^

/

FOR THE PEOPLE
FOR EDVCATION
FORSCIENCE 1

LIBRARY

OF

THE AMERICAN MUSEUM

OF

NATURAL HISTORY

4.W.N.I1

MEMOIRS AND PROCEEDINGS

MANCHESTER

LITERARY & PHILOSOPHICAL

SOCIETY.

{MANCHESTER MEMOIRS.)

Volume L. (1905-6.)

MANCHESTER :

36, GEORGE STREET.

1906.

off(i^/A//^^_ Ck:^^

NOTE.

The authors of the several papers contained in this
volume are themselves accountable for all the statements
and reasonings which they have ofifered. In these par-
ticulars the Society must not be considered as in any way
responsible.

contp:n'1's.

MEMOIRS.

INAUGURAL ADDRESS. By the President, Sir W. M.

Bailey pp, i — n

1. Note on the Buccal Pits of Pcripatus. By C. Gordon

IlKWiTT, B.Sc. With plalc pp. i— 8

{Issued separately., October zis't, igoj).

II. Some Convection Effects in a Heated Tube. By C. H.

Burgess, M.Sc .. pp. 1—3

(Issued separately, Dccctnber 23rd, igoj.)

III. Remarks on the Germinal Layers of Vertebrates and on the

Significance of Germinal Layers in general. By J. W.

JiCNKiNSON, M.A., D.Sc. J 4 Figs pp. i -89

(Issued separately, March 26th, iqob).

IV. Battack Printing in Java, with Notes on the Malay Kris and

the Bornean Sumpitan and Upas Poison. By John
Allan ... ... ... ... ... ... ... pp. i — 13

(Issued separately, April "jth, IQ06).

V. Report on the Recent Foraminifera from the Coast of the
Island of Delos (Grecian Archipelago). Part III. By
Henry Sidebottom. Plates I— I/. pp. i — 18

(Issued separately, April iqth, igo6).

VI. The Cytolngical Aspect of Parthenogenesis in Insects. By

C. Gordon Hewitt, B.Sc. Plates I^II. pp. i -38

(Issued separately, ISIay 2ist, igob).

VII. Total Solar Eclipses. (The Wilde Lecture). By Professor

H. PL Turner, D.Sc, F.R.S pp. 1—32

(Issued separately, May 26th, igo6).

VIII. On the Range of Progressive Waves of Finite Amplitude in

Deep Water. By R. F. Gwyther, M.A pp. 1—28

(Issued separately. May sist, igo6).

IX. Observations on a Captive Mole {Talpa europisa). By Lionel

E. Adams, B.A. pp. i — 7

(Issued separately. May 31st, igo6).

VI CONTENTS.

X. A New P'ern from the Coal Measures : Tiihicaulis Sulcliffii
spec. nov. By Marie C. Stopes, D.Sc, Ph.D. Plates

I— III. pp. 1—34

(Issued separately. June zgth, /go6).

XI. On the Difference between I'hysiological and Statistical Laws

of Heredity. By A. D. Darbishire, M.A. ... pp. i — 44

{Issued separately, July 6th, igo6).

XII. Notes. — On an Ailotropic Form of Arsenic and On the Estima-
tion of Arsenic when in Minute Quantities. By William
Thomson, F.K.S.E., F.I. C. ... ... ... ... pp. i — 9

{Issued separately, August 14th, /god).

XIII. Notes on the Palsartic Species of Coal Tits. By Francis

Nicholson, F.Z.S. With Coloured Plate pp. i — 21

{Issued separately, Attgust ibth, IQ06).

XIV. The Species of Ctenopteryx, a Genus of Dibrauchiale

Cephalopoda. By Dr. J. II. Ashwortii and Dr. W. E.
rioYLE ... pp. I — 8

{Issued separately, August 14th, igo6).

PROCEEDINGS.

Bailey, Charles, M.Sc, F.L.S. — Presentation to the Society's
Library of the Minutes of the meetings of the Manchester
Botanists' Association ... ... ... ... ... xiv

Barnes, C. L., M.A. — Seaweed less highly esteemed amongst the

ancients than the moderns ... ... ... ... ... vi

E.xhibit of Mr. T. E. Heath's stereoscopic charts of the stars... ix

Brothers, A., F. R.A.S. — Presentation to the Society of a portrait

of the late Rev. William Gaskell ... ... ... ... vi

Dl.XON, II. B., M.A., F.R.S. — Lantern exhibition of photographs

taken in South Africa ... ... ... ... ... ... vi

Faraday, F. J., F.L.S. — Foreshadowings of the conclusions of the

Cancer Research Committee ... ... ... ... ... iii

Morris-Airey, H., M.Sc. — On the variation of the electrical

resistance of Osmium with temperature ... ... ... vii

Nicholson, F., F.Z.S. — Presentation to the Society's Library of
Sturgeon's "Annals of Electricity, etc.," and "Annals of
Philosophical Discovery " ... ... ... ... ... xv

CONTENTS. VII

Nicholson, F., F.Z.S. — Exhibit of a specimen of the Fine Marten,

Ma7-lcs abietitni ... ... ... ... ... ... ... xvi

Presentation to the Society's Library of a volume of scientific

memoirs ... ... ... ... ... ... ... ... xxi

Ramsden, II., M.D. — Exhibit of a model illustrating the propaga-
tion of sound waves ... ... ... ... ... ... V

Stansfield, H., B.Sc. — Note on the behaviour of liquid films

formed from a solution of saponin in water ... ... ...vii — viii

SiOTES, Marie C, Ph.D. — An account of some recent researches

into the nutrition of the egg cell in certain plants ... ... iii — v

Stromeykr, C. E., M.Tnst.C.E. — On recent mysterious fractures of

steel plates ... ... ... ... ... ... xvii — xviii

Taylor, R. L., F.C.S., F.I.C.— On the origin of the salt in the

sea ... . . ... ... ... ... ... ... ix — xiii

Thorp, Thomas, F.R.A.S. — Exhibit of lantern slide photograph

of the total solar eclipse of 1905 ... ... ... ... ii

Communication on " shadow bands'' ... ... ... ... ii

General Meetings... ... ... .. ... ii, iii, v, viii, xiv, xvii, xviii

Aimual General Meeting ... ... ... ... ... ... xx

Special Meeting for the delivery of the Wilde Lecture .. ... xvi

Conversazione ... ... ... ... ... ... ... ... vii

Report of Council, 1906, with obituary notices of Sir |. S. Burdon-
.Sanderson, F.R.S., Professor Carl Gegenbaur, Dr. S. P.
Langley, Mr. R. Rawson, F.R.A.S., Mr. C.J. Heywood,
and Dr. George Wilson ... ... ... ... xxiii — xxxiv

Treasurer's Accounts ... ... ... ... ... ... xxxv — xxxvii

List of the Council and Members of the Society ... ... xxxviii — liii

List of the Awards of the Wilde and Dalton Medals and of the

Premium... ... ... ... ... ... ... ... liv

List of the Wilde Lectures ... ... ... ... ... ... Iv

Manchester Memoirs, Vol. I. (1905).

INAUGURAL ADDRESS

By the President^

Sir William H. Bailey,

October jjth, igoS-

I thank you sincerely for the great distinction you
have conferred upon me ; it is no small honour to occupy
this chair, and succeed such men as Dr. Dalton, James
Prescott Joule, Sir Henry Roscoe, Professor Boyd
Dawkins, and the long list of those distinguished in
the history of science and literature, who have preceded
me.

This Society, established in the year 1781, may be
considered the first Manchester Technical Institution for
the education, for the culture of its own members at its
own cost ; it is an Academy of adult persons. The
Society originated in a few gentlemen meeting in a weekly
club some time before the year 1781 who conversed on
Literature and Philosophy, as Science was then called.

Our early volumes of memoirs are of very great
antiquarian and historical interest. In them we may see
many questions that have agitated the minds of our
members of 125 years ago, and that many reforms our
members advocated, which no doubt influenced public
opinion, are now the law of the land.

The 1 8th Century was a brilliant period. It was the
seed time of England's commercial supremacy.

The mind of man was beginning to endow the body
of man. Leisure from the mere ministry to man's neces-

December, iQO^.

2 Bailey, Inaugural Address.

sities was being created. Intellectual and physical liberty-
were in the atmosphere.

Unknown to each other James Watt and William
Wilberforce were in partnership, for Science is the true
liberator, — the arts of peace are the real arts of war.

Man's command of the forces of nature, and the
genius of English mechanical inventors, were destroying
the chief handicraft trades.

We in England were then inferior to Holland, Spain,
the Low Countries, and Northern Europe, in most of our
manufacturing operations. We made worse iron than
Sweden and Holland. We imported our very fine yarns
from India, and we bought much warp and weft from
the Continent. We were slowly bringing into use our
first great textile invention, the fly-shuttle, which quad-
rupled the speed of the weaver, which had been invented
by Kay, of Bury, in 1733, and Crompton's mule for
spinning was being unwillingly adopted. Nevvcomen's
single-acting steam engine had been at work about three-
quarters of a century, chiefly for mining and pumping
purposes.

In dyeing, bleaching, and calico printing, we were not
superior to the Continent, and in paper-making — thanks
to the revocation of the Edict of Nantes — Portal, driven
from France, had introduced into England white paper
manufacture of a superior character, and we were a good
second best compared with the productions of the
Continent in printing, paper-making, and bookbinding.

In clocks and watches we were superior to all com-
petitors, for England was the birthplace of the chronometer,
and had taught mankind the truth to the hundredth part
of a minute. The great round globe had become a clock ;
the stars became the real milestones of the mariner, and
Greenwich his cosmopolis.

Manchester Memoirs^ Vol. I. (1905). 3

James Watt was busy with his new double-acting
engine, but he had doubts about the profit of it, and sent
his son to the foundry at Warrington of John Wilkinson
(who was a brother-in-law of Priestley) to learn book-
keeping and business routine, and this son, James Watt,
Junior, became one of the Honorary Secretaries of this
Society.

The first practical steamboat had been placed by
Jonathan Hulls on the Severn, the first of all steamboats.

It was in these busy times that a number of sincere
men of Manchester, Warrington, Bolton, and the neigh-
bourhood, mere day-dreamers, idealists and truth hunters,
who wanted to know the inner meaning of things, bound
together by this common object, met and founded this
Society.

The first volume of the memoirs is characteristic of
the desires and aims of our founders. After reading it
through, I nearly unconsciously repeated to myself the
lines of Browning's " Fra Lippo."

You've seen the world
— The beauty and the wonder and the power,
The shapes of things, their colours, lights and shades,
Changes, surprises, — and God made it all !

# * « *

This world's no blot for us,
Nor blank ; it means intensely, and means good :
To find its meaning is my meat and drink.

The laws and regulations of the Society enacted that
there should be two Presidents, four Vice-Presidents
and two Secretaries ; the subjects of conversation may
comprehend Natural Philosophy, Theoretical and Experi-
mental Chemistry, Polite Literature, Civil Law, General
Politics, Commerce and the Arts, but that Religion, the
Practical Branches of Physic, and British Politics be
deemed prohibited, and that the chairman shall deliver

4 Bailey, Inaugural Address.

his veto whenever they are introduced. I, with diffidence,
presume the practical branches of Physic does not mean
drugs and medicine, but surgery and operative work.
The difference between British Politics and General
Politics seems to me to be somewhat obscure, for in the
very first volume there is an article which is political and
British on " The Impropriety of allowing a Bounty to
encourage the exportation of Corn," to which I will
allude later.

Medals were awarded to those who read papers. It
they were of merit the award of a medal entitled the
Member to be enrolled amongst the list of honorary
members.

The founders of the Society were the chief Scientific
men of Manchester. Among the Honorary Members were
Erasmus Darwin, Dr. Franklin, Lavoisier, Dr. Priestley,
William Roscoe of Liverpool, the poet and grandfather of
Sir Henry Roscoe, the translator of the "Life of Benvenuto
Cellini," author of the " Life of Pope Leo the loth." and
also famous as the author of the life of Lorenzo de' Medici,
Doming Rasbottom, a kinsman of my family, Josiah
Wedgewood and others, equally distinguished in literature
and science.

The Reverend Samuel Hall has a paper in the first
volume which would have been anathematised by Mr.
Ruskin. The reverend gentleman shews clearly that the
taste for nature and beauty and fine arts has no influence
favourable or otherwise to morals. I hope that I may not
be called a Philistine if I say that I believe the reverend
gentleman's judgment is correct.

There is a useful paper by Mr. Thomas Barnes, who
says there are some sciences of importance to those who
wish to be supreme in manufactures. He also observes
that there are many machines in Birmingham which the

Manchester Memoirs, Fi?/. /. (1905). 5

Manchester mechanics know nothing at all about, and he
hopes shortly to see that the new fire engine (meaning
the steam engine) will be used in this district.

There is a foot-note informing the reader that a
machine for spinning cotton, probably Crompton's mule,
has been working for some time amongst the spinners in
Manchester, and there is another being erected for
grinding corn and in a state of forwardness near Black-
friars Bridge, London. Although the first great departure
in weaving, Kay's fly-shuttle, had been at work nearly
50 years, he proceeds to say that it is not known to a
single weaver in the Norwich trade.

He advocates with considerable force the creation ot
a school in the shape of a museum in which all the various
appliances known to man for spinning and weaving and
the mechanical arts should be exhibited, with a Professor
who knows all these things and is able to describe them
to the students, and who should be well versed in
mechanical and chemical knowledge. Students should
go to it after they had become accomplished in reading,
writing, and arithmetic, indeed, he advocated a secondary
technical school. The man in charge of it would be, he
says, a kind of general Oracle, who might be consulted
on mechanical movements which students might find
difficult to understand unaided. He wisely concludes, " in
a town like this Manchester, the opulence, and even the
very existence of which depends on manufactures, and
these again upon arts, machinery, and invention, a Public
Cabinet, devoted to this purpose would be at once of
general ornament and utility."

This remarkable paper was read on January 9th, 1782,
and as we know, with the exception of this Society, which
has been an Academy of Science, and the Manchester
Mechanics Institution, founded 40 years later, the teaching

6 Bailey, Inaugural Address.

of the first principles of science and the industrial arts in
this district was unknown and neglected.

There are in this first volume two papers by
J. Wimpey, " On Economical Registers " and " On the
impropriety of allowing a bounty to encourage the
exportation of Corn." In the paper on economical registers
which would be called statistics now, (indeed only about
1833, the year in which the first Statistical Society in
England was founded in Manchester, had the word become
common), he asserts that we took no census of the
population at this period, indeed, it was considered
wicked to take a census, and only in 1801 did we count
our population. We did not know, said Mr. Wimpey, the
amount of cash circulation in the kingdom, or the state of
the population or its health.

He shows the absurdity of paying four shillings the
quarter on all exportations of corn, especially when our
crops had been bad and corn was scarce for our people ;
he says, if we had an economical register we might
regulate the export of corn according to our necessities,
for we then exported more than we consumed, and
observes : " So keen is pursuit of private emolument, and
so ignorant and remiss is the Government, that they have
frequently given a bonus of fifteen per cent, to export
corn, when all they had in stock was very far short of
being sufificient to support their own people until the next
harvest." With indignation he says, "The Dutch by
buying this corn and sending it back had made millions
of money out of foolish England." He further proceeds
to say, " that our legislators have been very fruitful in the
invention of penal laws ; but in the measures of pre-
vention, which are infinitely more salutary, they are
either very inattentive or very barren."

He incidentally alludes to the unemployed and ne'er-

Manchester Memoirs, Fi?/. /. (1905). j

do-wells. He would give every man a medal with his
name on who was engaged in labour, and unless he could
produce it he would punish him ; but " liberty is in
such high esteem, fears are awakened, suspicions alarmed,
jealousies excited, lest any encroachment should be made
on the liberty of the subject, under the specious but
deceitful appearance of public good." Then he returns
to the corn bounties again, and insists that we ought to
have knowledge of our production, for "it is a question of
the greatest importance in this country, ignorance of
which has cost millions of pounds." He very wisely
continues, " No Government can be just whose support
and defence does not extend to the equal, and indis-
criminate benefit of the whole people," He denounces
the conflict between the landed interests and the com-
mercial classes, and tersely says, " There is a bias, for
where shall we find those who have no interests in the
laws they enact." He does not believe what is said in
Parliament, that a bounty on the exportation of corn has
benefited the landowner and the farmer, and added to the
riches of the nation. He points out that this export
bounty has been paid to the corn growers for 95 years,
and advocates that in times of great abundance the corn
surplus should be stored in public granaries, and in times
of scarcity it should be sold, and the extra price the growers
obtained in bad years would be better for the people and
the nation, than sending it out of the country and making
it so dear that people scarcely get any corn at all. It
appears that in 1795 the quartern loaf was one shilling,
and in 1800, one shilling and fivepence. In those days
working people never saw wheaten bread. This paper is
well worthy the study of those who are interested in
fiscal questions.

Mr. Wimpey is a thorough protectionist in other

8 Bailey, Inaugtiral Address.

respects. He agrees that it is proper to prohibit the
import of French silks into this country, and those from
our possessions in India were also prohibited under very
severe penalties.

The Laws of the Society were changed from time to
time, for in the volume for 1790 the prohibition of dis-
cussions on religion and politics is omitted from the
regulations.

In the early years of the Society the papers on
general literature and the drama, and on antiquities, were
more frequent than at the present time. Dr. Ferrier in
1790 contributed a paper on " Sterne," in which he shows
that much of the jocularity and wit of " Tristam Shandy"
has been lifted from Burton's " Anatomy of Melancholy";
this is no very severe charge, for Shakespeare's great art
consisted in polishing the gems, the diamonds and precious
stones he found, and his new setting of them is his
contribution to the wealth of the world. Montaigne
claims that he only desires to have the reputation due to
the man who has found the string to bind together the
flowers that he has gathered from classic gardens and
which he presents to his friends as a pretty posy.

As a Vice-President of the Library Association, I
have visited nearly every important Library in this
Kingdom. I desire to emphasize this, that in my opinion
ours is the best Scientific Library in the Provinces. Rich
by the productions of our own members, for we have the
records of the experiments of Dr. Dalton, Dr. Joule,
and Richard Roberts, Fairbairn, Professor Williamson,
Sir Henry Roscoe, William Sturgeon of the electro-
magnet, Dr. John Hopkinson, Professor Osborne Reynolds,
Dr. Schuster, Dr. Schunck, and other distinguished
history makers, for ever famous in the text-books and
records of science and mechanics, during the century and

Manchester Memoirs, Vo/. /. (igo$). g

a quarter of our existence. Further, we have the pro-
ceedings of many of the other Scientific Societies of the
country, the Continent, and the United States.

Not long ago, as a personal instance, I wanted more
information than that given in the biography by his
talented daughter, Maria Edgeworth, of the works of that
great inventor Richard Lovell Edgeworth, and I found
them all described in the proceedings of the Royal
Dublin Society, and other Societies on the shelves of our
Library.

In our " Memoirs " will be found records of the work of
our members in their investigations of the economics of
industry, the scientific uses of energy, the properties of
steam, the very elementary beginnings of electricity (that
for years was looked upon as a toy), and also those in
connection with the early manufactures, dyeing, bleaching,
&c., all of which have a fascinating interest to those who
are practically engaged in our industrial work, or to the
historical student.

The chief tools of the workshops of the world, not
only of those where steam engines, locomotives, and
steamships are built, but also of the textile factories of
the world, were invented in Manchester or within thirty
miles of it ; and our records contain the names of many
of these inventors who were members, for the men of this
County were the first to use steam power for spinning
and weaving, and for punching, cutting, shaping metal ;
they were to the front in changing the old handicraft
machines to suit the new steam power. Prominent among
them was that genius, Richard Roberts, who was one of
the founders of the Manchester Mechanics Institution and
always in the front rank in advocating technical education.
He was the most prolific of them all — his chief inventions
being the slide lathe, planing machine, and self-acting

lo Bailey, Inaugural Address.

mule for spinning cotton, and a number of others for
steamships, screw propellers, life-boats and armour-
clads, &c. Then we had Nasmyth, the inventor of the
steam-hammer, a man who loved science more than
money, for in his middle age he retired to his garden in
the South of England, and studied the lamps of heaven
with a great telescope he designed and made at Patricroft.
Mention should also be made of Sir William Fairbairn
and Sir Joseph Whitworth. Dr. Priestley, who discovered
oxygen, occasionally visited us. Count Rumford, who
tried to discover the mechanical equivalent of heat, was
amongst our honorary members, as was also Dr. Darwin's
grandfather, the celebrated Erasmus, author of the Botanic
Garden, and the poet and prophet of the future of the
steam engine, who with exultation wrote : —

Soon shall thy arm, Unconquer'd Steam ! afar
Drag the slow barge, or drive the rapid car ;
Or on wide-waving wings expanded bear
The flying-chariot through the fields of air.

There is no necessity to mention at length the great
work of our late illustrious members Dr. Dalton and
Dr. Joule, whose effigies in marble are in the entrance to
the Manchester Town Hall. Valuable records of the
Society will be found in a book edited by Dr. Angus
Smith, in which he summarises the work of our most
prominent members, the title of which is, " A Century of
Science in Manchester." I assisted in a few paragraphs
about Richard Roberts, whose friendship I much valued
when a young man. Dr. Angus Smith says in his preface,
" That the Society has made Manchester a scientific centre
for more than a Century, and has much disposed it to
seek a University, and has given it a right to demand
one, a right that has been conceded." And he goes on to
say that the Society has done its work for a century

Manchester Memoirs, Vol. I. (1905). ri

absolutely unaided by Government or other outside
assistance. Is it overstating the case to assert that the
Victoria University and the Manchester Technical School
are the direct results of the advocacy of the mennbers of
this old Society ?

The site of the Old Infirmary has been called Man-
chester's great opportunity. It is known to some of us
that members of the Manchester City Council are con-
sidering how this fine site may be used.

We have confidence that an Institution will be erected
worthy of the birthplace of Free Municipal Libraries, and
that literature, art, and science will be recognised in a
good, broad, generous and artistic manner.

Some of us are not without hope that space will be
found in this Institution for a Hall of Fame, many of
which are on the Continent, and one especially, I have in
my mind, at Munich, in which may be placed portraits
and sculptures that will cause us to hold in perpetual
remembrance the names and services of the illustrious
benefactors, inventors, and pious founders of this great
County of Lancaster, whose work, a perpetual endowment,
is in the class-books of the schools of the world, where
chemistry, electricity, and engineering science are taught.

Such a shrine would exalt the self-respect of every
thoughtful citizen, and be an inspiration to the imagina-
tion of every student, for, in the topographical distribution
of men of genius, this county has a rich record of inventors,
dramatists, statesmen, discoverers, poets and painters, and
moulders of the mind of man.

A noble company, the flower of men,
To serve as model for a mighty world.

Manchester Memoirs, Vol. I. (1905), No. 1.

I. Note on the Buccal Pits of Peripatus.
By C. Gordon Hewitt, B.Sc,

Demonstrator of Zoology in the University of Manchester.

(Communicated by Professor S. J. Hickson, F.R.S.).
Received and Read, October jrd, igoj.

The earliest detailed account of these structures* is
that given in a memoir on Peripatus capensis by Balfour ( i ),
who, after describing the characters of the outer and
inner pairs of jaws, adds (p. 222) — " A more important
difference between the two blades than in the character of
the cutting edge just spoken of, is to be found in their
relation to the muscles which move them. The anterior
parts of both blades are placed on two epithelial ridges,
which are moved by muscles common to both blades
(//. yiv\.,Jig. 11). Posteriorly, however, the behaviour of
the two blades is very different. The epithelial ridge
bearing the outer blade is continued back for a short
distance behind the blade, but the cuticle covering it
becomes very thin, and it forms a simple epithelial ridge
placed parallel to the inner blade. The cuticle covering
the epithelial ridge of the inner blade is, on the contrary,
prolonged behind the blade itself as a thick rod, which,
penetrating backwards along a deep pocket of the buccal
epithelium, behind the main part of the buccal cavity for
the whole length of the pharynx, forms a very powerful

* In Moseley's account (2) of this species the following words may refer
to these structures: " From the posterior part of the lateral surfaces of the
pharynx, a pair of small muscles which probably are protractors of the
pharynx, and serve to push forward the jaws,"

October 21st, igo^.

2 Hewitt, Note on the Buccal Pits of Peripatus.

lever, on which a great part of the muscles connected
with the jaws find their insertion. The relations of the
epithelial pocket bearing this lever are somewhat peculiar.

" The part of the epithelial ridge bearing the proximal
part of this lever is bounded on both its outer and inner
aspect by a deep groove. The wall of the outer groove is
formed by the epithelial ridge of the outer blade, and that
of the inner by a special epithelial ridge at the side of
the tongue. Close to the hinder border of the buccal
cavity (as shown in pi. x.v\.,fig. 12, on the right hand side)
the outer walls of these two grooves meet over the lever,
so as to completely enclose it in an epithelial tube, and
almost immediately behind this point the epithelial tube
is detached from the oral epithelium, and appears in
section as a tube with a chitinous rod in its interior,
lying freely in the body cavity (shown in pi xvi., figs.
13 — 16 le). This apparent tube is the section of the
deep pit already spoken of It may be traced back even
beyond the end of the pharynx, and serves along its
whole length for the attachment of muscles."

