of the twelfth century ; while the
townlands, the infima species, may reasonably be considered, at least in
part, the earliest allotment in the scale."

With the two last of these grades of classification we have nothing
to do here. But a word must be said regarding the third. The
baronial division does not indeed present any very difficult problem.
For though it be not easy to account for the adoption of the term
'* barony" as signifying the division of a county,^ seeing that it has
no such meaning in the territorial classification of Ghreat Britain, there
is no doubt that in general the baronies were successively formed on
the submission of the Irish chiefis, the lands of each chieftain consti-
tating a barony, and that they thus represent more nearly than any
other unit the ancient tribal territories. The origin of the parochial
system is much less easily traced; and the relation between the
diocesan areas and the provincial or county divisions is a subject
which might well engage the attention of some of our ecclesiastical
antiquaries.

The limits of the five kingdoms of what has been called the Irish
Fentarchy, into which Ireland was anciently divided, correspond
closely to those of the provincial divisions, as the latter were main-
tained down to the seventeenth century. They represent, as
Dr. Beeves has pointed out, '* a very ancient native partition," the
adoption of which in the twelfth century, for ecclesiastical purposes,
served to embalm a division of our island which, being based on no
great natural boundaries, must otherwise have perished. The five
provinces are shown separately as late as 1610 in Speed's map. For
it was not until late in the reign of James I. that Meath ceased to be

> ** The cause of the diflereaoe in name between the Irish baronies and English
hundreds has been thus accounted for : When the kingdom of Heath was granted
to the elder De Lacy, shortly after the arrival of the English, he portioned it out
among his inferior barons, to hold under him by feudal serrioe, and hence their
estates naturally took the name of baronies, which gradually extended itself to
aimilar subdirisions of other counties." See Hardiman*s *' Notes to the Statute of
Kilkenny," in « Tracts relating to Ireland," ii., p. 108.

[13»]

172 Proceedings of the Royal Irish Academy.

generally reckoned a separate province ; in popular usage it long re-
tained its provincial identity; and Boate, writing under the Common-
wealth, mentions the province as but lately merged in Leinster. The
Ulster of unsubdued Ireland was conterminous with the modon
province of that name, save that it included Louth — a fact oom-
memorated in the still existing incorporation of that county in the
See of Armagh and the northern ecclesiastical province — and that it
did not include Gavan. Ancient Munster differed from the modem
only by including within its boimds the territory of Ely (the
O'Carroll country), which, now represented by two baronies of the
King's County, forms a part of Leinster. Connaught included, in
addition to its present territories, the County of Cavan, and a part of
Longford ; while during the sixteenth century the earldom of Thomond
or County of Clare oscillated, as we shall see, at the pleasure of suc-
cessive deputies, between Munster and Connaught, giving to the western
province, in the periods of its association with it, a predominance it
has long ceased to enjoy. Meath, which is substantially identical
with the modem counties of Meath and Westmeath — ^it is practically
conterminous with the diocese of Meath — also embraced a considerable
portion of Longford ; while Leinster comprised the modem Leinst^
counties, less Louth, Meath, Westmeath, Longford, and the part of
the King's County specified above.

The first attempt at a division of L:«land into counties was, of
course, subsequent to the Anglo-Norman conquest, and is commonly
dated from the reign of King John. It is generally ascribed to the
tenth year of that monarch's reign ; but it does not appear that this
ascription, though doubtless substantially coirect, rests upon any
extant documentary authority of ancient date. It has been adopted,
however, by every writer, and Sir John Davies's account is as succinct
and accurate as any other : *' Tme it is that King John made twelve
shires in Leinster and Munster — ^namely, Dublin, Kildare, Meath,
Uriel or Louth, Catherlogh, Kilkenny, Wexford, Waterford, Cork^
Limerick, Kerry, and Tipperary. Yet these counties did stretch no
further than the lands of the English colonies did extend." Harris,
in his additions to Ware's account of the division of Ireland,^ asserts
and, indeed, elaborately argues, that the twelve counties attributed to
King John were really of earlier origin, and were, in &ct, part of an
earlier division effected by Henry II. Without a division into shires
and the appointment of sheriffs, Henry's grant to Ireland of the laws

1 << AntiquitieB of Ireland," ohap. v.

Falkinbr — The Counties of Ireland. 173

of England would, in his opinion, have been no better than a
mockery : ** For without sheriflfa, law would be a dead letter ; " and
without a shire there could be no sheriff. That there were sheriffs
in Henry's reign Harris considers proved by the language of a patent
to one Nicholas de Benchi, directed to all archbishops, bishops,
sheriffsj &c. ; and that shires were known in Ireland prior to the
tenth year of King John is shown by a patent of the seventh of
that reign, in which the County of Waterford is distinguished from
the City of that name. In further support of his thesis, Harris
also argues that the division of Connaught into the two counties of
Connaught and Roscommon is of earlier date than King John's
counties ; that Leiz and OfEaly were reckoned in Kildare, and other
portions of the Queen's County in Carlow, prior to the reign of Philip
and Mary ; and that there were unquestionably sheriffs of Down and
Newtownards, of Carrickfergus and Antrim, and of Coleraine, long
prior to the division of Ulster into counties under Elizabeth. But
though he would be a bold antiquarian who would venture to contro-
vert a proposition maintained by the erudition of Ware, the authority
of Ware's laborious editor is hardly so formidable. It may at least
be said that if the shiring of Ireland was really accomplished by
Henry II., aU substantial traces of it have perished ; and the historian
must be content to start with King John.

As has just been noted, there is no conclusive evidence now extant
of the formation by King John of the twelve counties traditionally
ascribed to him. And it is certain that though these divisions were
probably known as separate geographical areas, they cannot in several
instances, if in any, have formed counties in the modem administrative
sense till a date considerably later than King John's reign.^ For it
must be remembered that the earliest grants of territory by Henry U.
-were in the nature of counties palatine rather than of ordinary counties,
though the term *^ palatine " nowhere occurs in any early instrument ;
and of the twelve counties imputed to King John, five formed part of
the single liberty or palatine county of Leinster. In order to follow
the process of the development of our Irish counties, it is essential to
have regard to this fact and to the consequences flowing from it. It is
therefore necessary to consider the origin of the institution of counties,
and the difference, in the extent and nature of their respective
jurisdictions, between simple and palatine counties.

1 See Hardiman's *' Notes to the Statute of Kilkenny ** in <* Tracts relating to
Ireland,'* ii., p. 102.

174 Proceedings of the Royal Irish Academy.

The name and office of Count were derived from the Court of
Charlemagne, and the institiition of oounties in England is of earlier
date than the Norman Conquest.^ The creation of a count inTolTed
from the first a delegation of royal authority for legal and adminiB-
trative purposes, and the ordinary county had two courts — ^the King's
Court for criminal cases, and the Earl's Court for civil causes. But
the judicial officers and sheriJBb were in all cases appointed by the
Crown. Between a county palatine and an ordinary county the
distinction was broad and well defined. According to Blackstone,
*< counties palatine" — of which there were in England the three great
examples of Chester, Durham, and Lancaster, besides the smaller ones
of Hexham and Pembroke — '< are so called a palatio, because the
owners of them had formerly in those counties/Km reyalia as fully as
the King in his palace."* The Earl of a county was Lord of all the
land in his shire that was not Church land ; and his jurisdiction was
equivalent in all essential points to the jurisdiction of the King in on
ordinary county.' The /lira reyalia included a royal jurisdiction and
a royal seignory. By virtue of the first the Earl Palatine had the
same high courts and officers of justice as the King ; by virtue of the
second he had the same royal services and escheats, and could even
create barons, as was certainly done in Chester. Included in the
power to appoint officers of justice was the appointment of the sheriff;
and with the functions of the sheriff in the palatinate no King's
sheriff might interfere. And therefore, says Sir John Daviea, *' sudi
county is merely [absolutely] disjoined and separated from the Crown,
so that no King's writ runs there, except a writ of error, which being
the last resort and appeal is excepted out of all their charters."*

The origin of these immense delegations of royal power was of
course the inability of the Sovereign in early times to establish an
efficient administrative system throughout his realm; and the same
considerations which compelled resort to the palatine system in
England by the early Norman Kings, rendered necessary the applica-
tion of an analogous method of administration in Ireland by Heniy XL
In the case of England, where the central authority was strong, the
palatinates were limited to the march or border districts, as CSieeter

> Seidell's ••Titles of Honour," p. 694.
> Stephen's Blackstone, i., p. 131.
» Stubbs's •' Constitutional History," i., p. 363.

* Sip J. Davies's ** Reports des cases et matters en Ley," «« Le Case del Counlx*
Palatine de Weixford," p. 62.

Falkinbr — The Counties of Ireland. 175

on the Welsh and Durham on the Scottish or Northumbrian borders.
In the case of Ireland, the Crown having practically no authority, the
policy of Henry II. was to hand over the country to Strongbow and
his followers, with powers practically co-extensive with the powers
of the Crown, but subject to and excepting any grants of Church
lands. Only the sea-coast towns and the territories immediately
adjacent were reserved to the Sovereign. And, in fact, it was in these
latter districts, and in these only, that for a long period the authority
of the English Kings had any direct force in Ireland.

Accordingly, as Sir John Davies, with his usual insight, observes,
all Ireland was '' cantonised " by Henry II. among the persons of the
Kngliflh nation, who, ''though tiiey had not gained the possession of
one-third part of the whole kingdom, yet in title they were owners
and lords of all, so as nothing was left to be granted to the natives."
Of these grants at least three — ^those of Leinster to Strongbow, of
Meath to De Lacy, and of Ulster to De Courcy— were grants of royal
jurisdiction equivalent to palatinates; and most probably all were
intended to be such. It is dear at all events that the liberty of
Ijcinster was confirmed in right of Strongbow's daughter to William
Marshal, Earl of Pembroke, by King John, and that, on the division
of Leinster among the five co-heiresses of the latter, the five divisions
of Carlow, Kilkenny, Wexford, Kildare, and Leix were regarded as
separately enjoying, within their respective territories, the same
palatine privileges which had pertained to the undivided liberty of
Leinster. That Leinster was long regarded as preserving its palatine
privileges may be seen by the Statute 25 Ed. I., in which ''the whole
community of Leinster" is referred to as " lately but one liberty."

Of the remaining palatinates or liberties, Meath was divided
between the sisters of Walter de Lacy, of whom Matilda married
Walter de Greenville, and Margaret, John de Yerdon. The half known
as the liberty of Trim passed to the Crown through the marriage of a
descendant of Matilda de Lacy with Mortimer, Earl of March ; while
the second half, descending to the Talbots, Earls of Shrewsbury, was
resumed by Henry VIII. under the Statute of Absentees.^ Ulster,
originally granted to De Courcy, was re-granted by John to the
De Lacys, and descending through a daughter to the De Burghs, and
thence to the Mortimers, ultimately became vested in the Crown in the
person of Edward lY., as the descendant of Lionel Duke of Clarence.
Connaught, granted to the De Burghs, also technically passed with

1 Stat. 28 Henry YIII., cap. ill.

176 Proceedings of the Royal Irish Academy.

Ulater to the Crown ; though the rebellion of the younger branch of
the BurkeB, on the failure of heirs male of the elder, deprived the legal
title of the Crown of all effective force. The union of all these terri-
tories in the Crown of England is incidentally recognised in an Act of
Parliament of Henry VII.'s reign (10 Henry VII., c. 15), whicli,
reciting that *' the Earldoms of March, Ulster, the Lordships of Trim
and Connaught, bin annexed to our sovereign lord the King's most
noble Crown," mates provision for the better keeping of the records
of those ancient dignities, the title to which had been jeopardifled
by the loss of the muniments. This Act expressly refers to '* Eichard,
late Duke of York," as lord of Trim.^

The precise character of the jurisdiction conferred by King John on
the early Palatine counties of Ireland does not appear from any extant
documents. But if, as it seems reasonable to suppose, the later juris-
dictions conferred by Edward'III. were similar in their general scope,
its nature may be gathered from the records of the Palatinate of
Tipperary. The process of Quo Warranto by which James I. resumed
possession of Tipperary enumerates the courts and offices which existed
at the beginning of the sixteenth century, and which, doubtless,

^ Selden, ia hia ** Titles of Honour " (third edition, p. 694), has a reference to
the use of the name and office of Palatine Earl in Ireland, whicli seems to
state the facts with great accuracy : — '* The title of local Earl Palatine, as well as
of other Earls, occurs in the Becords of that Kingdom. But I do not beUeve that
any man was ever created into the title of Coufit Palatine there, or the County
expressly made a County Palatine by Patent ; but as in other countries, so here,
the enjoying of the title of earl (and sometimes of lord), together with a territory
annexed to that title, wherein all royal jurisdiction might he exercised, was the
original whence in speech and writing the title of Earl Palatine or Count Palatine
grew." This was written in 1614 ; and it is noteworthy that Sdden's view as to
the title of Palatine is confirmed by the Patent of Charles II. to the Duke of
Ormond in 1660 for the County Tipperary. Tipperary was an undoubted Pala-
tinate ; yet neither the Patent nor the Act of 2 (reorge I., cap. 8, by whicb it
was revoked, contains the term <* Palatine " ; but they speak only of the regal%tie$
and liberties of Tipperary.

The extent and character of the privileges of a county palatine or liberty of
England appear by the Charter of Edward III. to John of Ghaunt for the Palatinate
of Lancaster — a dignity which, owing to the prudent sagacity of Henry lY., has been
preserved in its ancient independence and prerogatives almost down to the present
day. Anxious that the hereditary honours of his dukedom should be secured to him,
even should fortune deprive him of a usurped crown, Henry, on attaining to the
throne, had an Act passed providing that the duchy of Lancaster should remain in
himself and his heirs in like manner as though he had never acceded to the royal
dignity.

Falkiner — The Counties of Ireland. 177

represented in all essentials the Palatine constitution of earlier times.^
The jurisdiction, authorities, and liberties set out in the Quo Warranto
of James I. were restored on the reconstitution of the Palatinate in
1662 in favour of James, first Duke of Ormond, with the exception
(which appears to have been a reservation common to all Palatine
grants) of the four pleas of arson, rape, forestalling, and treasure
trove, as originally reserved in the grant of Edward III.

In tracing the position of the Irish counties through the obscure

' The following are among the more important of the privileges vested in the
Earls of Ormond within their Palatinate : —

1. To have and to hold within the county of Tipperary one Curia Canetllariae,
commonly called a Chancery Court, and to make, appoint, and constitute one
CaneelUriuB, or officer of the same Court, commonly called a Chancellor, which
Chancellor, under colour of such his office, makes and causes to be made all kinds
of original writs and other processes in all actions, as well real as personal and
mixed, within the aforesaid county arising, occurring, or happening.

2. To have and to hold within the aforesaid county one other Court of Pleas of
the Crown of the said Lord the now King, and to make, appoint, and constitute
one other officer or Seneschallus, commonly called a Seneschal, and one other
officer or Justiciarius, commonly called a Justice, to hold Pleas of the Crown of
the said Lord the King.

3. And also to have and to hold within the aforesaid county one other Court of
Common Pleas held before the aforesaid Seneschal and Justice.

4. And also yearly to nominate, appoint, make, and constitute in the same
county one other officer, viz., one V%ceeom$9f commonly called a Sheriff, for the
custody of the same county, which sheriff makes execution of all writs, &c., issuing
and directed to the same sheriff from the four courts of the said lord the King held
at the King's Courts in the County of the City of Dublin, also from the Justices
assigned to take the assizes in the County of Tipperary aforesaid, as well as from
the aforesaid Chancellor, Justice, and Seneschal in the same county. And he holds
in the same county divers Courts of TWm, Leet^ and Curias Ctmitatui, called
County Courts.

5. And moreover to have and appropriate to themselves the power of granting
charters of Pardon, and ad parthnandum — Anglick^ to pardon — ^whatsoever persons
are suspected, accused, convicted, outlawed, condemned, or attainted of any felonies
and treasons by them within the aforesaid county in any wise done, committed, or

perpetrated And further to do and execute within the aforesaid county all

other things whatsoever which appertain to any Earl of any County Palatine to be
done or executed.

6. And also to make, appoint, and oonstitnte in the aforesaid county di?ers
other officers, viz., one or more Coroners, and one Escheator and one Feodary,
and one Clerk of the Markets, and one Sub«vicecomes, commonly called a Sub-
sheriff.^Fifth Beport of the Deputy Keeper of the Public Records of Ireland,
pp. 84-36.

178 Proceedings of the Royal LHsA Academy,

complexity of Irish administration under the Plantagenet Kings, tiie
only guide whom we may follow with any degree of confidence is the
Sheriff. The whole machinery of local or county administration in
Plantagenet times practically centred in the Sheriff, who united the
threefold functions of a civil officer in relation to the courts of law ;
of returning officer in relation to the election of parliamentary repre-
sentatives ; and of revenue collector in relation to the royal exchequer.
Owing to the destruction in the reigns of the two first Edwards of
most of the early records of the Kingdom of Ireland, the materials
availahle in regard to Plantagenet Sheriffs are unhappily meagre;
and the Act of Henry YII. just referred to indicates the paucity
of the records of several of the greater earldoms. But a study of
the Plea Rolls, Pipe Bolls, and Patent Bolls, as well as of the Planta-
genet statutes, so far as these survive, is not wholly fruitless ; and
the last-mentioned source is fairly rich in references to the functions
and office of the sheriff. An examination of these sources estahlishes,
at least negatively, the fact that from the time of King John to that of
the Tudors no new county was formed, or if formed that it did not
survive ; and that no Sheriff was created for any new district, with
the single exception of the suhdivision of the great territory of
Gonnaught into the separate districts of Gonnaught and Boscommon.^
It is impossible to say how much or how little of Gonnaught was
intended to be included in Boscommon, or precisely when the division
was made. But the separation is certainly as old as the thirteenth
century, and Boscommon is among the counties and liberties' whose
respective Sheriffs and Seneschals were directed by the Statute 25
Ed. I. (1296) to return to the ^'general parliament" held in Dublin in
that year ^'two of the most honest and discreet knights of each
county or liberty." This vagueness of the territorial divisions and of
the shrievalties associated with them was not confined to the western
province, but was characteristic of all the so-called counties of King
John. And this was especially so in the case of the Leinster counties,
whose south-western borders were probably in a state of continuous
flux. Thus in 1297 a list of Goroners of Kildare shows that county
to have included Offaly, Leix, and Arklow, and therefore to have

^ See Hardiman's << Statute of Kilkenny/' p. 106.

' The following is the enumeration in the Statute : — *' Likewise the SheriJSB of
Dublin, Louth, Kildare, Waterford, Tipperary, Cork, Limerick, Kerry, Coo-
naught, and Boscommon ; and also the Seneschals of the liberties of Meath, Weya-
ford, Katherlagh, Kilkenny, and Ulster.*' See Betham's ** Feudal Dignitiee," p. 262.

Falkiner — The Counties of Ireland. 179

extended far over its present borders into the modem counties of
King's County, Queen's County, and Wicklow.

The broad distinction which was drawn between counties ordinary
and counties palatine was reflected in the designation of the most
important office in their respectiye jurisdictions. In the county
proper that officer is invariably styled sheriff; but in the county
palatine he is as uniformly r^erred to as ''the seneschal of the
liberty." The distinction is clearly marked in a mandate of
Edwud m. to the Treasury of Ireland, which directs that '^ because
the liberty of Carlow has been taken into the King's hands,"^ the
writs of the King for execution should be directed to the sheriff of
Carlow, in place of the late seneschal of that liberty.'" It appears,
however, that a general jurisdiction lay in the sheriff of Dublin for
districts not clearly belonging to a specific county or liberty, or
wherever the seneschal of the latter should be found in default, as in
the case of Kildare prior to the Statute of 25 Edward I. In
18 Edward II. precepts were issued to the sheriffs of Dublin and
Meath to execute writs ''in spite of the liberties of Kildare and
Louth " ; but this interference with the general principle of palatine
independence was doubtiess exceptional and probably due to the
disorganisation resulting from the Bruce invasion. For so extensive
were the privileges of the liberties that, though the King might
and did appoint sheriffs within their limits, the authority of the
royal officers extended only to the Church lands, whence they were
known as sheriffs of the County of the Cross. Of such counties
there must originally have been as many in Ireland as there were
counties palatine' ; but with the gradual absorption of the palatinates
in the Crown, either by inheritance, as in the case of Ulster, or by
forfeiture, as in that of Wexford, they had all ceased to exist before
the reign of Henry VIII., except the County of the Cross of
Tipperary, which, being within the great Ormond palatinate, created
by Edwaxd III., survived till Stuart times.

' ThiB had been done by virtue of Edward III.'s arbitrary but temporary revoca*
lion of all franchiBes, liberties, and grants fonnerly made ui the Emgdom of Ire-
land— a measure doubtless intended primarily as an answer to the renunciation by
the Bourkes of Connanght of their allegiance to the Crown, and to the genend
disorganisation which followed the wars of the Bruces.

> Close BoU, 17 & 18 Edward III.

'In the list of Proffers and Fines of Sheriffs & Seneschals in the time of
EdwBzd III., Sheriffs of the Cross are mentioned for the Crosses of Kilkenny,
Tipperary, Carlow, Wexford, Kerry, Kildare, Meath, and Ulster.

180 Proceeding% of the Royal Irish Academy,

Whatever the precise origin of the counties so generally ascribed
to King John, there appears to he no douht that the writs either of
the king or of his palatines ran in all of them for a full century from
John's time, and that these counties represent the extent of the
effective predominance of English power down to the invasion of
Edward Bruce in 1315. Prior to that event some efforts seem to
have been made to extend the counties to Ulster, and to define
more accurately the limits of the Leinster counties. An Act of
25 Edward I. (1296), for the settlement of Ireland, enacted that
''henceforward there shall be a certain sheriff in Ulster, and that
the sheriff of Dublin shall not intermeddle henceforth in Ulster."
Meath was declared to be a county by itself ; and Kildare, which
had been regarded as a liberty of Dublin, was discharged from the
jurisdiction of the Dublin sheriff, and given an independent position.
But from the wars of the Bruce the English colony received a blow
from which it did not recover until the Flantagenets had been
replaced by the Tudors. The authority of the State, so far as it
was effective in the interior of the island, was exerted through the three
great earldoms of Ormond, Desmond, and Kildare, all of whidi date
from the fourteenth century. The area under the direct control of
the Crown was narrowed continually, until after a lapse of precisely
two centuries more the boundaries of the English Pale had shrank
to its lowest limits, and, in the quaint language of Stanyhurst, were
*' crampemed and crouched into an odd comer of the country named
Fingal, with a parcel of the King's land of Meath and the counties
of Kildare and Louth." Thus from the reign of Edward II. to that
of Henry VIII. the extension of the Irish counties was politically

That the shrinking of the English Pale had been accompanied by

^ The Pale at this period is thus descrihed in the State Paper of Henry YIII.
abeady referred to : —

'* Also the English Pale doth stretch and extend from the town of Dundalk to
the town of Derver, to the town of Ardee, alway on the left side leaying the
march on the right side, and so to the town of Sydan, to the town of Kenlys,* to
the town of Dangle, t to Kiloock, to the town of Clane, to the town of Naas, to
the hridge of Cucullyn,^ to the town of Ballymore,§ and so backward to the town
of Bamore,|| and to the town of Bathcoole, to the town of Tallaght, to the town
of Dalkey, leaving alway the march on the right hand from the said Dimdalk
following the said course to the said town of Dalkey."ir

• Kells. t Dangan. t Kilcullen. \ Ballymore- Eustace. || Rathmore.
ir "SUte Papers," Henry VIII., vol. ii., part iii., p. 22.

Falkinkr — The Counties of Ireland. 181

a parallel dizninutioQ of the interest in and knowledge of the country
poBsessed by the English Sovereigns may be sufficiently inferred from
the language used in 1537 in a ''Memorial for the Winning of
Leinster," addressed by the Irish to the English Council, which begins
by reciting that '' Because the country called Leinster and the situation
thereof is unknown to the King and his Council, it is to be understood
that Leinster is the fifth part of Ireland." ^ But from this period, never-
theless, may properly be dated the revival of English authority. In
1541 the resolution of the Sovereign himself to convert his long
nominal lordship of Ireland into an effective supremacy, was shown by
the Act constituting Henry YIII. King of Ireland ; and this was the
prelude to the adoption of that policy of converting the chiefs of the
Irish septs into the immediate feudatories of the Crown which led
directly to the conversion of the lands without the Pale into districts
cognisable by English law, and ultimately to their formation into
modem counties. Little, indeed, was done under Henry YIII. towards
defining the County boundaries, the only actual change in the map
being the severance of Westmeath from Meath by a Statute of 34 Henry
Vni. But though the proverb quoted by Sir John Davies continued to
hold good during the reign of Henry YIII. , that ' ' whoso lives by west of
the Barrow, lives west of the law," the area of the anglicised districts
steadily increased. The greater part of Leinster was in this and the
succeeding reign gradually won back to what was caUed '' civility'' ;
till towards the close of Elizabeth's reign the Pale was understood to
extend through all Leinster, Meath, and Louth.'

The first step in this process of restoration, and the first real
addition to the list of Irish counties made since King John's time, was
the formation of the King's and Queen's Counties in the time of Philip
and Mary.' The districts of Leix and OfEaly, the territories of the
powerful septs of the O'Moores and O'Connors, were, in that reign,
reduced to subjection, during the Yiceroyalty of the Earl of Sussex,
who, in the words of Sir John Davies, '^ took a resolution to reduce
all the rest of the Irish counties unreduced into several shires." Sussex
was the first of the Tudor Deputies to acquire a really systematic
personal acquaintance with the country he was sent to govern ; and
the accounts of his journeys through the provinces,' of which he made

» " State Papers Henry VIII.," vol. ii., Part iii.

'See **A Perambulation of LeioBter, Meath, and Louth, of which consiBt tbe
EngUflh Pale" in 1596. « Garew Cal.," iii., p. 188.

'See « Calendar of Garew Papers, I./' pp. 257, 265, 274, 330, 362

182 Proceedings of the Royal Irish Academy.

at least three, together with his reports to Mary and Elizabeth of the
results of his observations, are among the most valuable of the State
Papers of that age. Sussex proposed to divide Ireland into six parts,
viz., Ulster, Connaught, Upper Munster, Nether Munster, Leister,
and Meath ; and he enumerates in his Beport the countries which
these divisions respectively comprised. But though he appears to
have been the first to conceive any large plan for an efficient adminis-
trative settlement of Ireland, he was recalled before he had had
time to grapple effectively with that problem of the shiring of Ireland,
which he saw lay at the root of all real administrative reform. But
at least he made a beginning. It is worthy of remark, too, that
Sussex is the only Deputy who, in addition to creating fresh counties,
gave to his creations names not borrowed from the territories by which
they were constituted.^

In 1556 there was passed the Statute 3 & 4 Philip & Mazy,
Gap. II., '' whereby the King's and Queen's Majesties, and the heirs
and successors of the Queen," were declared entitled to the countries of
Leix, Slewmargy, Irry, Glenmaliry, and Offaly, and provision was
made for making these countries shire ground. After reciting that
these countries had been subdued in the previous reigns, bat had
rebelled and been again reduced by the Queen's Deputy, Thomas
Batcliff Fitzwalter, Earl of Sussex, the Statute proceeds thus: —
'* And for that neither of the said countries is known to be vrithin the
limits of any shires or counties of this realm, be it enacted that the
King and Queen, and the heirs and successors of the Queen, shall
have, hold, and possess for ever, as in the right of the Grown of
England and Ireland, the said countries of Leix, Slewmargy, Irry,
Glenmaliry, and Offaly." A further section provided that ** to the
end that the same countries may be from henceforth the better
conserved and kept in civil government, the new fort in Leix be from
henceforth for ever called and named Maryborough, and the countries
of Leix, Slewmargy, Irry, and part of Glenmaliry, be one shire
and county named the Queen's Gounty " ; and, similarly, that the
new fort in Offaly should be named Philipstown, and the country of
Offaly and part of Glenmaliry be called the King's Gounty.