I have not had the opportunity of examining sections
of these jaw -levers in P. cnpejisis, but have done so in
series of sections of P. balfouri, P. novae zeaiandiae and
Ooperipatus ovipanis (Dendy) and in every case the
chitinous rod was hollow and not solid as described and
figured by Balfour in the case of i^. cape?isis. The hollow
nature of these jaw levers has already been described by
Prof Dendy in his memoir on the oviparous species of
Peripatus (3). He assigns a respiratory function to these
hollow jaw-levers, which form the buccal pits, as will be
seen from the following description of them, which he
gives. Referring to O. oviparus, he says (p. 371): "In this
species (and probably in the others) there is a tracheal pit
immediately in front of the mouth, and a pair of very

Manchester Memoirs, Vol. L (1905), No. \. 3

large ones opening into the buccal cavity just behind and
close to the base of the inner jaw on each side, and
running backwards for some distance, at first just outside
the lateral nerve cords (yfig. 5, B. ir.) and then above and
just inside the salivary glands. These buccal tracheal
pits have a thick chitinous lining, and may be traced back
in a series of transverse sections very nearly to the level
of the second pair of legs. They give off along their
course and from their extremities an immense number of
very fine tracheal tubes. When the jaws are removed,
these enormously elongated tracheal pits may be pulled
out in connection with them, and the chitinous lining of
the pit appears to pass over into the chitinous covering of
the smallest accessory tooth of the inner jaw."

On comparing these two accounts it is quite evident
that they are descriptions of similar structures ; but there
is a striking difference in the characters and functions
ascribed to them. In Balfour's account there is a solid
chitinous jaw-lever in an epithelial pocket, to which are
attached " a great part of the muscles connected with the
jaws ;" whereas Dendy describes them as buccal tracheal
pits having a chitinous lining and giving off "along their
course and from their extremities an immense number of
very fine tracheal tubes." On the other hand their figures
are very similar.

In working over series of sections of P. balfouri and
P. novae zealandiae I suspected that Prof Dendy had
mistaken the striated muscle fibres which move the jaw-
levers for tracheal tubes and "enormously elongated tracheal
pits." I called his attention to this ; and he kindly sent
me his sections of Ooperipatus oviparus (from which his
description and figures had been made). An examination
of these convinced me that this was the case, and that
the supposed tracheal structures were muscles in 0. oviparus

4 Hewitt, Note on the Buccal Pits of Peripatus.

as Balfour had described them in P. capoisis, and as I had
found them to be in P. balfonri and P. novae zealandiae.

The muscle fibres which work the jaw-levers are
transversely striated, being the only transversely striated
muscle fibres which occur in Peripatus. The fine tracheae
are also striated, but on comparison it is found that the
striations of the tracheae differ considerably from those
of the muscle fibres. As Moseley showed (2) in the
first account which was given of the tracheae of Peripatus,
the tracheae are provided with a spiral thread, the indi-
vidual coils of which are fairly wide apart. Their walls
are extremely thin, and hard to detect, except with a high
power. The muscle fibres have a more dense appearance
than the trachea ; this is due to the fact that the striations
of the muscle fibres are very fine and close together, and
stain very readily.

I was unable to find in any of the sections examined
any apertures in the epithelial pockets into which the
chitinous jaw-levers fit, which might serve as means of
communication between the buccal pits and any tracheal
structures, supposing the latter existed in this region.
Certainly no tracheal pits were given off from these
epithelial pockets.

In the species examined the characters of these
structures are very similar, the only difference being in
the form of the jaw-lever, which in P. balfonri differs
slightly from those of /^. novae zealandiae and O. oviparns.
As Balfour shows, the cuticle covering the epithelial ridge
of the inner blade is prolonged backwards as a thick,
flat, chitinous rod. In the anterior region of this rod
the outer walls of the two grooves described by him meet
at the hinder end of the buccal cavity. These grooves
are lined by a thin cuticle, and when the fusion of the
outer walls of the grooves takes place, the cuticle also

Manchester Memoirs, Vol. I. (1905), No. \. 5

fuses, so that a hollow, chitinous rod communicating with
the buccal cavity is formed. In each of these species the
chitinous rod is flat and fluted, so that in section it has a
wavy appearance (see Figs, i, 2, le.). In P. novae zealandiae
and O. oviparus the inner wall of the rod is thin, while
the portion adjacent to the attachments of the muscles is
much thicker.*

The specimens of P. balfouri examined had these two
walls of equal thickness, as shown in the accompanying
figures. These walls are in all cases thick in the middle,
and gradually become thinner, as seen in transverse
section, towards their edges, where they join to form the flat,
hollow, chitinous jaw-lever. In some sections the rod has
a solid appearance, the two sides having collapsed. The
character of the epithelial pocket, into which the jaw-lever
fits, is the same in each species. It is made up of a single
layer of cubical cells containing large nuclei {^Fig. 2),
and is simply a deep pocket of the epithelium lining the
buccal cavity. In no section is the continuity of this
cellular layer broken, either at the edges or at the sides of
the chitinous rod. The ends of these epithelial pockets
are quite blind, and consist of a solid mass of epithelial
cells. The chitinous jaw-lever is also imperforate (except
in certain sections where the perforation is clearly an
artifact).

To sum up, the continuity of the cellular layer of the
epithelial pocket, and the imperforate nature of the hollow,
chitinous jaw-lever, render the presence of tracheal struc-

* The sections of P. novae zealandiae and P. balfouri were cut %ij. thick,
and stained with brazilin (4). This reagent stains the chitina deep, purplish
black, so that the thin inner wall is rendered perfectly visible. Unless this
thin inner wall is stained, it may be mistaken for an optical effect cau.sed by
the balsam and the inner wall of the epithelial pocket, consequently the jaw-
lever would appear to be a solid chitinous rod instead of a hollow chitinous
rod.

6 Hewitt, Note on the Buccal Pits of Peripatus.

tures, such as tracheal pits and trachea in connection
with them, highly improbable.

The muscles which work the jaw-levers are attached
to the dorso-lateral walls of the epithelial pocket, as
shown in the figures, chiefly about the anterior and
median portions of the rod, the larger muscles dis-
appearing towards the posterior extremity. They
consist of two series, a median series {^Fig. 2, liii) running
in a longitudinal direction enclosed by two transverse
bands of muscle {ini) which are attached to the greater
part of the dorso-lateral side of the epithelial pocket,
that is, to the whole surface of this side except in the
middle line (as shown in Fig. 2) and extend thence
dorso-Iaterally. The jaw-levers and their muscles dis-
appear opposite the beginning of the second pair of legs.

Prof. Dendy has pointed out to me that Purcell (5)
has shown that in the Attidae, a group of spiders, " by
far the greater part of the tracheal system is nothing
else but a pair of modified ectodermal tendons." If
ectodermal tendons can be modified in this way,* the
question naturally arises, can these hollow chitinous jaw-
levers of Peripatus be compared with ectodermal tendons
which in the Attidae have become modified, according to
Purcell, to form tracheal structures. I do not think that the
hollow chitinous jaw-levers of Peripatus can be regarded as
structures of a respiratory nature. They are simply
skeletal structures, flattened for the better attachment of
the muscles which control them, and hollow like many
structures of a similar nature found in other Arthropods.

* Korschelt & Heider (6) are inclined to trace back the origin of the
tracheae of the Arachnida to kings. Simmons (7) has found that the trachea
develop from foldings in the posterior surface of the third abdominal
appendage, which in their early stages are similar to those on the second
appendage, which give rise to the lung-books. He considers that tht;
trachea: are developed from a lung-book condition.

Manchester Memoirs, Vol. I. (1905), No. 1. 7

Summary. — It has been shown that the so-called
buccal tracheal pits of Ooperipatus oviparus, described by
Prof. Dendy, are really hollow chitinous jaw-levers ; the
striated muscle fibres which make up the muscles con-
trolling them, having been mistaken for tracheal tubes.

In conclusion, I wish to thank Prof Dendy for his
kind assistance, and for supplying me with specimens of
P. balfoiiri.

LITERATURE REFERRED TO.

1. Balfour, F. M. "The Anatomy and Development of

Peripatus capensis." Q./.M.S., n.s., vol. 23, pp. 213-
259> pl- 13-20, 1883.

2. MosELEY, H. N. " The Structure and Development of

Peripatus capensis." Phil. Trans.., vol. 164, pp. 757-
782, pi. 72-75, 1874.

3. Dendy, A. " On the Oviparous Species of Onychophora."

Q.J. M.S., N.S., vol. 45, pp. 363-416, pi. ig-22, 1902.

4. HiCKSON, S. J. "Staining with Brazilin." Q.J. M.S.,

N.s., vol. 44, pp. 469-471, 190T.

5. PuRCELL, F. "Note on the Development of the Lungs,

Entapophyses, Tracheae, and Genital Ducts of Spiders."
Zool. Anz., vol. 18, pp. 396-400, 2 figs., 1895.

6. KoRSCHELT, E., and K. Heider. " Embryology of Inverte-

brates." Vol. 3. London, 1899.

7. Simmons, O. L. " Development of the Lungs in Spiders."

Amer. Joiirn. Sci. (3), vol. 48, pp. 1 19-129, pi. 8, 1894.

8. Packard, A. S. "Text book of Entomology." London,

1903.

9. Sedgwick, A. "Monograph on the Species and Distribu-

tion of the genus Peripattis." Q./.M.S., N.s., vol. 28,
PP- 431-493' Pl- 34- -10, 1888.

8 Hewitt, Note on the Buccal Pits of Peripatus.

EXPLANATION OF PLATE.

Fig. I. Transverse section of P. baljouri, through the middle
of the first pair of legs.

Fig. 2. Enlarged view of transverse section of the jaw-lever
of the right side and its attachments, in the region of
the first pair of legs.

bp. Buccal pit. ep. Epithelial pocket, le. Chitinous jaw-lever.
hn. Longitudinal muscles of jaw-lever, pc. Posterior
lobe of brain. ///. Pharynx, sld. Reservoir of slime
gland. /;;/. transverse muscles of jaw-levers.

Fig. I

W(

K

* i

ip-

•

r. f ^^^-2

=f

Manchester Memoirs, Vol. I. (1905), No. 2.

II. Some Convection Effects in a Heated Tube.
By C. H. Burgess, M.Sc

Received and Read, November 14th, igo^.

W. C. D. Whetham, in his " Experiments on Ionic
Velocities " [F/ul. Trans. A, 1893, 337], half filled a V-tube,
of special construction, with a red alcoholic solution of
cobalt nitrate. A layer of a specifically lighter blue
solution of cobalt chloride in alcohol was placed above it.

On passing an electric current through the apparatus,
a number of purple lines were formed at the junction of
the two solutions and the lines travelled both with and
against the current. Whetham could explain this behaviour
only by the presence of coloured complex anions in the
solution.

During some similar experiments I filled the upper
portion of a U-tube with hydrochloric acid and the lower
portion with a specifically heavier solution of cobalt
chloride in the same acid. After standing overnight the
blue liquid had diffused up into the colourless, giving a
gradually shaded band. On passing a current of half an
ampere through the tube by means of carbon electrodes,
the band was resolved into a series of layers of gradated
shades, with a sharp line of demarcation between each.

To determine whether this was due in any way to
complex ions, the blue cobalt chloride solution was replaced
by hydrochloric acid, to which a red dye and a little
glycerine (to render it heavier) had been added.

After diffusion had taken place the step-like layers
were formed by the current.

December 2 jid, jgOj.

2 Burgess, Sowe Convection Effects in a Heated Tube.
This is seen in Fio. i.

Fiz. I.

Fis:. 2.

This indicated that the effect was due to heating, the
centre of the tube being hotter than the walls, which were
cooled by the air.

With a strong current it was possible to see the hot
coloured liquid rise in the centre, and spread out into a
mushroom shape. The cold liquid from the walls falling
underneath broke off the head of the column, giving a
new layer.

Water coloured by a dye was then placed at the
bottom of a vertical tube, with plain water above.

A platinum wire enclosed in a thin glass tube was
inserted down the centre.

Manchester Memoirs, Vol. I. (1905), No. ^. 3

The wire was heated electrically, and after a time a
number of striae made their appearance.

The}' gradually broadened, and formed a series of
layers which were sharp at the top, and then faded away
until the next layer was reached (Fz£: 2).

The effect is not quite so well marked as when the
liquid is heated directly, and so more uniformly by the
current.

The sharpness and permanency of these boundaries
seems to indicate that there are vortex movements due to
convection currents between them.

The pressure is greater at the walls, and the liquid
rises in the centre. Its place is taken by the cold liquid
from the outside, which rises in its turn, thus causing a
circular motion from the wire to the walls, between the
layers.

The breadth of the bands appears to depend directly
on the difference of temperature of the inside and outside
of the liquid.

Increasing the current sometimes even causes the
lines to become doubled, the original ones gradually being
effaced.

If the tube is heated simultaneously from the inside
and outside by coiling another wire round the tube, the
layers do not appear ; there is, however, a tendency to
set up large convection currents, which rapidly mix the
liquids.

As this is passing through the press, I notice an
abstract in Central Blatt, igo^, B II, No. 23 of a paper
by C. Christiansen, {Overs, o.d. Kgl. Dansk. Vidensk.
Selskab Fork, igos, 307-15,) who has obtained similar
results, and formed much the same conclusions as those
given above.

Manchester Memoirs, Vol. I. (1906), No. 3.

III. Remarks on the Germinal Layers of Vertebrates
and on the Significance of Germinal Layers in
general.

By J. W. JENKINSON, M.A., D.Sc.
Exeter College, Oxford.

(Communicated by Dr. F. W. Gamble.)
Received January 12th, igo6. Read January i^th, igo6.

INTRODUCTION.

That theory plays a part of predominant importance
in the acquisition of new facts is a matter of common
knowledge, and embryology has afforded no exception
to this rule. The cell theory, the recapitulation theory,
and theories of germinal layers have here not only
supplied a stimulus which has produced but also pro-
vided conceptions which have dominated and guided a
long series of developmental investigations. In the light
of these theories the embryologist has set himself to
complete the task already taken in hand by the com-
parative anatomist, the search for homologies, believing
that he has had in the homology of the germinal layers
an absolute and infallible criterion of the homogeny, or
community of descent, of the organs of the adult.

Now while it is true that the morphologists of the
older school and their pupils, to say nothing of the more

^ I have to thank Dr. Haldane, Dr. Ritchie, Dr. Bourne, and Mr.
Assheton for their kindness in reading through the manuscript of this
paper ; to Mr. Assheton I am particularly indebted for much friendly and
valuable criticism.
Alarch 26i/i, igo6.

2 Jenkinson, Germinal Layers of Vertebrates.

humble writers of text-books, continue to adhere to the
authority of long-established dogma, the student who
has attentively studied the embryological work of recent
years can hardly have failed to notice a growing feeling
of dissatisfaction and discontent with these older generali-
zations. The adequacy of the cell-theory to give an
explanation of ontogenetic processes has been openly
called in question in more than one quarter ; grave doubts
have been expressed of the validity of the fundamental
biogenetic law, while the difficulties which every con-
scientious observer experiences in trying to patch the old
theoretical garments of germ-layer hypothesis with the
new cloth of descriptive and experimental fact are
patent on every side. The time would then seem to be
ripe for a renewed, and a critical, examination of the
principles involved in those theories which have attached
a morphological significance to the primary cell layers
of the embryo, and all the more so in a country in which
no word has been spoken on the subject since the
publication of Francis M. Balfour's Comparative Embry-
ology twenty years ago. The views which will be
expressed in the sequel were forced upon me in the first
instance by a study of the formation of the layers in the
Vertebrata ; and it is, therefore, on a consideration of
these processes as they occur in this group that I shall
primarily base my argument. Very fortunately, the
recent publication by Oscar Hertwig in his HandbiicJi der
EntivickeliingsleJire der Wirbeliiere of a clear and com-
prehensive review of the facts has completely absolved
me from the necessity of giving more than the briefest
account of them, emphasizing only such points as are
needful for my purpose ; though I have taken the oppor-
tunity of discussing a problem which has been the
stumbling-block of more than a generation of embryolo-

MaiicJiester Memoirs, Vol. I. ( 1 906), No. 3. 3

gists, I mean the relation of the Amniote blastopore to
that of the Anamnia.-

Fatal, however, as I believe the evidence of Vertebrate
development to be to the common views, the testimony
available from other sources must also be taken into
account ; I have, therefore, supplemented my critique by
a consideration of the embryogeny — in particular of the
cell lineages — of various Invertebrates, and of the bearing
of all the facts of budding, regeneration, pathology, and
experimental embryology on the problem.

The conclusion to which I have come I have already
hinted at ; unacceptable as it may prove to many, it is
the onl}^ conclusion which, as far as I can see, is compatible
with the facts.^

Part I.— THE GERMINAL LAYERS OF THE
VERTEBRATA.

ANAMNIA.

As a type of the formation of the germinal layers in
the Anamnia, I will take the form which has been most
extensively studied, and with which I am myself most
intimately acquainted, the common English Frog {Rana
temporaria).

Let me say at the outset that I define these layers — ■
the ectoderm, the endoderm, and the mesoderm — solely
with reference to their destiny;* by the "ectoderm" I

- A complete literature of the germ-layers of the vertebrates will be
found in Hertwig's " Handbuch " Cap. 3, (Gustav Fischer, Jena, 1903) ;
to this the reader is referred for particulars of the memoirs quoted in the
present paper.

" Since this paper was written I have found that F. Braem, in a paper
entitled " Was ist ein Keimblatt ? " and published nearly ten years ago {Biol.
Centralblalt, vol. 15, 1895, P- 427— 443> 466—476, 491 — 506), came to
conclusions practically identical with my own. Braem's views are founded
mainly on the developmental phenomena of Bryozoa and Ascidia.

•'Braem's phrase deserves quoting: " Keimblatter," he says, *' sind
Organbildner." (p. 431.)

4 Jenkinson, Gerjuinal Layers of Vertebrates.

understand that layer or set of cells which will give rise
to the epidermis and its derivatives, such as hair, skin-
glands, feathers, to the nervous system and organs of
special sense, and to the stomodaeum and proctodaeum ;
similarly by the " endoderm " I mean the cells which will
provide the lining epithelium for the alimentary tract and
its outgrowths, the liver, lungs, and so forth ; and by the
" mesoderm " the source of the musculature, skeleton, and
connective tissue, the blood and blood vessels and the
urino-genital system.

THE FROG.
In the Frog the process of germ-layer formation
begins when segmentation is completed. The fully
segmented ovum consists of an animal hemisphere (rather
more than a hemisphere) of small pigmented cells, con-
taining but little yolk, and a vegetative hemisphere
of bulky white cells full of large yolk-granules. From
the centre of the animal hemisphere (or animal pole) a
straight line can be drawn passing through the centre of
the egg and the centre of the vegetative hemisphere (the
vegetative pole) ; this line is the egg-axis and about it the
^gg is radially symmetrical. In the normal position of
the Q.oa the axis is vertical. Whether, as Schulze has
maintained, segmentation has already conferred a bilateral
symmetry upon the &^^ is a point which for our present
purpose is immaterial. At the equator of the ^^^ are
cells whichiare in every respect intermediate between the
small animal and the large vegetative cells. In its interior
is a spacious segmentation cavity, hemispherical in shape,
placed axially but excentrically and nearer the animal
than the vegetative pole. Its roof is formed of about four
layers of small cells, the outermost of which is arranged
very distinctly as a cubical or shortly columnar epithelium^

Manchester Memoirs , Vol. I. (1906), No. 3. 5

its floor of about twenty layers of large polyhedral yolk-
cells.

The first sign of the formation of the germ-layers is
given by the appearance of the structure known as the
dorsal lip of the blastopore {Fig. i, A). This is a short,
deeply pigmented groove, placed parallel to the equator,
and a little below it (about 25") at one point in the
boundary between the pigmented and the unpigmented
portions of the &^'g. With the appearance of the dorsal
lip the bilaterality of the ^^^ becomes fully established,
if it was not already in existence ; the plane which
includes the egg-axis and the dorsal lip is the sagittal
plane of the future embryo.

The changes that now take place, as seen from the
vegetative pole, are as follows {Fig. i, B — D) : — The
groove begins to travel downwards over the surface of the
egg towards the vegetative pole, the area over which it
passes becoming covered by cells which are as deeply pig-
mented as those of the animal hemisphere. At the same
time the groove elongates, becoming crescentic ; in other
words, not merely one limited region of the boundary
between pigmented and yolk-cells, namely, the dorsal
lip, is involved in the process of overgrowth, but the
regions lying to the right and left of this, that is to say
the lateral lips of the blastopore, as well. As the dorsal
lip (the middle region of the groove) continues on its
course towards the vegetative pole, and as continually
fresh regions are drawn into the process at the sides, the
blastoporic lip becomes first semi-circular, and then three
parts of a circle, until, finally, when that region, the
ventral lip, which is diametrically opposite to the dorsal
lip also begins to grow down, it attains the form of a
circle enclosing the still uncovered portion of the vege-
tative hemisphere, the yolk-plug. The dorsal lip has now
moved down to, or a little beyond, the vegetative pole.

Jenkinson, Germinal Layers of Vertebrates.

The movement of the dorsal hp just described is due
to its growth over the surface of the egg, the axis of

Fig. I.

A — D. Views of the vegetative hemisphere of the Frog's egg during the
overgrowth of the blastopiiric lip.

j5', F. Views of the lower surface during rotation of the whole egg.

In all the figures the dorsal lip is on the upper side. In D the ventral
lip is just formed.

(Original.)

which has up to the present retained its original vertical
position. At this moment, however, the whole q^^ begins

Manchester Memoirs, Vol. I. (1906), No. 3. 7

to rotate about a horizontal axis in a direction which is
the opposite of that in which the dorsal lip had moved
{Fig. I, E, F) ; and this rotation continues — the circle of
the blastopore becoming smaller all the time — until the
dorsal lip has returned, rather beyond the point from
which it started, to the (new) equator of the Qgg. The
angle subtended by the area of the yolk surface which it
traverses — both before and during the rotation — is about
75'^, as estimated by Kopsch, and the angle through
which the whole egg rotates about 100°. It follows that
the present vertical axis of the ^gg, which will be the
dorso-ventral axis of the embryo, makes the same angle
of 100° with the original egg-axis, that the animal pole is
situated below what will be the anterior end of the embryo,
and that the antero-ventral half of the embryo is developed
over the animal, the postero-dorsal half over the vegetative
hemisphere. {Pig- 2, ¥).

It has been suggested from time to time that the
first movement of the blastopore, as well as the second,
is due to a rotation of the egg as a whole, the over-
growth of the pigmented over the unpigmented area being
only apparent, and the darkening of the vegetative hemi-
sphere due to the formation of pigment in the superficial
yolk-cells. Against this view it must be urged (i) that
Kopsch has seen the yolk-cells streaming underneath the
overgrowing lip ; (2) that if a small exovate is produced by
lightly puncturing the egg at the animal pole, this is after-
wards found in front of the medullary groove ; ^ (3) that
when the blastopore is experimentally prevented from
closing the anterior end of the embryo is seen to develope
at the animal pole ; and (4) that there is a cause, namely,
shifting of the yolk and consequent displacement of the

^ Mr. Assheton has pointed oiU to me that this evidence is not
concUisive, as the exovates may become detached and shift their position.

8 JenKINSON, Germinal Layers of Vertebrates.

Fig. 2.

Sagittal sections through the Frog's egg during the closure of the

blastopore. ,,

A. The dorsal lip just formed : i.z. zone of intermediate cells ; d.i.

dorsal lip.

Manchester Memoirs, Vol. /. (igo6), No. li. g

centre of gravity, to account for the rotation in the
second case, but no such obvious explanation of the
supposed rotation in the first.

The whole of the process which has been described is
no more nor less than the closure of the blastopore ; the
blastopore, which is indeed but a virtual blastopore at the
start, being that circular subequatorial line along which
the pigmented animal pass into the unpigmented vege-
tative cells. And clearly this closure is bilateral, taking
place as it does most rapidly at the dorsal lip, least rapidly
at the ventral lip, and at an intermediate rate at the lateral
lips in between. The examination of sections will now
show us that the closure involves (i) a movement of
the vegetative cells into the segmentation cavity together
with (2) an overgrowth and ingrowth of cells at the blasto-
poric lip resulting in the formation of a new cavity, for
which we will retain the time-honoured expression of
' archenteron ' ; and that during the process the material
for the germinal layers is brought into position and laid
down.

A sagittal section of the egg passing through the dorsal
lip at its first appearance {Fig: 2, A) shows the groove
placed about 25^ below the equator in the zone of inter-
mediate cells. The cells which immediately bound the

B. The dorsal lip further advanced. The yolk-cells beginning to
creep up into the segmentation cavity.

C. A slit-like archenteron formed {arch).

D. The dorsal lip has reached the vegetative pole ; the ventral lip
(&./.) is just formed. Mes. 2. mesoderm formed from yolk-cells pushed
into segmentation cavity ; y.p. yolk plug.

E. The egg is beginning to rotate. The segmentation cavity is
reduced, the archenteron enlarged.

F. Rotation completed. Mes. v. mesoderm at ventral lip ; s.c.
remains of segmentation cavity.

In all the figures the arrow marks the egg axis, the head of the arrow
the animal pole.
(Original. )

lo ]y.^¥:.IN'S>0^, Genni7ial Layers of Vertebrates.

groove are disposed radially about it ; this arrangement
marks the beginning of a process of overgrowth and
ingrowth which becomes more obvious as development pro-
ceeds {Fig. 2, B, C, D). It is then seen that a fold of small
cells has grown over a certain area of yolk-cells. This
fold consists naturally of two sheets, an outer and an
inner. The cells of the outer sheet resemble closely
the small pigmented cells of the animal hemisphere into
which they are uninterruptedly continued ; like the latter
they are arranged in about four layers, tljie outermost of
which is epithelial. At the lip of the blastopore the
outer passes into the inner sheet, the cells in the outer-
most layer of the former being gradually turned over
into the innermost layer of the latter. This inner sheet
also consists of several layers of cells, the innermost of
which is pigmented and epithelial, the remainder being
more irregularly disposed. The inner sheet forms the
outer, or, as it will be when the Qgg has rotated, the
upper wall of the slit-like cavity between itself and
the yolk-surface which is now covered up. This cavity
is the archenteron and the inner sheet of the fold is its
roof; its floor is the original vegetative surface of the
egg. This overgrowth and ingrowth of cells, with con-
sequent formation of an archenteric cavity, takes place
in an exactly similar manner at the lateral {Fig. 3, A)
and ventral {Fig. 2, E) lips ; and by the time that this
last has appeared the cavity extends over a considerable
area of the vegetative hemisphere, as far towards the
equator, in fact, as the line from which the lip of the
blastopore had originally begun to grow down. Now,
however, another process supervenes, namely, an upward
movement of the yolk-cells into the segmentation cavity.
The first indication of this is, as a matter of fact,
observable at an earlier stage (Fig. 2, B). Immediately

Manchester Memoirs, Vol. I. ( 1 906), No. 3. 1 1

over the dorsal lip the yolk-cells at the equator may be
observed creeping up into the segmentation cavity, along
the under surface of its roof As this movement begins
so it takes place most rapidly on the dorsal side, but it is
continued none the less laterally and ultimately ventrally
as well. The segmentation cavity is thus reduced to a
small space situated ex-axially, and on the ventral side
i^Fig. 2, E). The archenteron meanwhile is extended in
the direction of the animal pole {Fig. 2, D). This exten-
sion may be due to a ' splitting ' amongst the yolk-cells,
or (more probably) to an actual invaginatory process ;

Fig. 3.
Transverse sections of the Frog's egg.

A. Before rotation during overgrowth of the lips of the blastopore.
/./. lateral lips. The yolk-cells are creeping up into the segmentation cavity.

B. After rotation. The archenteron is widened. n.cJi. notochoid.
(Original.)

but in any case a greater change soon follows in its dimen-
sions. The yolk-cells which form its floor are now pushed
up towards the animal pole, and inwards towards the egg
axis, so that the cavity gains simultaneously in length,
depth, and breadth {Fig. 2, E, F ; and Fig. 3, B) ; it now
extends into the animal hemisphere, being separated by
only a thin partition of yolk-cells from the segmentation
cavity. The fate of the latter appears, as Professor Keibel

12 jE'iiKlNSON, Germinal Laj/ers of Vertebrates.

has informed me, to be variable ; in a certain rather small
percentage of cases it is stated to fuse with the archen-
teron ; otherwise it is obliterated by further immigration
of yolk-cells ; traces of it may then be seen below the
archenteric floor [Fig. 2, F).

The changes just described in the position of the
heavy yolk-cells afford the best explanation of the' rota-
tion of the egg ; for the centre of gravity being shifted to
the ventral side, the q^^ naturally turns about a horizontal
axis in the opposite direction.

We are now in a position to discuss the formation ^
of the germinal layers. With the exception of the yolk-
plug, the outer surface of the egg is entirely covered by a
sheet of small, pigmented cells, disposed in about four
layers, the outermost of which is epithelial. In part this
sheet consists of the original animal cells which formed
the roof of the segmentation cavity ; but it is also derived
in part from the outer sheet of the fold which grew down
at the lip of the blastopore. The sheet in question is of
course the ectoderm — as above defined ; dorsally it soon
becomes thickened to form the medullary plate.

The notochord and mesoderm have a double origin.
Antero-ventrally — that is, in the region of the animal
hemisphere — they arise from the yolk-cells which have
been pushed and folded back into the segmentation cavity.
Postero-dorsally — that is, sub-equatorially — they are dif-
ferentiated in the inner sheet of the fold, the roof of the
archenteron ; the notochord in the mid-dorsal line in
front of the blastopore, the mesoderm in the remainder
of the roof around and behind the blastopore. The roof
consists of several layers of cells, the innermost or lowest
of which is epithelial ; the outer or upper layers consist
of fairly closely packed polyhedral elements {Fig. 4, A).
In the middle line (in front of the dorsal lip) a strip or

Manchester Memoirs, Vol. I. (1906), No. 3. 13

rod of these cells becomes gradually distinct from two
lateral sheets {Fig. 4, B) ; this strip is the notochord, the
lateral sheets are the mesoderm. Further both notochord
and mesoderm become gradually separated from the
innermost cell layer, the mesoderm rather earlier than
the notochord (which led Balfour to speak of the hypo-

FlG. 4.

Differentiation of the roof of the archenteron into notochord, mesoderm,
and roof of the definitive gut in the Frog.
A, B, C. Three successive stages.

A. The roof of the archenteron consists of about four layers of cells.

B. The lowermost layer is distinct from the upper layers at the sides,
but not in the middle line {n.ch.).

C. Both notochord (//.c/;.) and mesoderm (w<?j.) are distinct from the
roof of the gut and from one another.

(Original.)

blastic origin of the latter) {Fig. 4, C). This inmost
layer persists as the definitive roof of the gut, and may,
with the yolk-cells which form the floor, be termed the
endoderm.

I Jenkinsov, Germinal Layers of Vertebrates.

It is clear that this portion of the notochord and
mesoderm is brought into position concomitantly with
the overgrowth of the blastoporic lip ; and that the two
sheets of mesoderm are confluent with one another
posteriorly in the similarly formed mesoderm behind the
ventral lip. The whole of this mesoderm is subequatorial
or, after rotation, postero-dorsal. Antero-ventrally the
mesoderm and the notochord are completed by a different
process ; as the yolk-cells are pushed into the segmenta-
tion cavity those which lie next the roof of the latter, next
the ectoderm that is, may be seen to be in a state of rapid
sub-division (yFig. 2, C — F : mes. 2){Fig, 3) A) ; a layer of
cells smaller than the yolk but larger than the ectoderm
cells is thus produced, placed on what will be the antero-
ventral side of the embryo (A';^. 3, B), and continuous with
the already described postero-dorsal mesoderm around
the equator.

The germ layers are thus definitely established ; at
the lips of the blastopore all three are, of course, continuous
with one another. It only remains, for us to discuss a
httle more minutely their origin from the two sets of cells
— the so-called primary layers — observable in the &gg at
the end of segmentation.

A very considerable portion of each germ-layer is
actually brought into being during the closure of the
blastopore. What is the origin of the cells which form
the fold by the overgrowth of which this closure is
effected ? Are they derived exclusively from the cells of
the animal hemisphere alone, as some (notably Lwofif)
have maintained ; or, while the outer ectodermal sheet is
to be traced to the animal cells, is the inner sheet merely
due to the invagination of yolk-cells as was first stated by
the author of the " Gastraea-theorie " ? As far as I can
see both these views are erroneous. The yolk-cells though

Manchester Memoirs, Voi. I. {igo6), No. ^. 15

not passive, since they move into the segmentation cavity,
are certainly not in a state of active division, and I think
we may say quite confidently that the roof of the archen-
teron does not come from this source. It consists of cells a
little larger than the cells of the animal hemisphere, but
resembling the latter closely in being pigmented and in
containing few yolk-granules of small size. This, however,
is by itself no adequate proof that that is their origin'' ; on
the contrary, a careful study of all the stages seems to
show that it is the rapidly dividing cells of the subequa-
torial intermediate zone which principally contributes
to the inner sheet of the overgrowth; not, perhaps, entirely
though, for inner and outer sheet are continuous at the
blastoporic lip, and many of the cells of the latter are
probably being perpetually rolled over into the former ;
the radiate arrangement of the cells seems a sufficient
demonstration of this. The outer ectodermal portion of
the fold is very largely derived from an extension of the
roof of the segmentation cavity ; the cells here are
certainly in a state of active proliferation, and as the total
thickness of the layer does not increase, the total area must
necessarily enlarge. This spreading of the animal cells
seems indeed to be directly concerned (mechanically) in
the first formation of the fold at the blastoporic lip ; for
if (as, for example, by growing the eggs in certain
solutions which affect primarily the yolk-cells and prevent
their movement into the segmentation cavity) the closure
of the blastopore is experimentally delayed, either the
roof of the segmentation cavity at once becomes thrown
into numerous wrinkles or the number of cell-layers
present in it is increased.

^ It may be pointed out in this connection that Samassa ( " Studien liber
den Einfluss des Dotters auf die Gastrulation II.", Arch. Ent.-mech. vol.2,
p. 370 — 393, 1896) has shown that if in the eight-celled stage the four vege-
tative cells are killed, a typical blastoporic lip and archenteric cavity are
subsequently developed from the animal cells alone.

1 6 jE'NKl'^^iSON, Gerjuifia/ Laj/ers of Vertebrates.

To sum up, the ectoderm of the Frog is derived from
the cells of the animal hemisphere, the endoderm is
derived partly from the intermediate cells, partly from the
yolk-cells of the vegetative hemisphere ; while mesoderm
and notochord have a similar double origin. It is
during the closure of the blastopore — the subequatorial
bounding line between the animal and the vegetative
cells — that all three layers are brought into their definitive
positions, and this closure is bilaterally symmetrical, taking
place most rapidly at the dorsal lip. A similar statement
may be made of the process of germ-layer formation in
the remaining Anamnia.

CYCLOSTOMATA.
Petromyzon.

The Q^% of the lamprey, like that of the frog, is micro-
lecithal and holoblastic. At the close of segmentation it
presents essentially the same characters, namely, an animal
hemisphere of small cells, which form a roof, and a vege-
tative hemisphere of large yolk-cells, forming a floor for
the excentric segmentation cavity.

The dorsal lip of the blastopore now appears and the
blastopore closes bilaterally as in the frog ; we have not,
however, in the case of the lamprey that exact knowledge
of the angle traversed by the dorsal lip. There is further
a very important difference between the two in that here
no ventral lip is formed {Fig. 5, A). Apart from that the
result appears to be closely similar in the two cases. The
overgrowth and ingrowth of cells produces an archenteron
which is said to open into the remains of the segmentation
cavity, and the material for the three layers is at the
same time brought into position. The ectoderm in its mode
of origin resembles that of the frog ; the mesoderm seems
here also to be derived from two sources {Fig. 5, B),

Manchester Memoirs, Vol. I. (1906), No. S. 17

partly from cells carried back with the overgrowth, in part
from yolk-cells pushed into the segmentation cavity ; this
at least is the account given by Scott ; other observers —
Gotte, Kupffer, Shipley — have attributed it to either the
one or the other alone. The fate of the roof of the
archenteron is, however, by no means identical ; whereas
in the frog, as we have seen, this layer becomes differ-
entiated into notochord and mesoderm above, and the
roof of the definitive alimentary tract below, its destiny
in the lamprey is to give rise to the notochord and the

Fig. 5.
Petromyzon (after Scott).

A. Sagittal section after the overgrowth of the blastopore lip ;
d.L, dorsal lip ; arch., archenteron.

B, C. Transverse sections showing formation of dorsal (d.m.) and
ventral (v.m.) mesoderm, and notochord (n.c/i.) ; m.t. (solid) medullary
tube.

notochord alone. The archenteron is here a very narrow
cavity, and its roof proportionately small. A median
groove appears running along the ventral side of the roof,
and the whole becomes folded off and lifted out from the
side walls of the archenteron as a strip, subsequently a
rod of cells, the notochord {Fig. 5, C). The lateral walls
then grow in towards the middle line, and, meeting,
complete the roof of the gut. The endoderm, therefore,
does not — as far as its mode of origin is concerned —
wholly and exactly correspond in these two types.

1 8 JE'NKI'NSO'N, Gervimal Laj^ers of Vertebrates.

Bdellostovia.

Our knowledge of the early phases of development in
the megalecithal and meroblastic Myxinoids is exceed-
ingly meagre, but Bashford Dean has shown that in

Fig. 6.
Baellostoma (after Dean).

A. Blastoderm at one pole of the ellipsoid egg ; op.r., opercular ring
of the shell.

B. Dorsal lip (</./.) beginning to grow down.

C. A little later : v.l, ventral lip.

D. Embryo formed : blastopore closing at vegetative pole.

Bdellostoma the ^^^ axis is identical with the major axis
of the ellipsoid &%%, and that segmentation produces a cap
of cells or blastoderm'' at one pole i^Fig. 6, A). At one
point in the edge of this blastoderm a dorsal blastoporic
lip appears, and the material for the germ-layers of the
embryo is laid down during the bilateral overgrowth and

^ Throughout this paper I have followed the customary usage in
employing one term, " blastoderm," to denote the cap of cells produced by
segmentation at the animal pole of large-yolked eggs, both in the Anamnia
and Amniotes ; and having regard to the similarity in origin of the cap of
cells in question in the two cases the use of a common term is justifiable.
In their subsequent history, however, as will be fully shown below, the
Anamnian and the Amniote "blastoderms" differ widely; and it might
therefore be advisable to retain the expression for one of these only if a
suitable word could be found for the other.

Manchester Memoirs, Vol. I. (1906), No. 3. 19

ingrowth of cells in this region (/^z^. 6, B, C). The yolk,
however, is not wholly covered by this process. As soon
as the body of the embryo is formed overgrowth ceases to
be bilateral ; all parts of the edge of the blastoderm —
that is, the lip of the blastopore, now take part in the
process, and the blastopore eventually closes at the
vegetative pole {^Fig. 6, D).

It is worth while noticing that in this case, where,
since the &g^ is ellipsoid, the egg-axis can always be
recognised, there cannot be the least doubt that the
movement of the lip of the blastopore really involves an
overgrowth, and is not due to a rotation of the egg as a

ELASMOBRANCHII.
The Elasmobranch ovum is megalecithal and mero-
blastic. At the end of segmentation there is found at the
animal pole a blastoderm lying on and continuous at its
edges with the unsegmented yolk. The blastoderm
consists of a superficial upper layer of columnar cells
arranged in an epithelium and a mass of loosely connected
lower layer cells lying in the cavity — the segmentation
cavity — which separates the upper layer from the yolk.
The lower la}'er cells are disposed principally in two
groups ; a group occupying almost the whole vertical extent
of the segmentation cavity in the anterior region of the
blastoderm, and a smaller group close to the margin of
the blastoderm at the posterior end i^Fig. 9, A). Below
the segmentation cavity is the yolk, unsegmented but
nucleated. Many of these yoke-nuclei are exceedingly
large, and a good number are, according to Riickert, de-
rivatives of accessory spermatozoa. Though small cells
are continually segmented off from the yolk to be added
to the lower layer cells it is doubtful whether any of the
yolk-nuclei play any part in the formation of the embryo ;

20 Jenkinson, Germinal Layers of Vertebrates.

those which remain in the yolk are probably concerned
solely with facilitating its liquefaction and absorption.

There is now produced at the edge of the blastoderm
posteriorly and in the middle line a fold or overturning of
cells of the superficial layer. The fold, which is the dorsal
lip of the blastopore, is slightly raised and covers over a
space — the beginning of the archenteron — between itself
and the yolk. By the continuation of the backward growth
of the fold {Fig. 9, B, C) (the whole blastoderm is also

Fig. 7.
Torpedo. A — C after Riickert, D — F after Ziegler (slightly modified).

A. Formation of the dorsal lip at the posterior end of the blastoderm.

B, C. Formation of the lateral lips (/./.) and caudal swellings (c.s.).
The extent of the archenteron, crescentic with a median prolongation, is
indicated by the shading.

D, E, F. Extension of the blastoporic lip to the anterior edge of the
blastoderm. Formation of the embryo in the hinder region.

Steadily spreading over the yolk) the archenteron attains a
considerable length ; its floor is formed of yolk into which
yolk-nuclei subsequently make their way ; its roof consists
of a columnar epithelium derived in part from the overturn-
ing of cells at the lip of the blastopore, in part possibly
from the posterior marginal lower layer cells. But while

Majichester Memoirs, Vol. I. (1906), No. 3. 21

this process is taking place at the dorsal lip, that is, at the
median posterior region of the blastoderm's edge, it is also
being extended, though in a far less degree, to the neigh-
bouring regions, the lateral lips, on the right and left
(^Fig. 7, A, B). The archenteron thus comes to assume a

'^^c^*^

Fig. 8.
Growth of the Elasmobranch blastoderm over the yolk and final
envelopment of the latter ; a.e. anterior edge which becomes v.l. ventral lip ;
y.b. yolk-blastopore. In A the median strip of yolk behind the dorsal lip is
seen bounded by the two lateral lips.

crescentic shape, with a median anterior prolongation ;
the latter underlies the embryonic portion of the blasto-
derm, the crescentic part is wholly extra-embryonic, and
remains very shallow, though it is subsequently prolonged
to the right and left round the edges of the blastoderm

22 JENKINSON, Germinal Layers of Vertebrates.

until a slight overgrowth is formed even at ^the anterior
margin.

('

c

c
o

OJ

<u

j^

c

o

%

o

«

^

S

>,

^

rt

2^

^

c

oJ

"Ij

J

w

o

^

3

;-i

o
bl)

J^

.G

■^^

(U

u

G

1)

.;C

rt

o

^

s

TJ

-S

;d

O

C

O

a

f^

•z

!±3

Hi

•<

•<

rt

"oj

• •s

3

1>

"o

-§

•v:

-il

s

1^

S

?

CD

E^

<u

1

o

,

X3

c

O

u

H

o

cS

3

«5"

•2.

<u

ii

•^

c3

O

s

.&*

til

C4

<L)

pj

I/)

c

.i-i

ci

rt

tS

_o

Ji

(U

0)

^

O

O

u

3

U)

^'

(U

O

rt

_d,

<J

a.

OS

,

-^ ^ =Q

Manchester Meinohs, Vol. l. (1906), No. 3. 23

With the overgrowth at the Hps of the blastopore the
material for the germ-layers is laid down. The super-
ficial layer is now the ectoderm. The mesoderm consists

Fin. 10.

A series of transverse sections through the blastoderm of ScyUitim in
the stage represented in Fig. 7 C to show the formation of notochord and
mesoderm. (Original.)

A passes through the caudal swellings : the mesoderm is seen arising
from the roof of the archenteron.

B. Just in front of the dorsal lip. Two mesoderm plates are now
separated from the roof of the archenteron except at the lateral lips (/./.).
Medially the roof of the archenteron is thickened for the notochord.

C. Further in front. The archenteron is divided into a median
embryonic portion and two lateral portions beneath the lips of the blastopore.

D. E. The notochord is cut out of the roof of the archenteron ; the
mesoderm is separated into two median embryonic {e.m.) and two lateral
extra-embryonic {ex.m.) sheets.

of two parts (Fzo^. 10), (i) two sheets of cells lying one on
each side of the median line over the embryonic portion

24 Jenkinson, Germinal Layers of Vertebrates.

of the roof of the archenteron ; posteriorly these sheets
pass into the caudal swellings — two thickenings at the
edge of the blastoderm, one on each side of the middle
line {Fig. 7, B, C ; Fig. 10, A) — where they fuse with
the roof of the archenteron, out of which they have
gradually been differentiated ; (2) the formation of
mesoderm is, however, not limited to the dorsal
h*p, but is carried on at the lateral lips, and,
as these extend forwards round the whole margin
of the blastoderm, at the anterior edge as well.

c.

Fig. II.
Two stages in the ventral closure of the gut in Scylluim. (Original.)
In B the ventral wall is seen to be formed by suture of the sides of the
archenteron : yolk and yolk-nuclei {y.n.) below.

This extra embryonic mesoderm (' peristomales ' of
Rabl) is naturally continuous in the caudal swellings
with the embryonic ('gastrales ') mesoderm first described ;
it takes part only in the formation of the area vasculosa.
The notochord is formed from a median strip of cells
which is cut out of the roof of the archenteron (Fig. 10,
D, E ; Fig. II, A); the process begins in front and proceeds
backwards. The exact origin of the lining epithelium of
the alimentary canal — the endoderm — is by no means so

Manchester Memoirs, Vol. I. (1906), No. 3. 25

certain. When the embryo begins to be folded off from
the hinder middle region of the blastoderm, the roof of the
archenteron is involved in the folding, and forms at least
the roof of the gut ; the floor, according to Balfour, is
derived from a layer of cells formed round the nuclei of
the underlying yolk. I must say that my own prepara-
tions do not support this view ; the yolk-nuclei are much
larger than, and totally unlike, any of the nuclei of the
blastoderm ; and it seems to me perfectly clear that the
gut is closed ventrally by suture of the sides of the archen-
teron as this becomes folded o^ {Fig. 11, A, B).

Up to the present it is the posterior edge or dorsal
lip which has been principally active, but now growth
ceases at that point, elongation of the embryo being
effected by the independent proliferation of the caudal
swellings {Fig. 7, F). The anterior and lateral margins of
the blastoderm, on the contrary, become exceedingly
vigorous and begin to grow over the yolk, the overgrowth
being accompanied, as stated above, by a slight marginal
invagination ; and eventually the anterior edge makes
the whole circuit of the yolk, passing round the vegetative
pole and reappearing behind the embryo as the ventral
lip of the small ' yolk-blastopore ' {Fig. 8, B — D). A
peculiar change has also taken place on each side of the
dorsal lip, the lateral lips immediately adjacent to it
having swung back until they bound a narrow median
strip of yolk {Fig. 8, A), by which alone the aperture at
the dorsal lip now communicates with the remainder of the
blastopore.

It is now possible for us to institute a series of com-
parisons between the processes we have observed in the
microlecithal and macrolecithal types. These com-
parisons are as follows : —

(i) The blastoderm of the Elasmobranch corresponds

26 JeNKINSON, Geinninal Layers of Vertebrates.

to the animal hemisphere of the frog's &gg, the
yolk to the vegetative hemisphere.

(2) Hence the edge of the blastoderm in the former

is equivalent to the subequatorial bounding line
between animal and vegetative hemispheres in
the latter.

(3) In both this bounding line becomes in its

entirety the lip of the blastopore, the posterior
point of the edge in the Elasmobranch being the
dorsal lip of the frog, the anterior point of the
edge the ventral lip.

(4) In both the blastopore closes bilaterally, growth

being greatest at the dorsal lip.

(5) In both the material for the germinal layers is

laid down during the process ; the notochord
lies in front of the dorsal lip, the mesoderm at
the lateral lips in two sheets continuous with
one another behind in the ventral lip.
The principal points of difference are two. First,
in the Elasmobranchs the closure of the blastopore is
divided (as in Bdellostoma, and also, as we shall see in
a moment, in the Teleostei) into two periods ; in the
first, the overgrowth is almost confined to the dorsal lip
and produces the material for the formation of the
embryo ; in the second, the yolk is gradually covered by
an extension of the blastoderm in which the lateral and
anterior margins are alone concerned. Secondly, in the
Elasmobranch a part only of the blastoporic lip is in-
volved in the formation of the embryo, the lateral and
ventral lips remaining wholly extra-embryonic,

TELEOSTEI.
The ^^^ of the Teleostei is, like that of the Elasmo-
branchs, megalecithal and meroblastic. Segmentation

Manchester Memoirs, Vol. I. (1906), No. 3. 27

leads to the production of a blastoderm lying upon, but
not continuous with, the yolk. In the latter are numerous
nuclei derived from the blastoderm in an earlier stage,
and, as a rule, from its edge, though occasionally
(Salmonidae) from the whole of its lower surface. These

Fig. 12.

Formation of the embryo and closure of the blastopore in Semmis
(after Wilson).