T\mt the Government of the Earl of Sussex contemplated a
further extension of the policy embodied in this Act appears from the

' The case of Londonderry is an exception to this statement more apparent than
real. In its first form, the County of Londonderry was known as Colenune,
taking its name from the well-known town of that name.

Falkiner — The Counties of Ireland. 183

Statute immediately succeeding it,^ '^ to convert and turn divers and
sundry waste grounds into shire ground." This Act provided for
the appointment of Commissioners ''to view, survey, and make
inquiry of all the towns, villages, and waste grounds of the realm
now being no shire grounds," with power to the Commissioners to erect
such districts into counties. Nothing was done in this short reign,
nor for some years afterwards, to give effect to this enactment. But
widely as the general policy of Elizabeth differed from that of her
predecessor, her attitude towards Ireland was in principle the same
as Mary's. The Statute (11 Elizabeth, Cap. 9), ''for turning of
countries that be not yet shire grounds into shire grounds," sub-
stantially re-enacted the earlier legislation.' And the task of giving
effect to these provisions was confided by Elizabeth in great measure
to the same statesmen who had devised them under Mary.

Though the actual delimitation of the counties was not finally
settled until, in the reign of James I., it was accomplished by Sir
Arthur Chichester with the assistance of Sir John Davies, the
business of shiring Ireland, in the sense of formally naming and
constituting the county divisions of Connaught, Ulster, and part of
Leinster under their modem designations, was practically the work of
the two last Tudor Sovereigns. Their policy was carried out by three
statesmen of eminence — the Earl of Sussex, Sir Henry Sydney, and
Sir John Perrot. And as in the case of the final measures taken in
the reign of James I. to perfect the county system we have been
provided by the chief agent of the work^ Sir John Davies, with a vivid
description of the proceedings, so in the case of the earlier and
tentative steps taken under Elizabeth, we have the advantage of an
authentic narrative by one of the principal actors. The part played
by the Earl of Sussex has just been noticed. Sussex was followed by
the gifted and valiant Sir Henry Sydney. Not only has that ablest
of Elizabethan Deputies left detailed accounts of his progress through
the provinces, but he has given in a memoir of his services in Ireland,
drawn up in 1583, a striking statement of the Irish policy of Elizabeth

> 3 ft 4 Philip and Mary, Cap. III.

' The preamble to both Statutes is worth quoting as showing the principle on
which this policy of shiring was based :— " Whereas divers and sundry robheries,
murders, felonies, and other heinous offences be daily committed and done within
the sundry countries, territories, cantreds, towns, and villages of this realm being
no shire ground, to the great loss both of the Queen property and of divers and
sundry her Highness true subjects of this realm, and to the bddening and
encouraging of many offenders. Be it enacted," &c.

184 Proceedings of the Royal Irish Academy.

in the first half of her reign, and a full summary of the proceedings
taken by him to reduce the backwoods of Ireland to shire ground.
The circumstances in which this memoir was written add to its
intrinsic value the piquancy of an interesting historical association.
Por the occasion of the narrative was the then approaching manii^
of the writer's son, Sir Philip Sidney, the chivalrous author of the
" Arcadia," to the daughter of Sir Francis Walsingham, a lady whose
fate it was to be successively the wife of Philip Sidney, of Bobert
Devereuz, the unfortunate Earl of Essex, and of the third Earl of
danricarde. The memoir was written primarily as an apology for
Sydney's inability to make a sufficient settlement on his son. Sir
Henry explained how his expenses as the representative of the Queei
in Ireland, and the neglect of the Sovereign to relieve his impoverished
fortune, had reduced him to a position of "bitiug necessity," which
prevented him make such provision as he desired for his mnch-loved
son. "Three times,*' wrote Sydney to "Walsingham, "her Majesty hath
sent me her Deputy into Ireland, and in every of the three times I
sustained a great and violent rebellion, every one of which I snbdued,
and with honourable peace left the country in quiet. I returned from
each of those deputations three thousand pounds worse than I went"^
Sydney's contribution to the formation of the Irish counties
consisted in the main in the shiring of Connaught. In 1566, in the
'first of his three Viceroy alties, he took the first step in this undff-
taking by providing efficient and permanent means of communication
between Dublin and the western province. " I gave order," he writes,
" for the making of the bridge of Athlone, which I finished, a piece
found serviceable ; I am sure durable it is, and I think memorable.**
A few years later a bridge over the Suck at Ballinasloe, " being in the
common passage to Galway,'* was constructed by Sir Nicholas Malby
at Sydney's direction. This was the necessary preliminary to say
effective assertion of English law in the remoter parts of the country.
It was followed by the division of Connaught into four of the five
counties of which it now consists, viz : — SHgo, Mayo, Qalway, and
Roscommon, with the addition of Glare. In his '' orders to be observed
by Sir Nicholas Malby for the better government of the provinee of
Connaught," issued in 1579, Sydney's reasons for this arrangement are
thus given: — ''Also, we think it convenient that Connaught be
restored to the ancient bounds, and that the Government thereof be

1 The accoimtB of Sydney's provincial journeys have been printed in the Ulrter
Archoeological Society's Joomal, vol. iii., et ttq.

Falkinbr— 2%tf Counties of Ireland. 185

under you, especially all the lands of Gonnaught and Thomond, being
within the waters of Shannon, Lough Bee, and Lough Enie." In
the same document suggestions are made for the appointment of ''safe
places for the keeping of the Assizes and Cessions." Sligo, fiures
(Burris hoole), Koscommon, and Ballinasloe, are respectively designated
as suitable county towns.^

Leitrim for the present was excluded. O'Eorke's country was
not reduced to a county until Ferret's time in 1583. But the
country of the O'Ferralls, called the Annaly, and the territory of the
O'Beillys, or East Breny, both of which, as already noted, were then
reckoned in Gonnaught, were formed into the modem counties of
Longford and Cavan.' East Breny was described at the time by Sir
N. Bagnal as **a territory where never writ was current," and
which it was almost sacrilege for any Governor of Ireland to look
into. The precise allotment of these counties among the provinces
seems to have been left open, for Sydney, as will appear in a moment,
was solicitous lest Gonnaught, which he had already extended in
another direction, should become disproportionately large.

The district of Thomond had always been reckoned a part of the
southern province. Indeed, the name signified North Munster, and
its people were a Munster people. But Munster was a troublesome
responsibility in Sydney's time ; and the Deputy, who was then form-
ing the system of Presidencies by which for the next seventy years
the provinces of Munster and Gonnaught were to be administered,
desired to reduce its importance.' He therefore ignored this ancient
division, and taking the Shannon as a natural boundary (the province,
if we exclude Leitrim, being thus, as the author of the '' Description of
Ireland " has it, '^ in manner an island"), he added this large territory
to Gonnaught. '^ Thomond, a limb of Munster, I annexed to the Pre-
sident of Gonnaught by the name of the Gounty of Glare," is Sydney's
concise summary of this important transaction.* In his instructions
to Malby, already quoted, the north part of the city of Limerick was
suggested as the " shire town," '' because a jury may be had there
for the orderly trial of all country causes." But the President was

> See 0*Flalierty*8 ** West Gonnaught," ed. Hardiman, p. 305.

* Sunez appears to have designed to add Cavan to Leinster rather than Ulster,
*<0'Beilly," he writes, "bordering upon Ifeath, and lying by situation of his
country imfit for any of the other OoTemments, is to be under the order of the
principal goyemor." Carew Calendar, L, 338.

> * ' Beasons for retaining Thomond in Gonnaught.* ' Caiew Calendar, iv., p. 47 1 .
4 Collins's Sydney Papers, i., 75.

B.I.A. PBOC, TOL. XXXY., 810. c] [14]

186 Proceedings of the Royal Irish Academy.

directed to choose some apt place in Thomond ; and Qnin, Eillaloe, and
Ennis were suggested as suitable.

We may pause at this point to consider the subsequent administn-
tive history of Thomond. It continued to be included, under its neir
designation of Glare, in the government of Connaught almost to the
end of Elizabeth's reign. It was then erected into an entirely distinct
division, and governed as a distinct entity under a separate Commis-
sion, by Douagh, Henry, and Bamaby, successive Earls of Thomond.'
In 1639, however, under Strafford's Government, it was arranged
that on the death of the last-mentioned earls the tenitoxy should be
reannexed to Munster; and though the ensuing disturbances delayed
the fulfilment of this intention, the County of Clare was finally
reunited to Munster at the Bestoration.

But to revert to Sir Henry Sydney. If he was successful in his
operations in the distant provinces of Connaught, he was less fortunate,
not only in the north, where, indeed, the conditions were hardly ripe
for such work, but in a district much nearer to the seat of his Govern-
ment. It is certain that the County of Dublin was originally much
larger than its present area indicates ; and it appears probable that it
anciently extended from Skerries, in the north, to Arklow, in the
south. It had been conterminous, in fact, as has been pointed out,
with the ancient Scandinavian kingdom of Dublin — a territory still
marked for us by the ecclesiastical division of the United Dioceses
of Dublin and Glendalough.' But the Danish rulers of Dublio
troubled themselves little about the interior of the country,' and it is
doubtful whether at any time prior to Henry YIII. the wild septs of
the Byrnes and Tooles, whose incursions in the neighbourhood of the
city Stanyhurst describes so graphically, had given even a nominal
recognition to the Norman or English power. In the thirty-fourth
year of that monarch's reign they are said to have petitioned the Lofd
Deputy and Council to make their county shire ground, and to call it
the County of Wicklow, but nothing came of the proposal.* Be tiiatas
it may, the sway of these Wicklow chieftains was exercised without
dispute down to Sydney's day right up to the near neighbourhood of
Dublin, and the inhabitants were ever, as Davies observes, '* thorns
in the side of the Pale." Indeed, it may be said that the whole

^ Libur Munerum Hibemice, Part II., p. 185.

' Haliday*8 ** Scandinavian Kingdom of Dublin," pp. 139 and il6.

s Stokes's ** Ireland and the Celtic Church," p. 277.

* Book of Howtb, p. 464.

Jb\LKiNEK — The CoiuiticH o/Irelfind, l87

countrj Bouth-west of Dublin, including large portions of Kildare,
Carlow, and Wexford, as well as the modem Wicklow, long remained
a rude "hinterland'' into which law and order seldom penetrated.
The State Papers are full of such entries as this of 1537 — *' Devices
for the ordering of the Kayanaghes, the Bjmes, Tooles, and O'Mayles
for such lands as they shall have within the County of Carlow and
the marches of the same county, and also of the marches of the County
Dublin, '' — which plainly show the unsettled state of these districts. In
1578, however, a Commission issued under the Act of 11th Elizabeth
and "the Bims' and Tooles' country, with the Glens that lie by
South and by East of the County of Dublin, was bounded out into a
shire, to be named and called the County of Wicklow."^ But though
this Commission was carried out, and the boundaries of the counties
defined by Sir William Drury, who succeeded Sydney as a Lord
Justice, the troubles of Elizabeth's latter years in Munster and Ulster
left little leisure to her Deputies to attend to the Wicklow septs.
The Byrnes and Tooles resumed their independence ; and in 1590, as
Sir George Carew wrote, " those that dwell within sight of the smoke
of Dublin" were not subject to the laws.' When Sir Arthur
Chichester came to complete the work Sydney hkd begun a genera-
tion earlier, of " adding or reducing to a county certain, every border-
ing territory whereof doubt was made in what county the same should
lie,'" he found that the mountains and glens of Dublin were almost
as far as ever from " civility," and contained such a multitude
of untutored natives that it seemed strange that "so many souls
should be nourished in these wild and barren mountains." The
flhiring of Wicklow was only finally accomplished in 1606, and it thus
fell out that the county nearest to the metropolis was of all the last
to be brought effectively within the scope of English government.

In connexion with this attempt towards the formation of the
County Wicklow, Sydney had also a project for dividing Wexford into
two shires, of which the northern part should be called Ferns. This
county, severed by the Wicklow mountains from the metropolis, had,
though less disturbed than its neighbours, been practically outside the
Pale.* The southern part of it, indeed, according to a "Description of

1 Fitnt of Elixabeth, No. 3,603, Iriah Becord Offico.
' Carew Cal., iii., p. 44.

* Sir J. Davies'* " Discovery."

* See Hore and Graves's <* Sociid State of the South-Eastero Counties in the
Sixteenth Century," p. 27.

188 Proceedings of the Royal Irish Academy,

the Provinces of Ireland," written about the year 1580, was " civil,"
'* that part contained within a river called Pill " (a name given to the
estuary of the Bannow) being inhabited by "the ancientest gentleman
descended of the first conquerors." But this district waa connected
with the capital by sea only, and the rest of the county was inacces-
sible. Sydney and Sir William Drury finding "that there were no
sufficient and sure gentlemen to be sheriffs, nor freeholders to make a
jury, for her Majesty," the project was let drop. Their successor,
Sir John Perrot, had the same object in view, and in a report to
Elizabeth, "how the natives of Ireland might with least charge be
reclaimed from barbarism to a godly Qovemment,"^ he gives a
picturesque account of the condition of the south-eastern counties and
the need which existed for providing a proper system of administration^
** The Birnes, Tooles, and Kavanaghs must be reduced." They are
" ready firebrands of rebellion to the O'Moores and O'Conors, and till
they be brought under or extirped, Dublin, Kildare, Meath, West-
meath, and the King's and Queen's County cannot be dear either of
them or of O'Moores or O'Conors, or of the incursions and spoils of the
McGeoghegans, O'Molloys, and other Irish borderers." But though
he stated the difficulty thus vigorously, Perrot, like Sydney, left
Ireland without doing anything effective to remedy it. Sir Henry
Sydney's last tenure of the office of Lord Deputy closed in 1578, and
for the next few years the Desmond rebellion perforce put a stop to
the work he had set himself to accomplish. It was not until the
southern rising had been crushed that Sir John Perrot, who, in 1584,
succeeded to the Irish Government, was able to resume the work.
Though this statesman is best remembered in our history in connexion
with the composition of Connaught, which was effected during his
administration, it is in relation to Ulster that his proceedings have
most interest in the present connexion. To Perrot belongs the
honour of having divided the northern province into divisions sub-
stantially corresponding to its modem counties, though twenty years
were to elapse before these divisions were generally recognised, or
before they became effective portions of the administrative machinery
of the country.

The story of the Anglo-Norman colonies of Ulster and the settle-
ment of Lecale, the Ards, and Carrickfergus, has never been fully
analysed, and to tell it is outside the purpose of this Paper. Hero it
must suffice to observe that the only counties in the modem sense of

^Sloane MS., 2,200, Brit. Mus.

t^ALKitJER— JAc Counties of Ireland. 189

the term which can be recognised as existing in Ulster before the
time of Elizabeth were Louth, which, as already noted, was anciently
accounted part of that province, and the counties of Antrim and
Down. The precise date at which the two last were constituted is
unknown ; but it appears by the ** Black Book of Christ Church" that
they, or at least certain districts bearing these names, had existed prior
to the reign of Edward II. From that time down to the settlement in
Antrim of the McDonnells of the Isles, under Henry VIII., little is
known of them ; but the two counties had been recognised as settled
districts by Perrot's time, and as such were distinguished by that
Deputy from the ** unreformed " parts of Ulster. In 1675 Sir Henry
Sydney had made a journey to Ulster with a view to dividing the
province into shires, but had failed to effect anything — an effort which
was referred to by Sir John Davies in his address as Speaker of the
Irish Parliament in 1613; when, congratulating the Commons on the
completeness of its representation, he observed, ** How glad would Sir
Henry Sydney have been to see this day, he that so much desired to
reform Ulster, but never could perfectly perform it."

Perrot's contribution to the shiring of Ulster was little more than
a settlement on paper of the boundaries of the new counties he desired
to create. It is best described in the language of Sir John Davies : —
" After him [Sydney] Sir John Perrot . . . reduced the unreformed
parts of Ulster into seven shires, namely, Armagh, Monaghan, Tyrone,
Coleraine, Donegal, Fermanagh, and Cavan, though in his time the
law was never executed in these new counties by any Sheriff or
Justices of Assize ; but the people left to be ruled still by their own
barbarous lords and laws." Perrot's work was of course interrupted,
and for the time rendered nugatory, by the rising of Hugh O'Neill ;
but it was so far effective that his division became the basis of the
subsequent allocation of the northern territories, which a few years
later followed the Flight of the Earls and the Plantation of Ulster.

Had affairs in England permitted the Government to bestow steady
and continuous attention on the affairs of Ireland, it is probable that
the work initiated by Sussex and Sydney, and so largely extended by
Sir John Perrot, would have been completed before the close of Eliza-
beth's reign. But Perrot was recalled in 1588, and the business of
shiring Ireland was arrested for nearly twenty years. With O'Neill
taking full advantage of the difficulties in which England was involved
by the struggle with Spain, and asserting his power effectively
throughout Ulster, the sub-division of the northern province remained
purely nominal, and even in the more settled districts much confusion
reigned. The result is seen in the discrepancies which appear between

l90 Proceedings of the Royal Irish Academy.

the various accounts which remnin to us of the division of Ireland at
this time. These exhibit considerable confusion, not only as to the
counties of which each province was made up, but even as to the pro-
vinces themselves. Thus Haynes, in his ** Description of Ireland,"
in 1598, states that Ireland is divided into five parts. He inclndes
Meath among the provinces, mentioning it as containing four counties,
viz.. East Meath, Westmeath, Longford, and Cavan, though he adds
that the last is by some ** esteemed part of Ulster.'* On the other
hand, in a survey printed in the Carew Calendar,* revised to the
year 1602, Longford is included in Connaught, while Cavan is not men-
tioned, and the completeness of the relapse of ITlster from " civility "
is shown by the description of that province as containing i^iree
counties and four *' Seignories."

Thus it was not until after the accession of James L, in the time
of Sir Arthur Chichester, that, in the words of Sir John Davies, " the
whole realm being divided into shires, every bordering territory
whereof doubt was made in what county the same should lie was
added or reduced to a county certain." The boundaries of the counties
forming the provinces of Connaught and Ulster were ascertained one
after another by a series of Inquisitions between the years 1606 and
1610, which confirmed in the main the arrangements tentatively made
by Perrot, though in the case of Ulster these were necessanly varied
in some important respects, particularly as regards Londonderry, by
the changes resulting from the Flight of the Earls and the Plantation
of the northern province. The enumeration of counties and provinces
in Speed's ** Description of the Kingdom of Ireland," in 1610, shows,
as already noted, that in that year the precise allocation of counties
among the provinces still remained vague and indeterminate in the
popular estimation. But Meath had by that time finally disappeared
from the list of provinces ; and though some years were to elapse ere
all the counties could be finally delimited, this process had been
practically completed when Sir John Davies left Ireland in 1616,
except in the case of Tipperary, where the exceptional conditions
created by the existence of the Ormond Palatinate long retarded the
final settlement.

Although Munster is of all the great divisions that which, if com-
pared with the original distribution imputed to King John, shows the
least alteration in its county system, the southern province has not
been without its vicissitudes in tiiis respect. In Perrot's time Munster
consisted of as many as eight counties, and the final settlement of

^ Carew Calendar, iv., pp. 446-454.

Fai.kinbr — The Counties of Ireland. 191

the Bix counties now embraced in it was, in fact, delayed tintil after
the other proyinces had assumed their present form. The shiiing
of Munster was effected chiefly through the instrumentality of the
provincial government known as the Presidency of Munster, which
was established by Sydney in 1570. No single act of Elizabethan
policy had more important or more satisfactory results than the insti-
tution of the Presidencies of Munster and Connaught; and as the
gradual demarcation of the counties of both proyinces as they now
exist was largely effected by their means, it seems desirable to give a
brief account of an institution which was devised by Sydney, as
Davies puts it, '* to inure and acquaint the people of Munster and
Connaught again with English Government."

The first idea of these instruments of administration was formed in
the time of Edward YI., when a scheme was devised for the appoint-
ment of separate Presidents for each of the thr.ee provinces of Munster,
Connaught, and Ulster. But although Sussex had a clearly defined
scheme for giving effect to this policy, it was not until Sir Henry
Sydney's first administration that, in 1565, definite shape was given
to it, or that the constitution of what for the next century were
known as the Presidency Courts of Connaught and Munster was formally
drafted. The Presidency not only included a President answerable to
the Lord Deputy, but a Council composed of prelates and nobles of
the province, and a Chief Justice vrith two Justices and an Attorney-
General, together with a Treasurer, Clerk of the Council, and other
administrative officers. In 1568 Sir John Pollard was nominated first
President of Munster, and in the year following Sir Edward Fitton
became President of Connaught. No President was appointed for
Ulster, the charge of which was confided, under a temporary Com-
mission, to a marshal, an officer whose duties were half-civil, half-
military. Pollard, however, never entered on his Government, and the
first acting President of Munster was Sir John Perrot, who, appointed
in 1570, was for six years a strenuous representative of the Crown
in that province.

It is a matter of great regret that the records of these Presidencies
have long since perished.^ They seem to have been lost in the

> See Pnndergast*a « Introduction to Cal. S. P. Ireland," James I., 1606-
1608, pp. zz.-zzzv. A volume called *'The Council Book of Munster"
BuiTivee in the Harleian Collection at tbe BritlBh MuMum (Harl. Col., No. 697) ;
but it only extends from 1601 to 1617. The " Infltructions for the Lord President
and Counoil of Munster," in 1616, have been printed in " Desiderata Curioea
Hibemica," vol. ii.

192 Proceedings of the Royal Irish Academy.

troubled times succeeding the rebellion of 1641, and the Presidential
institution itself did not long survive that cataclysm. Though Uiev
lingered beyond the llestoration, the Presidencies were not regarded
by the Duke of Ormond as necessary or efficient instruments of
government ; and in 1672, during the Viceroyalty of Lord Essex, they
were finally abolished. But though the Presidency system was not
destined to remain a permanent feature in the administrative system
of Ireland, its operation during the years first following its instita-
tion was unquestionably effective. In Perrot's hands, both as
President of Munster, and later when as Deputy he became
responsible for the whole country, it was largely utilized to effect
what was practically a fresh delimitation of the old counties of
Munster. In an old **note," probably dating back to the fifteenth
century, quoted by Perrot in his Report to Elizabeth, already cited,
the Munster counties are thus enumerated : ''In Munster there be
five English shires — Cork, Limerick, Waterford, Kerry, Tipperary ;
and three Irish shires — Desmond, Ormond, and Thomond." It will
be noted that the five former of these counties with Thomond or
Clare nominally make up the modem province of Munster. Ormond
represents Tipperary less the County of Cross Tipperary, and as
such still possesses a well-defined meaning. Desmond is a district
perhaps less clearly defined in the popular mind. It embraced a
large portion of East Kerry and West Cork, and at one time was
actually erected into a separate county. In 1571 a Commission
issued to Sir John Perrot and others, under the Statute 11 Eliz.,* for
the counties of Waterford, Tipperary, Cork, Limerick, and Kerry,
and the countries of Desmond, Bantry, and Carbery, and all countries
south of the Shannon in Munster, to make the country of Desmond
one county, and to divide the rest into such counties as may be
convenient." As a result of this Commission, Desmond became and
was long regarded as a distinct county, and its boundaries appear
from an Inquisition of 1606. But though Fynes Moiyson places
Desmond on the list of the Munster counties, stating it to have
been lately added, its separate identity is not invariably recognised,
though for a time it boasted that essential note of independence, a
separate sheriff. This, however, had disappeared before the close of
Elizabeth's reign, for Haynes writes in his account of Cork that that
county,* ** being the greatest in the realm, have been tolerated to have

i Fiant, Eliz., 1486. Irish Becord Office.

> *< The Description of Ireland in 1698," ed. by Rev. Edmund Ho^ s-^-i
p. 169.

Falkinek — The Counties of Ireland, 193

two sheriffB — ^tbe one particular in Desmond, the other in the rest of
the county — and this without any ground of law, but by discretion
of the L. Deputies ; the inconyenience thereof being espied, it had
been of late thought good that one sherifp should be for Kerry and
Desmond, and so two sheriffs in one county against law taken away.'
The amalgamation with Kerry appears to have been completed by
1606/ when Mr. Justice Walshe, in describing to Salisbury the
Munster Circuit of that year, mentions particularly the successful
union of Desmond and Kerry.

The dual representation of Tipperaiy in the list of Irish counties
was long a puzzle to antiquaries, and even an inquirer so diligent
and in general so accurate as Sir John Davies was misinformed on
the subject, notwithstanding the minute inquiries he appears to
hare instituted into the origin of what struck him as a curious
administratire anomaly. '' At Gashel," he writes in his account of
the Munster Circuit of 1606,' '< we held the Sessions for the County
" of the Cross. It hath been ancientiy called ' the Cross ' (for it had
" been a county above 300 years ; and was, indeed, one of the first that
" ever was made in this Idngdom) because all the lands within the
^'precincts thereof were either the demesnes of the Archbishop of
" Cashel, or holden of that See, or else belonging to Abbeys or houses
'' of religion, and so the land as it were dedicated to the Cross of Christ.
'' The scope or latitude of this county, though it were never great, yet
'' now is drawn into so narrow a compass that it doth not deserve the
" name of shire."

Davies' confusion as to the two counties of Tipperary, which con-
tinued to be separately represented down to Strafford's Parliament of
1634, was extremely natural in view of the limited information avail-
able when he thus accounted for the anomalous existence of the
County of Cross Tipperary. But, in fact, the duplication had really
originated in the Palatine system. To the accident which preserved
Tipperary as the last of the Palatinates was due the survival of Cross
Tipperary as the last of the counties of the Cross; and it will be
convenient here to trace the history of both jurisdictions. The
County Palatine of Tipperary was originally created by letters patent,
granted in 1 328 by Edward III. to James le Botiller, Earl of Ormond,
and confirmed by successive monarchs to that nobleman's successors
in the honours of the Butler family. The jurisdiction thus granted

> Cal. of '< State Fapen,'* Ireland, 1603-6, p. 673.
» Cal. of " State Papers," Ireland, 1606-8.

B.I. A. PROC, VOL. XXXT., SBC. c] [15]

194 Proceedings of the Ropal IriA Academy.

embraced tibe whole County of Tippersry, with the ezoeptumaf eeitsn
Ohturch lands, which constitiited, as was usual with Churdi land in
Palatine counties, a distinct shrievnlty under the ordinary jarisdictLon
of the King's Courts. In addition to these districts of the GrosB, there
was also excepted from the Palatine grant the district of Dough Am,
or MacBrien's country, adjacent to Eillaloe, which, long a debatable
land on the borders of ^e three counties ol Clare, limerick, and
Tipperary, was in 1606 joined by Chichester to liie County of the C^ofls
of Tipperary.

In 1621, during the wnidahip of the dan^^ter and heuesB of
Thomas, tenth Eari ol Oraumd, the Palatinate ol Tipperary was seized
into the Crown by James I. The County of the Cross apparently
remained unaffected by this exertion of the Boyal pvwogatiTOt and, as
already noted, it was represented in the Pariiament of 1634, though
the county proper appean to have returned no memben to that
assembly. The Palatinate remained in abeyance lor a period of foity
years, till alter the Bestoration it was reconstituted by QiarleB II.
in 1664, in favour of the first Duke of Omumd. The gnmt on
this occasion included both the old teiritory f^ the Cross, which never
thereafter returned memben to Pariiament, and the district of Doag^
Am, formerly excepted from the Palatine county. The liberties and
royalties ol the whole County of Tipperary were enjoyed by the
Butlers until the attainder in 1715 of the second Duke put an end
to the last Irish exan^le al these great medisofval juiiadictioofi.^
The Statute 2nd George I., cap. 8, *' an Act for extinguishing the
royalties and liberties of the County of Tipperary," by its second
sectiim enacted, '* that whatsoeyer hath been denominate or called
Tipperary or Cross Tipperary, shall henceforth be and remain one
county for ever, under tiie name of the County of Tipperary."