A, B. The material for the embryo is laid down by overgrowth at the
dorsal lip [d.l.) A, from the left side ; B, from above ; s.c, segmentation
cavity ; o.g., oil globule ; a.c, anterior edge of the blastoderm.

C, D. Final envelopment of the yolk by growth mainly of the anterior
edge, or ventral lip {v. I.).

The shading at the edge of the blastoderm indicates the formation of
extra-embryonic mesoderm.

nuclei lie in a superficial clear layer of the yolk, the
periblast ; they do not in any wise contribute to the
formation of any part of the embryo, but subserve simply
the liquefaction and elaboration of the yolk.

28 Jenkinson, Germinal Layers of Vertebrates.

Several accounts have been given of the process of
germ-layer formation in this group ; the best is, perhaps,
Wilson's work on the " Embryology of the Sea-Bass "
{Serranus). In Serranus the blastoderm consists of a

z>

Fig. 13.
Serranus (after Wilson).

A. The blastoderm lying on the nucleated periblast {par.) is cut
sagittally. Ingrowth of cells is just beginning at the dorsal lip {d.l.).

B. Ingrowth of cells at the anterior edge (ventral lip).

C. The same at one of the lateral lips.

D. Sagittal section after complete enclosure of the yolk ; k.v. Kupffer's
vesicle; y.p. yolk-plug ; 7n.p. medullary plate ; 11. ch. notochord ; end. endo-
derm ; other letters as before.

superficial epidermic layer of flat cells, covering three or
four layers of closely packed polyhedral cells below.
There is a slight segmentation or subgerminal cavity
between it and the yolk.

At one point — the posterior middle point — of the
edge of the blastoderm a thickening, followed by an

Manchester Memoirs, Vol. I. (1906), No.l^. 29

overgrowth and ingrowth of cells, takes place {Fig. 13, A);
the flat epidermic layer is not included in the process.
This point is, of course, the dorsal lip, the ingrowing sheet
of cells the roof of the archenteron ;® the latter consists of
two or three cell layers and terminates anteriorly with
a free margin. The ingrowth proceeds until sufficient
material has been laid down for the embryo, and the
archenteron has reached a good length. Meanwhile, a
similar, though much smaller, ingrowth has been taking

Fig. 14.
Serramis (after Wilson).
Transverse sections of the embryonic region.

A. The roof of the archenteron {arch.) is undifferentiated.

B. The roof of the archenteron is differentiated into notochord (n.ch.)
in the middle line, and mesoderm [mes.) and endoderm [end.) at the sides.

place round the whole of the periphery of the blastoderm ;
this portion is merely extra-embryonic mesoderm {Fig.
13, B, C). When the process has, with the formation of
the embryo, come to an end at the dorsal lip, the
remainder of the edge of the blastoderm continues to
grow with its lip-like edge over the surface of the yolk
until this is completely enveloped ; the anterior edge,
however, travelling faster than the lateral portions, sweeps
round the vegetative pole and appears behind the embryo

^ Wilson inclines to the view that the roof of the archenteron is pro-
duced by delamination as the thickened dorsal lip grows backwards.

30 JenkinsON, Germinal Layers of Vertebrates.

as the ventral lip of the circular blastopore (yFig. 12)
{Fig. 13, D). The closure of the blastopore therefore
falls into the same two periods and takes place in
essentially the same manner as it does in the Elasmo-
branchs.

Let us now return to the formation of the germ-
layers in the embryonic region. The superficial portion
of the blastoderm is the ectoderm. The endoderm, as
well as the notochord and mesoderm, is contained in
that sheet of cells — the roof of the archenteron — which
has grown forwards beneath the dorsal lip {Fig. 14, A).
In this sheet two narrow slit-like cavities presently

Fig. 15.
Scnanus (after W^ilson).
Formation of the alimentary canal {al.c.) by fusion, and folding of the
two endodrrmal plates {end.) ; s.n.ch. sub-notochordal rod.

appear, one on each side of the middle line ; each cavity
is parallel to the surface of the sheet, and soon extends
outwards to its edge ; the whole sheet is thus split on
each side with an upper and a lower plate, a median tract
remaining undivided {Fig. 14, B). This middle strip
then separates as the notochord from the two lateral
"portions. The latter consist now each of an upper meso-

Manchester Memoirs, Vol. I. ( 1 906), No. 3-

31

dermal and a lower endodermal plate. To complete the
alimentary canal {^Fig. 15) the two plates of endoderm
first fuse by their inner edges below the notochord ; the
outer edges then bend down, meet, and so close the
central wall of the gut.

A variation in the mode of formation of the notochord,
recalling the similar difference between the frog and the
lamprey, has been described by Henneguy in the trout ;
here the split in the roof of the archenteron passes right
across the middle line ; two cell plates are thus formed, a
lower, the endoderm, and an upper, which then becomes
differentiated into a median notochord and two lateral
mesodermal sheets.

It is clear that the close similarity which obtains
between the Teleostei and the Elasmobranchii, in respect
of the closure of the blastopore, extends also to the
mode of formation of the germinal layers,

GANOIDEI.

Our knowledge is here unfortunately very slight.

From the observations of Sobotta, Whitman and
Eycleshymer, and Bashford Dean, we know that in Ainia,
Lepidosteus, and Acipenser the segmentation of the ovum

Fig. 16.
Formation of the germ layers in Acipenser (after Dean).
A. Sagittal section. B. Transverse section of dorsal half of embryo.
Lettering as before.

32 Jenkinson, Germinal Layers of Vertebrates.

is intermediate between the holoblastic and the mero-
blastic types : the yolk becomes but very slowly divided.

The closure of the blastopore — during which the
material for the notochord and mesoderm is brought into

1/

Fig. 17.
Formation of the germ-layers in Amia (after Sobotta).

A. Axial section of the egg at the close of segmentation ; the yolk only
partially segmented.

B. Sagittal section after closure of the blastopore. Letters as before.

position and the archenteron formed — is bilaterally
symmetrical, exhibiting the usual preponderance of over-
growth at the dorsal lip. The blastopore becomes com-
pletely circular, and is provided with a ventral lip
{Figs. 16, 17).

DIPNOI.

With regard to the Dipnoi we are, however, thanks to
the researches of Graham Kerr and Semon, better
informed.

In Ceratodiis the &^^ is holoblastic. Closure of the
blastopore is bilaterally symmetrical, there is a ventral as
well as a dorsal lip, and a spacious archenteric cavity is
formed, which does not communicate with the segmenta-
tion cavity. The roof of the archenteron consists of a
broad cell plate {Fig. 18, A) which becomes divided
into a median strip, the notochord, and two lateral
mesodermal sheets. The vegetative cells, which line the

Manchester Memoirs, Vol. I (1906), No. 3.

33

floor of the archenteron, grow up beneath the mesoderm,
and, fusing in the middle hne below the notochord,

Fig. 18.
Ceratodus (after Semon).

A. Transverse section, showing wide archenteron.

B. The plate of cells forming the roof of the archenteron is being
undergrown by yolk-cells from the sides.

Lettering as above.

complete the dorsal wall of the alimentary canal {Fig.
18, B).

In Lepidosircn on the other hand, the o.^^ inclines
towards a meroblastic segmentation. The blastopore

Fig. 19.
Lepidosircn (after Graham Kerr).

A. Sagittal section during overgrowth of dorsal lip {d.l.), and forma-
tion of archenteron {arch.). Note that there is no ventral lip.

B. Transverse section, showing dorsal mesoderm (we-j.) brought into
position with the overgrowth of the lip of the blastopore, and notochord
{n.ch. ) split off the roof of the archenteron.

34 Jenkinson, Germinal Layers of Vertebrates.

closes bilaterally by an overgrowth of the usual character,

but there is no ventral lip {Fig. 19, A) ; on this side

the area of small animal cells is extended by delami-

nation from the yolk. Dorsally the mesoderm is carried

into place during the overgrowth of the blastoporic lip ;

ventrally it is differentiated from the vegetative cells

{Fig. 19, B). The notochord is split off the roof of the

archenteron (which then becomes the gut), as in the frog,

which this type indeed closely resembles in all respects,

except in the amount of yolk and the absence of the

ventral lip.

AMPHIBIA.

Urodeld.

In the Urodela, the process of germ-layer formation

Fig. 20.
Formation of the germ-layers in the Axolotl.
(Original.)
A."]^ Sagittal section at the time of closure of the blastopore.

B. Transverse section of the same stage. Letters as before.

C. Transverse section at a later stage, to show the conversion of the
whole of the archenteric roof into notochord. The lateral yolk-cells are
growing in towards the middle line.

Manchester Memoirs, Vol. I. (1906), No. JJ. 35

is practically identical with that which is observed in the
Anura, except in one in:iportant respect. In the bilateral
closure of the blastopore, the presence of a ventral as well
as a dorsal lip {^Fig. 20, A) and the formation of the
mesoderm from a double source, the two groups closely
resemble one another ; but while in the frog the under
layer of the roof of the archenteron persists as the dorsal
lining of the digestive tract, in the Urodeles the sheet of
cells in question becomes wholly converted with the upper
layers of the roof into the notochord, just as is the case in
Petrojiiyzon, and the roof of the gut is formed by an in-
growth of vegetative cells from the sides {Fig. 20, B, C).

GymnopJiiona.

The last group of the Anamnia whose germinal layers
remain to be considered is the Gymnophiona. We here
fbijlow Brauer's excellent description of HypogeopJiis.

In the Gymnophiona the o.^^ is so laden with yolk
that it nearly approaches the meroblastic type, and the
result of segmentation is what may fairly be described as
a blastoderm resting on a partially divided yolk. The
blastoderm comprises a superficial epithelium of columnar
cells, thinner and lower at the margin, and covers several
irregular layers of scattered cells, which are more abund-
antly supplied with yolk. The cavities between these
cells are equivalent to the ordinary segmentation cavity
{Fig. 22, A). Below the segmentation cavity is the yolk,
divided at its surface into cells, and containing scattered
nuclei throughout its substance. Immediately round the
blastoderm the surface of the yolk is also partially
segmented.

At one point — the posterior middle point — of the
edge of this blastoderm, the dorsal lip appears ; it
exhibits the usual radial arrangement of cells {Fig. 22, A).

36 Jenkinson, Germinal Layers of Vertebrates.

The dorsal lip quickly grows back, and produces a long
archenteron, which at its anterior end opens into the
segmentation cavity. The roof of the archenteron,
which seems to be derived entirely from the superficial
layer of the blastoderm, consists of a plate of columnar
cells, its floor of the partially segmentedyolk (^Fig. 22, B).

Fig. 21.
Gymnophiona.
A — D. Four successive stages in the closure of the blastopore and
formation of the ventral lip in Hypogeophis (after Brauer). The blastoderm
only is shown ; y.p. yolk -plug.

E. Diagrammatic section of the egg of Ichthyophis (after Sarasin) to
show the embryo resting on a partially segmented and unenclosed yolk.

The process of overgrowth is of course not limited to
the dorsal lip, but extends to the immediate right and
left. Surface views show that the transverse groove, the
outward sign of this lip, soon becomes crescentic ; the
horns of the crescent then grow not only backwards, but
towards the middle line as well, approaching one another

Manchester Memoirs, Vol. I. (1906), No. 3.

37

until they meet, and so form what is the ventral lip of the
now circular blastopore {^Fig. 21, A — D). In section it is
seen that there is a slight ingrowth at the lateral and at
the ventral lips of a plate of cells continuous with the

Fig. 22.

Hypogeophis (after Brauer).

Three stages in the closure of the blastopore.

A—C. Formation of the dorsal lip, archenteron, communication of the
latter with the segmentation cavity {s.c). formation of the yolk-plug and
ventral lip. Sagittal sections.

D. Transverse section showing the lateral lips of the blastopore (/./.).

similarly formed plate which forms the roof of the
archenteron in front ; beneath the plate is a slit-like
space, also, of course, archenteric ; in the midst of the
blastopore is the projecting typically Amphibian yolk-

38 Jenkinson, Germinal Layers of Vertebrates.

plug {Fig- 22, C, D). The resemblance of a section
through the blastopore to a section througli the circular
blastopore of a frog or newt is in fact complete. If the
section were continued, however, through the entire egg
a very striking discrepancy would be at once apparent, a
discrepancy which depends on the difference in the mode
of formation of the ventral lip.

In other Amphibia the ventral lip is developed from
that point in the subequatorial transition zone which is
diametrically opposite to the dorsal lip, and the whole of
the vegetative surface of the egg becomes consequently
covered up when the blastopore closes. Here, on the
contrary, the anterior edge of the blastoderm and a large
part of the lateral edges have no share whatever in the
ventral lip, which arises entirely by the fusion of the
extremities of the two lateral lips ; as a result the vegeta-
tive hemisphere remains entirely uncovered. This is
shown very well in Sarasin's figure of IcJitJiyopJiis
{Fig. 21, E). In other words while in the Anura and Uro-
dela (and all other Anamnia) the whole of the edge of the
blastoderm becomes converted into a blastoporic lip, the
posterior point being the dorsal, the anterior becoming
sooner or later the ventral lip, in the Gymnophiona only
a small posterior region of the blastoderm's margin ever
becomes active in this way, and this small portion gives
rise to the dorsal and the two lateral lips, which latter by
their fusion produce such a remarkable similitude of the
ventral lip of the other forms. The importance of this
fact for the correct understanding of the relations of the
blastopore to the blastoderm in the Amniota cannot
possibly be overestimated.

To return to the germinal layers. The superficial
layer of the blastoderm is now of course the ectoderm.
The plate of cells which forms the roof of the archenteric

Manchester Memoirs, Vol. I. (1906), No. 3. 39

cavity becomes divided, as it does in Ceraiodtis, into three
portions ; a narrow median strip, the notochord, and two
lateral sheets of mesoderm {Fig. 23, B) which are con-
tinuous with one another behind the yolk-plug by the cell
plate invaginated at the lateral and ventral lips. The
mesoderm has in fact precisely the same relations as in
the frog (or other Anamnian) at a similar stage.

Fig. 23.
Hypogeophis (after Brauer).
Differentiation of the roof of the archenteron into notochord and meso-
derm ; formation of the roof of the definitive gut {al.c.) by ingrowth and
undergrowth of yolk-cells.

Brauer states expressly that no additions are made to
either the notochord or the mesoderm from any other
source.

The roof of the gut is completed by upgrowth and
ingrowth of vegetative cells below the middle layer
{Fig. 23, B).

Let us now, before proceeding to the Amniota,
endeavour to express, in as brief a form as possible, the
principal facts of germ-layer development as they occur in
the Anamnia.

40 Jenkinson, Germinal Layers of Vertebrates.

It appears that —

(i) In all cases the edge of the blastoderm (applying
this term to both microlecithal and megalecithal
forms) becomes in whole or in part the lip of the
blastopore. In all cases the posterior edge be-
comes the dorsal lip, in all cases except the
Gymnophiona the anterior edge becomes the
ventral lip (unless this is absent).

(2) The blastopore closes bilaterally ; the closure

involves an overgrowth and ingrowth of cells
which is most vigorous at the dorsal, less vigorous
at the lateral, and least vigorous at the ventral
lip.

(3) During this closure an archenteric cavity is formed

and the material for the ectoderm, the notochord
and the mesoderm, and the roof of the archen-
teron is brought into its definitive position. The
notochord lies along the median line in front of
the dorsal lip, the lateral mesoderm sheets pass
into one another behind the ventral lip. All
three sets of cells ectoderm, notochord and meso-
derm, and roof of archenteron are continuous
with one another at the lips of the blastopore.

(4) Distinct discrepancies are observable in the

manner in which sets of cells which resemble
one another in origin are used in the production
of the germinal layers; this is particularly the
case with the roof of the archenteron, which
may form the whole, ox\ only a part of, or be
absolutely excluded from the definitive ali-
mentary tract.

We may now turn to the Amniota.

Manchester Memoirs, Vol. I. (1906), No. 3. 41

AMNIOTA.

Whereas in the Anamnia the blastoporic lip is formed
at the edge of the blastoderm in the Amniota the blasto-
pf)re lies wholly within the latter. The Amniote blasto-
derm^ consists of two layers an upper and a lower
(fiequently termed epiblast and hypoblast and identified
with the definitive ectoderm and endoderm). Well within
the margin of this blastoderm a blastopore (primitive
groove) is formed leading into an archenteron ; concomi-
tantly the material for the germ-layers is laid down. All
these structures are in many cases derived in the first
instance solely from the upper layer, though eventually a
connection with the lower layer is set up. The edge of
tlie blastoderm, which is entirely independent of the
blastopore, grows steadily over the surface of the yolk
finally enclosing it at the vegetative pole.

The whole process is far clearer in the Reptiles than
in either of the other two groups. They will accordingly
be considered first.

REPTILIA.

The principal authorities for the early development of
the Reptilia are Will, Wenckebach, Mehnert, Mitsukuri,
and Ballowitz. The best account is perhaps Will's
description of the formation of the layers in Platydactylus,
a Gecko.

There is distinguishable in the blastoderm at the close
of segmentation a circular or oval area placed excentri-
cally towards the posterior end ; this area is the embryonic
shield i^Fig. 24). The blastoderm consists of two layers,
an upper and a lower ; the upper layer consists of
cylindrical cells in the embryonic shield, of flat cells in the
surrounding region ; below it is the segmentation cavity.

'•*See Note 7, page 18.

42 Jenkinson, Germinal Layers of Vertebrates.

The lower layer is an irregular sheet of scattered rounded
cells, not arranged at present in an epithelium, and is
constantly being reinforced by the addition of cells from
the nucleated yolk beneath (^Fig. 25, A). For the lower
layer I propose — in order to avoid any morphological
implications — to employ the term " paraderm," first
suggested by Kupffer. Between the paraderm and the
yolk is a shallow cavity, the subgerminal cavity. At one
point in the posterior margin of the embryonic shield the
upper layer and the paraderm are continuous ; this point
of fusion is termed by Will the primitive plate {^Fig.
25, A,//.)-

The paraderm cells next arrange themselves in a flat
epithelium {^Fig. 25, B). At the same time a depression
makes its appearance in the primitive plate ; the anterior
margin of this depression is the dorsal lip of the blasto-
pore.

Seen from the surface tlie dorsal lip presents the
appearance of a transverse groove at the hinder margin of
the embryonic shield. The groove rapidly becomes cres-
centic, the horns of the crescent grow back, meet, and fuse
behind the primitive plate, which now corresponds exactly
to the Amphibian or rather to the Gymnophionan yolk-
plug {^Fig. 24, A — C). A study of sections brings out the
similarity more clearly still.

Beneath and in front of the dorsal lip there is produced
by invagination of the upper layer of cells a cavity which
rapidly increases in length until it reaches the anterior
end of the embryonic shield i^Fig. 25, C, D) ; the cavity is
proportionately broad {^Fig. 27, A). The roof consists of a
layer of columnar cells which at the dorsal lip turn over
in the ordinary way into the cells of the upper layer. The
floor is now, even in the region of the primitive plate,
perfectly distinct from the underlying paraderm. In

Manchester Memoirs^ Vol. I. (1906), No. 3.

43

Fig. 24.
Three stages in the formation and closure of the blastopore in Platy-
dactylus (after Will). The oval embryonic shield lies on the blastoderm ;
only part of the latter is represented.

.H

.2>

■ffiMSngs

Fig. 25.
Platy dactylus (after Will).
Sagittal sections of five stages in the development of the blastopore and
archenteron.

A. The upper layer and the paraderm {p.d.) are fused in the primitive
plate (/./.) at the posterior margin of the embryonic shield ; the latter is
distinguished by its columnar cells from the fiat cells of the surrounding
blastoderm.

B, C. Formation of the archenteron : dj. dorsal lip, y.p. yolk-plug,
tnes.v. mesoderm at ventral lip.

E. Fusion of the archenteric and subgerminal (j.^.c.) cavities.

44 Jenkinson, Germinal Layers of Vertebrates.

front it consists of a single layer of cubical cells ; behind
the dorsal lip it is thickened — the primitive plate — and
from the thickening there proceeds backwards a narrow
tongue of cells between the upper layer and the paraderm.
A transverse section through the blastopore shows the
mass of cells of the primitive plate flanked on each side
by a projecting blastoporic lip and sending out between
the upper layer and the paraderm two lateral sheets of
cells {Fig. 26, A).

h-i A

Fig. 26.

A series of four transverse sections through the blastopore and
archenteron of Platydactyliis (after Will).

A. Posterior section showing yolk-plug {y.p.) and mesoderm {mes.)
springing from the lateral lips (/./.). Underneath is the paraderm (p.d,).

B. Section in front of the yolk-plug but behind the dorsal lip.

C. Section in front of the dorsal lip ; the floor of the archenteron is
fused with the paraderm.

D. The floor of the archenteron, together with the underlying
paraderm, has come away, and the archenteric and subgerminal cavities are
in communication with one another.

The resemblance between these structures — ignoring
for the moment the paraderm — and those seen in the
Amphibian egg when the blastopore has become circular
is sufficiently obvious. The dorsal lip and the lateral
lips (there is no ventral lip in the Reptiles) clearly
correspond in the two cases ; the mass of cells in the
primitive plate embraced by these lips is the yolk plug ;
the cavity of invagination is the archenteron in which

Manchester Memoirs, Vol. l. (1906), No. 3.

45

floor corresponds to floor, and roof to roof; lastly the
sheets of cells projecting beneath the upper layer at the
sides of and behind the blastopore are the equivalents of
the mesoderm formed at the lateral and ventral lips in the
Amphibia.

PSESSSSSBlSSSg5gg55|gjSSros;

2>

Fig. 27.

Four stages in the formation of notochord, mesoderm, and endoderm
in Platydactyhis (after Will).

A. The floor of the archenteron is still intact.

B — D. The archenteron has come into communication with the
subgerminal cavity. Its roc)f is thickened in the middle line to form the
notochord (n.ck.) : the lateral parts of the roof are the mesoderm (mes.).
The paraderm is growing in from the sides to form the lining of the
alimentary tract.

From this comparison it follows of course that cells
which are the morphological equivalents of the yolk-cells
of the Amphibia are to be found in the upper layer of
the Reptilian blastoderm ; and that that layer cannot
be termed the ectoderm until the process of invagination
is complete.

4-6 Jenkinson, Germinal Layers of Vertebrates.

In other cases the resemblance may be just as striking ;
in the turtle, Trionyx, for example, the yolk-plug projects
in the characteristic Amphibian fashion i^Fig. 28).

The floor of the archenteron now fuses throughout
with the paraderm below ; and as soon as the fusion is
completed perforations begin to appear in the fused
layers. They seem to be unable to keep pace with the
general growth of the blastoderm, and to become
first stretched and then fenestrated. But to whatever
causes the perforation may be due, the floor of the archen-
teron with the underlying paraderm completely disappears,
and the archenteron then communicates freely with the
sub-germinal cavity {Fig. 25, E). The roof of the archen-
teron is now inserted by its edges into the surrounding
paraderm {Fig. 27, B). This fusion of the archenteron

^^I^^H^§-

Fig. 28.
Transverse section of the blastopore of 7rionyx (after Mitsukuri).
Lettering as before ; the yolk is dotted.

with the sub-germinal cavity is, as we shall have occasion
to see more fully later on, quite comparable to the com-
munication of the archenteron with the segmentation
cavity in the Gymnophiona ; the paraderm must then
be regarded as homologous with a part, but only a part,
of the yolk-cells in this group.

The median strip of the roof next thickens to form
the notochord, and separates from the two lateral portions
which then become the mesoderm {Fig. 27, B, C, D).
These lateral plates pass posteriorly, of course, into the

Manchester Memoirs, Vol. I. (1906), No. 3. 47

dorsal lip, where they are perfectly continuous with the
sheets of mesoderm produced at the sides of and behind
the blastopore. The mesoderm thus exhibits all the re-
lations which it has in the Anamnia.

The lining epithelium of the alimentary canal is
derived from the paraderm, which grows in from the sides
beneath the mesoderm and notochord towards the middle
line. The gut is subsequently folded off from this layer
in the ordinary fashion.

The notochord and mesoderm may therefore be said
to be laid down in the Reptilia, as we have seen them to
be in the Anamnia, at the lips of and during the closure
of the blastopore, a closure moreover which is as bi-
laterally symmetrical here as it is there.

We have only now to consider one or two interesting
points in some other forms.

An embryonic shield distinct from the surrounding
blastoderm has been observed in most cases {Cistiido,
Emys, Trionyx, Lacerta, Tropidonotus, Crocodilus).

A primitive plate is found in Lacerta and Tropido-
notus, but not in Emys.

In Lacerta, according to Wenckebach, the lip of the
blastopore is not the only source of origin of notochord
and mesoderm ; both receive additions in front by the
proliferation of cells of the anterior paraderm. A similar
double origin of the middle layer has been described in
several other forms ; it^may be compared with the double
origin of the mesoderm in Petroniyzon, the Frog and some
other Anamnia.