[No attempt is made hare to discuss the origin of the names of the Iiiah
counties. This may fonu the subject of a sepaiate inquiry.

The writer desires to express his obligations to the courteous officials of the
Irish Record Office, and especially to the Assistant Deputy Keeper, Mr. H. F.
Berry, m.r.i.a. He has also to thank Mr. Temson OroTes, c.e., for many useful
snggestioos. — C. L. F.]

1 See 6th Beport of the Deputy Keeper of PubUc Records of Iiehmd, p. 7, and
Appendix III, pp. 33-38.

[ 195 ]

XIL

NOTES ON THE ORIENTATIONS AND CERTAIN ARCHI-
TECTURAL DETAILS OF THE OLD CHURCHES OF
DALKEY TOWN AND DALKEY ISLAND.

Br JOSEPH P. O'REILLY, C.E,

[Plates Xni.-XVn.]

Sead FiB&UA&T 23, 1903.

Ths diurclies of Dalkey Town and Dalkey Island are of course alluded
to, or mentioned, in the different works treating of these localities, but
generally with relatively few details ; the dates of their foundations,
as well as the names of their founders, are apparently unknown. All
that can be ascertained as to their early history is to be obtained from
the records of Christ Church Cathedral, and from those of St. Patrick's,
to the Chapters of which these churches were given over by Hugh
de Lacy, who had received them in grant from Henry II. Both
churches date, therefore, from a period anterior to the Norman
Invasion. As to the saints or saint to whom they were dedicated, or
are mentioned as having been dedicated, there have been some doubts.
Seward's "Topographia Hibemica" (1795) says of the town: "This
vUlage in the reign of Queen Elizabeth, and a great part of the last
[seventeenth] century, before the port of Dublin was improved, was the
repository of the goods belonging to the merchants of Dublin. Here
are the ruins of a few old castles, places of defence against the
incursions of the pirates who at that time swarmed on the Irish
coast."

As regards the island, the work says : " It is so called [Dalkey]
from Dalkiy on account of the pagan altar there." There is no ancient
building on Dalkey Island but the ruins of a church.

In Carlisle's " Topographical Dictionary of Ireland" (1810), it is
stated : ** Dalkey liland, — ^Here are the ruins of a church."

Lewis's "Topographical Dictionary" (1837) says, as regards the
town : " The church is in ruins; it was situated in the village, and
appears to have been a very spacious structure,"

B.I.A. PBGC, VOL, XXnr., SBC. c. [16]

196 Proceedings of the Royal Irish Academy .

As regards the island : '* There are the ruins of a chnieh dedicated
to St. Benedict ; and ' kistvaens,' or stone coffins of rude workmanship
and great antiquity, have heen found near the shore."

D' Alton's "History of the Co. Duhlin," 1888, says (p. 882):
" While in the town, are the not uninteresting remains of an ancient
church, picturesquely situated at the foot of the Bochestown Hills,
and presenting a nave fourteen yards long, hy five hroad, and a ch(»r
eight by five, divided by a well-executed arch." Page 885 : " On the
shore, in a little rocky cove, the tourist will find a ready boat to
facilitate his pilgrimage to the island, where, surrounded by cli£b and
a frequentiy tempestuous sea, an ancient mariners' chapel was erected
and dedicated to St. Begnet or Benedict." Page 886 : " On it [the
island] is a doubtful remain, said to be the patron's church ; butcertabily
having nothing of the ecclesiastical aspect, unless perhaps a plain
gable belfry ; and wholly disconsecrated, even in the traditions of the
people, by its present uses." Page 887: "In 1178 Archbishop
O'Toole assigned to Christ Church (amongst several) the church of
St. Begnet of Dalkey, with aU its tithes ; and his grant was further
assured by letters-patent from Prince John. In 1200 the Archbishop
had a grant of a Wednesday market here [in the town], and an annual
fair to be held on St. Begnet's day."

"The Parliamentary Gazetteer" (1846), speaking of the island,
says : "A small old ruin on the island is usually regarded as having
been a church dedicated to St. Benedict; but though possessing a
belfry, it exhibits very distinct marks of simple domestic or dwelling-
house structure. 'Kistvaens' enclosing human bones are said to
have been found upon the island, and are regarded as vestiges of
Celtic or Belgic tribes of a very remote era."

Mr. F. Elrington Ball, in his " History of the County Dublin,"
1902, says as regards the churches (p. 79): Dalkey hJand, — ''The
ruined church — for such undoubtedly is the structure on the northern
end of Dalkey Island — ^is coeval with, and similar in construction to,
that of the Kill of the Grange. It has a primitive doorway and
window ; and its side walls project upon the end ones, as do those of
the Kill Church, forming pilasters." "The belfry is a later addition ;
and a fire-place and enlarged doorway and window in the south wall
were made by the workmen employed in the construction of the
Martello Tower, who used it as their dwelling " (Wakeman's "Primi-
tive Churches in the County Dublin," Journal B. Soc. Antiqq. of
Ireland, vol. xxi., p, 701 ; see also vol. xxvi., p. 415). "The churdi
[on the island], which is supposed to have been dedicated to St.

O'Reilly— OW Churches of Dalkcy Town and hland. 197

Begnet, the patron saint of Dalkey, indicates, by its state of preserva-
tion, use in the middle ages; but nothing is recorded of the history
of the island from the twelfth century, when it was giTen by Hugh
de Lacy to the See of Dublin, until the seyenteenth century, when it
was destitute of inhabitants, and used for grazing cattle.*' Page 81 :
**The church [of Dalkey Town] was dedicated to St. Begnet the
Tirgin, the patron saint of Dalkey, who is supposed to haye flourished
about the seyenth century, and whose festival falls on the 12th
November. After the English Conquest [second Norman Invasion ?],
it was assigned to the priory of the Holy Trinity."

The most important paper for the purposes of this present one^ is
that of Mr. Wakeman, cited by Mr. F. E. Ball. It appeared in
vol. xzi., 1890-91, of the Joum. Key. Soc. Antiqq. of Ireland, p. 697,
the title being: *' Primitive Churches in the County Dublin," by
W. F. Wakeman, Hon. Fellow, Hon. Sec. for the County Dublin.

After some introductory remarks as to the existence of early
Christian eella around Dublin, '' some of which have not hitherto
attracted antiquarian consideration," he says : '* It is a significant
fact that while several e$lla^ teampuUs, or cills, in the Dublin district,
are as generally ancient in character as any structure of the like
class to be seen in remoter provinces of Erinn, no architectural
connecting-link between them and churches of late twelfth- or even
thirteenth-century date can be discovered." '* The primitive churches
when not utterly dismantled or razed by Northern ravage were, in all
likelihood, left in ruinous neglect; and it would seem there exists
architectural evidence that it was not until some considerable time
subsequent to the overthrow of Danish influence, or, indeed, until
the Anglo-Norman settlement had commenced, that many of our old
parish churches were once more used as places of Christian worship."
P. 698 : '' When it was deemed necessary to enlarge the church, by
the addition of a chancel (a feature very rare in our earlier team-
pulls), they broke through the eastern gable, hacking an aperture,
the edges of which were then lined, in jambs and arch, with plain or
hammered stones. The added choir or chancel was simply built up
against the original east gable, and not bounded with it." '< Sur-
mounting the western gable, at the time of transition referred to, it
was customary to erect a turret with provision for one, two, or some-
times three bells." P. 701 : ^' I now draw attention to an old church
which still stands, almost intact, upon the island of Dalkey (see PL 11.,
fig. 1). This structure has long been regarded by Dublin people as
▼ery mysterious in character. They could scarcely fancy it a church ;

[16*]

198 Proceedings of the Royal Irish Academy.

and yet in all its features it presents characteristics which unmistak-
ably point to one conclusion, viz., that the structure is neither more
nor less than a slightly modified example of our oldest style of eill or
church,"

Its form is oblong — 27 feet 7 inches in length by 20 feet 3 inches
in breadth, external measurement. The walls average 2 feet 10
inches in thickness. Pilasters or extensions of the side walls are
found on the eastern and western ends. Similar features are observed
in connexion with a considerable number of our oldest churches,
such as Teampull McDuagh, in Arranmore, County G^way; on
St. M^Dara's Church in Inis M'Dara, off the coast of Connemara; at
Dulane, near Kells, County Meath; and, indeed, in many other
places. "Here they are 2 feet 7 inches wide, and project 1 foot
2 inches beyond the gables."

'< A fine fiat-headed doorway measuring 7 feet 3 inches in height
by 2 feet 8 inches in breadth at the top, and 2 feet 9 inches at the
base, occupies a position in the centre of the west end. The lintel
in this example is peculiarly massive" (see PI. II., fig. 2). ** Above
the western gable rises a somewhat clumsily-constructed beU-turret
containing a single aperture, the head of which is in a rather late
pointed form.*' '* It is quite evident that this campanile is a compara*
tively late addition." " Its aperture would have been completely
covered by the original roof, the pitch of which is indicated by traces
of mortar or cement which still remain." "A small flat-headed
window (see fig. 8, p. 702), placed high on the south side wall,
appears to be the only original light to be found in the building."
" The structure, indeed, bears evidences of alteration at various dates ;
but the principal change, no doubt, occurred in the second or third
year of the nineteenth century, when this curious and mysterious eill
was utilized as a dwelling-place by the Gk)vemment employes
engaged in building the Martello Tower, which was intended to
command the Sound of Dalkey, and much of the neighbouring coast.
I myself, some thirty years ago [^anU 1890], when residing in the
vicinity, was well acquainted with a truly ancient mariner named
Tom Doyle, who had assisted in the work." " He stated that the
church was used as a house by himself and fellows ; and that to make
themselves comfortable, and tiie building suitable for their occasions,
they had broken a doorway and window in the southern wall, and
constructed the still existing fireplace. He stated further that when
disturbing sods or scraws to be used in roofing material, the diggers
found human bones apparently of great antiquity."

O'Reilly— Old Churches of Dalkey Totcn and Island. 199

'^ Not far from tlie church, on the brink of the Sound, is a well
which the old people consider very sacred, and highly efficacious for
the cure of sore eyes. One relic of extremely early days may he
observed carved or picked out upon the natural undisturbed rock
which stands immediately facing the western gable. It is what
Bishop Graves styles an ' eastern ' cross, enclosed by a circle, and is
probably as early as the sixth or seventh century (see fig. 4).
Within the quadrants are raised pellets. Indeed, the figure is
extremely like some found on certain of the oldest remains for which
Inismurray and some districts of Kerry are famous. It is the only
rock-marking of its interesting class which I have seen out of the
West or South of Ireland."

'' The Church of Dalkey (Island), its details, and this cross, are
here, so far as I am aware, for the first time figured and described."

Practically all that was known up to their time concerning the
churches of Dalkey Town and Dalkey Island is given in the two
citations from Mr. F. E. Ball's work and Wakeman's paper cited by him.
This latter may indeed be taken as a text for the further considera«
tion of these two remarkable ruins.

As regards the name of the saint to whom both these churches are
said to have been dedicated, I am indebted to Mr. Ball's courtesy for
the communication of the following details to be found in the Beport
of the Deputy Keeper of the Public Eecords and Keeper of the State
Papers in Ireland (1896). There is given therein an Index to the
Calendar of Christ Church deeds, 1174 to 1684, contained in Appen-
dices to the 20th, 23rd, and 24th Eeports. It gives the following
forms of the name " Dalkey, ^^ and the indications of the mentions of
that name in the extracts given in the reports : —

Dalkey, Dalkaye, Dalkeya, Dalkie, Dalky, Gilbeknith, Killekenet,
Kilbekenet, St. Begnetes, St. Begnetts.

Entries— 51, 219, 379, 381, 2, 413, 415-6, 431, 779, 927, 1145,
1303, 1306, 1341, 1346, 1374.

Dalkey Church— 52-3, 379, 431, 440, 557, 1378.

Dalkey Churchyard— 1341.

In these extracts the Church of Dalkey Town is designated as
follows : —

51— Circa 1240— Chapel of Kilbekenet.

52 — „ „ — Church of Killekenet.

5»— 16th March. 1245— Church of Kilbek[enet].

379— 17th Septr., 1504— The Church of St. Begnet of Dalkey.

557 — Circa 1320 — A messuage in the tenement of Gilbeknith,

200 Proceedings of the Rof/al Irish Academtj,

the frontage looking from the sanctuary and lying on the weetera
side of Gilheknit Church.

(in darso) Quit claim of the land of the Church of Dalkey.

In the 24th Report, 26th May, 1892, there is the entry: —

P. 161, 1302— Lessors in No. 1298, and the Yicars choral of Holy
Trinity Church leave to Shane Eennay, alias Shane McBonaghe
" inclaune " of Saint Begnete's, Co. Duhlin, fisherman, a house and
land in St. Begnete's for 41 yean. Dated 8th Jany. 1565-6 & 28Ui
Eliz. (in doTBo) Dalkey, Kilbegnet.

P. 77, 1374— Lessors in No. 1298, lease to John Dongane, second
Remembrancer of the Irish Exchequer, a moiety of a messuage,
castle, orchard, and land, in St. Begnett's, alias Dalkey, Co. Dublin,
for 61 years. Dated 20th March, 1585-6 & 28th Eliz.

There will thus be remarked the great variation in the form of
the name applied to the church, as regards the name of the saint, and
the further fact that, according to the entry 1874, " St. Begnett's"
was at one time an dliter name for Dalkey Town.

As regards the St. Begnet to whom the churches are said to be
dedicated, and whose festival is mentioned as occurring on the 12ik
Nov., there is no such saint mentioned in Butler*s ''Lives of the
Saints," and none such under the date 12th November. The name
« Begnet " appears to be the diminutive form of Beg or Bee ; and the
question arises what particular saint of that name is thereby referred
to. In this respect the following citation from the '' Book of Obits
and Martyrology of the Cathedral Church of the Holy Trinity," with
introduction by Jas. H Todd, d.d. (1844), is of interest. P. xiv.
Nov. 12th, " St. Begneta or Begnait is not mentioned in the Martyr-
ology of JSngus. In the calendars of two ancient manuscript
breviaries, now in the Library of Trinity College, Dublin, she is styled
* Virgo non-martgr.^ One of these (B. 1, 3) belonged to the Church
of Clondalkin ; tfie other (B. 1, 4) to the Parish Church of St. John
the Evangelist, Dublin. The ancient church, now in ruins, on the
Island of Dalkey, near Dublin, is dedicated to St. Begnet; although,
in Lewis's ' Topographical Dictionary/ it is erroneously said to
have been dedicated to 8t, Benedict, Mr. D'Alton also, in his
' History of the County Dublin,' improves upon tMs mistake. He
•ays (p. 885) : * On the shore, in a little rock-cove, the tourist will
find a ready boat to facilitate his pilgrimage to the island, where,,
surrounded by cliffs, and a frequently tempestuous sea, an ancient
planner's chapel was erected and dedicated to St. Begnet or Benedict."

O'IIbilly— OW C/iurches of Dalkey Town and Island. 201

" Does Mr. D'Alton mean to say that * Begnet ' and ' Benedict '
were one and the same P In Alan's Register (folio 9 b) there is an
exemplification of an Act of Parliament held in Dublin on the Friday
next after the feast of St. Luke the EvangeHst, 22nd Edward IV.,
where it is enacted in favour of the Archbishop of Dublin : ^ Ordejne
est, et establie par auctorite du dit Parlement, que le dit Erchevasque
poet auer un marchee al dit ville de Dalkey annuelement, chescun
maresdye per ane, de Sepmaine en Sepmaine, et un jour de faire
Cestassauere le jour de Seyncte Begnet la Yirgine, continnuaunt III.
jours annuelment,' &c."

Starting from the statement that the St. Begnet in question was
a virgin, and presuming that the termination of the name is a dimi-
nutivey it may be asked what was the original form of the name of
the saint. The simplest would be Beca or Bega, and such a name is
found in Smith and Wace's '^ Dictionary of Christian Biography'*
(1877). Thus it gives (p. 300) Beega or JSeg^a, daughter of Gabhran,
virgin ; her festival on the 10th February. It is said (Colgan, '' Tr.
Thaum.," 121) that when St. Patrick was in East Meath, he left at the
Church of Techlaisran, in that county, two of his disciples, Be^a, a
virgin, and Lugaidh, a priest (Ap. 17th), probably brother and sister,
the children of Oauran, the latter (place) having the name of Feart-
Bige or JBs^a's Tomb. The same Dictionary gives (p. 304) the follow-
ing:— '*Bega, Beza, Beya, Begga, Bee, St. A Cumbrian saint of
whom nothing is clearly known, and whom the endeavours of the
bagiographers have only succeeded in investing with a history that
belongs to several other saints. According to Alban Butler, she was
an Irish saint (September 6th) and virgin who lived as an anchoret
in the seventh century, and founded a nunnery in Copeland. He also
mentions a place in Scotland called Kilbees after her. This is the
most reasonable account. According to the life of her, seen by Leland
(coll. iii., 36), after founding her monastery in Cumberland, she
removed into Northumberland and founded another north of the Wear ;
then to Hert, where she becomes identical with St. Heiu (Hmu), and
then to Tedcaster, winding up her career at Hackness, as identical
with St. Begu (Mon. Ang. iii. 676)."

** Begu and Heiu are well known from Bede, and were two different
persons, neither of them possibly identical with the Cumbrian saint.
Yet Suysken, in his conmientary on St. Bega (AA. SS. BoU., Sept. 2,
684-700), accepts this version as true. In default of an English

reer for the saint, she is next sought in Ireland and Scotland, and
the Aberdeen Breviary contains lessqns of two saints with either pf

202 Proceedings of the Royal Irish Academy.

whom she might be identified — (1) St. Bega, venerated at Donbar,
who lived in an island called Cumbria in the Ocean Sea as an anchoret,
visited occasionallj by St. Maura, and dying on September 3id, was
buried in her island, whence the Hector of Dunbfff, attempting to
fetch her remains, was driven back by a storm ; (2) St. Begga, an
Irish princess, given in marriage by her parents against her will, hears
of the Gospel as preached in England, flies to England to Oswald and
Aidan, and becomes the first abbess of nuns in England. She has her
home in a desert island, and, in her old age, resigns her abbey to
St. Hilda, under whose rule she ends her days (October Slst). After
460 years her remains were removed to Whitby (Brev. Aberd. pan
^stiv., fo. 145 and 136). Here are probably some reminiscences of
St. Heiu. She was probably a local saint of the eighth century. The
monastery bearing her name was founded as a cell to St. Mary's at
York in the reign of Henry I." This same Dictionary also mentioiu
'' St. Begha, Virgin, oirea a.d. 660, also called St. Bex and St BegagL
She left her home in Ireland on hearing of the flourishing state of
Christianity in Britain, and, in order to avoid a marriage intended for
her, fled into Scotland in a ship that was in waiting. She receired
the veil at the hands of Bishop Aidan in the reign of King Oswald in
Britannia, and ruled a community in a cell constructed by him in a
certain desert island. When St. Hilda returned from Gaul (Bede,
Eccles. Hist., iv., c. 23), St. Begha prayed that she might be freed
from the burden of government, and that St. Hilda might be conse-
crated Abbess in her stead, and this was accordingly done. After
many years she died in the odour of sanctity, attested by many
miracles at her tomb (Brev. Aberd. pars iEstiv. f. c. xxxvi). Bede
mentions a nun called Begu, in the monastery of Hacanos, thirteen
miles from Whitby, to whom the death of St. Hilda was revealed in a
vision (Eccl. Hist., iv., c. 23). St. Begha is honoured at Kilbagie
and Kilbucho in Scotland ; but her greatest foundation was within
the kingdom of Strathclyde at St. Bees/^which takes its designation
from her. It was founded in A.n. 656."

P. 305. <* There was a cell of this house at Nendrum or Mahee
Island in Down County (see "Description of Nendrum," by Rev.
William Reeves, d.d., 1845), and his Eccl. Antiq., 163, 190-199 for
the grant of the Island of Nedrum, or Nendrum, by Sir John de
Courcy in 1178 to the Priory of St. Bega do Copeland'' (Bishop
Forbes, '* Kal. of Scotch Saints," pp. 248-52).

In the " Proceedings of the Eoyal Irish Academy," voL viii.,
p. 258, there is a Paper by Dr. Wm. Bell (read by Dr. Beeves) on

O'Reilly— OW Churches ofDalkey Town and Island. 203

^' The so-called Eing Money in reference to many specimens in the
possession of the Eight Hon. the Earl of Londesborough, and more
•especially an Irish one with a movable Swivel Eing " (read Monday,
December 8th, 1862). In it, it is stated : '< St. Bega was the patroness
of St. Bees in Cumberland, iohere she Uft a holy hraeeUt, which was
long an object of profound veneration." A small collection of her
miracles, written in the twelfth century, is extant, and has been
published. In the prefatory statement of the compiler, we learn,
among other things, ''that whosoever foreswore himself upon her
bracelet swiftly incurred the heaviest punishment of peijury or a speedy
death." [May there not be some possible relation between the Greek
Cross on the rock in front of the church on Balkey Island and this
swearing on St. Bega's bracelet ?]

In Butler's " Lives of the Saints," under September 6th, St. Bega,
or Bees, V., it is said : " She was a holy virgin, who flourished about the
middle of the seventh century, led an anchoritical life, and afterwards
founded a nunnery in Copeland near Carlisle. Her shrine was kept
there after her death, and became famous for pilgrims. There is in
Scotland a place called Kilbees from her name, according to a note
of Thomas Innis on the Manuscript Calendar kept in the Scotch
College of Paris." (See Alford Annal., t. 2, p. 294. Monasticon
Angles. Suysken, t. 2, September, p. 694. Note : ^* She is honoured on
the f^i^nd Nwemher under the name of St. Bees.")

It may not be out of place to cite from Montalembert's ''Monks of the
West," vol. v., p. 247, where he speaks of her : " She was, according
to the legend, the daughter of an Irish King, the most beautiful woman
in the country, and already asked in marriage by the son of the King of
Norway. But she had vowed herself, from her tenderest infancy, to
the spouse of virgins, and had received from an angel, as a seal of her
'Celestial betrothal, a bracelet marked with the sign of the cross. She
escaped alone with nothing but her bracelet which the angel had
given her, threw herself into a skiff, and landed on the opposite shore
in Northumbria, where she lived long in a cell in the midst of the
woods. Fear of the pirates, who infested these coasts, led her after a
while further inland. What became of her? Here the confusion,
which is so general in the debatable ground between legend and
history, becomes nearly inextricable. (P. 250) What is certain,
however, is that a virgin of the name of Bega figures among the
.most well-known and long venerated saints of the north-west of
England. In the twelfth century, the famous bracelet which the
angel had given her was regarded with tender veneration ; the pious

204 Proceeding% of the Royal Irish Academy.

confidence of the faithful turned it into a relic, upon which usurpers,
prevaricators, and oppressors, against whom there existed no other
defence, were made to swear, with the certainty that a perjury
committed on so dear and sacred a pledge would not pass unpimished.
It was also to Bega and the hracelet that the cultivators of the soil
had recourse against new and unjust taxes with which their Lords^
burdened them."

In the ** Imperial Dictionary of Universal Biography" (1865)
there is a notice of her hy John P. Waller, LL.n., m^.i.a., Hon. Sec,
B. D. S., as follows: ^^ Bega, Saint, a native of Ireland, according
to Butler ; hut Dempter asserts that she was bom in Scotland, misled
probably by the earlier writers on hagiology, who are accustomed to
call Ireland ^ Scotia.' Be this as it may, she was a virgin of great
sanctity, and spent her life in retirement and devotion in Carlisle,
where she died in the latter half of the seventh century. A religious
house was established in her honour, and the 7th of September is
observed in memory of her. — J. F. W."

From the whole of these citations it may be concluded that, so far
as ascertainable, there is a tradition that the two churches were
dedicated to a St. Bega, or St. Begnet, '^ a virgin but not a martyr,"
whose festival is stated to have been celebrated on the 12th November.
It is equally clear that more than one saint and virgin bearing the
name of Begha, Bega, or Begge is mentioned in the ancient records
bearing on the subject, and that up to the present it has not been
possible to determine the particular St. Bega to whom the churches
were said to be dedicated, otherwise than by the date of the festival.
Kow there is no saint of this name having a festival in the month of
November mentioned in any of the works cited. If, however,, it were
allowable by way of argument to assume that there is a possible
confusion between the names ** Begnet" and "Benen," we have a
possible clue in the account given of St. Benignus or Benen, who
died the 9th Kovember, 468, of whom it is said in Dr. Healy's
'^ Insula Sanctorum et Doctorum," p. 95: *' Benignus, son of
Sescnin, Bishop of Armagh, died 9th November, 468," p. 95.. The
death of Benignus is thus noticed in the "Martyrology of Donegal":
*^ November 8th, Benignus, i.e. Benen, son of Siscnen, disciple of
St. Patrick, and his successor that of Primate of Armagh. He was
a virgin without ever defiling his virginity."

This would furnish a date for the festival very close to that
mentioned, viz., the 12th. But confining the question to the determi-
nation of the particular St. Bega or Begnet, to whom the church

O'Reilly— OW Churches ofDalkey Town and Island. 20&

of Dalkey was dedic-ated, the choice would lie between 8t Bega (1),
yenerated at Dunbar already referred to, whose festival is on the
Srd September^ according to Smith and Waco's ''Dictionary of
Christian Biography " ; or on the 7th Septemher^ according to the
notice of the ''Imperial Dictionary of Universal Biography."

Or on the 22nd Ifovemhery according to Allan Butler (under the
name of St. Bees).

Or St. Begga (2) mentioned already, whose festival would be on
the Slst October (old style or new style not stated).

Or St. Becga or Begga, d. of Gabhran, Y., whose festival is on
the 10th February.

Or St. Begghe^ Duchess of Brabant, daughter of Pepin le Yicux,
Mayor of the palace of Austrasia, who died in 692 or 698. She was
the mother of Pepin, called ^^ Sinrutal" After the death of her
husband she consecrated herself to the service of God, and founded in
680 the monastery of Andenne ("Art de verifier les dates"). It is
further said of this saint, that to her is attributed the foundation of
the " BegutneSf" an order of uncloistered nuns still existing in a
modified form in Ghent, Belgium. No date is mentioned for her
festival.

In a question involving so much uncertainty it is allowable to
offer a suggestion with a view to helping to clear it up. It is that
the name Bee^ Beg, or Bega may have had a titular or collective
signification, and have been attached to the heads of a certain female
Order ; or as a name for the whole Order, as in the case of the
'' Beguinee" just mentioned; or as in the case of the " Clairettes,^^
the name given to the Bemaidines (Littr6, "Dictionnaire de la L. Fr."),
wherein the termination seems to be a diminutive of the same
character as the " net" in " Begnet." This view would be to some
extent supported by the fact that in Old and Middle Age French the
word " Bee " was used with regard to women, as mentioned by
littre in his dictionary, under that word. Thus he says: §4°,
^'Minois":

" Tin sien valet avait pour femme
Tin petit bee, assez mignon."

La Foutaike, " Pat6."

" Tu voudrais me d6plaire, A moi, Crispin, k moi, que tu nommais
tojours • Ton Bee,' ton petit bee ? " (Hauteroche, "Nobles de Province,"
iv. 4).

The word "minois" is given by the same "Dictionary*' as

206 Proceedings of the Royal Irish Acadetny,

meaning, « par extension," '^ nne jolie fille." From tbis point of
view " Begnet " might represent either the Order collectively, or the
head of the Order or honse for the time being ; and the date of the
f estiyal might yaiy from place to place, and even apply to different
saints.

On the other hand, some consideration may be had of the Orienta-
tion of the church of Dalkey Town. In Chambers's *^ Encyclopaedia"
(1901), under the word "Orientation," it is stated : " The Orientation
of churches is not usually very exactly to the east ; and it is supposed
that the east end, in some cases, has been set so as to point towards
the place where the sun rises on the morning of the patron saint's day.
In other cases, the choir and the nave are not built exactly in a
straight line, the choir having thus a right inclination to one side,
which in the symbolism of the middle age, was supposed to indicate
the bowing of our Saviour's head on the cross."

Now the church of Dalkey presents the peculiarity of having the old
nave a more modem chancel orientated in slightly different directions
(PI. XIY., fig. 2). The older or western portion is orientated about
east 8^ 10' north ; while the eastern and modern end has a direction
of about east 9^ 30' north. Assuming that this or these orientations
were intended to point to the point of the horizon at which the sun
rose on the festival day of the patron saint, we have simply to see
to what dates in the year these northern declinations of the sun
correspond. This should take place at two different periods of the
year, the one on the passage of the sun from equinox to summer
solstice, and the other on his return southwards. In the first case,
there is an indicated north declination of from 8*^ 4' 24" to 9® 31' 47"
occurring between the 11th and the 15th April, In the second case
there is an indicated north declination of from 9^ 36' 38" to 8° IC
26" occurring between the 29th August and the 2nd September, Search*
ing among the different saints whose festivals occur about these dates^
we find mentioned St Benezet or little St. Benedict, of Avignon, who
died in 1184, and whose festival is kept on the lJ!^th April — a possible
solution if there were any equivalence between Begnet, Benen, and
Benedict, which, as shown, is denied by scholars such as Dr. Todd
and Dr. Joyce, not to speak of the difficulty of the St. Begnet in
question having been a virgin. As regards the St. Bega, or Begga, the
daughter of Oabhran, virgin, already mentioned, and whose festival
is given as occurring on the 10th February, it should be remembered
that she is mentioned as being the sister of Lugaidh, a priest whose
festival is on the 17th April.

O'Eeilly— OW Churches of Dalkey Town and IslamL 207

As regards the festivals mentioned as occurring between the £9th
August and the 2nd September^ the nearest in date would be that of
St. Bega {September Srd), of whom it is stated, as already mentioned,
that she was an Irish princess of the seventh century, was venerated
at Dunbar, who lived in an island in the Ocean Sea, and whose festival
is given by Allan Butler on the 6th September, and by Waller, in the
citation from the '^ Imperial Dictionary of Biography," as occurring
on the 7th September. But, as has been already pointed out, a
great confusion rests over the different saints known by this name.
At all events, if the Orientation of the church of Dalkey be taken as
having a connexion with the patron saint's festival, then it would
point to the St. Bega of Dunbar, whose festival is given as occurring
on the 3rd, or 6th, or 7th September, and relatively close approximation
to the dates indicated by the Orientations {iSQth August to 2nd
September), The question of the Orientation of the Chureh on Dalkey
Island will be discussed further on : it is sufficient here to say that it
is quite different from that of the church of Dalkey Town. The data
concerning the church on Dalkey Island are, as has been already
pointed out, very scant. The only thing apparently on record is its
transference by Hugh de Lacy to the Cathedral Chapter of Dublin, as
mentioned in Mr. Ellington Ball's account of it, already cited.
Nothing is seemingly known as to the date of its foundation, nor,
strictly speaking, as to the particular saint to whom it was dedicated,
except the tradition that it also was dedicated to St. Begnet, the
patroness of Dalkey Town church.

It might seem that any further description of it than that given
by Wakeman, already cited, would be superfluous; but the closer
examination of the ruin on the one hand, and meagreness of
documentary record concerning it on the other, justify a more careful
examination of the remains, were it only for the purpose of securing
a fairly correct plan and details of the structure. Moreover, the
question of the orientation presents an interest in this case also, and
being different from that of Dalkey Town church, it is presumable
either that the church may not have been originally dedicated to the
same saint, or that a different intention guided the founders in that
respect.

The position (PL XIY., fig. 6) occupied by the building is re-
markable, as shown by the east and west cross-section of the island
through the old church. It lies in a sort of depression on the central and
longer axis of the island, at a point where the ground offers an extent of
surface sufficiently level to allow of its being conveniently built on. It

208 Proceedings of the Royal Irish Academy.

also lies near the little rocky cove by which commnnication is still
had with the land ; and to the west of it, at a comparatively short
distance, stands the rock with the cross and circle described as Greek
by Dr. Graves, and mentioned and sketched by Wakeman in his Paper
already referred to. Towards the east the ground rises, as indicated
by the section, and nearly hides the structure on that side, since but
the summit of the roof and the points of the gables could be seen from
the sea. This disposition may have been intentional, with a view to
more completely hiding the building from the attacks of the sea-roveis,
who seem to have continually infested these coasts, and, indeed, those
of Ireland in general, more notably during the seventeenth and
eighteenth centuries, as would appear from the Public Beoords. In
the Beport of the Keeper of Public Kecords in Ireland (2nd May,
1888), p. 23, mention is made of ''the orders and letters concerning
principally the guarding of the coasts against pirates, Turkish and
other sea-rovers, with instructions to the commanders how to cany
out the orders of the State," and dating from 1631 to 1638.

The roof of the church, just overlooking the sea to the east and
north-east, would have afforded an advantageous position for a look-
out, from which to give warning by means of a bell to those on land*
In this respect it may not be out of place to cite the following foom
Chambers's ''Encyclopedia" of 1864 under the word "iAn-t^
Tower " :—

"The name is said to have been taken from certain Italian
towers built near the sea during the period when piracy was comnion
in the Mediterranean, for the purpose of keeping watch and giving
warning if a pirate ship was seen approaching. This warning was
given by striking on a bell with a hammer (ital., "martello"), and
hence the towers were called (" Torri da IfarUllo'*). Such a look-
out should, of course, have been constantly kept up, and necessitated,
therefore, the continuous residence of an outlooker or outlookers oa
the island. The position selected for the Oratory may also have been
influenced by the vicinity of the well, that of the landing oove, and
the relative shelter from the easterly and south-easterly storms
jifforded by the ground. In any case the position was weU selected
from all these points of view.

As is indicated by the section, the eastern horizon would have
been visible from a point at the height of the belfry; and it m
proposed to examine in this Paper the possibility and the probability
of the building having been either intended as, or at least used as, an
observatory for the determination, by direct observation, of the rising

O'Rkilly— OA/ CknrcheH ofLalkey Totrn and Is/and. 209

of the sun on the horizon, and thus to fix the precise period of the
■equinox and summer solstice — the former fundamentally necessaiy for
the correct determination of the Paschal Time.

The plan of the church was carefully made, and, for .^reasons here
unnecessary to explain, the dimensions were taken in metres and
centimetres, which can always he converted into English feet and
inches when required. It will he ohserved that the plan (PL XIV.,
fig. 1) does not indicate a very great precision in the laying out of the
foundations, and that measurements to a centimetre give the amount
of accuracy attainable.

The principal feature which strikes one on the examination of the
plan is the projections of the north and south side wall, beyond the
gable faces, so as to form what have been called, in the description by
Wakeman, '^ pilasters," but which, more properly, might be called
** anta '^ — a detail of form so characteristic of the more ancient styles
of Greek and Italian temples. These projections are not quite equal
at the four comers ; the two at the western end of the building are
practically equal in amount of projection ; but at the east end the
projections are unequal and somewhat greater than at the west end.

The door in the western gable is marked in its style as noticed by
Wakeman. The jambs are slightly inclined ; the breadth of doorway
between them is, at the top, under the lintel, 80 centimetres; while at
the ground, where a sill may have existed, it is 82 centimetres: thus
barely an inch, but determinable.

A character of the building, which docs not seem to have been
noticed, is the '' hatUr*^ of the walls, which may be observed on the
angles of the building, but more particularly on the jambs of the door,
which showed a thickness of wall of 98*5 centimetres at the floor,
and only 85*5 centimetres under the lintel. This batter, or inward
inclination of the walls, is fairly recognisable in the photographic
vignette placed at the end of the chapter (p. 80) in Mr. Elrington
Ball^s work already referred to. It would favour the presumption of
g^at antiquity for the building, and would, to some extent, account
for the resistance of the walls to the destructive action of time, wind,
and weather.

The masonry is very rough, and is composed of stones, which
seem to have been either surface-boulders or very weathered material
from some other structure. The sizes of the stones vary much, from
very large in the lower parts, to middling- and small-sized in the upper
parts. Of courses there are, strictly speaking, none, the stones
having been seemingly fitted to one another as they came to hand.

210 Proceedings of the Royal Irish Aciulemy.

with, however, a very abundaiit use of ''spawls/* and much intelHgence
shown in their use.

The mortar seems, indeed, to have been employed rather to secure
the '* spawls " than to bed the stones.

The material is, for the most part, of granite ; but here and there
may be noticed stones, more or less dressed, of limestone, evidently
dressed glacial boulders from the drift, and some slabs of mica-echi^
from Killiney shore. As ** spawls " were employed, pieces of granite,
mica-schist, andalucite-schist from Killiney shore, and even pieces of
the ** epidiorite " now found in Killiney Park, and described by the
author of the present Paper in the Proc, Boy. Ir. Acad. (3rd Series,
vol. vi., No. 1). As to the source which furnished the greater part of
the material, the rounded and weathered nature of which is so evident,
it may be recalled here that a dun, or fort, existed on the island prior
to Christian times. It is mentioned in " The Annals of the Four
^[asters," p. 6, as having been built, according to that authority, in the
age of the world 3501, by Sedgha, a Milesian chief of great renown.
This date would, according to the chronology of these anthoritieay
correspond to b.c. 1700. It may be assumed that its remains still
existed down into Christian times ; and there is therefore a certain
probability that the material employed in the construction of the
church was, to some extent, procured from the remains of this *' dm^^^
since so few loose stones or boulders are to be met with at present on
the island. It would certainly add to the interest attaching to the
present ruin if it were presumable that the materials employed there-
for had at one time formed part of the walls of that prehistoric
monument.

That the materials for the building of the dun itself were all pro-
cured from the neighbouring shore is hardly likely, and such would
imply the use of a size and style of coasting vessel, and skill in
handling it, that might with difficulty be conceded to the *' Milesians "
of B.C. 1700 ; but that some part of the material may have been so
transported is conceivable.

The only openings in the walls, besides the western door, are the
small window in the south wall described by ^akeman, and con-
sidered by him as original. He gives a sketch of it on p. 702 of the
volume containing the paper. This woodcut is so far incorrect as it
would lead to the impression that the jambs of this window are
vertical, or but slightly inclined. But the contrary is the case ; the
window is at 360 cm. from the present ground-level on the ontside ;
it has a single-stone lintel and sill. The breadth in the dear under

O'Beilly— OW Churches ofDalkey Town and Island. 211

the lintel is 27 cm., and on the sill is 80 cm., while the height of the
opening is about 55*6 cm. The inclination of the jambs is therefore
well marked, and more marked and distinct than in the case of the
door. The splay of the sides on the inside gives an interior breadth
of opening of 75 cm.

This window was seemingly intended to light a small room placed
above the floor, at the west end of the building, and of which traces
still remain on the interior surfaces of the walls. It seems to have
formed part of the original design. The larger window in the south
wall, situated at the south-east end, sketched and briefly described by
Wakeman, is by him considered as quite recent ; and he gives testi*
mony in support of that view. As the masonry has been exposed to
the action of the air and weather for at least a century or so, since
the period of the alterations referred to by him, it does not show with
marked evidence the certainty of this change, unless by the relative
smallness of the material employed on the sides of the opening and
the presence of the two sill-stones so strangely placed across the
opening (PI. XYI., fig. 1). That the space underneath these stones has
been the result of quite recent work, and is roughly a hole broken in
the wall, may be at once granted. But it is probable that there was
originally at the south-east end of this south wall an opening or
window of the same character as that still existing in the western
part of Dalkey Town church (PL XYI., flgs. 2, 3, 4, and 5).

This is a tall, narrow slit, so placed as to throw light on the altar
at the east end of the church ; possibly that of Dalkey Island was
divided towards the mid height by a cross-stone or sill, as in the case
of the Dalkey Town church window referred to, and of the two stones
remaining across this south-east end opening; the upper one was pro-
bably the middle sill of the original window. This is to some extent
suggested by a comparison of the relative distances of the two windows
in question from the respective south-east comers of the buildings. In
the Dalkey Town church the distance of the eastern vertical edge of
the opening from the south-east comer of that building is 186 cm. ;
whUe the same measurement in the case of Dalkey Island church gives
182 cm. — ^practically the same— and so far suggests that an entirely
new window was not broken in the wall, but rather that the narrow
light or opening, such as that of Dalkey Town church, was enlarged
towards the west side to its present breadth. It is proposed to dis-
cuss hereinafter the possible usage of this narrow opening for the
purpose of the determination of the periods of the solstices, by means
of the relative positions of the patches of light formed by the sun's

R.X.A. P&OC, VOL. ZXIV., SBC. C] [17]

212 Proceedings of the Royal Irish Academy.

rays, on the floor and wallB, for which object the nairowneea of the
opening would be more advantageouB than if it were wide, as at
present is the case.

The only other opening in the walls of the building is a small
square cavity, situated in the northern wall, quite near the north-
west interior angle of the building, and situated at a height of
273 cm. from the ground. Its dimensions are about 40 cm. by 40 cm.
It was originally evidently a recess in the nature of a cupboard, and
did not then extend through the thickness of the wall ; since on the
outside, as it presents itself at present, it is represented by a hole, the
place from which a stone had been forced out.

On the walls, in the interior, are to be seen the remaining traces
and patches of plastering, leading to the presumption that the greater
part of these surfaces had been so treated. On the interior face of
the west wall, this plastering shows the traces of a floor having onoe
existed at the height of 273 cm, from the present ground, and in
the south-west comer at this height, appears a rectangular space
measuring 185 cm. by 98 cm., marked .on the plastered sui&oe,
as if some article of furniture had been in position there (see
PI. XV., fig. 2).

The fire-place in the eastern wall is mentioned by Wakeman as
having been made by the workmen who took up their dwelling in tJie
old church during the building of thp Martello Tower. The reoesaed
space above it is probably original, and is unsymmetiical in its lines,
as regards the. vertical axis of the wall face. The workmanship \&
very rough, as is also that of the arching. There is a crack in this
face over this recess, as if there were a void space in the wall,
such as a chimney-flue.

There is every reason to suppose that the original rod was of
stone slabs, probably of the chiastolite mica-schists that outcrop on
Eilliney shore, possibly of the ordinary mica-schist to be found in
connexion witii the granite there. What was the form of the
termination of the eastern gable can only be a matter of oonjectme
at present; but it supported the roof at all events, and may have
presented an opening just under the ridge of the roof, or this eastern
gable may have carried a belfry or opening such as that still remain-
ing on the western gable. Wakeman considers this latter belfiy to
be a recent addition; but the appearance of the masonry hndly
supports that opinion ; while, on the other hand, the ihidrnessps ol
the two gable walls at the base, 91 cm. (about one-tenth greater than
that of the two side walls), would point towards the presamptioii

O'Beilly— OM Churches ofDaUcey Toum and Island. 213

that these two gables were intended to carry belfries or elevated
parts of that nature.

In the two elevations (exterior and interior, PL XTTT., fig. 1, and
PL XY., fig. 1 ) of the western gable herewith submitted are shown two
Mqnare holes in the sides of the belfry, one on either side of the belfry
opening, which may have been intended to receive the ends of the
purloins which supported the roof. These holes are thorough ; and I
have assumed the existence of, and sketched in dotted lines in the
longitudinal section (PL XIII., fig. 2), a projecting platform, supported
on these beams where they pass through these holes. I have done so in
order to bring out the idea already suggested, viz. : that observations
may have been made from this platform by an observer standing on it,
and looking through the existing belfiy opening, and a corresponding
one in the eastern gable, on to the horizon, which I assume to be
clearly visible from that point through such an opening. This implied
use of the belfry is to some extent supported by ^akeman's remark
that it must have been enclosed by the roof, which would precisely fit
it for such an application. It is not necessary to examine here the
size or form that such an opening should have had for that purpose ;
but there is ground for discussing the more general question of to
what extent and in what way were direct observations of the sun
and .stars made currently, in connexion with these ancient churches,
with a view to the determination of the festivals and hours of
service, of the equinoxes and solstices, and of the due fixation of the
paschal time and other festivals and seasons of the year.

That from the earliest periods of Irish history the division of time
into years, months, and days was known and employed, need not here
be discussed. Dr. Joyce, in his *' Irish Names of Places," voL L,
p. 200, chap, vi., discusses the names arising out of '^customs, amuse-
ments, and occupations," and says; "The Pagan Irish divided their
year, in the first instance, into two equal parts; each of these was
subdivided into two parts or quarters. The four quarters were called
Earraeh^ Samhradhj Fogkmhar^ and Q&imhridh [Arragh, Sowra,
Power, and Gevre] (Spring, Summer, Autumn, and Winter), which
are the names still in use; and they begin on the first days of February,
May, August, and November, respectively." Now such a division of
the year must have been based on some sort of actual astronomical
observations, and could only have been maintained by continual and
regular observations of the heavenly bodies which deteimine the
divisions of the year. Hence, there must have been at all times of

[17*]

214 Proceedings of the Royal Irish Academy.

Irish society, astronomers and places and methods of obsenration
sufficient for the requirements of the period. Moreover, observatories
or places of observation were requisite as matters essential to the
security of aggregated communities, and most essentially along the
coasts, on which incursions might be made by enemies or by piratical
adventurers. High points offering extended views in all or certain
directions would naturally serve as such observatories or look-outs ;
and in the case of buildings their highest points. Hence, places of
defence, fortifications, castles, &c., from the very earliest periods of
history have had, as part of their general scheme of arrangement*
elevated places or towers from which views, either of the heavens or
of the cotmtry in the neighbourhood, could be securely and advan-
tageously obtained. Dr. Joyce, in his *' Names of Places," vol. L,
p. 215, says : '' Look-out points, whether on the coast to command the
sea, or on the borders of a hostile territory to guard against surprise,
or in the midst of a pastoral country to watch the fields, are usually
designated by the word coimhsad (covade). This word signifies
^watching' or ' guarding' ; and it is generedly applied to hills from
which there is an extensive prospect."

^e should therefore expect to find corresponding arrangements in
the plans of the earliest monasteries and buildings intended to receive
Christian communities. In Smith and Cheatham's ''Dictionary of
Christian Antiquities," p. 1240, in speaking of the '* Canohitm^^ of
St. Euthymius, in Palestine, circa a.i). 828, it is stated : '' The whole
area was fortified with a palisade and wall, and further protected by
a strong tower forming the citadel or stronghold of the whole desert,
rising in the middle of the cemetery."

This tower just described was a very usual feature in the monas-
teries of the East, which, from their liability to attack from the
pted^tory tribes, assumed the character of strong fortresses. The
whole e$tabUahment was dominated by a lofty tower near the
entrance, like the keep of a Norman castle, placed under the patronage
of the Blessed Virgin Mary, St. Michael the Archangel, Apostles, or
sditit9^ to which the inhabitants might flee for protection when the
rest of ttie buildings had fallen into the hands of the assailants
(monasteries; (^ ]^(ount Athos). In some cases protection was still
further secured by the single entrance being made many feet above
the ground, only accessible by ladders, or by a bucket raised by a
windlass, e,y,, the monastery of St. Catherine on Mount Sinai.

At page 1243, the Dictionary says : ** The Irish and early Scotch

O'Bbillt— 0/tf Churches o/Dalkey Town and Island. 215

monasteries of tlie sixth and seventh centuries, snch as that of
Armagh and lona, followed the eastern model."

So far there is merely a presumption that these towers served for
look-out and ohservation purposes; but being construoted for the
safety of the community, this could only be secured by such continual
observation and outlook. Moreover, another important requirement
of the religious communities rendered such observations necessary,
more particularly that of the heavenly bodies — ^ihat was the division
of the hours of the day and of the night, for the regular oocupations
and offices of the community. The division of the day into hours
must have been in some way arrived at, and, moreover, announced
regularly to the members of the community. How the hours of the
day or night were marked in the pagan and early Christian times of
Ireland is not distinctly stated, so far as I can find. That the round
towers or '' doictheachs" served in some way for that purpose the
very name implies ; and yet Petrie barely concedes that they may
have been thus used. In the East, and in those latitudes wherein
the sun is generally visible during the day-time, the use of the
gnomon was common until the introduction of the '' clepsydra," and
later on of the deck ; but in a climate such as that of Ireland, the
sky of which is so frequently overcovered by clouds, and continuous
sunshine thus exceptional, means must have been found at an early
period of mechanically dividing the time of the day and of the night ;
and also a means of making known these divisions to the public, or
to those requiring this knowledge for their daily avocations.

In the *'Dictionnaire EncydopMique de la France" (Fh. le Bas,
Paris, 1843), under the heading ''Horloges," p. 485, the author
describes the wonderful clock or depsydra made for the Calif of
Bagdad, Haroun-al-Raschid, in a.d. 807, and presented by him to
Charlemagne ; and then continues : '' On a done ignor6 absolument
jusqu'au 12"* si^de, la division du temps par le moyen des roues
denties, et des pignons qui s'y engrSnaient. Ce n'est que depuis ce
temps, qu'on a commenc6 & fabriquer, pour les doches des 6glises, des
grandes horloges, qui fonctionnaient au moyen d'un poids attache &
la plus grande roue et faisant aller tout le m^canisme. Des ouvriers
intelligents perfectionn^rent ensuite cet appareil, en y ajustant un
rouage correspondant k un marteau, qui frappait sur un timbre sonore
les heures indiqu6ee par le cadran. Ce perfectionnement devint d'une
grande utility, et pour les monastires, od avant son introduction il
faUlait que les religieux proposassent des gem pour observer les itoUes

216 Proceedings of the Boyal Irish Academy.

pendant la nuit afin d^itre OMrtie dee heursi de Voffiee^ et pour les TilleB,
oii les crieurs faisaient connaitre la marche du temps, usage qui se
conserve dans plusieurs provinces."

'' On a & tort fait descendre jusqu'au IS^' siiole et meme jusqu'au
14"^ sidcle, I'inveiition des horloges sonnantes; elles se troaTent
ddjfl cities dans les statuts de I'ordre de Citeanz, leunis vers I'ann^e
1120. On Toit en effet dans ces statnts un article par leqnel on
defend toutes sonneries de doches, m^me & I'horloge, depnis la messe
de jeudi saint, jusqn'i ceUe dn samedi saint; un autre artLde aussi,
qui enjoint au sacristan de regler Phorloge de sorte qu*elle sonne et
qu'elle l'6yeille pendant Thiyer, avant matines ou avant les noc-
turnes/' &o.

This article distinctly points out the obserration of the stars
during the night for the fixation of the hours of office in the manas-
teries of the early Christian period, and such observation implies an
observatory or part of the building capable of being so applied, such
as a tower or elevated part dominating the surrounding parts of tiie
building, any trees in the proximity, and having a free and extended
view towards the horizon. Let it be remarked, enpassani, that such
conditions are presented by the highest story of the round towers.
Another and more important requirement for the early Ghristaan
churches called for such regular observations and for corresponding
observatories. It was that of the correct determination of the Easter
time. It is a matter of history the difference that subsisted for nearly
two centuries between the Churches of Ireland and England, and that
of Home and the East, as regards the proper period for the cele-
bration of Easter. Dr. Healy, in his Insula Sanctorum et Ihet^rmm
(1890), p. 238, says: ** Of course the system of computing the date el
Easter in use in Ireland and in England, at the beginning of the
seventh century, was that which was introduced by St. Patrick him-
self, and which he acquired in the schools of France and Italy. From
[p. 284] the very beginning, however, much diversity of practice
existed between tiie Churches of the East and West, and even between
some Churches in the West itself, in reference to the date of Easter
Day." He then gives an account of the results of the Synod of Aries
in this regard, of the Nicene Synod of A..n. 825, and of the reference
to the Church of Alexandria for the exact date thereof, and its notifi-
cation to the Boman Church, by which it was finally made known to
the other Churches. He says : '' The Alexandrian usage ultimately
prevailed, but was finally accepted in the Western world only about

O'Ebilly— 0/rf Churches o/DaOcey Town and Island, 217

A..D. 580, when explained and developed by Dionysius Ezigans.
This, the correct Bystem, therefore lays down tluree principles : — (first)
Easter Day must he always a Sunday; never on, hut next after, the
14th day of the moon; (secondly) that the 14th day, or the full
moon, should be that on, or next after, th$ vernal squtnox ; and (thirdly)
the equinox itself was invariably assigned to the 21st of Mardi.
Whilst, however, the continental Churches aimed at uniformity after
a troublesome experience of their own errors, the Irish and British
Churches, practically isolated from their neighbours, tenaciously dung
to the system introduced by St. Patrick."

Xhis citation is made in order to show the importance attached to
the question in the early Churches, the differences that existed between
them, the effect of their isolation from the continental communities,
and the intimate dependence of the exact date on that of the vernal
equinox. It is true that from a very early date a cycle of years was
adopted, and brought into use for the purpose ; but it is dear that the
actual observation and determination of tiie vernal equinox were not
the less necessary as a check on, and a control of the computation ;
and hence in the Western churches, and more particularly in those of
Ireland, such means of observation must have been provided for, and
have been employed from the time of St. Patrick.

In support of this point of view, it may be interesting and useful
to dte the inscription which appears on the pavement floor of the
north transept of the Church of St. Sulpice in Paris, in connexion
with the meridian line traced on that floor. It is thus referred to in
Baedeker's ** Ghiide to Paris," 1891, p. 252 :— '< St. Sulpice, Transept.
— On the pavement here a meridian line was drawn in 1748 with the
signs of the Zodiac. It is prolonged to an obelisk of white marble,
which indicates the direction of the north ; while towards the south it
corresponds with a dosed window, from a small aperture of which a
ray of the sun falls at noon on the vertical line of the obelisk." This
description, rather curt and wanting in detail, does not suffidently
describe these details, nor show the significance of this remarkable
piece of sdentific work. Not only is the meridian plane clearly and
sharply defined by a ribbon of brass, inlaid on edge into the floor, but
it is marked with signs of the Zodiac on the floor, and on the gnomon
or obelisk situated in the north transept, and serving to indicate by
horizontal lines traced thereon the positions which the ray of sun-
light, coming from the south transept window, occupies at the various
periods of the year corresponding to the inscribed signs thereon.

218 Proceedings of the Royal Irish Academy.

These refer to the equmoxes and solstices, as clearly pointed out by
the inscription, which is as follows (the lines of which run across the
meridian line) : —

Onomon Astronomicum

ad Gertum paschalis

jEquinoctisB Ezplorandum.

Quod 8. Martyr Episcopus Hyppolitus

Advisus est, Quod coDcil. Nics&ain

PatriarchsB Alexandrine, Bimandavit,

Quod Patres Constantienses et Late

ranenses, solicitos habuit. Quod inter

Eomanos Fontifices Gregrorius XIII

et Clemens XI incredibile Lahore et
I

Adhibitan Feritorum Astronomorum
industria conati sunt. Hoc Emulator,
Stylus iste, cum sub Ductum Lin. men
diana, punoto ^quinoctiaH certis
Periodorom, — Solarium indicibus.