In Eviys on, the other hand the paraderm takes no
part in the formation of any portion of the embryo what-
ever, but gives rise to the lining of the yolk-sac alone.
The roof of the archenteron in this form splits into two
sheets, an upper, which again becomes sub-divided into

48 Jenkinson, Germinal Layers of Vertebrates.

a median notochord and lateral mesoderm, and a lower
which becomes the epithelium of the gut. Such dis-
crepancies recall the similar diversities amongst the
Anamnia.

AVES.

The conditions observable in the Birds are very
readily derived from and very easily understood in the
light of those which obtain in the Reptiles.

The primitive groove is simply a laterally compressed
blastopore. In front of the anterior end — the dorsal lip —
the notochord is produced i^Fig. 30, A) ; to right and left
of the notochord are the sheets of mesoderm which,
springing from the sides — the lateral lips — of the primitive
groove i^Fig. 29, B), are continued into one another at its
posterior end, where there may be an actual ventral lip
{^Fig. 30, B). The archenteric cavity has, however, in most
cases disappeared ; between the sides of the primitive
groove, which, even in those cases in which it is most
reduced, exhibit the characteristic structure of blastoporic
lips {^Fig. 29, B), is merely a mass of cells — the representa-
tive of the yolk-plug — fused with the paraderm beneath ;
and the belated 'neurenteric canal' is the sole vestige of the
archenteron and the communication which we have seen
to be effected between it and the sub-germinal cavity in
the Reptiles.

In some cases, however, described by Schauinsland^
the archenteric cavity is better developed, and the blasto-
pore quite similar to that of the ReptiIia(/^^>. 30). The primi-
tive streak and groove invariably originate in the upper
layer (^Fig. 29, A), fusion with the paraderm being merely^
secondary ; only after the germ-layers have been formed,
can the upper layer be described as ectoderm. The
paraderm always gives rise to the alimentary tract ;
whether it also provides cells for the anterior extension of

Manchester Memoirs, Vol. I. (1906), No. 3.

49

the notochord and mesoderm, as in Lacerta and some
other Reptiles, is perhaps an open question ; this was

3

Fig. 29.

Two stages in the formation of the primitive streak and groove in the

Chick ; transverse sections.

(Original.)

A. 12 hours ; proliferation of cells in the upper layer of the blastoderm.

B. 15 hours ; formation of lateral lips to the blastopore (primitive
groove) ; a diminutive yolk-plug is seen between them ; mesoderm is
spreading from the sides of the groove between the upper layer and the
paraderm, and the whole is fusing secondarily with the latter.

B
Fig. 30.
Sagittal section of the blastoderm of the Sparrow (after Schauinsland).
A. Anterior half. B. Posterior half.

In front of the dorsal lip the notochord has been produced ; below it is
a slight archenteric cavity.

There is a slight ventral lip with a posterior tongue of mesoderm
{mes.v.) behind it. The cells of the primitive streak {p.s.) have fused with
the paraderm [p.d.).

Stated by Balfour to be the case in the chick, but denied
by Kolliker.

50 Jenkinson, Germinal Layers of Vertebrates.

MAMMALIA,

The egg of the oviparous Monotremata is megalecithal
and meroblastic. At the close of segmentation there are
two layers in the blastoderm — an upper layer and a
paraderm ; beyond this, nothing of importance is known
of the germinal layers.

The Qgg of the Placental Mammals, on the other hand,
is so small as to be almost alecithal ; but there is abundant
justification for the view that these Mammals are descended
from forms with large-yolked eggs, a justification which
is largely based on the mode of formation of the germinal
layers, which very closely resembles that observed in the
Reptiles and Birds, especially in the former.

The germinal layers, however, only begin to be formed
after another — quite peculiar — process has taken place ;
andasthis process has — in my opinion, quite erroneously —
been confounded more than once with ' gastrulation,' or
a part of gastrulation, I may perhaps be allowed to state,
or rather re-state, the view which I have been led to adopt
on this matter.

The fully segmented ovum consists of an outer layer
of cells surrounding an inner mass. The outer layer —
known as the 'trophoblast,' from the part it subsequently
plays in the formation of the placenta — is the homologue
of the false amnion, strictly speaking of the ectoderm of
the false amnion, of the Sauropsida ; the inner mass
contains within itself all the material necessary for the
production of the true amnion, the embryo, and the yolk-
sac. A cavity next appears separating the inner mass
from the outer layer at all points except one, the embry-
onic pole, and the ovum is now termed a ' blastocyst ' ;
at the same time the inner mass becomes differentiated
into an ' embryonic knob ' attached to the trophoblast.

Manchester Memoirs, Vol. I. ( 1 906), No. 3.

51

and a layer of flattened cells, the paraderm, which begins
to grow round the blastocystic cavity. Segmentation
therefore is followed in the Placentalia by the separation
of the elements of the trophoblast from those destined to
give rise to the embryo and the remainder of its foetal
membranes, and this ' precocious segregation ' seems to
have occurred phylogenetically during the gradual loss of
yolk which the egg of these Mammals has undergone.

Fig. 31.
Formation of the germ layers in Mammals.

A. Vespertilio (after Van Beneden). Sagittal section showing the
archenteron opening to the exterior behind the dorsal lip, and communicating
with the subgerminal cavity {s.g.c.) below. A small portion of the floor
remains.

B. Transverse section of the embryonic area of the mouse. (Original).
The roof of the archenteron (which will become the notochord) is inserted
by its edges into the paraderm ; from its sides spring two lateral sheets of
mesoderm.

From a part of the embryonic knob the true amnion
is eventually formed, in a manner which varies in different
types. The rest of this knob becomes the upper layer of
the embryonic area, and is the equivalent of the
embryonic shield of the blastoderm in the Sauropsida.

The development of the germinal layers now proceeds.
As in the Sauropsida, the necessary material is brought

52 Jenkinson, Germinal Layers of Vertebrates.

into position by the bilateral closure of a blastopore and
simultaneous formation of an archenteron. These
processes originate here, as in the Birds, exclusively in
the upper layer, though subsequently a secondary
fusion with the paraderm occurs. The blastopore
may be laterally compressed — a primitive groove — and
the archenteric cavity practically absent or reduced to
a very narrow ' neurenteric ' or 'chorda-canal'; but in
some cases — the best instance is Vespertilio (van Beneden)
{^Fig. 31, A) — the blastopore and archenteron are as well
developed as in the Reptiles. The archenteron always
comes into secondary communication with the subgerminal
blastocystic cavity. In front of the dorsal lip the noto-
chord is formed in the ordinary manner from the roof of
the archenteron or mass of cells — ' Kopffortsatz ' of the
older authors — which is its representative ; from the sides
of the blastopore spring the lateral sheets of mesoderm
confluent, as in other cases, with one another behind
{^Fig. 31, B). According to many authors, the blastoporic
lips — the primitive groove — are the sole seat of formation
of notochord and mesoderm, but Heape has described in
Talpa, and Hubrecht in Sorex and Tar sins, the formation
of an anterior portion of both notochord and mesoderm
from the paraderm ; '" and the latter author has further
observed the origin of a peripheral ring of mesoderm
from the lower layer in the same two genera. This
peripheral mesoderm has been recorded by Bonnet in the
sheep, where, however, it is denied by Keibel.

The paraderm — growing in from the sides — seems to
complete the endodcrm in all cases.

^" Mr. Assheton informs me that this also occurs in the Rabbit.

Manchester Memoirs, Vol. l. {igo6), No. ^. 53

We can now summarize the principal facts for the
Amniota as follows :

(i) A bilateral blastopore is formed within the
blastoderm and towards its posterior end.

(2) The blastopore leads into an archenteron, which

is also bilateral, its extent being greatest beneath
the dorsal lip. The archenteron is developed
wholly from the upper layer.

The primitive groove of Birds and Mammals
is a laterally compressed blastopore, the ' neuren-
teric canal ' (' chorda - canal '), a rudimentary
archenteron.

(3) The floor of the archenteron fuses with the para-

derm secondarily, its cavity secondarily with
the subgerminal cavity. The roof of the archen-
teron is then inserted by its edges into the
outlying paraderm.

(4) The notochord is formed from the roof of the

archenteron in front of the dorsal lip, the meso-
derm from the same layer at the lateral lips, the
two sheets of mesoderm being continuous behind
the blastopore. Additions may be made to the
notochord and mesoderm from the paraderm.

(5) The lining of the alimentary canal is usually,

though not always, a derivative of the paraderm.

(6) As in the Anamnia, discrepancies are observable

in the destinies of sets of cells of similar origin.

It only remains for us to consider the possibility of
deriving the Amniote from the Anamnian condition.

54 Jenkinson, Germinal Layers of Vertebrates.

THE RELATION OF THE BLASTOPORE OF THE
AMNIOTA TO THAT OF THE ANAMNIA.

The first serious attempt to get over the difficulty was
made by Balfour. Balfour held that the primitive groove
of the Amniota represented only the dorsal portion of the

^ 2)

Fig. 32.

A. Diagram to illustrate Balfour's view of the relation of the primitive
groove of Amniota (d.b.l.) to the edge of the blastoderm or lip of the yolk
blastopore {y.b.l.). The two are connected by a posterior suture (the two
dotted lines) ; y. yolk.

B — D. Diagrams to illustrate Rabl's view of the relaticn of the
Anamnian to the Amniote blastopore (after Keibel).

B. Amphibian egg with closed blastopore and ventrally placed yolk-
mass ; the latter is bursting through the ventral body wall at x.

C. Result of augmentation of the yolk.

D. Protamniote condition reached by still further enlargement of the
yolk. The yolk is black in all three figures ; y.c. yolk cells.

Anamnian blastopore {Fzg. 32, A, d. b. /.), the ventral
portion being found in the edge of the blastoderm {y. b. /.) ;
the two portions he supposed to have been originally

Manchester Memoirs, Vol. I. (1906), No. 3. 55

united by a posterior suture behind the primitive streak,
which suture he thought might have originated in the
same sort of way as the suture which connects the dorsal
lip with the yolk-blastopore in an Elasmobranch {Fig. 8, A).
The paraderm Balfour compared simply to the yolk-cells
or yolk of Anamnia, the upper layer to the animal cells
or 'ectoderm,' and made the most of the secondary con-
nection between the primitive streak and the paraderm,
whereby the necessary continuity of all germ-layers in
the lips of the blastopore was effected.

Now to this theory there are two grave objections ;

(i) The upper layer contains elements — yolk- plug
and floor of archenteron — which must undoubt-
edly be homologized with some of the yolk-cells
of Amphibia ;

(2) The edge of the blastoderm in the Amniota does
not behave as the lip of the Anamnian blasto-
pore ; at it there is no rolling over of cells and
no formation of mesoderm.

It may be added that the attempt made by Duval to
show that the primitive grove originated at the edge of
the blastoderm has completely broken down. The
blastopore is from the first moment of its appearance
completely within the blastoderm.

Another solution was offered by Rabl (yFig. 32, B — D).
Recognising the resemblance between the Amniote primi-
tive groove and the circular Anamnian blastopore and
that the edge of the blastoderm was a new formation,
Rabl suggested that the increase of yolk which had taken
place in the transition from the Amphibia to the Protam-
niota had finally ended in the rupture of the body-wall
on the ventral side, with the result that in the Amniota
the body-wall of the Amphibian embryo with its circular

56 ]-E^Kllii?>0'i^, Ggrminal Layers of Vertebrates.

Fig. 33.
Diagrams of the relation of the Amniote blastopore to that of the
Gymnopkiotta and of that of the Gymnophiona to that of other Anamnia.
(Original.)

Manchester Memoirs, Vol. l. {igo6), No. ^. 57

blastopore is represented by a blastoderm lying on the
surface of the yolk and including the primitive streak.

Here again we have to urge a fatal objection. As
will occur to everyone, and as Keibel has already pointed
out, increase of yolk must take place at the vegetative
pole while the &gg is still in the ovary, and therefore long
before gastrulation can have come to pass and the yolk
assumed its ventral position. An increase in the yolk
could only delay (as in the fishes) or modify (as in the
Gymnophiona) the closure of the blastopore ; whereas it
is the already closed blastopore with which Rabl begins.

Keibel himself has put forward a theory — which I
cannot discuss here — that gastrulation in the Amniota
takes place in two phases, the second of which is
expressed in the primitive groove ; rejecting Rabl's
hypothesis he returns to the earlier view of Balfour and
Duval. There is still one more suggestion, that brought
forward by Ziegler, which must be briefly noticed. Ziegler,
indeed, comes near the mark in taking the Gymnophiona
as his starting point, though he misses it very widely in

In all the diagrams the (Anamnian) blastoderm is dotted, and that part
of its edge which becomes a blastoporic lip represented by a thick black line.

A. The Frog.

(i) At the end of segmentation.

(2) When the blastopore is circular.

(3) After rotation.

B. Lepidosiren.

(i), (2), and (3), similar stages.
There is no ventral lip.

C. Gymnophiona.

The yolk is increased, the blastoderm smaller, the region of its edge
which becomes a blastoporic lip more limited than in Lepidosiren ; but a
ventral lip is formed in (2) by union of the lateral lips.

D. Amniota.

The blastopore is formed as in Gymnophiona at the edge of the
embryonic shield (the inner darkly dotted area) ; the blastoderm is enlarged
by the addition of a surrounding belt of yolk-cells (outer lightly dotted area).

58 Jenkinson, Germinal Layers of Vertebrates.

making the Gymnophionan correspond to the whole of
the Amphibian blastopore, and in returning to Rabl's
erroneous conception of the ventral augmentation of the
yolk.

So far then no theory has been propounded which, in
my opinion at any rate, affords an adequate explanation
of the facts. And yet the explanation is obvious enough.
In the Gymnophiona [Fig. 33, C), as we have seen
already,

(i) The blastoderm is a circular area of columnar
cells resting upon and surrounded by a partially
segmented yolk.

(2) At the posterior margin of this blastoderm a

dorsal lip is formed and lateral lips quickly
follow ; the lateral lips then grow back encir-
cling a small area of the yolk behind which they
meet and fuse to form a ventral lip to the now
circular blastopore. In this process the anterior
margin of the blastoderm is wholly unconcerned.

(3) The archenteron which is formed at the blasto-

poric lip, opens into the segmentation cavity.

(4) Notochord and mesoderm are derived from the

ingrowth at the lip of the blastopore, while the
endoderm arises from the yolk-cells.

And now let us describe the same stages in the
development of a Reptile i^Fig. 33, D).

(i) The embryonic shield is a circular area of
columnar cells resting upon a paraderm and
surrounded by a zone of flattened cells.

(2) At the posterior margin of this embryonic shield
a dorsal lip is formed and lateral lips quickly
follow ; the lateral lips then grow back encircling

Manchester Memoirs^ Vol. I. (1906), No. 3. 59

a small area of the outer zone of cells behind
which they meet and fuse to form a (virtual, or
in some cases an actual) ventral lip to the now
circular blastopore. In this process the anterior
margin of the embryonic shield is wholly
unconcerned.

(3) The archenteron which is formed at the blasto-

poric lip opens into the sub-germinal cavity.

(4) Notochord and mesoderm are derived from the

ingrowth at the lips of the blastopore, while the
endoderm arises from the paraderm.
I venture to think that, considering the difference in
the amount of yolk present in the two cases, a closer
parallelism could hardly have been expected ; viutatis
nmtandis the description of either might be applied to
the other type.

It seems clear then that

(i) The embryonic shield of the Amniota is the
representative of the blastoderm of the Anamnia.

(2) The marginal zone of the upper layer of the
Amniota, together with the paraderm, represents
the yolk-cells or nucleated yolk of the Anamnia
{Fig. 32, D).
In passing from the Gymnophiona — which I suppose
are as nearly related as any living group to the ancestors
of the Amniota — to these higher forms, we have, therefore,
only to suppose that with the enlargement of the vegeta-
tative hemisphere of the egg segmentation has become
restricted, not to the blastoderm alone (as in the fishes),
but to the blastoderm and those circumjacent and sub-
jacent cells which in the Gymnophiona are partially
segmented from the yolk. In the most primitive Reptiles
the paraderm cells are still crowded with yolk, and still

6o J EN KIN SON, Germinal Layers of Vertebrates.

retain a connection, in the primitive plate, with the cells
of the upper layer. In the higher forms — the Birds and
Mammals — this similitude of the paraderm to Amphibian
yolk-cells is lost, as well as the connection between the
two layers in the primitive plate. The upper layer fuses
only secondarily with the paraderm when the primitive
groove and streak are developed.

There still remains an accessory problem. What is
the relation between the Gymnophiona and the rest of
the Anamnia ? The Gymnophionan blastopore though
apparently precisely similar to that of the frog corresponds,
as we have seen, only to a part of the latter. It is clear,
therefore, that the former must be derived from a blasto-
pore in which the anterior edge of the blastoderm takes
no part and which is consequently devoid of a ventral lip.
And very fortunately such a blastopore is to be found in
a form which not only is less megalecithal than the
Gymnophiona, but which may be presumed to stand not
far from the direct line of Amphibian descent ; I refer, of
course, to Lepidosiren {Fig, 33, B), which may be taken
back in its turn, if necessary, to Petroviyzon. The increase
of yolk that has taken place between the Dipnoi and the
Gymnophiona has simply resulted in the restriction of the
process of germ-layer formation to a perpetually diminish-
ing region of the blastoderm placed symmetrically about
the posterior margin or dorsal lip {Fig. 33, C). These
lateral lips — fusing behind in a ventral lip — have then
produced a circular blastopore superficially resembling
that of the remaining Amphibia {Fig. 33, A). The relation
between the two is, of course, just that which Balfour
supposed to exist between the primitive groove of the
Amniota and the Anamnian blastopore ; but it hardly
needs, I hope, to be pointed out that the present theory
differs ioto coelo from his ; for the representative in the

. Manchester Memoirs, Vol. I. (1906), No. II. 61

Amniota of that portion of the edge of the y\namnian
blastoderm, which in the Gymnophiona is excluded from
participation in the lip of the blastopore, is to be sought
not in the edge of the blastoderm but in the margin of
the embryonic shield.

It is interesting to notice that in various Anamnia —
Petromyson and Rana — the same union of the archenteric
and segmentation cavities which is observed in Gymno-
phiona and Amniota also occurs.

Before concluding this discussion I ought to glance
at the view which Oscar Hertwig has recently published.
Without attempting any more serious solution than is
contained in the hint that the clue to the problem must
be sought in the Gymnophiona, Hertwig expresses it as
his opinion that the separation of the Amniote blastoderm
into upper and lower layers represents in a measure a
process of gastrulation though the cavity of invagination
is entirely absent. To the archenteron of the Amniota,
on the other hand, he refuses a complete homology with the
similarly named cavity in the lower forms ; it is, according
to him, merely a ' Mesodermsackchen ' involved solely
in the production of the middle layer and the notochord.
Although it is true that the homology in question is — if
destiny as well as origin be taken into consideration —
usually, though not always, an incomplete one, still for
the purposes of comparison of one and the same stage in
development throughout the series it may be regarded as
complete ; for the upper layer contains in the floor of the
invagination cavity and the primitive plate cells which
can and must be compared with the cells which line the
floor of the archenteron and form the yolk-plug in the
Anamnia.

62 Jenkinson, Germinal Layers of Vertebrates.

The broad features of germ-layer development in the
whole group of Vertebrata may now be summed up in
the following formulae.

(i) By the germinal layers are to be understood the
ectoderm — the source of the epidermis and its
derivatives, of the nervous system and the organs
of special sense ; the endoderm — or lining
epithelium of the alimentary tract and its
diverticula ; and the mesoderm (including the
notochord) — the layer which gives rise to the
musculature, the skeletal and connective tissues,
the vascular system and the urogenital organs.

(2) The material for the ectoderm, notochord and

mesoderm, and roof of the archenteron is brought
into position during an overgrowth and in-
growth of cells which takes place at the lip of
the blastopore during the closure of the latter.

(3) This closure is bilaterally symmetrical, taking

place more actively at the dorsal lip than at any
other point, and leads to the formation of a
bilateral archenteric cavity, the extent of which
is greatest anteriorly and least posteriorly,

(4) At the lips of the blastopore the ectoderm, the

roof of the archenteron, and the notochord and
mesoderm are continuous with one another.

(5) The notochord lies medially in front of the dorsal

lip ; the mesoderm sheets spring from the lateral
lips and pass into one another posteriorly behind
the ventral lip.

(6) {a) The blastoporic lips are formed from the

margin of the blastoderm ^Anamnia) or of an
embryonic shield lying wholly within the blasto-

Manchester Memoirs, Vol. I. (1906J, No. 3. 63

derm (Amniota). The dorsal lip is always
formed at the posterior margin.

{b) In all Anamnia, except the Gymnophiona,
the ventral lip is developed immediately (microle-
cithal forms), or after a considerable interval
(megalecithal forms), from the anterior margin
of the blastoderm.

(c) In the Gymnophiona and Amniota the
ventral lip is always developed by the fusion of
the lateral lips, the anterior margin of the blasto-
derm (or embryonic shield) taking no part
whatever in the process.

(7) Considerable discrepancies exist in the manner in
which the endoderm is formed in different types.
This may arise from the roof only, the floor
only, or from both roof and floor of the archen-
teron, or from the paraderm.

It is these anomalies which lead us naturally to a
critical consideration of the morphological value of the
germinal layers.

Part II.— THE MORPHOLOGICAL SIGNIFI-
CANCE OF THE GERMINAL LAYERS.

The phylogenetic speculations which have so long
dominated our conceptions of development are all directly
traceable to a hypothesis first enunciated by Ernst
Haeckel, thirty years ago. This hypothesis is the
' Gastraea-Theorie.'

There occurs, according to this theory, in the ontogeny
of all multicellular animals, a two-layered form, the Gastrula,
produced by invagination or some modification of that
process from a one-layered form, the Blastula. The two

64 Jenkinson, Germinal Layers of Vertebrates.

layers of the Gastrula are the primary germ-layers, the
ectoderm and the endoderm, and are homologous or
morphologically similar throughout the series ; the cavity
of the Gastrula is the archenteron or primitive gut, and
invariably becomes the alimentary canal ; the opening of
the cavity is the primitive mouth. With the mesoderm
the theory is as such not concerned. Further, this Gastrula
is a recapitulation in ontogeny of a phylogenetic form,
the Gastraea, the common ancestor of all the Metazoa ;
to this primeval form the living Coelenterates are most
nearly allied. Before proceeding to examine the cre-
dentials of this creed, we may glance briefly at some of
its later successors and descendants.

The planula theory of Lankester — put forward almost
contemporaneously — closely resembles the hypothesis of
Haeckel, differing only from it in substituting for the
Gastrula a Planula or Gastrula without a mouth, in which
the primary endoderm arises by delamination instead of
invagination. The theory was intended to meet the
difficulty presented by the variability in the fate of the
blastopore.

Passing over a few years we find Balfour" returning
to a belief in the Gastrula as a form which ' reproduces
'with more or less fidelity a stage in the evolution of the
' Metazoa permanent in the simpler Hydrozoa.' Still later
developments of the Gastrula theory, restricted, however,
to the Vertebrata, are those propounded by Lwoff" and
Hubrecht.*' Lwoff identifies the two primary layers — the
animal and vegetative cells — of the vertebrate embryo
with the ectoderm and endoderm of the Coelenterate or
Gastraea, but part only of the archenteric cavity with the

11 "Comparative Embryology," vol, 2, ch. 13, 1881.

12 Bull, soc.imp. nat. AIoscou, n.s., vol. 8, pp. 57, 160, 1894.

13 " Furchung und Keimblattbildung bei Tarsius spectrum," Verh. k.
Akad. Wetensch. Amsterdam, (2), vol. 8, no. 6, 1902.

Manchester Mei?ioirs, Vol. I. (1906), No. 3. 65

digestive tract of these. Two distinct processes, according
to this author, have hitherto been confused under the
head of gastrulation. One — which is alone represented
in Invertebrate forms — is an invagination of the primary
endoderm into the segmentation cavity ; the other —
pecuHar to the Vertebrates — is the overgrowth and in-
growth of a ' dorsal plate ' of cells (the roof of the
archenteron) produced entirely from the primary ecto-
derm and taking place most actively at the dorsal lip of
the blastopore. The ectodermal dorsal plate gives rise
to notochord and mesoderm, while the alimentary tract is
derived entirely from the invaginated endoderm cells. It
must be pointed out that, quite apart from the difficulties
of the Gastrula theory, of which it is of course a modifica-
tion, Lwoffs hypothesis is very seriously contradicted
by the fact that the roof of the archenteron — his ' dorsal
plate' of 'ectoderm' — in many cases either contributes to
or forms the whole of the digestive canal. To argue, as
Lwoff actually does, that in these cases primary endoderm
ceils have found their way into the roof of the archen-
teron is, it is superfluous to point out, destructive of the
universality of the ' ectodermal ' origin of the latter, for
which, as a matter of fact, a good deal might be said.

Hubrecht's theory resembles that of Lwoff in attribut-
ing to the ' dorsal ' or ' notochordal ' plate the sole function
of notochord and mesoderm formation ; but differs from it
in seeing in the segmentation cavity, with which, as this
embryologist urges, the archenteron in all cases becomes
confluent, the primitive Coelenterate gut, the archenteron
itself being then regarded simply as an ectodermal stomo-
daeum.

Now underlying the Gastrula theory and its modifica-
tions there are two fundamental assumptions, assumptions
which must be vindicated if the theory is to be maintained.

66 JenkinSON, Germinal Layers of Vertebrates.

In the first place, it is postulated that the primary layers
of the Gastrula are completely homologous, that is, homo-
phyletic or homogenous, throughout the Metazoan series ;
and in the second, that they bear certain constant and
invariable relations to the organization of the adult on
the one hand, and to that of the reputed ancestral form
on the other. Haeckel recognises this perfectly, ' Der
' vvesentliche Inhalt dieser Gastraea-Theorie,' he says
' beruht auf der Annahme einer wahren Homologie (ohne

* welchen die Gastraea-Theorie nicht haltbar ist) der primi-
' tiven Darmanlage und der beiden primaren Keimblatter

* bei alien Thieren mit Ausnahme der Protozoen. "

Further, since, as is held, it is phylogeny which is
repeated in and ultimately determines the form which
.ontogeny takes, any doubt which exists as to the homo-
logy of an organ in the fully developed form may be
completely set at rest by a reference to the germinal layer
in which it arises, the homologies of the germ-layers in
their turn being determined by reference to their mutual
anatomical relations or, when possible, their mode of origin
in the segmenting ^^^. It is origin then, and not destiny, if
we push these theories to their logical conclusion, which is
the final criterion of the homologies, whether of the organs
of the adult or of the layers of the germ.

The first question accordingly which we have to
answer is, are the so-called primary layers always homo-
logous? Let us consider this in the light of the facts
which vertebrate development has taught us.

At the close of segmentation the microleicithal verte-
brate ovum consists of an animal hemisphere of small ceUs
and a vegetative hemisphere of large cells, identified
by Haeckel with the primary ectoderm ahd endoderm
respectively. In the megalecithal eggs segmentation is

^^ Jenaische Zeitschr., vol. 8, p. lo, 13, 1874. • .

Manchester Memoirs, Vol. I. (1906), No. 3. 6"]

confined to a blastoderm, but in this we distinguish again
an upper and a lower layer of cells ; as far as their relative
positions are concerned, the upper layer corresponds very
closely with the animal hemisphere ; the lower layer,
together with the yolk, to the vegetative hemisphere in
the former case. There thus exists a very fairly complete
morphological similarity or homology between the primary
germinal layers throughout (we except, of course, the
Placental Mammals, in which any attempt to identify the
outer layer and inner mass with the primary germ layers
has long ago been abandoned), though it must be borne
in mind that this rests merely on the structure of the fully
segmented &^%, not on anything that we know of the
origin of these layers during segmentation.

To the second question, however, is the relation which
these primary layers bear to the parts of the embryo in
the next stage constant and invariable, we can most cer-
tainly not return an affirmative answer ; for while in the
Anamnia the floor of the archenteron and the yolk-plug
are derivatives of the primary * endoderm,' in the Amniota
they are of purely 'ectodermal' origin ; nor is it even
possible to say that the cell-layer which forms the archen-
teric roof is homologous throughout, its ' ectodermal ' and
^ endodermal ' origin having been equally maintained.

If we prefer to compare the primary layers not with
the structures visible in the next stage, but directly with
the organization of the adult, our difficulties are just as
great. For now the alimentary tract of Teleostei (and
Elasmobranchii ?) is ectodermal, that of Ceratodus,
Gymnophiona, and Amniota ' endodermal,' that of the
Frog and Lepidosiren of mixed origin. It is abundantly
clear that each of the primary germinal layers, as defined
by Haeckel, though constant in its structural relations
' throughout the series, may be, and often is in its destiny,
exceedingly diverse.

68 jENKlNSOlsi, Ger;/iiua/ Layers of Vertebrates.

But, it will no doubt at once be said, in the Vertebrates
the Gastrula must be looked for, not at the end of seg-
mentation, but in that stage in development when the
blastopore is already closing. Though this is neither
Haeckel's two-layered Gastrula (except in the case of the
* Disco-Gastrula ' of the Teleostei), nor the Planula of
Lankester, we will^nevertheless examine the hypothesis in
its new form. In this stage the Vertebrate embryo
possesses a blastopore leading into an archenteron, the
archenteron has a floor and a roof, and from the lips of the
blastopore the notochord and mesoderm have been pro-
duced between the archenteron and the ectoderm. There
is a very close similarity between all embryos at this stage^
and we may — if we disregard for the moment the
differences in their mode of origin — fairly describe their
parts as homologous. But do these homologous parts
exhibit that absolute constancy in their relation to adult
structures which the theory demands ? By no means ;
the ectoderm truly has the same fate in all cases, except
when, as in the skin glands of Amphibia, it gives rise to
muscles, or, as in the thymus of Elasmobranchs, to blood-
corpuscles, and so has the^mesoderm, except when it gives
rise to enamel ; but the roof and floor of the archenteron
vary widely in their subsequent behaviour. The roof may
form notochord only, or notochord and mesoderm only,
or both these, and either the roof or the whole of the
definitive gut. The floor may form a part or the whole of
the lining epithelium of the digestive tract, or be
excluded from it altogether. It will be urged, of
course, that the archenteric cavity nevertheless becomes
the alimentary canal ; but here the Amniota involve
us in great difficulties ; so little a part does the archen-
teron play in the formation of the lumen of the gut that
Hertwig has felt constrained to call it a ' Mesodermsack-

Manchester Memoirs, VoL I. (1906), No. 3. 69

chen,' and Hubrecht a ' Stomodaeum,' looking for the
primitive Coelenterate gut in the cavity of the blastocoel.
But, perhaps, it will be maintained, this is a mere caenoge-
netic adaptation, and the embryo does possess in this
stage to all intents and purposes a primitive gut opening
by a blastopore, an ontogenetic type of frequent, though
not — as witness the Planula theory — of universal occur-
rence. The desired homologies are now sustained only
by explaining away as caenogenetic every inconvenient
fact, which is suicidal ; by refusing to define a cavity by
the nature of its walls, which is absurd ; and by ignoring
dissimilarities in origin, though origin is in every other case
held to be the sole ultimate test of community of descent,
which is inconsistent. Such reasoning must surely be the
very last refuge of the destitute phylogenist.

In fine, if we establish homologies, as we are perfectly
entitled to do, on the basis of the structure which the ^^^
presents at the end of segmentation these homologies are
without constant reference to those resemblances which
appear in the next stage during the closure of the blasto-
pore ; and these, in their turn, bear no necessary and
definite relation to the structure of the adult forms.
Either origin or destiny may be taken as decisive of
homology ; but origin and destiny do not of necessity
coincide. The appeal to destiny must either rob the
supposed recapitulatory stage of all independent signi-
ficance or involve us in a vicious circle, while any attempt
to make origin the ultimate criterion will inevitably end
in the subversion of that homology which we are bound
to predicate of the adult organs through and through.
The Vertebrata then afford no support to the Gastraea or
to any other theory which attaches a morphological,
phylogenetic, or recapitulatory significance, a value for the
determination of adult homologies, to the germinal layers.

yo Jenkinson, Gervdnal Layeis of Vertebrates.

We have now to consider whether the embryology of
other groups promises any better prospect of success.

The two-layered — Gastrula or Planula — stage is, of
course, of widespread occurrence in all the Invertebrate
groups. Without pausing at present to enquire into the
origin of these two primary layers, let us merely follow
out their fate in a few cases, and in particular the fate of
the endoderm. Frequently, it is true, the endoderm gives
rise quite simply to some part of the alimentary canal ;
but it has always been a source of great difficulty that
the region of this canal which arises from the endoderm
is so exceedingly variable in extent. While it may and
generally does happen that the major part is endodermal
in origin, there are several cases amongst the Arthropods
in which only a very small midgut is derived from this
source, the remainder of the digestive tract being stomo-
daeal and proctodaeal. Apparently then, unless develop-
ment misleads us, ' there are,' to quote Balfour, ' instances
in which a very large portion of the alimentary canal is
phylogenetically an epiblastic structure.' ^^ Balfour felt
this so strongly that he suggested that there was a border
land between the epiblast and the hypoblast which might
be developed indifferently from either. Worse, however,
is to follow. The endodermal midgut, already reduced
in these cases almost to a rudiment, may in others
entirely disappear.

In the Cephalopoda, the alimentary canal, according
to Watase,^" (though this is denied by Korschelt,") is
formed entirely of the stomodaeal and proctodaeal in-
vaginations, the yolk-sac epithelium — which is unani-

'^■' "Comparative Embryology," vol. 2, ch. 13.

'^ Stud. Biol. Lab. Johns Hopkins Univ., vol.4, P- 163 — 183, 1887 — 90.
. ^^ " Festschrift fiir Leuckart," Leipzig, 1892, p. 347 — 373.

Manchester Memoirs, Vol. I. (1906), No. 3. 71

mously identified with the primary endoderm — being
wholly excluded from it. Watase, indeed, goes so far as
to say on the strength of this result that the digestive
systems of a squid and a snail are not homologous.
Earlier observers, it may be added, had also failed to
trace the endoderm into the gut. And this is not a
solitary example. In 1884 Witlaczil ^'^ described the
origin of the alimentary canal in Aphis from stomodaeum
and proctodaeum alone ; Voeltzkow'^ has maintained the
same of Musca ; Lecaillon"" of various Coleoptera; while
Heymons,^^ it is interesting to note, though denying to
the yolk-cells any share in the development of the gut
in Orthoptera, has shown that in Lepidura and Cmn-
podea the digestive cavity arises in the fashion required
by theory from the primary endoderm.

Physiologically it is not difficult to understand these
anomalies ; they find a parallel, for example, amongst the
Vertebrates in the behaviour of the yolk-nuclei of Elasmo-
branchs and Teleostei, structures set aside for the
elaboration of the yolk but playing no part in the
production of the embryo. Morphologically, however,
they offer obvious and insuperable difficulties to the
ordinary theories of gastrulation.

Nor is it merely in the development of the digestive
cavity that this want of harmony between origin and
destiny makes itself so painfully felt. Irregularities of
this kind are exhibited by all the germ-layers, and often
in the most startling and irrefutable manner. For

'^ Zeitschr. wiss. ZooL, vol. 40, p. 559 — 696, 1884.

^^ Arb. Zool. Inst. IViirzbnrg, vol. 9, p. i — 43, 1889.

-" " Rech. sur Toeuf et devel. embiyon. de quelques chrysomelides."
Theses Facult. sci. Fat-i?, ser. A, no. 299, 1898 ; abstr., Zool. Centralblatt,
vol. S, p. 813 — 6, 1898.

■^^ Sitzungsb. Gesell. Naturf. Freunde Be}-lin, 1897, p. ill — 123;
Sitzungsb. Akad. Wiss. Berlin, 1894, pt. i, p. 23 — 27.

72 Jenkinson, Gerviinal Layers of Vertebrates.

fortunately it is possible, in the 'cell-lineages ' of many
Molluscs, Annelids, and Turbellarians, by tracing back
an organ or layer to a single cell to establish its origin
beyond cavil or question. It will be sufficient for the
purposes of this argument to cite the most striking
discrepancies.

In the three groups just mentioned the segmentation
of the &%% proceeds upon a very regular so-called ' spiral '
plan {Fig. 34). The first two divisions — meridional,
and at right angles to one another — produce four cells

Fig. 34.
Diagram of the formation of the first (la — \d), second {za — 2d), and
third (3a — yl) quartettes of micromeres from the macromeres {A — D) in the
egg of a Turbellarian, Mollusc, or Annelid.

lying, roughly speaking, in one plane. Of these four cells,
one is left, one is anterior, one right, and one posterior ;
they are known as the macromeres, and termed respectively
A, B, C, and D. Each macromere now buds off towards
the animal pole, and in a right-handed direction a small
cell or micromere ; the four constitute the first quartette,
and are termed (following Conklin's notation) \a, lb, \c,
and id. A second quartette {2a, etc.) is given off left-

Manchester Memoirs, Vol. I. (1906), No. 3. 73

handedly, a third quartette (3(T, etc.) right-handedly, and
a fourth (4^, etc.) left-handedly again. The micromeres
themselves obey the same law of alternating direction
of cleavage in successive divisions. These four quartettes
and the residual macromeres contain the material for
the germinal layers. In one case {TetJiysf'^ a fifth
quartette of micromeres is produced.

Let us now follow, first, the fate of these cells in
different cases.

The first quartette becomes the ectoderm of the pre-
oral region, including the apical organ, when present, the
cerebral ganglia, and the whole or a part of the prototroch.

In the second quartette the posterior blastomere, 2d,
is often very large, and known as the first somatoblast ;
it gives rise to the ventral plate of ectoderm. The cells in
the remaining quadrants {2a, 2b, 2c) usually also give rise
to ectoderm, forming sometimes the stomodaeum or
oesophagus {Capitella)^^ sometimes contributing cells
to the prototroch. There are, however, many cases in
which mesoderm also arises from this group of cells ; in
Crepidula ^* derivatives of 2a, 2b, 2c produce larval
mesoblast, in Unto-'" a descendant of 2a, in Aricia"^^
possibly 2c and 2d ; in Aplysia depilans (according to
Georgevitch^^) 2c and 2d are the primary mesoblasts, and
in Dreissensia"-^ mesenchyme is derived from the descend-
ants of all four cells ; while in the Turbellaria, according to
Hallez^® and Lang''", the blastomeres of this quartette are

'^^ Viguier, C, Arch. Zool. Exp. ser. 3, vol. 6, p. 37—58, 1898.

*» Eisig, H., Mitt, Zool. Stat. Neapel, vol. 13, p. 1—267, 1899.

2* ConUlin, E. G.,Journ. Morph. vol. 13, p. I — 204, 1897.

*^ Lillie, ¥.,Journ. Morph., vol. lO, p. I— 84, 1895.

'^o Wilson, E. B., Biol. Led. Woods Holl, 1898, p. 21—42.

2'^ Anat. Anz., vol. 18, p. 145 — 174, 1900.

^^ Meisenheimer, J., Zietschr. wiss. Zool., vol. 69, p. I — 137, igoi.

^' " Contributions a I'histoire naturelle ties Turbellaries." Lille, 1879.

** "Die Polycladen des Golfes von Neapel. Leipzig, 1884.

74 ]eis!KINSON, Gervii;m/ Lajers of Vertebrates.

devoted to the production of mesoderm alone, though
Wilson^' has stated that in Leptoplana some ectoderm is
formed from them as well.

Similarly, while the cells of the third quartette are
usually devoted to ectoderm formation — ectoderm which
may surround the blastopore or provide the lining for the
stomodaeum or proctodaeum — there are many cases in
which a part, for example 3<; i, yi i in Capitella^'^ "^b i,
T,c I in Physa^f derivatives of 3«, 3^, 3c in Podarke^^ and
of all four quadrants in Dreissensia^^ "^d in Teredo^^ and
Cyclas^\ or the whole, as in the Turbellaria — though again
this is denied by Wilson — of these blastomeres contributes
to the larval or permanent mesoderm.

In the fourth quartette the posterior cell ^d is as a rule
set aside for the exclusive development of the mesoblast
bands ; but in many cases this blastomere contains endo-
dermal elements as well {Nereis^^, Crepidiila, Aplysia,
Podarke, Dreissensid) ; while in the Turbellaria, and
apparently in Teredo^ it is purely endodermal. In
Capitella, on the other hand, it gives rise not to the meso-
blast bands, but to the larval mesoderm, together with
some ectoderm, and in TetJiys it is stated by Viguier to
be restricted with the other cells of the same quartette to
the ectoderm.

The remaining cells of this quartette (4^, Afi, 4c) with
the residual macromeres (i\, B, C, and D) constitute the
endoderm. Here there are fewer exceptions although, as

3^ Loc. cit.

'^ Eisig, H., Loc. cit.

32 Wierzejski, A., Biol. CentralbL, vol. 17, p. 388—394, 1897.

=>* Treadwell, A. 'L.,Journ. Morph., vol. 17, p. 399—486, 1 901.

3^ Meisenheimer. Loc. cit.

»• Hatschek, B., Arb. Zool. Lnst. Wicn, vol. 3, p. 1—42, iSSi.

2^ Stauffacher, W., Jenaische Zeitschr., vol. 28, p. 196 — 240, 1894.
W.'lson, E. '2,., Journ. Morph., vol. 6, p. 361 — 480, 1892.

Manchester Memoirs, Vol. I. (1906), No. 3. 75

just mentioned, 4^, 4/7, 4^ in TetJcys are ectodermal and in
Cydas and Teredo the macromeres in these three quad-
rants as well ; (in the two latter genera the first division
is so unequal, as also the following division of the larger,
posterior cell, that A, B, and C were termed by Stauffachcr
and Hatschek micromeres in contrast to the larger
macromere D| In Clepsine no endoderm is derived
from D^l

There is clearly then no necessary regularity in the,
subsequent behaviour of cells which in origin are identical;
and if now conversely, we trace back identical organs to
their origins precisely the same lack of agreement will
become manifest.

The prototroch, for example, though always ecto-
dermal, is not limited to a particular set of cells : for
while it is true that certain cells in the first quartette —
derivatives of la. 2. — \d.2 — always take part in its forma-
tion, it frequently happens that these are reinforced by
secondary trochoblasts derived from the second quartette
{Crepidula, Trochus,^'^ Ainphitritef' etc.). The stomodaeum
again, usually formed from the second, may {Aremco/a/^
arise from the third, or sometimes {Capitella and
IscJinochitonY from both second and third quartettes.
Nor does the ectoderm itself even escape these variations ;
ordinarily derived from the first three quartettes of
micromeres its formation may (Turbellaria, though not
according to Wilson) be limited to the first quartette, or-
on the other hand extended to the fourth ( Tethys) and
even to some of the residual macromeres {Cyclas and
Teredo).

^^ Whitman, C. O., Quart. Journ. Micr. Set., vol. i8, p. 215—315, 1878.

*" Robert, A., Arch. Zool. Exp., ser. 3, vol. 10, p. 270—513, 1902.

*^ Medid, Journ. Morph., vol. 13, p. 227 — 326, 1897.

*- Child, C. M., Arch. Ent.-inech., vol. 9, p. 587 — 709, 1900.

*^ Heath, H., Zool. Jahrb. {Altai.), vol. 12, p. 567—656, 1899.

76 Jenkinson, Germinal Layers of Vertebrates.

In the origin of the mesoderm the anomaHes are even
more serious. In the majority of cases it is, truly, derived
from 4<y, though even then, it must be remembered, this
cell may contain endodermal elements ; but it may and
frequently does happen that the mesoderm receives
additional elements from other sources. In Crepidnla
descendants of 2a, 2b, 2c, in Unto of 2a, in Aricia either
of 2c, 2d, or of 3(:, 3*^, in Physa of 3(5, 3^ produce larval
mesoderm ; in Dreissensia and Liniax^^ mesenchyme
arises from both second and third quartettes ; in Capi-
tella 3^. I and -^d. i are the pole cells of the mesoblast
bands, 4<3^ producing merely larval mesoderm. Long ago
Lankester*^ showed that in Pisidiuin the ectoderm of the
Gastrula could give off mesenchyme cells, and now Meisen-
heimer and Tonniges*^ have demonstrated that in Di'-eis-
sensia, Limax, and Paludina, while the mesoblast bands
originate from d^d in the typical manner, the material for
the heart, pericardium, kidneys and gonads is derived from
the ectoderm later on. In Aplysia, on the authority of
Georgevitch, though not of Carazzi,*'' 2c and 2d are the
primary mesoblasts, in TetJiys $d instead of 4<^, while in
the Turbellarians {pace Wilson) the second and third
quartettes alone produce mesoderm and mesoderm alone.

Lastly, the endoderm is in Cyclas and Teredo confined
to quadrant D, while in Tethys it is only formed after a
fourth quartette of ectomeres has been given off

Here, then, in a series of forms in which we can apply
the test of origin in the most decisive and unequivocal
manner to questions of homology, that test breaks in
pieces in our hands ; here, as in the Vertebrates, origin

** Meisenheimer, J., Zeitschr. wiss. ZooL, vol. 62, p. 415 — 468, 1897.
*6 PhiL Trans., vol. 165, p. 1—48, 1875.
*® Zeitschr. wiss. ZooL, vol. 6i, p. 541 — 605, 1896.
* '' Georgevitch (/. c. ) describes A. depilans ; Carazzi {Anat. Ans. , vol. 17,
p. 77 — 102, 1900), A. limacina.

Ma?ichester Memoirs, Vel. I. (1906), No. 3. J J

and destiny do not in any cell, or set of cells, invariably
and necessarily coincide ; homologies founded on the
former may be the diametrical opposite of comparisons
based upon the latter, and in this dilemma it is to the latter
only that we can continue to adhere without reducing
comparative anatomy to an absurdity.

The facts of descriptive embryology might well be
deemed sufficient of themselves to warrant us in thus
finally relinquishing any hope of retaining the morpho-
logical or phylogenetic significance which for so long has
been attached to the germinal layers. Corroborative
evidence is, however, forthcoming from other quarters,
from the phenomena of budding and segmentation, from
pathology, and from the results of experimental embryo-
logy. Let us briefly consider these.

We have seen that in the normal development of the
fertilized ovum there comes a moment, at any rate in
some forms, when it is possible to detect three layers or
sets of cells which, however much they may vary in their
origin, bear constant relations to certain definite parts of
the adult organization by reference to which they may be
defined. Even this constancy is, however, not absolute ;
muscles, for example those of the skin-glands in Amphibia,
may be derived from the ectoderm, and enamel is said
occasionally to arise from mesodermal tissues. In the
mode of reproduction termed budding this inconstancy
may be much more marked.

The difficulties presented by the anomalous behaviour
of the germ-layers (or rather of their representatives) in
the budding of various Ascidians, is well known to
zoologists. The young bud is a two-layered sac, the outer
layer being formed in all cases of ectoderm ; the inner
layer, on the other hand, may be a diverticulum of the

78 Jenkinson, Germinal Layers of Vertebrates.

atrium of the parent (Botryllidae), in which case it is
ectodermal, or may arise from the dissepiment of the
stolon {Clavellina, Didemnum), in which case it is a
derivative of the epicardium, that is to say of an organ of
endodermal origin. Now, in spite of this diversity, the
behaviour of the inner layer is by the concurrent testimony
of Seeliger, Oka, Pizon, and Hjort^^ essentially the same
in all cases ; it gives rise, in fact, to the pharynx and
alimentary canal, to the atrial cavity, and, according to
Hjort, also to the hypophysis and nervous ganglion of the
bud,though a different mode of origin of the last-mentioned
is described by the other observers. While therefore in
the first case an organ, the alimentary tract, which in the
embryo is derived from the endoderm, is in the bud of
ectodermal origin ; in the latter organs, the atrium and
nervous system, which are ectodermal in the larva are
developed in the bud from the other layer.

The relation of gemmation to the germ-layers in the
Ectoproctous Polyzoa is just as great a source of per-
plexity. After segmentation, a two-layered stage is
reached, either by unipolar immigration or a process
wliich is reducible to this. The inner layer so formed —
comparable as far as its origin goes with the primary
endoderm of other forms — never takes any part in the
development of the adult alimentary canal ; although in
those few cases amongst the Gymnolaemata in which
the larva possesses a gut, this organ apparently arises
from these cells, the remainder becoming mesoderm.
Where the larva has no gut most of this inner mass
disintegrates and is ultimately found, together with the
degenerating larval organs, in the ' brown' body ; ' a small

** Seeliger, O., Sitzungsb. Akad. Wiss. Wien, voj. 85, pt. i, p. 361 — 413,
1882.

Oka, A., Zeitschi: wiss. ZooL, vol. 54, p. 521 — 547, 1892.

Pizon, M. A., Ann. Sci.Nat. (Zool.), {•/), vol. 14, p, i — 386, 1893..

Hjorc, J., AnaL Anz., vol. 10, p. 215 — 229, 1895.