On the floor between the two transepts occurs the following
inscription (the lines of which also run across the meridian line) : —

Opus. D. 0. M. Sacrum,
elaboravit
Scientiarum AcademisB nomine et consi
liis G. Gl. le Monnier Ejusdem Acad, et
London, Socius. Ad. ^quinoctise Autumnali
et in Hienmali Solstitia absolvit An.
Rep. Sal. i MBCGXLIII.

O'Ebilly— 0/rf Churches of Dalkey Town and Inland. 21»

At the side door of the south transept entrance is placed the
following inscription on a slab let into the wall : —

Ohliqiiitas EceMptin: Maxima.
23°: 2J8': 40"

Faii par Cldmle Langlois

Ingr, aux Gai^/ries du Louvre

MDCCiXLIV

This remarkable piece of work demonstrates, by its arrangement
and inscriptions, the traditional acknowledgment of the Papal and
Patriarchal admonitions as to the observation and detennination of the
equinoxes and solstices, in view of the correct definition of the paschal
time, and of the festiyal connected therewith. That similar arrange-
ments may still exist in other churches and cathedrals on the
Continent, particularly in those of Rome, there is reason to believe; and
considering the influence that the Continental ecclesiastical customs
had on the early churches of Ireland, it is presumable that in many
of these some such arrangement was provided for.

The examination of the south-eastern opening of Dalkey Town
old church seems to me to point to such a use of the beams of sunlight
which may pass through the upper and lower compartments of tiiis
window, of which an elevation, section, and interior elevation are
submitted herewith (PL XYI., figs. 2, 3, 4, and 5). It will be observed
that the opening is divided towards the middle of its height by a
cross-piece or sill ; exteriorly, this and the lower sill are of roughly-
fashioned slabs of granite; but the middle siU, while showing a
granite slab exteriorly, presents on the inside a mica-schist slab, which
naturally offers relatively smooth surfaces and sharp straight edges.

Now the thickness of the wall, taken in conjunction with the
height of the opening in the clear, determines the conditions under
which sunbeams are able to pass through these openings, and the
forms determined by the beams of light on the floor or opposite wall.

220 Proceedings of the Royal Irish Academy.

It is quite clear that during the winter months, or those during
which the meridian sun's eleyation above the horizon is low, sun-
light can penetrate into the church by both compartments of the
window, and show itself on the floor, or on the opposite wall, in the
form of two more or less rectangular parallelograms or patches of
light separated by a bar.

It is also evident that for a certain meridional elevation of the sun
above the horizon, the upper portion of the window will not allow
any sunbeam to penetrate which could still penetrate by the lower
one, and that finally for a still higher elevation of the midday sun —
that is, during the summer months — ^the meridian sun could send no
beams into the church by either part of this window.

The dimensions that were taken of these openings allow of a
sufficiently dose determination of what these different elevations may
be, and therefore of the periods of the year that would be indicated
by the appearance of both patches of light, of one only, or finally of
neither one nor other, on the floor or wall of the church. Prom the dia-
gram section herewith submitted (Fl. XVII., fig. 3), it may be seen that
the angle of incidence of a beam of sunlight, for which it would cease to
penetrate, or would be '' extinguished," in the upper compartmsit of
the window, is 44°; while in the lower part of the window the
corresponding angle is 52°. Now these angles of incidence of sunlight
would occur twice in the course of the year, for each compartment,
accordingly as the sun moves from one solstice to the other. For the
angle of 44°, the date would be 9th April and 4th September; while
for the angle of 52°, the dates are 2nd May and 11th August (I
have here to acknowledge the kindness of Sir Bobert Ball, p.k.8.,
of the Observatory, Cambridge, for these determinations.)

It has already been pointed out that, taking into consideration the
direction of orientation of Dalkey Town church, the St. Begnet or
Bega to whom the church was dedicated was the virgin, venerated at
Dunbar, whose festival is given on the 6th September by Albaa
Butler, and on the Srd September by another authority. It may
therefore be assumed that the upper compartment of this window
was arranged so as to give notice or warning of the arrival ol the
dedicatory sainfs festival.

It is probable that the under compartment of the window was
intended to give some such warning or notice as regards some other
festival.

Presuming, as has been already advanced, that the south-east
window of the Dalkey Island church was originally similar in style

O'Beilly— 0/tf Churches of Dalkey Toum and Island. 221

to that still ezifiting in tbe church of Dalkey Town, it may be
presumed that it was designed to fulfil the same objects — that is,
to admit sunlight and mark the arrival of some certain period of the
year or festival day, or even that of the equinox. This double mode
•of observation of the latter — ^that is, by direct observation towards the
horizon from the height of the belfry, on the one hand, and by the
incidence of the sunbeams through the compartment of the window,
•on the other — ^in no way contradict, but rather supplement one
another; since in our climate the horizon may be covered at sunrise,
and the sky quite clear at mid-day. Enough, however, has been said
to point out the interest that the forms, positions, and dimensions of
the different openings of this class of ancient church in Ireland
present, and to justify the proposition that a more careful examination
.and measurement of the still existing '' eilU " or oratories should be
made in the expectation of very interesting and instructive results
being furnished thereby. As the church on Dalkey Island is stated
to have been dedicated to St Begnet, as well as that of Dalkey Town,
it might be expected that its orientation would in some way concord
with such dedication. As already mentioned, however, the orientation
in this case is nearly due east and west, the difference or error
of direction therefrom being about 8^, as determined by a hand-
•tiompass. It might be asked is this error due to defective observation,
or to imperfect means of tracing the east and west line, or rather was
the direction as existing so intended from the foundation. It has
been already remarked that the position of the building is such that
the eastern horizon cannot be seen from its actual site, on account of
the ground rising towards the east, as shown in the section (PI. XYI.,
fig. 6) ; hence, if the orientation were made by actual observation of the
aun on the true equinoctial day, and if his appearance above the ridge
of rocks lying to the east of tiie site were awaited for the tracing of
the line of orientation, there should be an error of at least some degrees
to the south of the correct east and west direction intended ; and such
is actually the case. Hence, it is reasonable to presume that a true
.east and west orientation was intended; and the eiror of about 3® is
quite in harmony with this view. Were the error to the north, it
would be more difficult to reconcile with such intention, and there
would be grounds for assuming that it was designed to refer to some
feast-day happening close to the equinox (such as that of St. Benedict,
the patriigrch of the western monks), 21st March.

The presumption that the orientation was intended to be due east
and weaty and that the observation of the sun for equinox was one

222 Proceedings of the Royal Irish Academy,

of the objects for which, the building was intended, receiveB a certain
support from the relation of position of the church to the rock-faoe,
on which is cut the Greek cross, mentioned and figured bj Wakeman
in his Paper already cited. This rock is situated at 1 1 * 1 5 m. (about 14
yards) west of the western door ; not, however, due west thereof, but
north of west, a certain number of degrees (PI. XYII., fig. 1). It had
long occurred to me that the Irish crosses showing a circle with cross-
arms might have some connexion with solar observations, serving, for
instance, as a means of determining the position of the sun at certain
periods, or rather fixing certain periods by the shadow of the pillar
and cross at certain positions and elevations of the sun. I was there-
fore led to examine attentively the position of this cross relatively to
the east and west direction of the building. On the ground that the
determination of the direction of the setting sun is equally important
as that of the rising sun, if accurately determined, for the fixation of
the solstices or equinoxes, it might be assumed a priori, that some
means would be found to ensure this determination ; and on examining
the position of the Greek cross in question relatively to the plan of
the building, it was found that a line passing through the north-west
edge of the building, due east and west, passes nearly througlL the
centre of the cross ; and probably if very exact measurements of the
orientation and of the position of the cross relative to the sides of t^e
building were made, this relation would be brought out more markedly.
It would seem as if, when the building was completed, and obeerva-
tions from the belfry height could be made on to the horizon, leading
to the recognition of the correct east and west direction, the cross
was cut as a fiducial point which, with the north-west edge of the
building, gave the true east and west direction.

The lineal measurements given in this Paper are in centimetres;
and, perhaps, it may not be out of place to here offer an explanation
of the use of this unit of measurement in this case, rather than of
English feet and inches, usually employed in this country for such
purposes. The explanation is simply that the author had been
continually in the habit of employing the metre in connexion with
geological and stratigraphical studies and measurements, and therefore
continued to use it when measuring buildings. In G<»usequenoe of
this use of the metre, a very interesting observation has resulted, and
a very important question arises.

Having been, when in Northern Spain, continually under the neces-
sity of converting the *'vara" or Castilian yard (the unit of measure-
ment of the country) into metre units and vice versa, the relation of the

O'Reilly— 0/rf Churches ofDalkey Town and Island. 223

^'vara" to the metre became familiar to me. When, therefore, I
came to examine the metric measurements of the Dalkey Island
ehnrch, I soon observed that the '' yara " unit seemed to have been
that employed in the construction of the building.

In O'Shea's '' Guide to Spain and Portugal," p. 109, he gives the
foUoving table of '' Reduetion of Varas into Metres*^ : —

iTara » 0*835

7 varae = 6-845

2 varas = 1-670

8 „ *» 6-680

3 „ = 2-606

9 „ - 7-516

4 », » 3-840

10 „ « 8-360 (oorrectly 8-8489)

6 „ « 4176

11 „ « 9-186

6 „ - 6010

12 „ = 10-020

Examining the pilasters or '' ants " of the building, the following
were the thicknesses found : north-east, 86 cm. ; south-east, 82*5 cm. ;
north-west, 83 cm. ; and south-west, 85 cm. ; the mean value of these
is 84- 1 cm. ; a close approximation to 83*5 cm. ; that is, a ** vara."

Examining then the horizontal dimensions of the building, the
breadths interiorly of the east and west gables give the following
measurements (two for each end) : 416 cm., 418 cm., 417 cm.,
416 cm., of which the mean is 416*75 cm., or approximately
417-5 cm., that is, " 5 varas," according to the above table, with a
difference of 0-75 cm.

The breadth of the opening of the western door at the sill was
found to be 82 cm., probably intended for a ''vara," so that the
spacing on the inner side of the gable shows the intervals of 167 cm.,
82 cm., and 166 cm., that is 2 varas, 1 vara, 2 varas.

The length of the south wall on the inner face is 626-5 cm.
(mean value), equal to 7^^ ** varas " (or 626-3 cm.).

The same dimension taken on the interior face of the north side
wall is 619-5 cm., which differs sensibly from that of the south wall,
and may be taken as the result of imperfect construction, since the
two diagonal measurements differ; that from the south-east to the
north-west measuring 750 cm. » 9 '< varas" (7-515 m.) ; while that
from north-east to south-west measures only 746 cm., that is a differ-
ence of 4 cm. (almost negligible, in a building so primitive and so
ruined). The lengths of the north and south side walls on their
exterior faces measure : for the north wall, 881 cm. ; and for the
south wall, 879 cnu ; this last approximates to the '' vara " measure-
ment of 10^ varas » 876*7 cm.

224

Proceedings of the Boyal IrM Academy,

The measurements for the heights cannot offer any satisfactorj
terms of comparison, since neither the gahles nor any one of the
comers of the pilasters remain in a state sufficiently complete to allow
of any satisfactory comparison with ''vara" measurements. The
small window in the southern wall, sketched hy Wakeman, and so
markedly " pelasgic " in the character of its inclined jamhs, preflents
the following dimensions : — ^breadth of opening at top, 27 cnu ^ i Taia
(27-8 cm.) ; breadth of opening at sill, 30 cm. (27 cm. + i = 30 cm.) ;
height, mean value of the two sides, 55*5 cm. s f 'Wara" (55*6 cm.).
The western doorway presents on the basement course an opening
of 82 cm., as already remarked, or approximately 1 ^'vara"; this
opening under the lintel is only 80 cm., or a diminution of about -A*;
while the height from the basement course to the lintel under-aur£Boe
is 208 cm. »2i varas (208*72 cm.).

To bring out more distinctly Uie <' vara " relations of the different
measurements, it may be convenient to present them in a tabular
form as follows : —

Meanyiaae.

Vaiaviaiie.

DifloraBDes.

CM.

Pilasters: breadths ... 860^

Do. do. ... 82-5
Do. do. ... 830 '

CM.

8410

ex.
83-5

Of.

0*6

Do. do. ... 860

E. & W. gables— interior (416.0'
faces : measaremente of U18.0
Do. do. ... 417.0 f

416*75

417*6

0*76

Do. do. ... 4I6.0J

Spacing on inner face of (167*0
west gable. I 820

167*6

0*5

83-5

1-5

Do. do. ... 166-0

167-6

1-5

Inner face of south wall : length

626*5

626-3

0-2

Inner face of north wall : length

619*5

626-3

6*6

S.W./NJS. diagonal

760 0

751*5

1-6

N.W./S.B. diagonal

746-0

761*6

6-6

North wall: exterior face

881*0

876*7

4-8

South wall :— exterior face

879*0

876-7

2>3

Small window in south wall :

breadth at top

27*0

27*8

Do. do. atsiU

30-0

27*8 + 1/9 « 30*9

Do. height

66*5

66*6

Western door-way:

opening on basement coarse ...

82-0

83*5

1-5

Do. under lintel faoe

80*0

83-6-1/41

Height

208*0

208-72

0*72

The mean value of these yarious differences is 1*82 cm., or a little
more than a half inch English measurement. *

O'Reilly — Old Churches ofDalkey Town and Island. 225

These seyeral cencoidances between the '' vara " values and the
metric measurements, found for different parts of the Dalkey Island
church, can hardly be fortuitous, and go far to support the assump-
tion that Spanish masons, or builders, used to the Spanish unit of
measurement, were engaged on the building of this church. This
assumption would be quite in harmony with the remarks from
Gheethamand Smith's " Dictionary of Christian Antiquities," cited
in the paper '' On the mode of ringing or sounding bells in the early
churches of Northern Spain and of Ireland" (Proceedings, Eoyal Irish
Academy, third series, vol. yi., p. 490), as to the points of resemblance
between certain very ancient oratories or churches of Northern Spain
and those of Ireland, It would also go to demonstrate the activity
of the commercial relations between Spain and Ireland in ancient
times, and the f roquentation of the safest ports of the Irish coast by
Spanish and Continental traders, most probably for fishing purposes,
and for the trade in salt, amongst other objects.

A still moro interesting question is raised by the considera-
tion of these measuroments ; it is that of the units of length which
provaihd in Ireland at various periods of its history. Up to the
prosent, it has been customary to give the measuroments of monu-
ments, no matter what their age or naturo, in standard English feet
and inches. For practical purposes this is perfectly intelligible ; but
it is not to be supposed that all the monuments of this country wero
laid down as rogards dimensions in the units of measuroment now
currontly in use. It is evidently presumable that various units pre-
vailed from time to time, according to the culture and customs of the
predominating races, and that most certainly the use of British units
of measurement did not generally provail until long after the Norman
invasion.

In O'Gurry's Lectures, vol. iii. (Lecturo XIX., '' On Buildings,
Fumituro, ftc, in Ancient Erinn"), frequent mention is made of
dimensions of buildings in feet ; but no indication is given as to the
absolute length of the foot mentioned, or as to the standard implied;
and the reader, accustomed only to the curront English foot, naturally
reads it into the measuroments cited by O'Cnrry and others. The
rocovery of these ancient units is most desirable, and should be
attempted, however arduous the task may prove to be ; and it can
only be brought about by the carof ul measuroment with such a com-
mon unit as the metro of all our monuments, still sufficiently well
proserred to allow of such measuroments being satisfactorily taken,
and the comparison of these with such units as aro known to have

226 Proceedings of the Royal Irish Academy.

^reyailed in former times. This task was undertaken and carried out
to a definite conclusion by Nenrton, as regards the Egyptian cnhit or
cubits nsed in tiie constiruction of the great monoments of tiist
country, and by other eminent sayants as regaxds the nnits of Pexsia,
Babylonia, Oreece, and Italy. It presents, therefore, a field of study
which has been cultivated by men of the greatest learning, and as
necessary for the proper understanding of the histories of the oounkie^
mentioned as displayed in their monnments, works, and utensils.
The study of the ancient eiUs or oratories of Ireland, from this point
of view, would, I beg leave to submit, fomish data of the ▼err
highest historical interest, and merits, therefore, the encouragement of
the Royal Irish Academy.

EXPLANATION OP PLATES.

Platb XIII.

Fig. 1.— Westem elevation of Dalkey Island Church.
Fig. 2.— E. to W. yertical aection of same.

Platb XIV.

Fig. 1.— Flan of old Chuioh on Dalkey Island.

Fig. 2. — Plan of St. Begnet's Church, Dalkey Town: showing orientation.

Platb XV.

Fig. 1.— Interior elevation of western gable wall of Dalkey Island Church.
Fig. 2. — Interior elevation of eastern gable wall of Dalkey Island Church.

Platb XVI.

Fig. 1.— Interior elevation of S.-E. window of Dalkey Island Church.
Fig. 2.— Interior elevation of S.-E. end window of Dalkey Town Church.
Fig. 3.—- Vertical cross-section of this, N.-S.
Fig. 4. — Elevation of same, facing south.
Fig. 5. — ^Plan of same.

Fig. 6. — ^E.-W. vertical cross-section of Dalkey Island, to show lie of gioond, snd
visual from Belfrey of Old Church towards eastern horixoiu

Platb XVII.

Fig. l.—Phm of Dalkey Island Church, to show relation of orientation with Gmk

cross on rock to the W.
Fig. 2. — ^Elevation of rock bearing Greek cross, Dalkey Island.
Fig. 3.— S.-E. opening in western end of St. Begnet's Churoh, Dalkey Town:

N.-S. section, showing the angles of extinction of rays of sunlight

Froc HlAcaxl Vol XX!V Sect. C

Plate XIII

GV/»tt4SonallU>

Proc R I Acad Vol XXIV Sect.C.

Plate XIV

Pt 'c HLAca.a Vol X/.iV S'ect C.

Plate XV.

?^?W

I^^^^T^I^

^.^|^f*¥

c;v/«st i S<m« hth

Proc. Ri.Acaa. Vol XXIV. Sect C.

Plate XVI.

Wi9,'''

i
1

T '

J /

Proc.-Ri Acaa. Vol XXIV Sect. C

Plate XVII

Y\ in

Fig 2

L-^-iT^r- ^

L=i=x;_x^-t=l ^

J._i i-_l

1 rrr-ift

'^ \v..t A i-

[ 227 ]

XIII.

THE FIRST MOHAMMADAN TREATIES WITH CHRISTLiNS.
Bt STANLEY LANE-POOLE, M.A., Litt.D.

Bead April 27, 1903.

TwE early treaties of the Arabs are important documents in the
history of Islam. They show ns, upon evidence that cannot be
disputed, the policy adopted by the conquerors towards the
vanquished ; and they enable us to understand in some degree
the causes which contributed to the spread of the new religion.
There is a very widely spread misconception on this subject. It
is frequently alleged that Islam was ^propagated by the sword.*
Carlyle's rejoinder, 'First get your sword', was only a partial
answer to the accusation; for though the religion of Islam must
have possessed other attractions to draw men to it in its hour of
weakness, when there were no swords on its side, yet it would be
quite natural that, when the faith had been embraced by many
thousands of fighting men, the argument of the sword should be
employed to bring others to the confession of the creed. Indeed,
if it is held that there is but one road to salvation, it is at least
arguable that forcible methods would be justified in saving men
even against their own wills. But, as a matter of history, Islam was
not * propagated by the sword.' The Eor&n never enjoins any such
principle. It does indeed exhort Muslims to ' fight in the path of
God with those who fight with you,' but adds, 'if they desist, God is
forgiving and merciful ; ... let there be no hostility save against
transgressors.'* 'Unprovoked war is clearly contrary to the letter
and spirit of the Eur-&n ; but war against the enemies of el-Isl^,
who have been the first aggressors, is enjoined as a sacred duty;
and he who loses his life in fulfilling this duty (if unpaid) is pro-
mised the rewards of a martyr. ... Of such enemies, if reduced
by force of arms, refusing to capitulate, or to surrender them-
selves, the men may be put to death or be made slaves; and the

♦ Koran, ii., 186-9.

B.I.A. PliOC, VOL. XXIV., SBC. c] [18]

228 Proceedings of the Royal Irish Academy.

women and children also, under the same drcumstanoea, may be
made slaves : but life and liberty are to be granted to those who
surrender themselyes by capitulation or otherwise, on the conditicm
of their embracing el-Islim, or paying a poll-tax, unless they have
acted perfidiously towards the Muslims.'* In other words, unless
there were exceptional circumstances of treachery or inveterate
hostility, the invariable terms offered by Muslim generals were com-
prised in the simple formula * Embrace Islam, or pay the poll-tax.'
As this tax on non-conformity was not more than two dinars, or about
a guinea, a head per annum, and was levied only on able-bodied men,
and not on the aged or women or children, it was scarcely heavy
enough to induce many to become converts on purely eoonomical
grounds.

There is no justice in the charge against IsUlm that it was
' propagated by the sword ' ; but it is easy to see how it arose. The
Arabs made vast conquests, and the majority of the people they
conquered became, sooner or later, Muslims ; therefore, it is argued,
Islam owed its extension to the sword. But this is to confound two
distinct things. The Arabs were inspired to a new life and a common
enthusiasm by Islam, and in their unprecedented union they set out
to conquer ; but the motive of conquest was gain, not proselytixing,
and the sword was wielded by an expanding people, inspired, it is
true, by the new faith, but not for the purpose of imposing it on
others. Arab statesmen indeed clearly recognized the ^ct that the
more converts were made to Islam the less would be the revenue
from the non-conforming poll-tax ; and as the Arabs have never been
indifferent to money, this consideration formed a check upon a too
zealous propaganda.

The early Muslim treaties are an irrefragable proof of the accuracy
of what has been said about the terms offered to non-Muslim subjects.
We have several records of early treaties of peace with Christians.
The first is with the city of Jerusalem in 636 (▲• h. 15), the text of
which will be given later on. An earlier convention, of which' the
text is not preserved, was made on the surrender of Damascus in the
previous year, by which every male adult who did not become a
Muslim was to pay annually one dinar (10«. 6d.) and one measure of
com from each field. Of the first treaty made by 'Amr ibn el-'Asl,
the conqueror of Egypt, with the Christians, we fortunately posBees
the complete text, as in the case of the Jerusalem treaty. The ]

* Lane, 8$lect%on$ from th$ Kttr-dn, 1st ed., 70, 71.

Lanb-Poole — Mohatnfnadan Treaties with Christians. 229

of tbe witnesses are given, as in the Jemsalem document, and the
name of the scribe is appended. The two treaties run on similar lines,
and contain not only practically identical clauses, but even absolutely
identical words and phrases. Since we find the name of *Amr ibn
el-^sl among the witnesses to the Jerusalem treaty, it is easily
understood that he carried its terms with him, in memory or in
writing, when he invaded Egypt, and that he endeavoured to accord
to the Christians of Egypt the same terms, mutatis mutandis, as the
Caliph 'Omar had, in his presence, accorded to the Christians of
Jerusalem.

The following text of the Treaty of Miar (or Egypt) is from
Tabarf s Annalss^ in de Ooeje's edition, part i., pp. 2588-9 : —

J«^*>i ' f* J^3 (^>^ (h4^^ j^-^^j ^^y^3 (^=^^ f^^-*''

yj^ '^j^ ^^^^3 ^j^, ')ff^ ^ ^^ij M-op? J *^ ^^dA

*UJs^ U*, ^'U Ji* <li VjJb iTj^^V r«*^ J ^'^ crtJ
*ii iU^ '»^' jf J IU< *j^ U* ^ UilU- ^ ^>u

230 Proceedings of the Royal Irish Academy.

^J !2i*i ^ c>^ L^ "V;* '*^^ ^^J ^b ^*^ ^'^ Ir^ c;'
'»Ui\ xofli^ ^\ Jju^ j^\ Jf^ ifj;!^ ^^ ijcLfi ij\kyj ^ \y

Fonotf lectianes : —

a. C Jr^-Jj, Co ^jU^Jj. *. IH et IK c ^^^

C tjcSLiJ 9 IK mox iyj^ ^»^xfeL-j )• • (?. C et IK s.p., CJo ^,^ ]
C mox t|>^TW . d, IH om. #. C om., IK . ^ „ * . /• C ajjU ,
IH* ^U , IH' ^ jU 5 sed J loco rasurae. y. Co ^^|^ S-'yJt ;
IK rursus iyji. . A. Co om. «. Co i^A* . k, IH add. ^ •
IK haec verba inde a J^ J om. /. IK om. m. IH add. ijyt^^
n. IK j^ et mox liy«jc#j . <>. IH ^^ Uo\ . 1?. IH* jo^^y
apud IH' pimctum litterae * erasum est.

Translation : —

' In the name of Gt>d, the Compassionate, the Merciful.

'This is what 'Amr ibn el-'AsI granted the people of Mifr
in pledge of security for their persons and their religion and
their goods, and their churches and their crosses and their lands
and their waters : there shall not be taken from them anything of
this, nor diminished.*

* Or as de Sacy renders it, more freely, ' on n'attentera li lean droits
relatiyement k aticune de oes cboses, et on ne leur fera ^prourer aaoun tort.*
MSmoires de Vlmtitut {dead, de$ in$et\ et beUei-lettrei), r,, 85 if. (1821).

JjAVK-FooLK—Mohammadan Treaties with Chrutiam. 231

' And the garrisons shall not settle among them.

' And [it is binding] upon the people of Mi^ that they pay the
pell-tax when they come into this Treaty of peace and the overflow
of their river has subsided — fifty millions. I

* And [binding] on them is what their robbers oommit.

' And if any of them refuse [to come into this Treaty], the sum of
the taxes shall be cut down for them [who are liable for it] in pro-
portion to them : and our obligation towards those that refuse is quit.

* And if their river has less than its full rise, then the sum
[of taxation] shall be reduced for them in proportion.

* And whoso of the Bomans and garrisons shall come into their
Treaty, for him is the like as for them, and on him is the like
[obligation] as on them.

'And whoso refuses and chooses to go away, he shall be safe
till he reaches his place of security or departs from our dominion.

* What is [laid] upon them is by thirds, at every third draw-
ing a third of what is [laid] upon them.

' For what is in this writing [stands] the pledge and warranty of
C^od, and the warranty of His Prophet, and the warranty of the
Khalifa, the Commander of the Eaithful, and the warranties of the
Faithful.

' And [it is prescribed] for the garrisons who consent [to this
Treaty], that they shall assist with* so many head and so many
horse that they be not plundered or hindered from commerce to and
fro. Witness E2-Zubeyr and 'Abdallah and Mohammad hia sons.
Wardan wrote that and was present.'

* Do Sacy renden this * foiimir tant d'hommes ei tant de chevaux, moyeniuuit
quol on ne portera point la guerre chei euz ; on quoi ila aeront dispenafa de
r obligation de faiie la guerre saer6e.' The verb Lji may be aettye or paanve.
The whole clau«e is obscure.

232 Proceedings of the Royal Irish Academy.

For purposes of comparison I sabjoin the Jemsalem Treaty* :—

^^^ ^Uj\ Jjb\ ^j^^\ y^\ j^cx. Si JOc ^Ja^^ ^ ^A>

* >■ ^ * ^ -

(^^ 'A L5^ ^U, A-^ ^ y^l^ '<>li ^ ^j^ ^^

Ate

y^'Ujjl >^ J* U Ji* iAaj ^\ 'j|» j»«;^ j»l5\ yy^
l»j^^ ^ <1U, A-ii; ji^, ^J »Uj1 JaI ,j^ t^^ ^j iij^^

"c^^-j (^ L^j (•«-**'^ v^ u**-'^ (^^ (*^j r«*rf ij^j

jL »lA ^j li^\ ^^ AAA Jjh\ ^ U Js«« *J* J j»« ^ 'li
'juMc; ,j^»^ "(ifi^ ^ji ^ *i\i 'i^\ J\ ^frj *^ err? r^ f

y w, ^ y ^

* T. 1., r p. 6, f 1^. 1. For a similar Treaty with Lydda, »ee Tabwi.