Manchester Memoirs, Vol. I. (1906), No. 3. 79

portion is, however, reserved for the mesoderm of the
adult. In the Phylactolaemata, on the other hand, the
whole of this layer becomes mesoderm, and the cavity
which it circumscribes is shown by its subsequent
behaviour to be coelom. The larval organs, when present,
bear no relation to the corresponding organs of the adult,
but degenerate wholly and pass into the brown body.
The organs of the first polypide are formed from a bud
which is produced by an invagination of ectoderm covered
by a layer of mesoderm ; from the ectodermal layer are
developed not only the epidermis with the nervous
system but the whole of the digestive tract as well.^"

It is obviously exceedingly hard to reconcile such
facts with ordinary germ-layer theories ; though an
attempt to save the situation has been made by showing
that the polypide bud arises (in the Phylactolaemata only)
near the original pole of immigration of the inner mass,
and that there is a complete gradation from those types
in which the inner mass gives rise to the larval gut and to
the mesoderm through those in which the endodermal
portion degenerates in an early stage, to those in which
there is never at any time any endodermal portion at all.

The behaviour of the layers during the regeneration
of lost parts is as irregular as it is in budding. This has
been particularly shown to be the case in the regeneration
of the head and tail in various Annelids. The accounts
given by different observers, however, of the changes that
take place do not completely accord with one another.

According to Rievel^° [OpJiryotrocha, Nats, Allolobo-
phora, LuDibriciis) a mass of ' granulation tissue ' of mcso-

*" Korschelt, E. und K. Heider, " Lehrbu;h der vergl. Entwick. d.
wirbellosen Thiere," Jena, 1893-. , .

6 0 Rievel, H., Zeitschr, w/ss. Zoo/., vol. 62, p. 289-341, 1897.

8o Jenkinson, Germinal Layers of Vertebrates.

dermal origin is formed over the cut surface. This
becomes covered by an overgrowing layer of epidermiSj
and subsequently perforated by the forward (or backward
in the case of the regeneration of the tail) growth of the
blind end of the gut ; after the fusion of this with the
epidermis a new opening — mouth or anus — is effected.
It will be noticed that in this case the original stomodaeum
and proctodaeum are replaced by endodermal epithelia.
Haase's^^ description of the processes that occur in
Tubifex is very similar. MicheF', on the other hand,
states for a very large number of Oligochaeta and Poly-
chaeta — and von Wagner^^ {Lnmbriculus) and Hepke^^
{Nats) are at one with him in this matter — that the cap of
reparation tissue which developes over the wound is due
largely if not entirely to an active proliferation of the
epidermis. All three observers agree in deriving not only
the new nerve-cord but also the new mesoderm — including
the ' ectodermal ' nephridia — from this mass of cells ; but
differ in their description of the regeneration of the gut.
According to Michel the intestinal aperture made by the
cut never closes, but persists as the anus. Von Wagner
also derives the lining of the regenerated portions from
that of the old alimentary tract ; the gut closes first and
then grows through the reparation tissue to fuse with the
epidermis and open to the exterior ; there is only a very
slight invagination at the aperture. Hepke, on the con-
trary, describes the forward (or backward) growth of a
solid cord of cells derived from the epidermal reparation
tissue which, acquiring a lumen, opens into the old gut at
one end and to the exterior at the other.

^^ Haase, H., Zeiischr.zviss. ZooL, vol. 65, p. 211-256, 1899.

** Bull. Sci. de la France et de la Belgigtie, vol. 31, p. 245 — 401, 1898.

5 2 Zool.Jahrb. {Anat,), vol. 13, p. 603-682, 1900.

5* Zeitschr. wiss. ZooL, vol. 63, p. 263-291, 1898.

Manchester Memoirs, Vol. I. (1906), No. 3. 81

By the testimony, therefore, of all these authors
except Hepke, the original stomodaeum — which includes
the pharynx — and proctodaeum are replaced in the
regenerated parts by epithelia of ectodermal origin ; by
Hepke's account these epithelia are ectodermal but replace
(apparently) not only the stomodaeum and proctodaeum,
but originally endodermal portions of the digestive tract
as well. In either case it seems we are involved in a
contradiction. With regard to the new formation of the
mesoderm there is more agreement ; it is derived from the
epidermis, a fact which, once more, is out of harmony
with its embryonic development.

It should be added that Dendy ^^^ has shown that in
Antedon the visceral mass can be regenerated partly from
the mesoderm, partly from the ectoderm of the calyx ;
the new alimentary canal arises from the latter.

In this connection it may be mentioned that cases occur
in which tumours resembling one another in every respect
may, apparently, have been derived in one instance from
the ectoderm, in another from the mesoderm, or even the
endoderm. The most remarkable instance of this is,
perhaps, that quoted by Hansemann^^ where a malignant
growth of the epithelium of the gall-bladder had assumed
the structure of an epidermal cancroid. Many pathologists
have, therefore, given up the attempt to classify the
abnormal growths by reference to their developmental
origin.

Last of all it remains for us to discuss the bearing on
our question of a very important series of facts which the
experimental embryology of recent years has made known
to us.

6*« Stud. Biol. Lab. Owens College, vol. i, p. 299—312, 1886.
5^ " Die mikroskopische Diagnose der bosartigen Geschwiilste," p. 23,
Berlin, 1897.

S2 JENKINSON, Genuma/ Laj'ers of Vertebrates.

The experiments which most immediately concern us
are those instituted by Oscar Hertwig, Driesch, and other
embryologists on the segmentation of eggs submitted to
pressure, and on the behaviour of isolated blastomeres ;
experiments which in the first instance arose out of a
critical enquiry into the pretensions of the Roux-Weis-
mann hypothesis of preformation. This hypothesis, as it
is hardly necessary to point out, is a modern resuscitation
of the famous theory of evolution which was destroyed by
Wolff more than a hundred years ago, and postulates a
necessary predetermination of definite parts of the ferti-
lized ovum, or of its nucleus, for the production of particular
organs of the embryo. Segmentation in this case is a
qualitative process ; all that it has to do is to separate the
already different parts, and so make manifest that structure
which was invisibly present before ; it is, as Roux phrases
it, a ' Mosaikarbeit'

Now at bottom the theories which have been criticised
above are also attempts to throw back the structures of
the adult into definite predestined parts, not necessarily
of the unsegmented, but at least of the segmented ovum ;
they are in fact essentially preformationist,^'' and, more-
over, the logical outcome of that doctrine. For, if develop-
ment is strictly a process of 'self-differentiation' pro-
ceeding wholly from causes resident in each part and
independent both of other ' internal ' conditions — such as
mutual position — and of 'external factors' — such as the
physical and chemical composition of the environment —
then the potentiality of each part is predetermined and
limited, similarly constituted ova will develope through
similar stages into related forms, while any dissimilarity
occurring at any stage will lead to the production of
organs that are not homologous and forms that diverge,
and, since the prime cause of differentiation — the structure

^^ This has also been pointed out by Braem {I.e. p. 435).

Manchester Memoirs, Vol. I. (1906), No. % ^i

of the fertilized ovum — is itself a heritage from a long
line of ancestors, each individual will of necessity repeat
in its ontogeny the history of its descent. If, on the
other hand, there is no necessary constancy, even within
the limits of a single species, in the fate of cells of identical
origin, such a regularity need certainly not be claimed for
an entire class.

The question is thus narrowed down to a perfectly
plain issue, which may be brought to the test of experi-
ment ; with the preformationist hypothesis germ layer
theories must stand or fall.

We owe the pressure experiments, to which allusion
has been made, to Driesch^^ and Hertwig.^® The former
showed in Echinus, the latter in the Frog, that by this
means the nuclear spindles could be displaced, and the dis-
arranged nuclei made to pass into other than their normal
blastomeres, without affecting the normality of subsequent
development. In Hertwig's experiments, it is interesting
to observe, nuclei pass into vegetative which are ordinarily
allotted to animal cells, and vice versa. Wilson's^^ experi-
ments on Nereis are of a like nature. By pressure the
egg is made to segment into a flat plate of eight cells,
all of which, when the pressure is released, behave as
macromeres. Eventually cells are found in the lining of
the gut, which ought to have contributed to the ectoderm of
the first quartette.

The results of the experiments on the development of
isolated blastomeres fall into a regular series. At one end
of this stand the Vertebrates and Amphioxus, the Asci-
dians^^" and the Coelenterates ; here an isolated blastomere

^^ Zeitschr. -iviss. ZooL, vol. 55, p. I — 62, 1893.

^* Arch. mikr. Attat., vol. 42, p. 662 — 794, 1893.

59 Arch. Ent.-mech., vol. 3, p. 19—26, 1896.

^^"■'Dnt?,ch,ll., Arch. Ent.'iJieck., vol. i, p. 398— 411, 1895. E.G.
Conklin has, however, recently shown that in Cynthia there is a pre-
localization of definite organ-forming substances in the egg {Jotirn. Exp.
ZooL, vol. 2, p. 145 — 221, 1905).

$4 ]'El!iK\NSO'^, Germinal Layers of Vertebrates.

behaves from the first as an entire ovum, segmenting
normally and producing a whole embryo. It may be
remarked at this point that Hertvvig^" failed to obtain the
half-embryos of the frog which Roux produced by killing
one of the first two blastomeres ; that Schulze''' has made
double monsters by inverting the Qgg in the two-celled
stage ; and, most satisfactory of all, that Herlitzka"'^ has
succeeded in isolating the first two blastomeres of the
newt, each of which then developed into a complete larva.

In Amphioxus Wilson*^" was unable to obtain a
gastrula from a ^ blastomere, though a ^ blastomere
readily gastrulated. But in Hydroids Zoja*'^ found that
even a xV blastomere, a cell which is probably either
' ectodermal ' or ' endodermal,' would give rise to a normal
embryo.

In the Echinoderms ^^, however, although eventually a
normal whole larva results, the segmentation of an isolated
blastomere is partial, a curved plate of cells being produced
which only gradually rolls up to form a hollow blastula. It
is only from |, \, or ^ blastomeres that perfectly normal
gastruljE can be obtained ; but the | gastrulae are of course
developed from either animal or vegetative cells. In this
connection — the point which for us is of especial interest —
Driesch has shown that a normal pluteus can be formed
from the eight animal cells alone, the eight vegetative cells
alone, or even the four macromeres alone of the sixteen-
celled stage.

The Nemertines, as Wilson''" has recently stated,
resemble the Echinoderms ; segmentation of the isolated

8 0 Loc. cit.

8^ Arch. Ent.-mech., vol. i, p. 269 — 303, 1895.

*" Arch. Ent-mech., vol. 2, p. 352 — 366, 1896 ; vol. 4, 624 — 654, 1S97.

^'^ Joiirn. Morph., vol. 8, p. 579 — 638, 1893.

"* Arch. Ent.-iJiech., vol. I, p. 578—594, 1S95 '■> ^o'- 2, p. i — 31, 1896.

'^° Driesch, 11., Zeitschr. wiss. ZooL, vol. 53, p. 160 — 184, 1892 ; vol. 55,
p. I — 62, 1893; Arch. Ent.-mech., vol. 4, p. 75 — 124, 1897; Mitt. Zool.
Stat. Neapel, vol. il, p. 221 — 253, 1895.

Manchester Memoirs, Vol. l (1906), No. 3. 85

\ox \ blastomere is partial, but its ultimate development
total. Further, complete Pilidia may be reared from
either animal or vegetative fragments.*'''"

These forms lead us naturally to the Ctenophora and
Mollusca, the extreme term at the other end of our series.
Here the segmentation of the isolated blastomeres is,
as in the Echinoderms, partial ; the resulting embryo is,
however, partial too. In the Ctenophora"^ the product of
a \ blastomere is a little more than a half-larva, and
eventually regenerates the missing half ; but in the
Mollusca*'- — one of the groups we may notice with a
' mosaic ' type of segmentation — not only does each blasto-
mere segment as though the remaining blastomeres were
there, but no regeneration of the remaining parts ever
takes place.

If we had before us these results alone, it would be
easy to draw from them a strong confirmation of the
evolutionist theory. When, however, we remember
Wilson's experiments on Nereis — ^just another of these
'mosaic' forms — and when further we take into con-
sideration the great mass of evidence which the behaviour
of the isolated blastomeres of the other groups presents
to us, it is impossible, to say the least, to accord to the
preformationist hypothesis a universal validity. On the
contrary, this hypothesis finds in this evidence, as Driesch
puts it, its formal contradiction.

"^ Arch. Eiit.-viech., vol. i6, p. 411 — 458, 1903.

*® "According to Zeleny, larvae from \ animal blastomeres are devoid of
an archenteron, from g vegetative blastomeres of an apical organ {Joiirn.
Exp. ZooL, vol. I, p. 293 — 329, 1904).

«'^ Chun, C, " Festchr. f. Leuckart," Leipzig, 1892, p. 77 — 108. Driesch,
H. and Morgan, T. H., Arch. Ent.-mech. vol. 2, p. 204 — 224, 1896.

^^ Crampton, H. E., Arch. Ent.-mech., vol. 3, p. 1 — 16, 1896. The
results recently obtained by Wilson on Dentalhmi and Patella are similar
i/oin-/!. Exp. ZooL, vol. i, p. i — 70, 197 — 266, 1904).

86 Jenkinson, Germinal Layers of Vertebrates.

These clearly are cases in which, at certain stages,
the blastomeres are interchangeable, their destinies as
yet not fixed ; in determining their fate other factors,
their mutual position or rather the influence they exert
upon one another and — as experiment is abundantly
showing — the conditions of their environment, come into
play.

And what experiment has thus shown to be true of
the individual finds an obvious parallel in nature in the
formation, under the stress of varying internal or external
causes, of homologous organs from cells or layers of
unlike origin. In the one case as in the other the same
end is attained by paths that are diverse ; one of these
may be a recapitulation ; all cannot.

On all sides then — and we have now examined the
morphological theories of germinal layers from every
possible point of view — the facts forbid us to see in these
elementary organs of the embryo that definite pre-
determination for the performance of certain ontogenetic
functions which the hypotheses we have been criticising
demand. The germinal layers are not sets of cells of
universally identical origin which necessarily and in-
variably give rise to certain fixed parts of the adult
organization, but merely convenient terms for the
primordia of the structures of the adult.

Similarly constituted ova may and do — as we have
seen in the various groups of the Vertebrata, and still
more clearly in the spiral cleavages of Turbellarians,
Mollusca, and Chaetopoda — segment and gastrulate in a
precisely similar manner, and give rise to cells which in
origin are alike ; and this similarity in the segmentation
or gastrulation of the ova of related forms we may, if we
insist on retaining the word, perhaps still call ' recapitula-

MancJiester Memoirs, Vol. I. (1906), No. 3. 87

tion,' though it is not a recapitulation of any adult
ancestral type, but merely a repetition of similar onto-
genetic functions by cells which have inherited a similar
structure.*^' In destiny, however, such cells may be
exceedingly diverse : ' Furchungsmosaik,' as Driesch has
it, 'braucht kein Mosaik der Potenzen zu sein." It is
only within comparatively narrow limits that origin and
destiny can coincide.

Nor is the failure of embryology to provide an in-
fallible criterion of homology evident in the history of
the germ-layers alone. In organogeny we are often
reduced to an exactly similar impasse; the oviduct of
the Elasmobranchs, to take an example, bears the closest
resemblance anatomically to the same organ in the air-
breathing Vertebrates ; and yet devclopmentally it is
totally dissimilar, arising from the pronephric half of the
segmental duct, while in the other forms it has nothing
to do with any part of the excretory system. Morpholo-
gists, as a matter of fact, adopt the embryological evidence
when it suits them and ignore it when the facts are
inconvenient. In discussing that vexata qucestio the
homology of the ear-bones, for instance, Gegenbaur,™
while unhesitatingly accepting the equivalence, as based on
development, of the Mammalian stapes with the Saurop-
sidan columella and of both with the hyomandibular of
Fishes, just as unhesitatingly passes over as a caenogenetic
modification the origin of the Amphibian columella from
the auditory capsule.

This, of course, is the usual apology offered ; it is the
one which Haeckel himself suggested to meet difficulties

^* Such a repetition as that here indicated is what I understand O.
Hertwig to mean by his ' Modifikation des biogenetischen Grundgesetzes '
" Handbuch der Entwicklungslehre der Wirbelthiere." Einleitung, p. 56, 57,
Jena, 1901.

7 0 " Vergleichende Anatomic der Wirbelthiere," vol. i, p. 440, S96 sqq.
Leipzig, 1898.

88 ]E'^VilN^O^, Genninal Layers of Vertebrates.

which he clearly realised, and it may perhaps be admitted
when the exceptions are few and the rule general. But
when, as in germ-layer development, exceptions to the
rules of theory meet us at every turn, when everything
has to be explained away as caenogenetic — how much?
or rather how little ! of the palingenetic is left !

In the germ-layers, at least, between the conflicting
alternatives of origin and destiny, there is no media via.
To cleave to origin is to plunge into a quagmire of
absurdities ; to follow destiny is to abandon all hope of
finding any ultimate criterion in development, and to
return to that older conception of morphological simi-
larity or homology which is based simply on identity of
anatomical relations extending over a large series of forms
in the same stage of their life history.

And in this direction serious embryological thought is
steadily trending. Though some still seem to halt between
two opinions not a few — notably Driesch, Hertwig, Braem,
Child, Conklin, Treadwell, Morgan — have definitely re-
jected the ontogenetic criterion of homology and refused
any morphological significance, any phylogenetic value to
the germinal layers.

But though thus divested of the claims falsely set up
on their behalf the germ-layers remain, from another
point of view, the morphogenetic, structures of paramount
importance. The aim of the experimental embryologist
is to give a causal account of the sequence of develop-
mental phenomena, regarding development as one of the
functions of the organism to be studied by the ordinary
physiological methods ; and the problems which confront
him in this effort are, in the main, two. The first is to
describe in accurate terms the influence exerted upon the
embryo by its environment ; the second is to determine
the mutual relations which subsist between the parts of

Manchester Memoirs, Vol. I. {igo6), No. 3. 89

the embryo and between the parts and the whole ; he has
to discover, in a word, what are the external and what
the internal factors which govern the process of differentia-
tion. Differentiation sets in with the separation of those
elementary embryonic organs which we are accustomed
to speak of as the layers of the germ ; it is in a precise
physiological study of these organs that we must look for
the clue to one of the greatest of biological problems, the
problem of the epigenetic evolution of the complexity of
the adult from the apparent simplicity of the fertilized
ovum.

Manchester Memoirs, Vol. I. (1906), No. 4.

IV. Battack Printing in Java, with Notes on the
Malay Kris and the Bornean Sumpitan and
Upas Poison.

By John Allan.

Received and Read, January i6th, rgo6.

In these days of extended foreign travel there are
very few parts of the world which at some time or
another are not visited by travellers. It was my good
fortune to spend the greater part of the year 1905 in
journeying through some of the lands which are far out
of the beaten track, and in which there is much with
which we are unfamiliar. Of the Malay Peninsula, Java,
Borneo, and Sumatra, much has been written at various
times, but there are still many things of more than
ordinary interest pertaining to these far off countries and
their inhabitants about which little and, in many cases
nothing, is known.

Many of the native industries are certainly worthy of
greater attention than has so far been bestowed upon
them, and, certainly to us in Manchester, none of these
should be of greater interest than the process of Battack-
ing which has been practised by the natives of India
and the Further East for centuries, and which exhibits
the simplest form which calico-printing can possibly
take.

The commonest articles of wearing apparel amongst
the 30 million native inhabitants of Java, indeed through-
out the whole of Malaysia, are the Battack-printed
sarongs and slendangs and pieces of the same printed
material are almost universally used as a covering for the
head.

April ytk, igo6.

2 Allan, Battack Printing in Java.

The sarong, as a portion of the attire of both men
and wonnen amongst the Malays, is worn in many ways.
It is of such a size and shape, that, at the desire of
the possessor, it becomes trousers or petticoat, shirt or
overall, or even an article of bed-clothing, and it is the
only form of bathing-costume which the natives use.

The Malay is a true conservative in the matter of
dress, for, centuries ago, the sarong was worn as it is
to-day. In the very heart of central Java, there rises in
an open plain a massive pyramid of dark-grey stone, a
chaos of cupolas and spires, surmounted by a high
central dome. This is the Boro-bodoer, probably the
oldest Buddist remains in the world. It is an ancient
pile, dating back to the 7th or 8th centuries of this era,
and on its sculptured walls are to be seen figures of men
and women wearing the sarong as it is worn to-day.
Whether the sarongs of that time were battacked or not,
it is impossible to say, but it is known that a similar
method of printing on cloth was in use in Southern
India over 500 years ago, the wonderfully coloured
^' palampoors " of Madras being produced then, and even
now, by a process which is a refinement of the present
battacking of Java.

Until comparatively recent times the fabrics used for
clothing were entirely of native manufacture, but the
ease-loving Malay has found the product of Lancashire
looms equally satisfactory for his purpose, and the labour
which he is saved by using it counts for much in his
eyes. In the most out of the way districts one finds
Lancashire cottons in common use, but there is still a
good deal of the native-made product to be found in the
inland districts.

The first stage in the conversion of white cotton into
battacked goods is naturally the removal of the size with

Manchester Memoirs, Vol. I. (1906), No. 4. 3

which they are impregnated before leaving this country.
This is effected by frequent washings in the rivers, and
subsequent exposure of the wet pieces to the sun, the
washing and exposure being continued until the whole of
the starchy and saline matters are completely removed.
The pieces are then finally dried, cut up into sarong length,
and are ready to be passed into the hands of the printers.
Wash-day is looked upon as more or less of a necessary
evil in this country, but amongst the Javanese this con-
dition of things is quite reversed, for water seems to
possess a fascination for them, a considerable part of their
time being spent in the warm streams with which the
country abounds ; and in washing articles of clothing they
always take the opportunity of standing in as deep water
as they can conveniently work in.

Before entering into details, I might explain here that
there are really two methods of preparing battacked goods.
One of these may be described most simply as a process
of colour painting, in which the design is actually drawn
on the cloth by hand. This work is almost entirely carried
out by women, and, as might be expected, involves the
expenditure of an enormous amount of time and labour ;
and, as a result of this, the process is confined to the pro-
duction of very expensive articles of attire, the wearing
of which is restricted to the wealthy, who are a very small
proportion indeed of the native population. Since it is
almost an impossibility to imitate such work by a mecha-
nical process, I do not intend to enter into any of the
details of carrying it out, except to say that the colours
are applied by means of small tubes provided with orifices
of varying shape, the tubes having a small cup on their
end, which forms a reservoir in which the colour is con-
tained. In many cases the process is combined with the
second method of battacking, and from the description

4 Allan, Battack PrtJiting in Java.

which I shall give of that it will easily be seen how such a
combination is possible. Details of this form of battack-
ing are to be found in "Die Indische Batikkunst," by
Rouffer and Jagnball, a book of 5 vols., of which 4 have
up to the present been issued, and which is worthy of the
study of those interested in this subject.

The second process of battacking is more one of dye-
ing than of printing, the whole fabric being immersed in
the dye-bath, but prior to this the parts which are not
intended to be dyed are protected by a wax preparation,
which has been placed on it in such a way as to form a
design. In the coarsest cloths this design is made by
freehand drawing with a short thick brush, similar to what
is used in this country for rough stencilling, the work
being usually done by women who sit on the ground
with the cloth stretched out on their knees. The
wax which is used for the process is a mixture of
paraffin wax and beeswax, or a much-adulterated Japan
wax which is shipped specially to the country for the
purpose. The wax is kept melted in a pot on a small
charcoal fire, round which three or four of the women are
seated, and to judge from the continuous chatter which is
going on, there must be as many subjects for conversation
in Malaysia as there are at an afternoon tea-gathering at
home.

The amount of work of this class is much less, how-
ever, than that where male labour is employed in
stamping the resist patterns on the cloth with a metal
die, in exactly the same way as the block-printing of
calico was carried out before the introduction of the
roller printing machine. In this case, the workman, like
the women, is seated on the ground, with his legs under-
neath a large padded board, on which the white cotton
is spread out. Having coated the die with wax, he

Manchester Memoirs, Vol. l. (1906), No. 4. 5

presses it on the cloth, working continuously from left to
right until the whole cloth has been covered and the
design is complete.

The blocks which are used are made of thin strips
of sheet brass inserted in a wooden back, in such a way
that their edges produce the running patterns which are
invariably used in battack work. They act in use in
exactly the same way as the type does in ordinary letter-
press printing. Several blocks are used to work out
the complete design, and the work of any particular
battacking establishment can be recognised in different
patterns of sarong by the appearance here and there of
the same block.

The wax used for pattern stamping is a different
composition from that employed in the brushwork, as
the latter is so thin that it would run too easily from the
block and smudge the design. Tlie block wax is a
mixture of ordinary resin and paraffin wax, or some of
the varnish-gums which are collected by the natives.
The wax is melted over a charcoal fire, and after being
dipped the block is freed from the excess of wax by
being pressed on a pad which is kept warm by being
placed in close proximity to the melting fire.

Having been completely covered with the design in
this way the cloth is hung up for a short time, so that
the wax may become thoroughly hard before the process
of dyeing is commenced.