Lanb-Poole— JfoAammo^ten Treaties with ChriatiaM. 233

Variae lectumet : —

a. Mod]. Af}ULdL h, Modj. et Soj. \^j» If^JU** ^- Modj. et
Soj. 1|}U <^* Soj. ^^^iljj. #. Modj. et Soj. Uj^ ; Cod.

Leid. Ib^ . /. Codd. \y^Jj ; Modj. et Soj. ^yb/j. g. Modj.
Jlc J A. Modj. et Soj. ^y^ . i. Modj. et Soj. y . *. IH
oni. ; Buppl. e Modj. /. Modj. et Soj. ; codd. om. Ac a • m. verba
spuria? n. Modj. et Soj. \^ . o. Soj. et Modj. om. p. Modj.
etSoj., AijU ^ Modj. ajuc. r. IH' ^.^h:* ; Soj. Jbi^r,
cm. Ji>. . «. Modj. et Soj. t, Soj. clUjj • r. Modj. et Soj. om.

TVamlatien : —

' In the name of God, the Compassionate, the Merciful.

* This is what the servant of Qod, ' Omar, Commander of the
Faithful, gave to the people of Jerusalem [Tli^] in pledge of security:
he gave them security for their persons and their goods and their
churches and their crosses, and its sick and its sound, and all of
its religion : their churches shall not be impoverished or destroyed ;
nor shall [aught] of it be diminished, neither of its appurtenances
nor of its crosses nor of anything of its provisions ; and they shall
not be forced against their &dth, and not one of them shall be
harmed.

^ And none of the Jews shall dwell with them in Jerusalem.
' And [it is binding] on the people of Jerusalem that they pay the
poll-tax as the people of el-Medain pay it ;

* And that they expel the Bomans and robbers from it [Jerusalem] :
and whoso of them goes forth he shall be safe as to his person and
property until they reach their place of safety ; and whoso of them

234 Proceedings of the Bayal Irieh Academy.

stays, he shall he safe and on him [is hinding] the like of that which
[is hinding] on the people of Jerasalem, a poll-tax«

' And whoso of the people of Jemsalem prefers to go away, himself
and his property, along with the Bomans, and leave their churches
and crosses, they shall he safe in person [and churches and crosses?]
until they reach their place of safety*

' And whoso of the people of the land was in it [Jerusalem] hef ore
the fighting, if he wish'to settle, on him [is hinding] the like as what
[is hinding] on the people of Jerusalem, a poll-tax, and if he wishes
to depart with the Bomans or to return to his own people, nothing
shall he taken ^m them [i.e. in poll-tax] until the harvest is reaped.
And for what is in this writing [stands] the pledge and warranty of
God, and the warranty of his prophet, and the warranty of the
Ehalifas, and the warranty of the faithful, provided they pay what
is due of the poll-tax. Witnesses to that, Ehalid ihn WelTd, and 'Amr
ihn el-'AsT, and 'Ahdu-r-Baliman ihn Auf, and Mu awiya ihn Ahi-
Sufran. And wrote and ^as present [x], year 15.'

The dose similarity hetween the two documents will he seen at
the first glance. In hoth we find an assurance of security for the
person, goods, religion, churches and crosses, of the conquered people.
In hoth we have the imposition of a poll-tax on those who do not
conform to Islam. In hoth we have the undertaking that a dominant,
or once dominant, people shall not dwell among them, in the one case
the Jews, in the other the Romans. In hoth the Romans, as rulers,
are to depart, yet if any of them choose to remain as suhjects, they are
to enjoy the same privileges and hear the same hurthen of tax as the
native Christians. Both end with pledges of warranty and the names
of witnesses, and the formula jAt^j (..^c^* 'wrote and was
present,' only the name of the scrihe is given in the Egyptian treaty
hut not in that of Jerusalem. It is evident that we have here two

Ix4NE-PooLB — Mohanimadim Treaties with Christians. 235

formal doonments, drawn up on a standard model ; and I do not think
there can be any doubt of their textual accuracy, subject to minor
▼ariations in different manuacripts. These variants I have appended
to the texts.

It will be noticed that the Egyptian Treaty, with which I am
chiefly concerned, does not in so many words impose a capitation
tax ( Ji^>. ) at so much a head, but states a fixed tribute of fifty
milHons. It does not say millions of what coin, but it must evidently
be dirhems. Abu-SaUh,* writing about a.d. 1200, says that 'Amr
imposed an annual tax of 26} dirhems (i.e. 2 dinars), on all, but
made the rich pay three ardebbs of wheat in addition, and this
is the universal tradition. The conditions annexed, that the tax
is to be paid after the inundation, i.e., in harvest time, and that it is
to be reduced if the Nile is lower than the arerage, seem to point to
a tax upon land-produce; but if, as is dear from all authorities, there
was only one tax, by whatever name it was called, it would in any
case fall upon the land in an agricultural country like Egypt ; and as
at the conquest the whole population was Christian, the Arabs
forming an insignificant minority, the poll-tax would in reality be a
land-tax. In fact there is no evidence that any land-tax was imposed
at the conquest (except at Alexandria), beyond the statement that
three ardebbs were levied from the richer class. The land-tax
( ^ji^ ) was imposed somewhat later. It seems probable therefore
that the fifty millions (of dirhems), equal to three and one third
million dinars, represent a rough guess at the sum which would be
produced by a poll-tax of two dinars a head on adult males. It was,
as a matter of fact, too low an estimate, for the poll-tax soon
brought in twelve millions; but at the time of the treaty, when
only a small part of the country was subdued, and most of the Delta
was still in Boman hands, it was impossible to take an accurate

* Church4s and MomuUrUt of Ejiypi, ed. and tr. EyetU, f. 22a.

236 Proceedings of the Royal Irish Academy.

census of popnlatioii. It is recorded that *Amr only raised one
million dinars in the first year, two millions in the second, and four
in the third year of his occupation of Egypt ; and however we may
distrust this geometrical progression, it indicates at least that at the
beginning the revenue from the poll-tax was incomplete.

Another point in the Treaty which is of the first unportanoe
relates to the Boman gairisons. 'The gairisons shall not settle
amongst them ' : but ' whoso of the Bomans and garrisons shall come
into their treaty, for him is the like as for them.' I wish to draw
particular attention to these clauses because my translation differs
from all previous versions. Hitherto the word mib i^y has been
translated ' the Nubians,' and in my History of Egypt in the Middle
i Ages I followed the received version. But the introduction of
Nubians into a treaty made with the people of the Egyptian Delta at
a time when the Arabs had not even penetrated into Upper Egypt
struck me from the first as curiously unnecessary. We read nothing
in history about Nubian influence or Nubian settiements in Egypt, at
least since the Ethiopian dynasty of thirteen hundred years before.
A passage in Tabarl* set me on what 1 think is the right path. He
says, in reference to 'Amr's arrival at Heliopolis ('Ayn Shems), ' aad
the dominion (or rule, cJl* ) ^^ between the Copts and the IM.*
This apparent omission of the Bomans as the ruling power points
clearly to some other meaning of Nub, It could not be stated
seriously that the government of Egypt was shared between Copts
and Nubians. The phrase puzzled the copyists, for two transcripts
(I H) have a marginal note to en-Nub^ 'perhaps the Bomans,'

Now I need scarcely explain to you that Arabic MSS. seldom
give the vowel-points, and that a word without vowel-points may
mean several different things. Nub certainly means Nubian, but put

* De Goeje*8 text, i. 2687.

ItATSTB-PooLK-^Mohammadan Treaties with Christiana. 237
the vowel-sign fetha over the ^ , and it becomes nuab ^.^j , the
plural of nauba. In the text of the treaty in two places we find the
Taziant en-^umba i)yi\ in place of lm!\ ; and ifJ3\ occurs through-
out in the text of the treaty printed by de Sacy from Abu-1-Mahasin
(quoting Ibn-Kethir) in his Seeend Mhnoire sur la nature $t lesrevolutiam
du droit de proprUU territoriale en Hgypte,* Nauba means primarily a
' turn ' ; hence what is done in turns or takes turn-about, a ' sentinel,'
a ' guard' ; and so it comes to mean a ' garrison.' This last meaning
is common in later literary Arabic ; and Dozy cites it, s. y. (^ J , as
used by el-Bekrl in the eleyenth century. I believe, therefore, that the
true translation of h^ and im^A\ in the Treaty is ' garrison ' and
' garrisons.' This rendering makes the whole document intelligible.
There was no reason to suppose that the Nubians were disposed to
settle in Lower Egypt ; there was certainly no foundation for the
statement that they shared the dominion with the Copts ; and there
seems no object in connecting them closely with the Romans. But as
soon as you substitute ' garrisons ' for ' Nubians ' the whole sense
becomes clear. ' The garrisons shall not settle among ' the people of
Egypt : this was the chief desire of the Copts, for whom the Roman
garrisons were the symbol and agents oft that Melekite or ' Chalce-
donian ' persecution which had made the Roman rule intolerable to
the monophysite church to which the great majority of Egyptian
Christians belonged. Tet, if the Roman soldiers chose to become
peaceful citizens, they might enjoy the privileges of the treaty and pay
the poll-tax : ' whoso of the Romans and garrisons shall come into
their [i.e. the Copts'] treaty, for him is the like as for them, and on
him is the like as on them.' It would be quite unnecessary for the
treaty to lay down such a rule for the Nubians, whose inclusion was
at that time scarcely probable. Precisely the same policy is laid down
in the Jerusalem Treaty, which enacts that the Romans are to depart,

* Mem. d$ VImtitut {Aead. du inter, et beUes-iettres) r, 1 if.

238 Proceedings of the Boyal Irish Academy.

but that, if any of them prefer to stay, they shall be protected, and
have the same conditions as the people of Jerusalem : l« JjL« aJ^j
*LLJ Jjbl ^£. ; compare the words in the Treaty of Mijr, jJL* ili

^»|J^ Lc JjLc aJxj a^ to •

The last clause relating to the garrisons, which comes like an
afterthought between the citation of the warranties and the names of
the witnesses, is not very intelligible. It is translated by Weil* in
the following sense : ' And [it is binding] upon the garrisons who
consent [to this Treaty] that they shall help [the Uuslims] with so
many men [/t^. heads] and so many horses that they [the Nub] be not
attacked nor hindered from trading coming and going.'f I do not see
what other meaning can be made out of Uvftj; • £ven if we disregard
the vowel-point, and take the word as the Ist form imperfect of ^^ ,
instead of the I Vth of _^ , \^0M would mean ' scout ' instead of
* help,' and would come to much the same thing. Ljib again maybe
taken either as active or passive — ^to ' attack ' or ' be attacked.' The
clause may be understood to provide for a limited escort of friendly
Romans to protect the caravans trading between Egypt and Syria ;
but such a provision appears extremely improbable. The Arabs would
scarcely trust the Bomans with sufficient forces to guard the caravan
routes, and would undoubtedly prefer to guard (or plunder) tlie
commerce themselves. It is not dear from the text whether it wis
the trade of the Romans or the trade of the Arabs that was to be
protected ; but if the former, the clause would seem to suggest that
the Romans were to be allowed a small force in self-defence ; and
this appears to be the more probable interpretation of the sentence.

I have called this document the Treaty of Misr throughout, not
the Treaty of Egypt, because, although Misr means ' Egypt,' it also
means the middle capital of Egypt, successively known as Moaphis,

* OctehiehU der Chaiifin, i. 112. t De Saoy*8 rendeiiag is given abovs, p. Ul>

Lanb-Poolb— Jf^Aamnuu/oii Treaties with Christians. 23^

FuBtat, and Cairo. That there was at the time of the Arab conquest
a dty called Misr, and known to the Greek historians as Babylon, the
successor of the partly ruined city of Memphis, is evident from all the
authorities, though its extent is dpubtful. We know only that it was
dominated by the fortress of Babylon which gave its name to the city
in both earlier and later times, and supported by at least two other
fort8« To judge by other treaties, such as those of Damascus,
Jerusalem, and Lydda, it was the custom of the conquering Arabs to
make treaties with a dty, not witb a country as a whole. It may be
urged against this view, that the amount of tribute is altogether out
of proportion to a single city, and must refer to Egypt at large ; and
the reference to ' lands and waters ' also suggests a wider meaning
than Misr the dty. But the same occurred in the case of the Treaty
of Lydda, which was made to include the neighbouring people of
Palestine, but was formally contracted with the town of Lydda. I
think 'Amr made the treaty with the capital of the Copts (ignoring
the as yet unconquered Roman capital, Alexandria), and made the
capital responsible for all the rest of the country. Tabarfs phrase,
however, * So the people of Misr, all of them, entered into that and
accepted the Treaty, and horses were collected,' clUj .J J^tU
Jj-ac\l^ ^^..^.MS^^j ^XJ] tjLij J^ yA* Jjb\ , seems to imply a
general acceptance. The double meaning o£ Misr is a perpetual cause of
confusion, and it would be rash to insist on either interpretation.

It is abundantly evident, however, that this was a treaty witb the
Copts, not with the Komans. The Roman garrisons are mentioned,
but only in a subordinate manner. The people of Misr, not the
Roman army of occupation, still less the emperor Heraclios, were the
contracting parties on the other side. As there is no indication in
the treaty itself that the Romans were consulted in the matter, we

240 Proceedings of the Royal Irish Academy,

must conclude that this treaty was made behind their backs ; that it
was a compact between the Copts and the Arabs without the
authority of the Roman garrisoui though these had the option of
accepting the same terms. Mr. A. J. Butler, in his recent learned
work on The Arab Conquest of£yypt, labours under the extraordinary
impression that this treaty is really the treaty of capitulation of the
Eoman garrison of Alexandria. His words* are : ' But the text of the
treaty is actually given by Tabari, who by a strange confusion calls
it the Treaty of *Ain Shams, instead of the Treaty of Alexandria.'
Mr. Butler unfortunately gives a very inaccurate translation, and
then appends the curious footnote: 'This treaty is premrved by
Ibn Khaldiin, who quotes it from Tabarl ; but it does not seem to
occur in Tabart's extant account of the conquest of Egypt; see
Zotenberg's edition, vol. iii, pp. 461 seq.* Mr. Butler's valuable
work is vitiated in many places by his references to the Persian
abridgment of Tabarl, which not only does not contain a great deal of
the most important passages of the original Arabic work, but intro-
duces errors by compression, and even adds mere legends from Peraian
tradition. As we have seen, the original Arabic text of the treaty
does occur in de Qoeje's edition of Tabarf ; but it is not there called
the Treaty of 'Ayn Shems, and it could not possibly refer to the
capitulation of Alexandria, According to the earliest, indeed the
almost contemporary Christian authority — though unhappily we
possess it only at third hand, and in a distractingly dislocated order —
John of Nikiu's Chronicle, cited by Mr. Butler from a translation of
the Ethiopic version of the Arabic translation of ihe Coptic or Gre<:k
original, the capitulation of Alexandria included an armistioe of

« Arab Conque$t rf EgypU Oxford : at the Clarendon PiMi, 1902, p. 324.

Lanb-Poolb — Mohammadan Treaties with Christians. 241

eleven months, at the expiration of which the Boman garrison of
Alexandria was to quit the dty and depart by sea ; no Boman army
was to return or attempt the recovery of Egypt ; hostages were to be
given by the Bomans for the due execution of the treaty ; and the
Jews were to be allowed to remain at Alexandria. There is not
a word of all this in the Treaty of Misr ; and it obviously has no
connexion with the capitulation of Alexandria.

It was, as is evident from its contents, a treaty with the Copts of
the city of Misr as against the Bomans, rather than with the Bomans.
The questions now arise, when could it have been concluded, and by
whom? Now the Arab historians — ^upon whom alone we have to
rely for events between the capture of the city of Misr and the fall
of the fortress of Babylon, for there is a gap here in John of Nikiu's
Chronicle — are full of reports of negotiations between the Copts and
the Arabs with a view to a peace, which was strongly opposed by
the Boman garrison in the fortress, then the chief position of Boman
power at the apex of the Delta. According to Tabail, after the
Arabs had reached Babylon, there came to meet them, on the part of
of el-Mukawkis (the name they give to the governor of Egypt), a
patriarch (jdth(Mtj catholicus) and a bishop, who, after some fighting
( s Jj\i )) were invited by *Amr to discuss terms : this was beforo
reinforcements had reached the Muslims, and *Amr and his 4000 men
were apparently in a precarious position. The discussion was of a
friendly nature in regard to the Copts, for whom, it was said, the
Prophet Mohammad and the Muslims had always entertained a kindly
feeling, on the atavic ground of the Arabs' descent from the Egyptian
bondmaid Hagar* *Amr offered the usual terms : those who embraced
Islam should be the equals of the conquerors and enter the universal
brotherhood of the Muslims, and those who refused should pay the

242 Proceedings of the Royal Irish Academy.

poll-tax (/isya). The two ecclesiastics were disposed to accept, and
returned to el-Mnkawkis to rep<»i; the negotiations. But Aretion
( ^^^fl?;^ in the Arabic, happily emended as ^^^^ ^7 ^^* Butler
— ^Aretion had previously been governor of Jerusalem), the Roman
governor of the fortress, rejected the proposals absolutely and gave
orders for an attack. The patriarch and bishop then said to the
people of Misr: *We will try to ward off evil from you, but we
cannot return till four days hence.' They had to go to Alexandria,
one presumes, to consult el-Mukawkis, and apparently they broo^t
him south with them, for he was present in the fort when the Arabs
laid siege to it. Meanwhile *Amr joined by reinforcements under ex-
Zubeyr and others camped at Heliopolis ('Ayn Sh^ns). Tabail does
not mention the battle of Heliopolis by name in this connexion; but
he afterwards speaks of the encounter of 'Amr and el-M ukawkis at
'Ayn Shems, and John of Nikiu gives a detailed account of the battle,
which, he records, was followed by the fall of Tendunyas (Umm
Buneyn), a fortified place on the site of the later mediaeval Maks
and the modem Ezbeklya quarter of Cairo ; and this involved the fall
of the city of Misr, which is recorded by John of Nikiu merely in the
heading of a chapter. Tabarl goes on to relate how the people ol
Misr, alarmed at the approach of the Arabs, entreated their rul^
( ft^L* ) to make terms with them, but he refused ; ' and this was
the fourth day' ; so there was a battle ( ^jb Jjlii ), and after the
victory ez-Zubeyr scaled the wall and opened a gate — Tabail does
not say of what city or fortress — whereupon the people came to sue
for peace, and the Treaty of Misr was concluded.

The mention of the fourth day, when the patriarch and bishop
were expected to return, points to an armistice, and shows that the
Romans were awaiting the return of the ambassadors. It is not

Lamb-Poolb — Mohanmuulan Treaties with Christiana. 243

lecoided that the ecclesiastics took part in the treaty, bnt Tabaif !
mentions their reappearance immediately afterwards to arrange about '
the prisoners. Who they were it is impossible to say. TabarT and
other Arabic writers give them the impossible names of Abu-Maryam
and Abu-Maryam; and Mr. Butler regards Abu-Maryam as a cor-
ruption of Abu-Miyimln, which itself is an Arabic penrersion of
Benjamin. Is it possible that Benjamin, the monophysite patriarch
who had been driven into hiding by Gyrus, the Melekite patriarch of
Alexandria, but who was still aliye, and was afterwards reinstated,
came out of his retreat near Kus in Upper Egypt to help his people
to throw over tbe Roman yoke ? Or was Abu-Maryam Cyrus himself ?
TabarT's story fits perfectly with the contents of the treaty, which
is thus shown to be a treaty with the Egyptian people against the
wish of the Boman army of occupation. Tbe authority of Tabarl as
a careful compiler of attested traditions is very great, indeed almost
absolute in Muslim acceptation : and this story rests on a chain of
traditionists running up from es-Sarl through 8hu*eyb and Sey f to
Abu-Haritha and Abu-'Ottiman (|_^.«^ ^^ jj?-J^ /J^ \^-^^

A\ Iti j^Lolft ^\^ fi^U. jI\ ^ <— f^^ j^ )' I* is not a record
to be lightly set aside.

The most widely accepted story of the surrender, and the most
detailed, is given by el-MakrTzl.* It must not be inferred from the
lateness of Makrlzl's date (he wrote about 1420) that his account is
necessarily of little authority. He was a laborious compiler from
good sources ; and he had at his disposition manuscripts of early works
which have since disappeared. His account rests upon traditions
which may go back — some certainly do— to early times, and it is

* Khitat, i. 289-294.

R.I. A. PKOO., YOL. XXIT., SBC. C] [19]

244 Proceedings of the Royal Irish Academy.

confirmed by much earlier writers. MakiizI first relates the storj
in brief, and then gives a detailed narrative. Probably these two
accounts come from different sources, for it was the usual habit of
Arabic chroniclers to set down the different accounts one after the
other with little or no attempt to reconcile them. These two
accounts in Makrizl, however, agree. The short account says that
after 'Amr had been reinforced by a body of 12,000 men under es-
Zubeyr he laid siege to the fortress; that ez-Zubeyr scaled the wall
and captured the fort ( .nr- ),* and seized a gate ; and that el-
Mukawkis in alarm sued for peace, which was concluded on the basis
of a tribute of two dinars a head from the Copts. His is practically
Tabarl's account. The longer narrative relates how el-Mukawlds,
after a month's fighting, discouraged by the perseverance and energy
of the Arabs, left the fortress of Babylon in company with the
leaders of the Copts, and took up his position in the opposite island,
now called er-Bawda, cutting the bridge of connexion. Then fearing
that the fortresses would &11, he opened negotiaUons with the Arabs.
He urged that the Bomans were far more numerous and better
equipped than the Muslims ; that the Nile was high and hemmed in
the invaders ; and that their wisest course would be to come to terms
before the Komans overwhelmed them. Hia object was evidently to
get easy terms before the decrease of the inundation set the Muslims
free for wider operations. 'Amr kept the envoys two days and nights,
and then sent them back with the usual alternatives : embrace Islam
and be our brothers ; or pay the poll-tax and be our inferiors ; or
•else fight till Ood decides the issue. £1-Mu)uiwkis asked the envoys
to describe what they had seen during their two days' visit to the

* Not necessarily the Castle of Babjlon.

Lane-Poolb — Mohammadan Treaties with Christians. 245

MnalimB' camp ; and they answered, ' We saw a people who love,
•eyerj one of them, death more than life, and set hnmilitj above
pride, who have no desire or enjoyment in this world, who sit in the
dnst and eat upon their knees,* and their commander is like all the
rest; yon cannot distinguish the strong from the weak, nor the
master from the slave.' This report increased the dread which the
Arahs inspired. The negotiations were continued on the island of er-
Bawda ; but el-Mukawkis could obtain no modification of the terms.
Fighting with the garrison of Babylon was accordingly renewed ; but
finally el-Mukawkis persuaded the people that resistance was hope-
less, and 'Amr's terms were accepted — a poll-tax of two dinars a head,
except from old men, children, and women, with three days* mainten-
ance for the Muslims.

In spite of superficial differences, Makrizl's story tallies with
Tabari's. In each there is the contrast between the willingness of
the Egyptians to treat and the stubborn resistance of the Boman
garrison. In each we find the capture of a fort and gate to be the
decisive event which hastened the conclusion of the Treaty. In each
it is essentially a treaty with the Copts, not with the Bomans, though
Bomans who submitted were included. Makrizl's statement that
the negotiations took place during high Nile, coupled with the remark
that they began after there had been a month's fighting at the
fortress, though it does not agree with Jabarl's '* four days," shows
that this treaty must have been made about October, 640. It cannot
therefore refer to the final evacuation of Babylon, which is definitely
£xed at 9 April, 641. The capture of the fort, ^^^.o^K must evidently
be distinguished from the fall of the castle, j^\ , and must represent

* Mr. Butler's translation of mj^j Ac • ' on horseback,' is obTiously a
jnistake. *

[19*]

246 Proceedings of the Royal Irish Academy.

only a partial lodgment of the Arabs in the fortifications or even in a
neighbouring fortress. It has generally been assumed that there va&
but the one castle of Babylon to bie taken ; but it is clear that this was
but a part of the fortifications of Misr. We hare already seen that,
according to John of Nikiu, Tendunyas (Maks) was a fortified post;
and it is possible that ez-Zubeyr's scaling of the fort may refer to
what Mr. Butler describes as the second capture of Tendunyas.
Makrlzi mentions another fortress besides Kasr esh-Shema' (the well-
known fortress close to Cairo which is generally identified with the
Castle of Babylon) ; and this other fortress, which was situated on a
rocky hill to the south-east of Kasr esh-Shema', and was within the
city, was particularly called the fortress or palace (^ ) of Babylon.
Remains of this other fortress may possibly be represented by the
massive walls on the southern part of the hill, afterwards known as
•Antar's Stable.*

We have seen, therefore, that this Treaty of Misr was concluded
between 'Amr and el-Mukawkis on behalf of the Copts about the
month of October, 640. It was a treaty of surrender for the whole
country, but the Roman garrisons remained unsubdued. Hence the
clause ' The garrisons shall not settle (or dwell) among the people of
Misr,' a clause to which the Romans were obviously no party.
Makrlzi, however, now enters upon a fresh division of the subject,
introduced by a fresh chain of tradition,! dating back through Ibn
Lahl'a to Yahya ibn Meymun. According to this tradition,
el-Mukawkis stipulated for the Romans that they might choose
whether they would stay in Egypt on the some terms as the Copts,
or whether they would rather go to their own country, which they

♦ See Lane, Cait^ Fifty Years Ago, 146, 147.
t Khiiat, i. 293.

Lai«e-Poole — Mo/Mfnmadan Treaties with ChriBtiam. 247

were free to do, including the Bomans of Alexandria and the parts ronnd
about. And it was agreed in writing that el-Mukawkis should write
to the emperor to inform him of what he had done ; and if he accepted,
the treaty was good. Heraclius's reply was naturally a repudiation
of the treaty. He pointed out the small numbers of the Arabs
compared with the Romans and the Copts, and ordered hostilities to
be resumed. Upon this el-Mukawkis, convinced that resistance was
useless, went to 'Amr, and begged of him three things : first, ' do not
break faith with the Copts, but count me as one with them, and on
me be binding what is binding on them, for my word and theirs
agreed upon what thou didst covenant, and they are fulfilling towards
thee what thou wishest ; secondly, if the Bomans after this sue for
peace, make no peace with them tOl thou hast made them confiscate
and slaves ; . . . and thirdly, I beg of thee when I go to my rest to
have me buried at St. John's at Alexandria." And 'Amr agreed
to these requests. It is true that HakrizT, in another part of his
work,* gives the same three requests of el-Mukawkis in slightly varied
words, on the authority of Ibn 'Abd-el-Hakam, in connexion with
the conquest of Alexandria. Such confusions are unhappily too
common in regard to many events in the Arab invasion of Egypt.
But the three requests, to whichever date they belong, show clearly
enough that el-Mukawkis and 'Amr held by the Treaty of Misr
which had been concluded with the Copts, and that the Romans put
themselves outside the treaty. The ninth-century writer, Ibn 'Abd-
el-Hakam's account of that treaty, as cited by MakrTzT, closely agrees
with what has already been related, and the learned geographer

• Ibid,, L 163. Mr. Butler says, * Here we get back to an earlier Tersion' :

but Ibn 'Abd-el-Hakam ia a hundrod years lat^r than Ibn Lahfa (t 164 a.h.)

The latter moreoyer waa a f amoua traditioniit, as well as chief Kadi of Fustat.

' • • • •

248 Proceedings of the Royal Irish Academy.

Ylkut, a Greek, who wrote his great work in 1225, uses the iden-
tical words employed by Makrizi in parts of a practically identical
narrative, so far as it goes.* He adds that the Treaty was made by
el-Mukawkis for the Copts and the Eomans ; but while the Bomans
had the choice of assenting to it or not, according v^ their emperor
should decide, the Copts had no such ehoiee ( .L^ ^JLi kJiSt U\ •)•

This treaty with the Copts, which both the Arabs and the Copts
upheld against the Eomans, is, as we have seen, ascribed by MakrizI,
and by much earlier writers, such as Ibn *Abd-el-Hakam and el-
Beladhurl, to el-Mukawkis as representative of the Copts. Makiizi
describes him as over both the Copts and Romans f y \^. a« J1 . . •

ijJLLoJI Al^r * (•4'^9M ^il^ )-t ^^^ ^^ Mukawkis was has been
a puzzle to all historians. He is called, by Arabic writers, either
'the Boman ' ,v«iilf ^^ ' ^^ Greek ' jV^^J^ > ^^^ ^ name is gives
either as * George son of Miiul ' Lu* ^ <^SLp* » or * son of
Eurkub ' c^y ^ • Tabarl and Beladhuri give bim no name
beyond el-Mukawkis — a word which is explained as meaning ' ring-
dove' in Arabic, but which is probably not Arabic — and John of
Nikiu does not mention him by name. Professor Karabacek{ interprets
the names as George, son of Mina Parkabios, makes him both strategos
and pagarch, and thinks the title Mukawkis may represent the Greek
/icyavx^T^ — & title, however, which he has invented on a rather loose
analogy with titles, such as ^Ffio^oTaros, found in papyri of the Boman
period. Mr. MlLneg identifies him with George the prefect mentioned
by John of Nikiu. Professor Bury|j follows KarabaSek, but not in the

• Mu'jam-el-Buldan, 8.v. t\Ljd\, iii- 894-5. f Khitat, i. 290.

X Pap, Efxhernog lieifter, i. 1-1 1 .
§ Sffypt undei' Soman Jiule, 224.
II Ed. of Gibbon, v. appendix, 540.

Lane-Poole — Mohammadan Treaties with Christians, 249

acceptance of the asanmed Ghreek title /ic/avxijs. Finally, following
the lead of the Fortagnese scholar Fereiray Mr. Butler, in his Arab
Conquest of £gypt^ after an elaborate examination of the authorities,*
has come to the conclusion that el-Hukaw]^is was none other than
Cyrus, the Melekite patriarch of Alexandria.

The evidence he relies upon for this theory consists partly in state-
ments by Coptic writers; partly in coincidences between acts attributed
to el-Mukawkis by one set of historians and acts attributed to Cyrus
by another set of authorities. The statements of Coptic writers are
these:

1. Seyerus, bishop of TTshmuneyn in the latter part of the tenth
century, in an Arabic work on the lives of the patriarchs, which
has not yet been printed, says, ' When Heraclius had recovered his
territories, he appointed governors in every place. To us in the land
of Egypt Cyrus was sent to be governor and patriarch together.'
Referring to the ten years' persecution of the monophysites, he says,
' These were the years during which Heraclius and Al Mukaukas
were ruling Egypt ' ; and again, ' When the ten years of the reign of
Heraclius and the misgovemment of Al Mukaukas were over.' He
speaks of 'the misbelieving governor, who was both prefect and
patriarch of Alexandria;' and he makes the ex-patriarch Benjamin
speak of 'the time of the persecution which befell me when Al
Mukaukas drove me away." It should be added that the Greek
historian Theophanes (9th c.) also makes Cyrus at once patriarch and
prefect.

2. The Coptic Synaxarium, quoted by Am^lineau, says, 'The
Mukaukas was head of the faith of Chalcedon, and hajl been made

* Arab Conqustt of Egypt, App. 0, 508-526. He uses the Ethiopic vocalization
Mukawkas, instead of the Arabic Mukawkis.

250 Proceedings of the Boyal Irish Academy.

ruler and patriarch over Egypt ' ; and the Ethiopic Synazariuin con-
tains the wQrdfi ' The Mjikaiikas, that is to say, the governor and
archbishop of Alexandria and all the land of Egypt.'

3. In the Coptic life of Samuel of Kalamun, of which a tenth-
century fragment is preserved in the Bodleian, and of which the
original would appear from internal evidence to have been composed
before the death of Cyrus, a curious story is told of the patriarch's
visit to a monastery ; and incidentally he is described as niCA.*VXIOC
nence-rrO^pXHeniCKOnoc, or *the kauehios, the false arch-
bishop.' In this Coptic word — ^if it be Coptic — kauehios Mr. Butler,
following AmSlineau and Pereira, sees the original of the title
Mukawkis. The explanation is a case of ohseurumper obscurius^ for
no satisfactory meaning has so far been found for kauehios ; and ICr.
Butler himself hazards three distinct conjectures — ' Caucasian,*
* Cholchian,' and * paederastian/ The obscurity of the meaning,
however, does not affect the argument; if kauehios be the origissl
of Mukawkis, then this Coptic document makes Mukawkis and Cyrus
one person. But it is far from certain that kauehios is the Coptic
original of the Arabic or Arabicized title Mukawkis.

Supposing these translations to be accurate, and supposing the xbb.,
which are chiefly late, to be faithful transcripts of early authoritative
documents — a matter which I am not qualified to decide — these
extracts taken together show that Cyrus and the Mukawkis were one
and the same person in the opinion of the writers. This can hardly
be contested. The only question is whether the writers were
authoritative. Severus was ignorant of Coptic, and not Tery trust-
worthy,* and he wrote late in the tenth century, later by a hundred

* Butler, Arab Conquut^ ziy, xtu.

luLsi&'VooLE^Mohammadan Treaties mth Chrieiiafis. 251

yean than Beladhurl, and fifty or sixty yean than Tabari. By
himself, I do not think his evidence counts for much. The Synaxaria
are thus described by Mr. Butler:* 'Every [Coptic] church has
specially attached to its service a book called in Coptic *' synaxiLr,'
i.e. oiivajapiOK, or lives of the saints, from which a portion is often
read at matins, in accordance with a very ancient custom sanctioned,
far instance, at the third Council of Carthage in 397 A.n. This book
corresponds closely to the passional of our English churches, from
which the lessons at matins were sometimes taken, or to the martyr-
ology, which was read at the end of prime-song. The synax^ is
confined within the sacred walls, and then is no copy of it in any
private person's possession. It has, of course, been nndered into
Arabic for use at service : and the legends printed at the end of this
work, which an from the Arabic venion, will serve to give an idea of
the miraculous traditions to which the faithful still listen with un-
questioning nverence.' This does not give a very high position to the
synaxaria as historical authorities ; but, as in the case of Severus, it is
possible that genuine historical data may be included among much
legendary garbage.

Such is Mr. Butler's positive evidence. The coinddenoes upon
which he also relies an the statements on the one hand that Cyrus,
on the other that el-Mukawkis was Governor of Egypt under
Hdraolius ; the statements of the Greek historians and John of I^ikiu
that Cyrus made peace with the Arabs, and those of the Arabic
historians that el-Mukawkis made peace with them. But these
coincidences may be explained by the hypothesis that el-Mukawkis
was the sub-governor who made the peace, and Cyrus the patrianh

• Coptic Churches of Eyypt, ii. 259, 260.

252 Proceedings of the Royal Irish Academy.

and supreme governor who accepted his subordinate's arrangement and
reported it to the emperor.

The whole question really turns on the respectiye credibility of the
two or three Coptic authorities and the whole series of Arabic historians.
Now Mr. Butler himself admits* that ' the historical value of these
Coptic documents is not very great. The writers were set upon
recording matters of Church interest — ^the more miraculous the better
— and their minds were almost closed to the great movements of the
world about them.' And referring to Severus, he adds that this
historian mentions ' that he had recourse to some Copts to get Greek
and Coptic documents turned into Arabic, as the two former languages
even then were unknown to most Christians. This is interesting as
showing the state of decay reached by Coptic and Greek, and as showing
Severus' own ignorance of both languages. Indeed the evidence as
regards Coptic is so remarkable as to seem barely credible.'f It is
clear, then, that the Copts as a rule got their historical information
through the Arabic. In studying Arabic chroniclers Severus would
find that el-Mukawkis made a treaty of surrender to the Arabs ; if he
read Tabarl, as he probably did, for Tabarfs work was a standard
authority in the I^timid library at Cairo, and Severus was a persomt
grata at the I^timid caliph's court, he would also find that a
eatholieos came to 'Amr and treated for peace. He might naturally
put the two statements together, and being a Jacobite bishop not
averse to believing every evil of a ' Chalcedonian ' patriarch, he might
very well saddle Cyrus with the shame of betraying Christian Egypt
to the Muslims. As soon as we realize that the Arabic sources were
older than Severus, and were probably under his eye, and that he

• Arab Coftquest^ x. t iW</., idt.

Lanb-Poolb — Mohammadan Treaties toith Christians. 253

could not read any langaage but Arabic, it is easy to see hour he
might perrert or nusunderstand the sufficiently confosed and obscure
narratives of the Arabic chroniclers. Whether the same argument
would apply to the Synazaiia depends upon their dates, on which we
have at present no information.

If we had nothing but these Coptic and Ethiopic data to go upon,
the identification might perhaps be taken as proved. But when we
look at the long series of Arabic writers, not only those whose works
survive, but many who are dted by survivors, but whose original
writings are lost, and when we fail to find the slightest hint that any
one of them suspected el-Mukawkis and Cyrus to be the same person,
I confess that their evidence, negative as it is, seems to me over>
whelming. How is it that not one of them says that el-Mukawkis
wuH a priest, much less an archbishop ? Why do they give him- the
name of George son of Mliul or son of Kurkub, if his real name was
Cyrus ? Why does Abu-I^Qih, who was a Christian, and wrote about
1200 A.!)., state that Heradius placed the government of Egypt
under ' Oeorge the son of IfTna el-Mukawkis,' and also cite ' the book
of el-JaxuLh ' for the fact that ' the bishop of the Romans at Misr and
Alexandria was named Cyrus * ? How is it that not a single historian
of Egypt, Muslim or Christian, has ever said in so many words
' el-Mukawkis was a title or nickname given to the patriarch Cyrus ' ?
It is incredible that such an identity — surely a striking fact if true —
should have escaped them all. And against this solid wall of negative
evidence that no Muslim historian, no Christian historian, not even
the almost contemporary John of Nikiu, mentions this identity, are we
to accept two jottings in two church office-books, the date of which
is not given, and a not very definite incidental statement of a tenth-
century Copt who did not know Coptic ?

^54 Proceedings of the Royal Irish Academy.

Moreover, if el-Mukawkis was Cyrus, who was not sent to Egypt
omtil 631, what becomes of the misaion which the Prophet liohammAd
sent in 628 to < el-Mukawlds, lord of Alexandria'?* Mr. Butler
thinks that this is merely a case of applying a later name to an
earlier governor by mistake ; but it must be remembered that in reply
to Mohammad's mission, el-Mukawkis sent him presents, including
two Egjrptian girls, one of whom, Mary, was received into the
Prophet's hanm and bore him a son. There was every reason for
preserving accurately the name of the man who gave a wife (or
concubine rather) to the Prophet ; and Mary herself and her fellow-
slave would not be likely to forget it or to fail to make it known.
The Mukawkis of 628 may very well be the same person as ths
Mukawkis of 640, but he cannot be Cyrus.

Apart from this silence of the chief authorities, the inherent
improbability of the hypothesis must be considered. Cyrus was
patriarch and civil governor, but not military prefect : yet we find
him (if he be el-Mukawkis) commanding at the battle of Heliopolis.
When the treaty was repudiated by Heradius, el-Mukawkis, accord-
ing to the Arabic tradition (reported by so early an authority as
Ibn-Lahf a), threw in his lot with the Arabs ; but Cyrus, according to
the Greek historians, was recalled to Constantinople and castigated.
That he should have returned at all to Constantinople, knowing what
he had to expect, after making his peace with the Arab conqueror,
seems preposterous. Cyrus finally came back to Egypt, and arranged
the capitulation of Alexandria in October or November, 641 ; he had
now accomplished the insidious plan attributed to him by Mr. Butler,
and he lived five months longer: why do we hear nothing of his

* Tabari, i. Id59-6U Ibn-Hiaham, Wiistenfeld'a trans., 318.

Lane-Poole— ifoA/imma^ii Treaties with ChtHsiians. 255-

reward for bis treachery from hiB Arab allj? On the contrary,
according to Mr. Butler, tbe only request made by CyruB to 'Amr was
apparently refused. Certainly the ambitious patriarch took little by
bis treachery, if indeed treachery it was. Looking at the transaction
in the cool light of history, it has more the aspect of wise submission to
the inevitable.

Admitting, as we must, that Cyrus was recalled and reprimanded
for concluding the Treaty of Misr, is it necessary to hold that he was
the sole negotiator ? Supposing that the catholicos who according to
Tabarl came to 'Amr and treated for peace was Cyrus, we are told
that he went away to report the negotiations to el-Mukawkis. Now
if el-Mukawkis was the military prefect, or eames limitis Aegffpti* it
was essential that he should be consulted by the civil prefect before
peace could be concluded. According to Mr. Butler, who follows the
indications of John of Nikiu, Theodoras the military prefect was at
Alexandria at the time of the Arab invasion ; was then brought ta
Babylon by Cyrus ; and commanded at the battle of Heliopolis. I^ow
this is exactly what is related of el-Mukawkis by Tabarl. El-
Mukawkis was absent from Babylon when the catholicos was treating
with 'Amr. He appeared at Heliopolis, where the catholicos also
appeared after the battle. He was the commander who corresponded,
so far as we can see, with the military prefect. So far as the Arabic
evidence goes, except for his names, el-Mukawkis may have been
Theodorus.

This only illustrates the extreme doubtfulness of any identification
of the mysterious Mukawkis. Until further evidence is obtained,

* This is the Imter title of the military commander formerly styled dux Aegypti,
See Milne, Egypt under Boman Sui€, Note Yxiit 215, and cp. 181.

256 Proceedings of the Royal Irish Academy.

such as we may hope for from the constantly increflsmg disooTeries
of papyri of the Boman period, it seems rash to attempt to fix his
identity. That he was a military gOTomor of high rank, and that he
<»ncluded the first treaty hetween Muslims and ChiistianB in Egypt,
with the concurrence of the patriarch Cyrus, is all that can he
affirmed with certainty.

[ 257 ]

XIV.

SOME MONUMENTS OF THE LA T^NE PEEIOD RECENTLY
DISCOVERED IN IRELAND.

By GEORGE COEFET.
Read Kotucbbb 9, 1908.

[Plates XVm.-XXII.]

Thk late Sir A. W. Franks was the first to distinguish in a definite
manner the antiquities of what he called the Late Celtic Period.
The term ''Late Celtic" was introduced by Franks with reference
to Britain, to denote the period preceding the permanent occupation of
that island by the Romans, dating from about 200 b.c. to the middle of
the first century ▲.!>. It cannot be strictiy applied to the Continent
4>r to Ireland. Franks' conclusions were published in Kemble's
^'HorsB Ferales" in 1863. He then wrote that in this class of
antiquities " the British Islands stand unriyalled ; a few ancient
•objects, analogous in design, may be found in various parts of the
Continent, and more extended researches in local Museums may bring
4>thers to light, but the foreign contributions to this section are
scanty when compared with those in our own country."^

Since that was written our knowledge of the antiquities of the
period has been greatly extended, especially on the Continent. The
^'foreign contributions" are no longer scanty; and although the
magnificent Late Celtic shields found in the rivers Witham and
Thames are still unrivalled, the foreign finds far exceed in number
those of Britain. This we should naturally expect to be the case, for,
speaking generally, the style may be described as Gaulish, and repre-
sents on the Continent the period of the historical Celts dating from
about 400 B.C. to the Roman conquest of Gaul.

On the Continent the period is known as " La Tdne,'* so called
from the site of a Helvetian oppidum on the Lake of Neuchatel — La

> " Horn Fenlet,*' p. 172. Franks took a wider view of the subject in s later
paper, Ardueologia, vol. xlv., p. 265.

258 Proceedings of the Royal Irish Academy.

T^ne : the Shallows — ^where this class of antiqnitieB first attracted
prominent notice.

A threefold division of the period is now recognised into Early,
Middle, and Late La Tdne (or, as M. Eeinach has proposed, forhreTity,
La T^ne I., La Tdne II., and La Tdne III.), dated approximately:
Early, 400-250 b.c. ; Middle, 250-150 b.c; Late, 150 b.c., to the
beginning of the Christian era.^

In Early La Tdne the treatment of the ornament is much freer
than in the later periods, and the influence of classical elementB,
especially the Greek anthemion, may be traced. In the middle and
late periods a progressiye geometrical conventionalization is apparent,
nntil in the late period the classical elements are completely absorbed
or are eliminated.

In England the Late Celtic style was submerged by Boman art, but
not wholly destroyed ; it reacted on Roman art locally, and re-emerged
as a native style in Saxon times, reinforced from Ireland and Scotland.
In Ireland its history is continuous into the Christian period.'

It has been a habit of mind with English archaeologists to regard
the periods in Ireland as later than, and the styles as derived from,
Britoin. This view was expressed in an extreme manner in a recoln-
tion passed by the Society of Antiquaries of London, November 28tb,
1901, in connexion with the recent controversy over the gold
antiquities found at Broighter, County Londonderry. The resolution
contained the statement that these antiquities, which were ascribed to
the close of the La Tdne period, were '' remains of the art of tbe
ancient Britons," and had "only an accidental connexion with
Ireland."*

We need not take this attempt to make archaeology by resolution
seriously. The general impression, on the subject of which the reeo-

1 Tisohler, '< Uber Gliedening der La T^e*Periode/' in CorrospondaDS-Blatt
der deutacben AnthropoIogiBchen GesellBobaft, 1885, pp. 157-172 : Hooteliitf:
Cong Pr^hifltoriquei, Paris, 1900, p. 363. Also p. 427.

' In Graul important political changes appear to bave occnired between Ls T^
II. and La T^ne III. Continuity of burial customs is broken, inhumation and chariot
burial is replaced by cremation. A similar difference has been noted in Briuia-
Tbe earlier class of Late Celtic interments are represented by burials sneh si st
Arras. Yorksbire, where tbe skeleton was laid with the chariot and hoiMB, and tk^
later by the Late Celtic um-field at Aylesford, Kent. See " Note sur To^tiiiB
de Bibracte/' Congr^s Prfibist., Paris, 1900, p. 418; and Evans, "LateWtk
Um-field at Aylesfoid," Archaeologia, vol. lii., p. 386. We tare not yet "1
information from Ireland on this branch of the suljeot.

* The Times, Nov. 29, 1901.

CoFFRY — Monuments of the La Tine Period. 269

Intion is an expression, is based on the assamption that Ireland being
more remote from the Continent than Britain, was less within reach
of Continental influence in early times. I have combated this view
elsewhere.^ The fact that Danubian types, such as the conical caldron
and late Bronze Age swords, are so well represented in Ireland, in
itself contradicts the assumption. The geographical argument must be
used with caution and knowledge. Trade does not necessarily follow
the lines of nearest geographical contact. It is chiefly determined by
the objects desired, and conyenienoe of transit and of market centres.
The frequent intercourse between Ireland and Oaul in early Christian
times, fifth-seyenth centuries, need not be insisted on. The chief
point of landing appears to have been the river Loire. The central
lands of France, to which the Loire gives ready access, were much
frequented by the first Irish Christians. We hear of them at Auxerre,
at Autun (close to the ancient Bibracte), at Luxeuil.' It was from
Nantes that St. Columbanus was deported to Ireland in a ship <' qua
vezerat eommereia eum JZiB^mia."* In Roman times Ireland was
believed to lie between Britain and Spain, and is mentioned as
"favourably situated as regards the Gallic Sea."^

The Hallstatt sword can be traced westward across Gaul, and has
lately been found as far west as Poitou.' We have possibly an
indication here that the Loire was a point of departure for Ireland as
early as the end of the Bronze Age.

It is not, however, the purpose of this Paper to discuss the question
of trade routes, but to describe a new class of La T6ne monuments
recently discovered in Ireland, the first examples of La Tine carving
in stone, I believe, which have been brought to light.

Some two years ago Lord Walter Fitz Gerald showed me a rubbing
of a stone he had discovered at Mullaghmast, in the.County Kildare.
The carving on the stone was of the form we are accustomed to call
in Ireland trumpet-pattern, and I was at once struck by its early
character. As the stone had been removed from its original position,
I urged on Lord Walter the importance of securing such an interesting^
monument for the Museum. He gladly undertook to do so if possible.
At his instance the stone was presented to the Koyal Irish Academy

1 Journ. B.S.A.I., 1896, p. 28.

'Margaret Stokes, Tiana. B.I.A., voL xzx., p. 286.

s Reevee's ** Adamnan," p. 37.

« Tacitus, " Agricola," cap. 24.

* Bevue Arcb6ologique, 4 ser., vol. ii. (1003), p. 56.

K. I. A. PBOO., VOL. XXIT., 8BC. c] [20]

260 Proceedings of the Royal IrUh Academy.

\tj the daughtera of the late 8. W. Haughton, of Carlow, ownen of
the property, and has been placed in the National Collection.

Last year Mrs. Coote, of Garrowroe, Boecommon, sent me photo-
grapha of a stone at Castle Strange, near Koscommon, which she
thought might be of interest. Here was another stone carved with
trumpet-pattern. This time the La Tene character of the ornament
was unmistakable. This stone appeared to me to be so important that
I forthwith determined to visit it, and make a cast of it for the
Museum. Mr. Coote gave me eyery assistance ; and with his help 1
was able to take a mould of the stone in plaster, from which a cast
has been placed in the Museum. I should mention that Mr. John
Byrne, the present tenant of Castle Strange, spared no trouble for lu,
and most kindly undertook the paoking and forwarding of the mould
after we had Ic^

I had heard some time previously of a stone near Loughrea, in the
neighbouring county of Galway, which was said to haye carving on it
of spirals. I had endeavoured to get a photograph of the stone, but had
not been suocesaful.

On seeing the stone at Castle Strange, I lost no time in visiting
the one at Loughrea, on the chance that it might be of the same dass.
I was surprised to find that it was the most remarkable example of
the three, richly carved with La Tine ornament in bold roliei Mr.
Dolphin, the owner, readily consented to a cast being taken, which has
been placed in the Museum.

It vnll be convenient to describe these stones in reyetae order of
disooyery. I shall theref oro take the Loughrea stone first.

It stands in front of Mr. Dolphin's house at Turoe, about three
miles from Loughroa, Ordnance map, 6-inch sheet 97. It was moved
to its prosent position by Mr. Dolphin's father some fifty years ago.
A small fort a short distance to the west of Turoe House was pointed
out to me as the place from which the stone had been taken. It was
said to haye been inside the fort. Subsequently an old man, said to
be the oldest inhabitant of the locality, brought me to the exact spot,
as well as he could remember, from which it had been taken. This
proved not to be within the fort, but some distance to the west of the
fort, towards the bottom of the slope on the top of which the fort is
placed. The old man's rocollection was'quite clear that the stone wai
outside the fort. Thero is, therefore, no reason to suppose that the
stone had any connexion with the fort.

In its prosent position it stands 4 feet aboye ground, and measures
3 feet and 2 feet 4 inches at the sides. It is an erratic boulder of

CJoFFBY — Monuments of the La Tine Period. 261

granite. The earring is very distinct and well preserved. The orna-
ment does not require description ; it is folly shown in the accom-
panying illustrationB (Plates xix., xx., zzi., fig. 1). These are from
photographs of the cast in the Museum, painted in parts to bring out
the pattern. An untouched photograph of the stone itself is shown
(Plate XYin.).

The carving of this stone is, I think, distinctly early. The treat-
ment of the ornament is free, not constrained geometrically as in late
La TSne, especially the examples found in Britain and Ireland. The
derivation from the Oreek antiiemion can still be traced.

The fret pattern is rarely associated on the same object with
La T^ne ornament. In the preceding or Hallstatt period, the fret
occurs frequently. It is usually simplified to plain rectangular forms.

Fig. 1.— Fret-pattems from bronxe vesaelB, Hallstatt cemetery ; the fret is also
found on Hallstatt pottery.^

Again, on a sword-sheath of La Tdne I., found at Halstatt, we find
the simplified fret.' The simplified fret, often in the form of fragments
consisting of single steps, occurs also on the pottery found in the
Gaulish cemeteries of the Mame. There it appears to be a pottery
tradition, and the period is abundantly established as La Tine I.-II.,
chiefly La Tine I. The higher forms of the fret are not found on the
Mame pottery ; so that these single-step patterns, or fragments of
meanders, may be considered as a feature of the Gaulish style in that
district, fig. 2(1-6).* Several examples of these fret-forms have been
found in Ireland, and must, I think, have reached our island as early
as the close of the Hallstatt period, or in La Tine I.

We see the form on a bronze spear-head found near Boho, Co.
Fermanagh, fig. 8, in which there appears to be a mixture of Hall-
statt and La Tine elements. Good examples of this class of fret were
also found in the crannog of Lisnacrogher, Co. Antrim, associated
with swords of La Tine I. They occur on the bronze bands of

1 Von Saken, '< Bas Grabfeld von HallsUtt,*' Pis. 23 and 26.
'Munro, '' Bosnia- Henegoyina," fig. 161.

s Morel, "La Champagne Souterraine," Pis. 6, 19, 20, 41 ; Moreau, << Albom
Caianda," iii., PI. 133 ; Bevue Arch^ologique, 3 s., zli. (1902)» p. 196.

[20#J

262

Proceedings of the Royal Irish Academy.

Bpear-shafts, fig. 4. The blades of the spears were of iron.' An iion
spear-head, found at Corofin, C!o. Clare, likewise shows the fret-form ;
the borders of the openings in the blade are inlaid with bronze,* fig. 5.
The fret-pattern at the bottom of the .Turoe stone, it will be
observed, is similar in treatment to those on the Mame pottery. It is,
in fact, the same as fig. 2, no. 4, bi*ought closer together. Again, the
scroll-work may be compared with that of the torques figured by
Morel, pi. 37. The feeling of the work is very similar. A relation-
ship of the highest interest is thus established between the Turoe stone
and the style of the Mame district. Our knowledge of the La Tene

^T^

•lilhl^iL

Kg. 2.

period in Ireland is still very imperfect. We cannot say whether
Tischler*s classification holds good for Ireland or not, or how far we
have to allow for survivals. But the preceding considerations preclude,
in my opinion, a late date being assigned to this stone, and induce me
to place it as early at least as La T^ne II., or possibly the later half of
La T^e I.

The Castle Strange Stone. — This stone is also an erratic, of the same
class of granite as the Turoe stone. It is at present in the demesne

^Journal B. Soc. Ant. Jr., vol. xvi., p. 395. One of the sword-aheftthB from
Lisnacrogher is figured by LindenBchmit, **Alt. u. h. Vorz," iiL, HefL iii, Taf. S,
from the specimen in the British Museum. It is erroneously stated to be from
the Thames.

2 This specimen is the property of Mr. Mark PattenKm, of Gorofin, and has not
been published before.

GoTvnY—Monumenta oj the La Tine Period.

268

of Castle Strange, a few yards to one side of the principal avenue.
There are no remains near it with which it can be associated, and it
has probably been moved from its original site and placed beside the
avenue as an ornamental stone. No traditions are attached to it. I
could learn nothing about it, save that it had been in its present
position as long as the oldest people remembered. Its dimensions are
3 feet by 2 feet 8 inches by 2 feet. The carving is not in relief,
but incised. The under side is not carved ; the natural surface of
the stone has there been left untouched. Figure 6, an end view of
the stone, shows the form of the under side : the drawing is from a
photograph taken when the stone was raised to examine the under

Pig. 3 (1).

Fig. 4.

side. The style of the ornament is similar to that on the Turoe stone,
and it must be referred to the same period. The illustrations (Plate
X2I., figs. 2, 3) are from the cast in the Museum. The stone has
suffered somewhat from weathering on the top and at one side.

The MuUaghmoit Stone. — This stone is a compact limestone. It
measures, in its present condition, 3 feet by about 1 foot 3 inches at
the sides. It is approximately square in section. The history of the
stone, as far as Lord Walter FitzGerald could ascertain, is that at the
time the Haughton family demolished the FitzGerald castle of Mullagh-
mast, which formerly stood in a field on the hill called " Oldtown,"
this stone was found in the walls, and it was then removed to the
haggard of the farm of Prospect House, on MuUaghmast hill, which
was built out of the materials of the castle.

264

Proceedings of the Boyal Irish Academy,

Pieces of the stone haye been broken off at each end, as if an
attempt had been made to sqnare it at the ends for building. There
is nothing to show whether this was done at the time the castle was

Fig. 6.

Kg. 6 (i).

Fig. 7.

built or when the stone was removed to Prospect Farm. If the stone
was built into the wall of the easUe at the time it was erected, it was
probably done with the idea that some virtue or power attached to the
stone on account of its ornament.

The carving of this stone is later in treatment than the preceding

CorrEY—Manumenta of the La line Period. 266

stones. It is more geometrical, the inciBed patterns in particular
being of the interlocked &scheme order frequently found on our early
Christian monuments.

When I first saw a rubbing of this stone, I was on that account
inclined to regard it as belonging to the Chriatian period, and con-
jectured that it might be a practice-piece or specimen of work. On
seeing the stone I abandoned that opinion. The monument is complete
in itself, and there is nothing of a Chriatian character about it to
connect it with the Christian period.

It is necessary to describe the earring in some detail. On the
slanting top of the stone a characteristic Celtic whorl of the triakele
type is carved in relief; it is probably symbolic in intention. The
principal face, which we will call a, is carved in relief, with the
exoeption of the bottom panel. The upper half of the face ia occupied
by a pointed oval panel enclosing two spirala, set obliquely on the
stone. The treatment of the spaces in this panel is characteristic of
La Tdne III. Below this panel is a band of ornament which has
unfortunately suffered much from weathering and injury ; the late
La Tdne character of the design is, however, apparent. Below this is
a curious zigzag fret, and below that is a panel divided x-wiae by
inoiaed lines. The stone is a good deal injured, and the surbce flaked
off at the left side, but the restoration of the design, fig. 7, is
probably fairly correct.

Taking the faces in order from left to right, face h is the next.
Nearly half thia face has been broken off at the right side. The
deaign appears to have been an oval panel filled with incised trumpet
pattern. At the top and at the bottom traces of carving in relief are
noticeable, and the border framing the oval was also in relief.

Face e haa lost nearly all its carving. Large pieces have been
split off it, leaving only a portion of the upper end intact. Here a
single spiral is carved in relief. The carving on this face waa, no
doubt, chiefly in relief, and it should be noticed that it is opposite to a,
also carved in relief.

face d. — Here again we have an arrangement after the manner of
face b. The centre is covered with incised trumpet-pattern, while
above and below the carving is in relief. The upper carving consists
of a triangular panel, injured by cuts where the stone has been used as
a sharpening stone ; the lower is a good piece of La T^ne ornament.

It is not certain whether or not the incised patterns are finished,
or were intended to be carved in relief. But however that may be,
there is no doubt that, notwithstanding their late look, they are

266 Proceedings of the Royal Irish Academy.

oontemporaiy with tlie rest of the camng ; the omament of the stone
is a completed scheme. The hands round the base show that it iru
intended to he set upright, and was in fact a stele. As already stated,
there is nothing Christian about the monument. The incised panels,
however, lead into the Christian period, when the use of the trumpet-
pattern organized as a system of interlocking spirals becomes freqae&t
for the filling of panel spaces. The Mallaghmast stone must, there-
fore, be placed towards the end of the pre-Christian period in Irekod,
or in the overlap of the Pagan and Christian periods.

It may now be asked, what light do these stones throw on the
general question of a La T^ne period in Ireland ? I tiiink it may be
claimed that they show that the La T^ne style had taken deep root in
Ireland before the Christian period, and that the La T^ne antiquities
found from time to time are not to be accounted for simply by Inde
or raid from Britain and the Continent.

That the La Tine style was widely spread in Ireland I have held
to be probable, chiefly on the ground that the derived La T^ eras-
ment, which forms so marked a feature of early Christian ornament in
Ireland, presumed an extensive use of the La Tdne style prenons to
the introduction of Christianity. These stones go far, I think, to
confirm that opinion. It must be borne in mind that the ancient
inhabited sites of Ireland have not been excavated. The great period
of Tara and of Emania, in our heroic literature, was from a centoiy or
two B.C. to the third century A.n. From Emania, some very heautifnl
La T^ne brooches are known, and there can be no reasonahle doubt
that if either of these sites were excavated, numerous antiquities of
the La Tine style would be brought to light.^

In conclusion, I desire to express my indebtedness to Mr. A.
McOoogan, of the National Museum, for much kind assistance in the
preparation of the photographs to illustrate this paper.

1 The fibuk, «Horae Ferales," PI. xxi., fig. 1, ia sUted to be from NaTin Bath,
Go. Meath. Tbis is an error for the Nayan Bath, near Armagb, the site of the
ancient Kmanis (on Eamhmn s nBamkuin, pronounced Navan, and now known as
the Navan Bath, or Navan Fort). Two other examples from the aame site ars ia
the National Collection. Wilde's Catalogue B.I.A., p. 668. Petrie CoDectioD,
No. 612.

( 267 )

XV.

*'THE ANCIENT F0KT8 OF IKELAND." BEING SOME
FURTHER NOTES ON A PAPER OF THAT NAME,
ESPECIALLY AS TO THE AGE OF MOTES IN
IRELAND.

By THOMAS JOHNSON WESTROPP, M.A.
Read Notbmbbb 30, 1903. Published Jakuabt 23, 1904.

Whsn I laid before the Academy a Paper' on the extenrive and as
yet most imperfectly known subject of the ancient forts of Ireland,
I was well aware of the many limitations of my work. I was also
prepared for the detection of many errors in its pages, and have been
the more agreeably surprised at the consideration it has received from
other antiquaries. I would confine this paper to giving certain cor-
rigenda and addenda of my own had not one criticism been published,
which, though friendly, affects not the details, but the broad deduc-
tions of one section of my paper. I wish therefore to reply to this
one point, lest my silence should be misinterpreted ; for I believe the
following facts will justify and bear out my views in the above-
mentioned paper.

It has been stated that, from being unacquainted with an essay by
a certain English antiquary, I have adopted the view of the pre-
Norman and, in some cases, prehistoric origin of Irish motes, the fact
being (it is alleged) that they are confined to the English Pale and
were only of Norman origin.

First, to avoid error — for the word " mote " or " moate " is
sometimes applied by Irish antiquaries to the low rath or liss — ^I use
the word ** mote " exclusively -for the high flat-topped ** mount," with
or without a lower side-platform or '' bailey," and girt with one or
more rings and fosses. When without the base court, I use the term
** simple mote " ; otherwise, the term *' complex mote."*

1 TraziB. R. I. A., xxzi., p. 679. I have olao dealt with the mote question in
a paper read Noyember, 1902, before the R. 8. A. I.

* As only one Thing mote is recorded at any of the Danish settlements, I do
not deal with any but the residential motes in this paper.

R.I.A. PHOC, VOL. XZnr., SBC. c] [21]

268 Proceedings of the Royal Irish Academy.

Let me first state that I was (necessarily) folly aware of the trend
of antiquarian opinion as to the exclnsively (or at least nsoally)
Norman origin of the English motes. I had read the essay alluded to*
hefore my paper even went to press ; hut I decided not to alter or
add notes in press, nor to modify my statement as regarded Irish motes,
for the following reasons. These, so far as I can see, preyent the
English theory from heing as sweepingly applied to Ireland as has
heen done on very slight authority in the aforesaid essay. Hy reasoos
are, first, the term " English pi^le " is most variant, vague, and mis-
leading ; but even allowing it to include all districts (even when
occupied for the shortest time by the Normans, from 1170 to 1250)
the motes exist outside it , and are non-existent in very impoitsmt
districts colonised by the Normans. It is only hy ignoring aU Irish
field-work and history that this base of the theory as applied to Ireland
could be maintained. Turning from more genenil objectionB to
details, let us note' : —

1. Motes do not occur at the recorded sites of many important
early Norman castles, e,g, Kells, Eillare, Ardfinnan, Limerick, Tris-
tledermot, Imleach, Earkinliss, Iniskefti, Castro-Iconing, Kilmallock,
BiiT, Bindown, Athlone, Carrickfergus, Gaoluisge, Carlingford,Thurles,
Croom, Dungarvan — all earlier than 1217. Motes are not found on the
chief manors, and '* vills" of the great colonies in Eastern limericV
Central Connaught,* and Cork, or in the important settlementB in
Thomond.

2. While they occur in places never held by the Normans or not
occupied by any castle during the earlier generations after theiuTEsion,
*.e. before 1250. Several of these are noted in sections 128 and 134.

3. Several motes, and those of the first importance, occurring at the
site of Norman castles, represent forts recorded as at those places long
before the Norman invasion. For example : the sheet-anchor of the

1 Mrs. Armitage, in Joum. Soo. Ant. Scotland, xzziv (1899-1900), p. 276.

^ I use the short fonoB as usual— C« S. P. I., Calendar of Documents (or State
Papers) relating to Ireland ; B. S. A. I., Roy. Soc. Antiqq., Ireland.

3 Castles of Esduen, Castleconnell, Wethney, Croom, Askeaton, Caatk Bobeit
Goer, Castle Kobert DoondonneU, Newcastle, Caheroonlish, Adare, and Kil-
mallock ; or the vills of Aney, Bruree, Bathkeale, Mahoonagh, Athlacca, and
Corcomohide. In this large district there are only two small and probably
sepulchral tumuli near Aney. Neither Shanid nor Kilfinnane %urea among the
early castles ; but motes are found there alone.

* See for this colony the important papeni by Mr. H. T. Knox, in Joum.
B. S. A. I., xxxi., p. 179, &c. ; and xxxUi., p. 68, &c.

Wbstropp— J%« Ancient Forts of Ireland. 269

theoiy, BO far as English writers apply it to Ireland, is the fact of the
making of motes at Slane and Trim in County Meath, as mentioned in
*' The Song of Dermot and the Earl." No mote remains at Trim ; and the
Slane mote was levelled soon after its construction in 1176. Slane
has a fine simple mote on the hill-top near the Ahbey. But the ** life
of St. Patrick," by Murohu Macci Mactheni (who was a friend of
Aedh, Bishop of Sleibhthe, before 698, and which work is preserred in
the Book of Armagh, 807-812), mentions great earthworks and fosses
on the Hill of Slane, and evidently near St. Patrick's Gamp. Now
the abbey is supposed, on early tradition, to occupy the site of
St. Patrick's foundation, and bears his name. The medisBval castle
stood down the slope, near the Boyne, where the present castle
stands,' and possibly there (and not at the Abbey) did Flemyng
make his mote. Mactheni says that, even in his day (some 500 years
before the Norman invasion), the Slane earthworks were attributed
by ''a fabulous story" to the slaves of Feccol Ferchertni, a pre-
Christian prophet of *'Bregia."*

The Normans made a castle of earthworks, palisades, and a long
wall at Downpatrick or Dim da leathglas in 1177. But the Annals
of Tighemach, who died in 1088, mention ** expugnatio Duin leath
glaise " under 496. The Annals record the storming of the same fort in
783. The ^'Annals of Ulster " mention it in 1009 : '< Dun da leathglas
was burned both the/ortre$$ and a third of the town (the lay part) by
lightning." Under its other name Kathceltchair, it figures in the
pre-Norman *' Book of Leinster," and the earlier lives of St. Patrick,
its legendary founder belonging to the earlier heroic cycle of the Bed
Branch heroes.' As will be noted, Jocelin of Fumess, before 1 186, attri-
butes this fort to a period earlier than St. Patrick, and accurately
describes Dun da leaUiglais as a *' neighbouring mote " (monticulus)
near St. Patrick's Church at Down, <' surrounded by marshes of the
sea.

The Normans built a castle at Knockgraffan, County Tipperary,
in 1 192.* The place possesses a fine complex mote, with the ruins of
a stone castle in its bailey. But the fort of Ghraflan is reserved to

> See the maps in the Down Surrey, where the Flemyngs' Castle ie ihown in
detaU.

■Mactheni (Ed., Bev. A. Barry, 1896), p. 19.

t <• Book of Leineter," p. 118.

« Ann. F. M., noticed in 0. 8. P. I. vol. L, No. 169, as gnmted to W. de Burgo,
1201-2.

270 Proceedings of the Royal Irish Academtj.

the King of Munster in the ^' Book of Bights,"^ which claims to be
of the fifth century, dates in its present form from its compilation or
revision before 902, and is found in pre-Norman manuscripts. The
fort, in fact, is connected in early legend with the mother of the
pre-Christian King Fiacha Muillethan.

Naas, County Kildare, was fortified by the Normans in 11 86. It has
a mote which figures both in early legends (such as the pre>Norman
Dindsenchas) from 277, and in the Annals from the fifth to the
ninth century. It is mentioned as the <'Dun of Naas," in the
" Tripartite Life" (tenth century),' as being visited by St Patrick,
who camped ** on the green of the fort to the east of the road " ; "to
the north of the fort is his well." As the chief fort of the Kings of
Leinster it was deserted in 904.

These legends, at least, prove the forts to be long pre-Norman.

These examples from four out of the five ancient provinces (th^ne
being, so far as I am aware, no case of a mote and early Norman
castle "coming together" in Connaught) may suffice to show my
reasons for adopting the view that soiiie residential motes in Ireland
are pre-Norman and even prehistoric.

4. Such motes occur in places where no early Norman castle is
known to have stood; but where in some cases early forts are
recorded.'

Again confining ourselves to a few examples : —

The great mote of Kilfinnane, County Limerick, with triple fosses
and rings, is evidently (from the identification of the surrounding
great forts of Clare, Duntrileague, &c., &c.) the Treada na riogh or
triple fort of the kings named in the " Book of Rights," <mU 902.*

The complex mote with two great fosses at Donaghpatrick,
County Meath, appears in the " Annals of Tighemach," in 746, as
being stormed ; and six of the prisoners taken in it were crucified.

The "rath" of Magh Adhair, County Clare, lies some miles
distant from the bounds of the almost nominal borderland of the

» ** Leabhar na gCeart" (Ed. 0 Donovan), pp. 87-89.

'Ed. WMtley Stokes, p. 185.

' Those who hold the Norman origin of Irish motes ought first to establiBh
records of castles being built (say) before 1250, at the vast majority of forts,
in the long list of motes named in my paper, pp. 708-712. Our records and state
papers at least are silent. Mrs. Armitage's remarks suggest an acquaintanoe vith
Wright's * Louthiana/ but no knowledge of motes outside the Pale.

* Loe. eiU

Westrofp — The Ancient Forts of Ireland. 271

English colony in Tradree.^ It figures as the place of inauguration
of the Dalcassian princes from 877, when King Flan Sunach was
defeated on its green, to 1313, and is a good example of the simple
mote. This is not only mentioned in the Annals, hut in the
pre-Norman '* Wars of the Gaedhil with the Oaill."

5. The type of such motes is prehistoric. It occurs in Austria
and Bosnia with Early Bronze Age ** finds." A hronze axe was found
near the mote of Dromore, County Down, and early um-hurials in
the mote of Skeirk, Queen's County, not in the mount, hut in the
'' hailey." I fully recognise the great difficulty in Ireland (if not
elsewhere) of disentangling the sepulchral elements. For example,
there is no reason either to douht that a pre-existing sepulchral
tumulus may have heen adapted for residence and defence hy the
addition of a hailey and fosses, or that the practice of hurial in resi-
dential forts was so little unusual, that the discovery of sepulture in
a mote (as at Greenmount) in no way disproves the residential nature
of the earthwork.'

6. If English antiquaries are right in applying the fact of the
ascertained Norman *' origin " of English motes to Ireland, there
should he evidence forthcoming in the abundant records' of the early
Norman colonies. This is not so : save for the '' motes " of Trim
and Slane— and I may add a third example (not given by the Eng-
lish writers), the mote at Boscrea — the evidence rather runs to the
contrary. English antiquaries have apparently made no use of the
most obvious and, in this matter, most reliable authority, Giraldus
Oambrensis. He was a contemporary, a relative of some of the chief
actors in the Norman invasion of Ireland, and visited the country
during the events he records in 1183 and 1186. He mentions the
erection of many forts and camps : the Normans use an ancient fort,
or make fortifications of sods, and boughs, stakes, &c. ;* but he only

^ Save for the ahort-liyed Castle of Quin, 1280-86, the nominal Engliah lands
north of Dromoland, and beyond Finlough, were nncolonised and lay waste. See
C.S.P.I., 1287, and '* Wan of Turlough." No Engliah castles, save Quin, Claio,
and Bunzatiy, are recorded in that part of Thomond ; no mote occurs at their sites
or in the English settlement.

' For burial in vahout types of forts, see my paper, Trans. B. I. A., section 44.

* The making of only one mote appears (so far as I have found) in the great
mass of records cited in the " Cal. Doc. relating to Ireland," and at the Dublin
Record Office.

*Giialdus' (Ed. Bohn) "Topography," p. 194; "Conquest," Book x.,
sections xi. and xiii.

272 Proceedings of the Royal Irish Acaden^y,

once mentionB the motes, not as built in his day, bnt as haiing bea
made before 838, by Turgesins, who ''erected castles; . . . they were
surrounded with deep ditches, and very lofty, being also round, and
most of them haying three lines of defence."" If his contemporaries
made similar structures at all the places where motes and Normaa
castles exist, his silence is very unaccountable ; if the pre-existmg
motes were, like the raths and cahers of earth and stone, utilized by
the Normans, his silence speaks very plainly indeed. The evidence
of Jocelin also tells against the exclusively Norman origin of oar
greater motes. He was a monk of Fumess, and wrote in the time,
and at the suggestion, of Thomas, Archbishop of Armagh (1181-1201).
He probably compiled his work before 1186, as he does not allude
to the translation of the remains of the three Patrons, at Down, in
that year. He mentions *' a work called a rayth," ».^. •* a wall,"
and other earthworks ; but his one allusion to a mote is to attribute
it to the fifth century or earUer. He tells how the hostage of Biehn
was starved and ill-treated by his detainer, and of his liberatioii
by St Patrick. The saint then placed the broken chains, as a
remembrance, '< one in a place at Down, where now is erected the
church of St. Patrick ; and the other on a neighbouring mote (monti-
culus) surrounded by a marsh of the sea," which was still called in
Jooelin's day, Dun da leathglas.' Seeing how hastily made and
easily destroyed were the motes of Slane and Trim ; how the Boecrea
mote was *' run up " so hastily, that the leave of the Bishop of
Eillaloe, on whose lands it stood, could not be obtained before its
completion ^ (though only thirty miles distant from his see) ; we cannot
readily believe that even these motes were structures suoh as are
found broadcast all over Eastern Ireland in and outside Norman
territory, and rarely elsewhere even in the early English colonies.
In view of the continuance of *' fort "-making, both of the 8t«»e
caher and the ring mound with fosses, down to very late times, I am
theoretically inclined to believe in the late construction of motes in

1 Ibid,^ ** Topog.," oh. zxxvii. and xzxviii.

» Jocelin's ** Life of St. Patrick, chapter zxxvii. : *« In loco ubi nmie in Dna
sdificata est eocleeia S. Patricii ... in monticule • vicino oiienmcloBo palode
pelagi ... a catenia conlractii yocabulum, scilicet Dun da leathglaa, lortitaa
ett.'* See also " Eoclemastical Antiquities of Down, Connor, and Dromore,"
Br. ReoTes, 1847. Note that with Jooelin <'rath" meant <<nunpait": and
"dun," **a mound or mote."

«C. S. P. I., vol. i., No. 2760.

WtmBon—The Ancient ForU of Ireland. 273

Ireland. I do not deny that some of these motes may have been
made by the Normans ;^ but the evidence is (so far as I have found in
the records or on the field) non-existent in any save three cases, and
those haVe left no trace. I merely show that the attempt to include
Ireland in any theory requires local study and local knowledge,
and that the ^'rule" laid down for Irish motes by some English
antiquaries ought not to be held ^' proved by its exceptions," though
the exceptions are endless.

The facts certainly show the necessity of great caution before
theories founded on facts lying outside this island can be sweepingly
applied to monuments within its shores, and accepted without further
examination. ^

The *' confusion" between sepulchral tumuli and motes also calls
for further notice. In my paper (section 128) I wrote of simple
motes, ''It is very easy to confuse this form with tumuli; but the
mistake is of less moment that certain defeusive motes contain burials,
and certain sepulchral motes have been adapted for fortification."
This has naturally called forth criticism, which leads me to add a
little to what appears above on p. 271. The *' confusion" exists in
the monuments themselves. We have some reputed sepulchral mounds
girt with fosses and rings evidently for residence. The '' mound" of
Donaghpatriok, and that of Horristown Biller (so familiar to travellers
from Dublin to Eildare), ar^ round-topped' ; but in each case we find a
large, and evidently residential, entrenched annexe or '< bailey." This,
and the allusion in our history to the capture of Donaghpatriok, show
that, even where the mote is not flat-topped, we cannot lightly declare
it to be sepulchral. The mote of Magh Adhair is the traditional grave
of a mythical Firbolg prince, the ''rath," and eventually the mound
of inauguration, of more historic chiefs. Here we have a complete
confusion of tomb, residence, and thingmote in one earthwork. So
&r as I can find, there is no evidence for the existence of thingmotes
at any Norse colony, except at Dublin. This being so, may not this
latter mound have been an earlier fort used by the " Danes," or even
a sepulchral tumulus, like those at Clontarf and other places round
the city ? In view of all this, I should have " darkened knowledge,"

^ The only casee 1 have collected where a recorded Korman castle stands near
or at a mote, and at which no pre-Nonnan mention of a fortress \a discoverable,
amount to eight. I have twenty-seven early oastle-ntes from Oiraldus, the
Annals, and State Papers, where no record or traces of motes remain.

' So, however, are the defensive motes shown on the Bayeuz Tapestry.

274 Proceedings of the Royal Irish Academy.

and staited erroneous theories, had I yentnred to distinguish between
these structures, and to lay down boundaries where the old mearingB
were lost, as some would have had me to do.

One other objection is made — ^that I regard the forts of the Irish
types over central Europe as the work of one (presumably Celtic)
race. This is met by a passage in my paper oTerlooked by my critics
{he. ciLy p. 580) : — " I use the term * Celtic ' as a mere symbol for
the types prevailing in Ireland. Many forts of these types were most
probably constructed by tribes to which even the loosest users of the
term would never think of applying the word ' Celtic' "

ADDEirnA AKD COSBIGSKBA.

I may add the following additions and corrections to my paper,
giving first the page of Transactions, then (in brackets) that of the
reprint : —

Page 598 (15). As Duncriffin appears to have been near the sea,
but on the side of Howth, next Meath, it was more probably the
destroyed fort near the martello tower, above the harbour, than the
'' Dun Hill," and was certainly not the promontory fort at the Bailey,
which bears its name on the Ordnance Survey.

Page 618 (40), page 678 (100). The wall of Orimspound is 10,
not 20, feet thick ; and the well is, I am told, merely the inflow of a
stream.

Page 620, note (42). For "Doronman" read "Downman."

Page 626, note (48). For *« 902 " read " 802."

Page 642, note (64). The description of this monument has since
been published by Mr. P. Lynch in ** Journal Royal Society Anti-
quaries, Ireland, vol. xzxii., pp. 380-332.

Page 644 (66). Add the Island KiUeen, County Waterford, as con-
taining an ogham-inscribed pillar of a descendant of NetaSegamon.

Page 648 (70). The fort of Langough is now thickly overgrown ;
and long reaches of the foundations cannot now be traced.

Page 693 (115). Add to section 85. The great stone fort of
Oughtmama, on the hill south of Corcomroe Abbey, Clare, is over
700 feet across.

Page 696 (1 18), and opposite page after << Dunoonor," for '* Inis-
here," read '* Inishmaan."

Page 698 (120). Add to section 96 : << Middens in Cashlaun Oar
and Cahercommaun have since yielded bones of deer and oxen, and
(in the latter) ironffragments."

Wbstropp — The Ancient Forts of Ireland.

275

Page 701 (123). Kilbradran. The fort is stone-faced, 100 feet in
diameter, with cnrved annexes to the north, west, and south-east ;
of these the western is the chief, with a deep fosse and earth-works,
6 to 11 feet high, and is 50 to 70 feet across its '< half-moon"
girth. The north annexe is mach defaced, 118 feet across, with fosse
and earthworks, 5 feet high. The eastern annexe is nearly levelled ;
two ancient roads wind round it to the central caher ; it is 187 feet

The fort of Kilbradran, County Limerick.

across, and its earthwork is rarely over 4 or 5 feet high, the fosse
being nearly filled up.

Page 704 (126). Add to the promontory forts of Waterford, *« Island
Hnbbock * entrenchment.' It has two deep fosses and a mound, and
is on a sheer headland."

Page 706 (127). Add to ** descriptions" of Dunnamoe that by
Kev. CsBsar Otway, in "Erris and Tirawlej (1841)," p. 67.

B. I. A. PROC, VOL. XXIT., SBC. c]

[22]

276 Proee&dinffs of the Royal Irith Academy.

Page 70S (180). Add to the simple motes— " X^VMTMi, Sbtnid
Castle (0.8. 19)/'

Page 711 (laS). 8kirk or Skeffk is placed by mistake smong the
simple motes. The mount is 16 feet 6 inches hi^, and 43 feet
across the top ; the annexe is 6 hei high, and 160 feet by 220 feet;
the fone 12 to 14 feet wide ; one of the piUar-stones is still standing.

Page 713 (196), line 1. For "within the ciicuit" read ''about
the drcnit."

Page 717 (139). Section 156, for '' ancient forts " read '* ancient
roads " ; and add, " The view is most probable in the lines in Eeny,
Limerick, and Waterford." The aboTO misprint took place after
the proofs had left my hands ; and it completely reverses the meuing
of the whole paragraph.

In section 158, Mercator shows the ''Radnffe" as having a
central mound, with a fosse on each side.

Proc. R.I.Acad., Vol. XXIV. Sect. C.

Plate XVIII.

LA TEN E MONUMENTS. -The Turoe Stone.

Proc. R.I.Acad., Vol. XXIV. Sect. C.

Plate XIX.

Ppoc. R.I Acad., Vol. XXIV. Sect. C.

Plate XX.

(/>

(/}

H
Z
UJ

D
Z
O

Ppoc. R.I.Acad., Vol. XXIV. Sect. C.

Plate XXI.

Fig. 1.— Top of Tupoe Stone.

Fig 2.

Figs. 2 and 3 — The Castlestrange Stone.
LA TENE MONUMENTS.

•^•i****.

IC:

2044 090 857 582

?A-

-- >f

t :< f^-

'«../

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Proceedings of the Royal Irish Academy"

See other formats