So far as I have been able to observe, it is the practice
of the Javanese never to use more than two colours in
this class of sarong printing, these being indigo and the
brown dye obtained from mangrove bark, but by dyeing
in both of these colours they are able to produce a black.
The result of this limited use of colours is that this battack
work is characterised by a white, blue, brown and black

6 Allan, Battack Printing in Java.

combination. The cloth is first dyed with the indigo,
and then before being immersed in the brown dye is
freed from the old wax and reprinted with wax on those
parts which are to be protected from brown, some of
which are the original white, and others the blue, where
this is to be left of the pure indigo shade.

The indigo which is used is of native manufacture,
and is always in the form of a moist paste. The dye-
bath is prepared by diffusing a quantity of this paste in
water and reducing it with waste molasses from the sugar
factories by fermentation, a process which is similar to
the bran fermentation vat so largely used in wool dyeing
in our own country. In another factory, I have seen, the
solution is kept alkaline with a mixture of wood ashes
and quicklime and gently warmed. The preparation of
the indigo bath in this way is interesting, as it is similar
to a method used in this country for fixing indigo in
printing, by means of a reducing mixture of glucose and
caustic soda.

To obtain the full depth of blue that is required it is
usual to dip and expose the cotton to the air several
times, the process therefore taking some little time to
complete.

In preparing the brown dye the mangrove bark is
first broken into small pieces, and then boiled for some
time with water, so as to obtain the whole of the soluble
extractive matter. After removing the bark, the solution
is concentrated to a very thick brownish black syrup, and
kept in stock for use as required. As a rule no mordant
is used with this brown dye, which is essentially of the
cutch type, although in this country when dyes of this
character are used, both copper sulphate and bichromate
of potash are employed as fixing agents. To fix this dye
the natives simply expose the fabric to the air for some

Manchester Memoirs, Vol. I. {\go6), No. 4. 7

time, the process of dipping and exposing being con-
tinued until the required shade has been obtained. The
dyed material is then freed from wax by thorough
washing in wood ashes, and after being ironed becomes a
finished sarong or slendang.

Many sarongs are now printed in this country and
also in Holland, but the native still prefers the home-
made article, and is prepared to pay double the price for
it, that he will pay for an imported one. Perhaps his
faith in the superiority of the locally produced article is
not unjustified, for he has had frequent experience of the
fact that the colours of European-made sarongs are of a
very fleeting character, and it must be admitted that a
colour which will stand constant wetting and exposure to
a tropical sun must be fast in more than the ordinary
sense. If the trade in sarongs exported from this country
is to increase, this point must be more fully attended to
than it has been in the past.

Malay Kris.

To mention the Malay Kris is to recall at once count-
less acts of piracy and bloodshed in Eastern seas, in
which this weapon has been used with deadly effect.
The name is invariably associated with a zig-zag shaped
double-edged dagger, from 12 to 16 inches long, but,
though this form is very frequently met with, it is not the
only one in which the kris is made. The zig-zag kris is
purely a weapon, but in the countries throughout Malaysia
where the kris is so universally carried, the native finds it
necessary, or, at any rate, a source of great convenience
to use his weapon as a cutting tool, and for this reason,
the kris is frequently made with a straight blade, with or
without a double cutting edge. The single-edged blades,
however, are usually heavier and longer than the true

8 Allan, Bat tack Printing in Java.

kris, and are distinguished from it by the name of
kulewang. For jungle-clearing, there is probably no
better tool than the kulewang if the Bornean parang
ilang be excepted.

The kris blade is usually characterized by beautifully
damascened markings, though many plain blades are
met with, but these are not esteemed so highly by the
natives, since the plain blades are not so strong, nor is
the labour entailed in their manufacture so great as is
the case with the damascened blade.

The kris is fitted with a wood or bone handle, usually
carved into grotesque shapes of men or animals, but with
no hand-guard other than a triangular widening-out of
the base of the blade.

In using the weapon, a favourite stroke is an upward
blow with the point entering beneath the shoulder blade
or a downward stroke with the weapon thrust between
the collar bone and first rib.

Until comparatively recent times, the metal which
was used for kris manufacture was entirely native iron,
but of recent years, with the extension of commercial
relationships with European countries, nails, hoop-iron
and other forms of scrap have come to be employed —
not, however, it must be acknowledged, to the improve-
ment of the weapon. How long iron has been made in
Malaysia it is impossible to say, but the process of manu-
facture is similar to that still used in India and China,
and it is probable that the art was imported from one or
other of these two countries, as there has for a long time
been a close commercial association between them and
the Malay Archipelago.

The furnaces in which iron is at present made by the
natives are built up from clay to a height of 3 or 4ft.,
and from 8 to loft. in diameter, the whole furnace

Manchester Memoirs, Vol. I. (1906), No. 4. 9

being bound round with rattan or split bamboo to prevent
its coming to pieces, owing to the extensive cracking
which takes place when it is in use. At the top of the
furnace, the walls are about 2ft. thick, but the square or
circular opening of 4 or 5ft. diameter gradually tapers to
about i^ to 2ft. at the hearth. A blast of air is supplied
to the furnace from two cylinders of wood fitted with
feather-edged pistons, to which rods are attached, and
which a workman operates by sitting between the
cylinders and working the pistons alternately up and
down. The blast is carried in bamboo tubes and is
admitted to the furnace through clay tuyers, which are
formed by moulding clay round small pieces of bamboo
and then burning the bamboo out in a fire, whereby the
clay is at the same time baked.

The ore which is used consists of nodules of haematite
usually obtained from a river-bed, or else a clay iron stone,
which is surface mined. Both of these ores are roasted
in wood fires and are then charged into the furnace with
a large excess of fuel. No flux of any kind is used in the
charge, and as is always the case where this is not done,
the process is a very wasteful one since the impurities in
the ore are fluxed off by combination with oxide of iron
which otherwise would have been reduced to metal. The
slag is drawn off at regular intervals, and in the end a
spongy mass of about loolbs. of steely iron is obtained,
the quality of which is determined by the proportion of
fuel to ore in the original charge, and the temperature
which the furnace is allowed to acquire in working.

The mass of iron is freed from slag by hammering,
but always retains some which shows itself in the form of
small flaws in the metal, when this is worked up.

A plain kris is made by simple forging from the iron
thus obtained, but the damascened form is made by

lo Allan, Battack Printing i?i Java.

welding together selected pieces of metal which approxi-
mate to our wrought iron and soft steel in composition.
No less than 5 strips of metal are welded together to
make the blade, the central and two outer ones being of
steel and the two others of wrought iron, the strips of this
metal having previously been bent up into a kind of
serpentine ribbon. The various pieces of metal are welded
together at one operation, and the welding temperature is
such that the more fluid steel is forced between the
corrugations of the wrought iron ribbons producing the
damascened structure which is so much desired. From
this compound piece of metal the blade is shaped by
rubbing down on a stone, and finally, to more fully
develop the damascening, it is pickled by being boiled for
two or three days in rice water to which some sulphur
and nitre obtained from the soil have been added.

The sheath for the kris is formed from solid wood
and has a large curved cross-piece at the top resembling
a sword guard, which, owing to its size and shape, enables
the wearer to carry the weapon in the folds of his sarong
without cords or straps of any kind.

BORNEAN SUMPITAN AND UPAS FOLSON.

It is impossible to take up any book describing the
habits and customs of the Dyaks of Borneo without
finding some mention of the deadly poisoned darts of
the sumpitan or blowtube which is universally employed
by them both as a means of obtaining food and in their
many intertribal wars.

The sumpitan is a tube formed from a hard brownish
black wood called by the Dyaks " tapang " and is about
7 or 8 feet long having an external diameter of about an
inch and an internal diameter of about a quarter of an

Manchester Mcj/ioij's, Vol. I. (1906), No. 4- 11

inch. A peculiar method is adopted for boring the tube
of the sumpitan. The wood having been carefully
selected is roughly hewn into a cylinder of the required
length but usually at least three inches in diameter and
is then hung up in a vertical position in the centre of a
stout four cornered scaffolding in such a way that its
lowest end is just over the workman's head. The tool
which is used for boring consists of a long iron rod with
a chisel-edged end, which the workman directs with
upward strokes against the centre of the hanging log.
The iron drill is fitted with a moveable wooden handle
which is moved downwards as the hole is driven further
and further into the log. The operation, as might be
expected, is a slow one, but in spite of the primitive
tool which is employed and the method of using it, it is
astonishing how accurately the centering of the hole is
maintained. To finally smooth and polish the tube a
piece of rattan, which closely fits it, is passed through it
and this is worked backwards and forwards until the
required finish is obtained. The rough blowtube is now
carefully cut down to the required thickness and the
exterior rounded and smoothed off by scraping. Usually
the sumpitan is used as a spear as well as a blow tube, a
heavy spear head about 12 or 14 inches long being
lashed to the end with thin strips of rattan in such
a way that there is no interference with its use as a
means of projecting the small darts of which the native
carries a supply in a bamboo case slung at his waist.

The darts are seemingly harmless weapons, since
they are merely small pieces of the central stem of a
palm leaf from six to eight inches long and no thicker
than an ordinary knitting needle. One end of the dart
is fitted with an inch- long plug of soft pith which closely
fits the tube of the sumpitan, whilst the pointed end is

12 Allan, Bat tack Printing in Java.

smeared over for about an inch of its surface with the
poison.

There are two kinds of the so-called Upas or more
correctly Ipo poison in use in Borneo, but the one most
frequently employed is prepared from the sap of the
Antiaris toxicaria or Upas tree, whose leaves are
popularly supposed to distil a poisonous dew and under
whose shade it is fatal to rest. Needless to say these
beliefs have not the slightest foundation in truth, but of
the deadly effects of the prepared poison there is not the
slightest doubt. Injected into the system by means of an
arrow, sickness rapidly comes on followed by convulsions
and death. It has fortunately not been my lot to see its
effect upon the human subject, but birds fall almost as
if they had been shot with a gun and a wild pig pierced
with only one arrow was dead in less than 20 minutes
after being struck.

To prepare the poison the native makes an incision
in the bark of the tree and collects the white milky sap
on a plantain leaf This is exposed to the sun until it
has been concentrated to a thick brown syrup, after
which the leaf is folded up and hung above one of the
cooking fires until further evaporation has reduced the
sap to a sticky brown mass. In this state the poison can
be kept for some time and when wanted for use is
dissolved \\\ the juice of the " tuba " root, well known for
its property of stupefying fish, or failing this, then tobacco
juice or even lemon juice may be employed, the dried
poison being mixed with the solvent to the consistency
of a thin paste into which the arrows or weapons to be
poisoned are dipped.

The active principle of the poison is the glucoside
antiarin, the name of this as well as of the tree being
derived from the native name of the poison, Upas Antjar,

Manchester Memoirs, Vol. I. (1906), No. 4. 13

which distinguishes it from the otlier arrow poison, Upas
Rajah or sometimes Tjettik. Of the manufacture of the
latter poison and its properties, I can only report in-
formation given to me by the natives as I have neither
seen it made or employed. I have been told that it is
prepared by boiling Tjettik roots with water and from
the description of its effects it must be somewhat similar
to the South American curare poison which is of the
strychnine family.

Manchester Memoirs, Vol. I. (1906), No. 5.

^c/ J I yju^oq^)

V. Report on the Recent Foraminifera from the
Coast of the Island of Delos (Grecian Archi-
pelago). Part III.

By Henry Sidebottom.

Received and read February ijlh, igob.

LAGENID^E.

LAGENINtE.
Lagena, Walker and Boys.

*Lagena globosa, Montagu, sp.

Entosolcjiia globosa (Montagu), Williamson ('58), p. 8,
pi. I, figs. 15, 16.

Lagena globosa (Montagu J, Reuss ('63), p. 318, pi. i,
figs. 1—3.

Lageiiulhia globosa (Montagu), Terquem ('76), p. 6"/,
pi. 7, figs. 3, 4.

Lagena globosa (Montagu), Brady, ('84), p. 452, pi. 56,
figs. 1—3.

Many of the specimens are nearly globular in shape,
others more or less pyriform and irregular. Frequent.

*There are present a number of tests that are elon-
gate, more or less pyriform, slightly compressed, with
stellate aperture and no entosolenian tube. Mr. Millett,
in his Malay report, suggests that many of these forms are
nothing more nor less than arrested growths of Nodosaria
and Polynnorphiiia.

* The asterisk denotes that this species occurs at Palermo.

April jQih, igo6.

2 SiDEBOTTOM, Foraviinifera from the Island of Delos.

*Lagena botelliformis, Brady, van (PI. i, fig. i).

Lagena botelliformis, Brady ('84), p. 454, pi. 56, fig. 6.

The contour of the Delos specimens so closely resem-
bles Brady's figure in the above reference, with the
exception that the orifice is situated at the end of a
produced neck, that I think it may be considered a varia-
tion of that species, in preference to treating it as a variety
of L. laevis. L. laevis does not occur in the material
examined. Six specimens were found, only one at
Palermo. Very rare.

*Lagena ampulla-distoma, Rymer Jones (PI. i, figs. 2,3).

Lagena vulgaris, var. ampulla-distoma, Ry. Jones ('72)>
p. 62>, pi. 19, fig- 52.

L. ampulla-distoma (Ry. Jones), Brady ('84), p. 458,

pl- 57, fig- 5-

L. ampulla-distoina (Ry. Jones), Millett (:0l), p. 5,
pl. I, fig. 5.

This occurs in three forms. In the four largest
specimens the test is roughened all over, as in fig. 2 ; in
the smaller ones only half of the test is rough, and the
shell deposit takes the form of very short blunted spines,
which have a tendency to coalesce and run into lines as
they approach the clear part of the shell, as in fig. 3.
Whilst in the smallest examples (and they are very minute)
the shell is covered with protuberances which show a still
greater tendency to run into lines — in fact, in one case
they have nearly done so. All the forms are very rare.
More frequent at Palermo.

* Lagena striata, d'Orbigny, sp.

Oolina striata, d'Orbigny ('39), p. 21, pl. 5, fig. 12.
L. striata (d'Orb.), Brady ('84), p. 460, pl. 57, figs. 19,
22, 24, 28—30.

Manchester Memoirs, Vol. I. ( 1 906), No. 5. 3

Lagena striata (d'Orb.), Brady, Parker and Jones ('88),
p. 222, pi. 44, fig. 28.

The examples are small, and the striae become very
faint between the body of the test and the neck. More
typical examples are found at Palermo, where this species
is also more plentiful.

* Lagena sulcata, Walker and Jacob, sp.

Lagena striata, Williamson ('48), p. 13, pi. i, figs. 6, 8.

L. vulgaris, vzx. perlucida, Williamson ('58), p. 5, pi. i,
fig. 8.

L. vulgaris, var. striata, Williamson ('58), p. 6, pi. i,
fig. 10.

L. sulcata (W. & J) Brady ('84), p. 462, pi. 57, figs.
23, 26. Very frequent.

* Lagena sulcata, var. interrupta, Williamson.

Lagena striata, var. interrupta, Williamson ('48), p. 14,
pi. I, fig. 7.

L. vulgaris, var. interrupta, Williamson ('58), p. 7, pi. i,
fig. II.

L. sulcata, var. interrupta (Williamson), Brady ('84),
p. 463, pi. 57, figs. 25, 27, Frequent.

*Lagena semistriata, Williamson. (PI. i, figs. 4, 5.)
Lagena striata, var. /3, semistriata, Williamson ('48),
p. 14, pi. I, figs. 9, 10.

L. vulgaris, var. semistriata, Williamson ('58), p. 6,
pi. I, fig. 9. Very frequent.

The Delos examples of these three forms run into
each other. Fig. 5 shows how closely some specimens of
L. semistriata approach L. laevis, the stris being repre-
sented solely by very minute tubercles on the bottom of
the test. Z. laevis not being represented in the Delos

4 SiDEBOTTOM, Foraviinifera from the Island of Delos,

gatherings, it appears that this example (fig. 5) is rightly
placed under L. seinistriata ; it also has a likeness to
Mr. Millett's L. clavata, var. setigera (:oi, pi. 8, fig. 9), but
the shape of the test is not that of the type, and the end
is not cup-shaped, but round.

The number of costai or stride in L. seinistriata varies
very much, some of the exam.ples having as few as six.

*Lagena variata, Brady.

Lagena variata, Brady ('84), p. 461, pi. 61, fig. i.
L. variata (Brady), Millett (:0l), p. 7, pi. i, fig. 7.
Six very fair examples of this rare species occur ; all
are irregular in shape. Very rare.

*Lagena lineata, Williamson, sp.

Entosolenia lineata, Williamson ('48), p. 18, pi. 2.
fig. 18.

E. globosa, var. lineata, Williamson ('58), p. 9, pi. i,
fig. 17.

Lagena lineata (Williamson), Reuss ('62), p. 328, pi. 4,
fig. 48.

The examples are typical. Rather rare.

*Lagena striatopunctata, Parker and Jones.

Lagena sntcata, var. striatopunctata, Parker and Jones,
('65). P- 350, pi. 13, figs. 25—27.

L. striatopunctata (P. & J.), Brady ('84), p. 468, pi. 58,
figs. 37, 40.

L. striatopunctata (P. & J.), Millett (:0l), p. 489, pi. 8,
fig. 6.

In all cases the neck is bent to one side, the number
of costse being generally seven. The test is cylindrical
with rounded bottom. About forty were found. Rather
frequent. Frequent at Palermo.

Manchester Memoirs, ]''ol. I. (1906), No. 5. 5

'''Lagena hexagona, Williamson, sp.

Entosoknia squamosa, var. Jiexagona, Williamson ('48;,
p. 20, pi. 2, fig. 23.

Entosolenia squamosa, var. hexagona, Williamson, ('58),
p. 13, pi. I, fig. 32.

Entosolenia squamosa, var. scalar if ormis, Williamson
('•58), p. 13, pi. I, fig. 30.

Lagena favosa, Reuss ('62), p. 334, pi. 5, figs. 72, 77,.

L. geometrica, Reuss ('62), p. 334, pi. 5, fig. 74.

L. hexagona (Williamson), Brady ('84), p. 472, pi. 58,
figs. 32, IZ-

The Delos examples of this elegant foraminifer vary
considerably, both as to the size of the hexagonal
markings and the shape of the test. Frequent.

*A variety is present which has an elongate and
tapering test, terminating in a short neck. This form is
the smallest of the variations. Very rare.

*Lagena laevigata, Reuss, sp. (PI. i, fig. 6.)

Fissurina laevigata, Reuss ('50), p. 366, pi. 46. fig. I.

Lagena laevigata (Reuss), Balkvvill and Millett ('84),
p. 80, pi. 2, fig. 6.

L. laevigata (Reuss), Brady ('84), p. 473, pi. 1 14, fig. 8.

Several varieties of this common species are present,
and the orifices vary. In some cases the orifice is merel)-
a short slit on the median line, in others the apertural end
is slightly produced, and the mouth shows itself at one
side of the median line.

*The form figured has a short neck and the usual
entosolenian tube. In all the varieties there is a very
small ring at the aboral end, and where this projects a
little, as it does in most cases, they approach L. acuta,
Reuss. Frequent.

6 SiDEBOTTOM, Foraminifera from the Island of Delos.

Three specimens of the ordinary kind occur in
trigonal form.

*Lagena laevigata, Reuss, sp. var. acuta, Reuss, sp.

(PI. I, figs. 7, 8).

Fissurina acuta, Reuss ('62), p. 340, pi. 7, fig. 90.

F. apiculata, Reuss ('62), p. 339, pi. 6, fig. 85.

Lagena laevigata (Reuss), var. acuta, (Reuss), Millett
(:0I), p. 494, pi. 8, fig. 16.

This apiculate form of L. laevigata is frequent.

*One of these chosen for illustration, fig. 8, is inter-
esting from the fact that the outline is inequilateral.
This latter variety is rare,

*Lagena lucida, Williamson, sp. (PI. i, figs. 9 — 12).

Entosoknia inarginata, var. lucida, Williamson, sp. ('58),
p. 10, pi. I, figs. 22, 23.

Lagena lucida (Williamson) Balkwill and Millett ('84),
p. 80, pi. 2, fig. 7.

In the Delos examples of this species the form of the
test is very variable, as will be seen by reference to the
figures. In one form only is the apiculate variety present,
fig. 12, and this has the entosolenian tube attached, in the
other varieties it is short, straight, and free. Frequent.
All the varieties occur at Palermo.

* Lagena fasciata, Egger, sp. (PI. i, figs. 13 — 16.)

Lagena fasciata (Egger), Reuss ('62), p. 323, pl. 2,
fig. 24.

L. quadricostulata, Reuss ('70), p. 469.

L. quadricostulata (Reuss), Brady ('84), p. 486, pl. 59,
fig. 15.

Mr. Millett, in his Malay report (:oi, p. 495), says :
" Taking L. annectens^ Burrows and Holland ('95, p. 203,

Manchester Memoirs, Vol. I. (1906), No. 5. 7

pi. 7, fig. 11), two curved bands appear on each side of the
shell. In L. faba, Balkwill and Millett ('84, p. 81, pi. 2,
fig. 10), these bands are slightly raised, whilst they
become costae in L. quadricostulata, Reuss, L. fasciata,
Egger, and L. meyeriana. Chapman ('94, p. 706, pi. 34,
fig. 7). These bands may or may not unite at the base
of the shell ; Dr. Egger's examples show both conditions,
whilst in the only known specimen of L. meyeriana, the
costae, although continuous, are recurved, and form a
sinus at the aboral extremity."

The peculiarity of the Delos examples is that the
grooves, or shallow sulci, have their edges raised above
the body of the shell.

In the varieties figs. 14, 15, 16, 17, this feature is well
marked, whilst in fig. 13 there is a doubt as to its being
present, owing to the fineness of the markings, and I
have been unable to satisfy myself on this point. In
only two or three cases are the grooves continuous at
the base of the shell. This range, with few exceptions,
has a short entosolenian tube which is straight, and
therefore not attached to either face of the test. This
tube is often split at the end, instead of having the well-
known trumpet orifice.

The Delos forms are evidently closely allied to
L. annectens, Burrows and Holland, and future research
may connect them with L. alveolata, Brady. Rare to
very rare. Only the forms represented by figs. 14, 15,
16, occur at Palermo.

Lagena fasciata, Egger, var. carinata, nov. (PL i.,

fig- 1;)-

This variation has a central keel ; the test is very
finely pitted all over, and so has not the same transparency
as we find in the other varieties. A short, stout, straight

8 SiDEBOTTOM, Foraminifera Jvovi the Islmid of Delos.

entosolenian tube is present in all cases. Frequent.
Examples occur frequently of a more compressed and
clear form which has the keel almost confined to the
aboral end, the sides of the tests feebly carinate, the
mouth slightly produced and carinate at its sides. The
costs in these cases are exceedingly delicate.

*Lagena staphyllearia, Schwager, sp. (PI. i, figs. i8^

19, 20).

Lagena staphyllearia (Schwager, sp.) Millett (:0l),
p. 619, pi. 14, fig. 2.

L. staphyllearia (Schwager, sp.), Flint ('99), p. 307, pi.

54, fig- I.

All the specimens are compressed. Fig. 18 shews
this form with the basal spines separated from each other,
test carinate as in L. marginata. Fig. 19 shews it with
what may be considered as four spines joined together,
test non-carinate. Intermediate forms occur, connecting
these two varieties. Respectively frequent and rare. A
single specimen with two spines, one on either side of the
test, was found (fig. 20).

*Lagena quadrata, Williamson, sp. (PI. i, figs. 21, 22,
and PL 2, figs, i, 2, 3.)

Entosolenia marginata, var. quadrata, Williamson ('58),
p. II, pi. I, fig. 27.

Lagena laevigata, var. quadrata (Williamson), Wright
('86), p. 324, pi. 26, fig. 9.

L. quadrata (Williamson), Egger ('93), p. 331, pi. 10,
figs. 7^, 79.

L. quadrata (Williamson), Brady ('84), p. 475, pi. 59,
figs. 3, 16, and pi. 60, fig. 5.

Charming variations of this species are present in fair
numbers. *Fig. 21 has the mouth placed in an immature

Manchester Meuwirs, Vol. /. {igo6), No. ^. 9

hood, and is rare. *Fig. 22 has faintly frosted-looking
bands as in L. hicida and is very rare. PI. 2, fig. i, has
the fine lines running round the edges of the test as in
some varieties of L. fasciata, also very rare. *Fig. 2 has
a short neck, sides not quite parallel, and perhaps might
be treated as a variety of L. laevigata. *Fig. 3 is partially
carinate, and agrees very closely with Brady's "Challenger"
drawing ('84, pi. 59, fig. 16). This form in the Delos
gatherings is the most numerous of the quadrate
varieties, but I hardly think it worthy of a varietal name.
Frequent.

*Lagena marginata, Walker and Boys.
Oolina conipressa, d'Orbigny ('39), p. 18, pi. 5, figs.

I, 2.

Fissurina carinata, Reuss ('62), p. 338, pi. 6, fig. 83,
pi. 7, fig. 86.

Lagena marginata (VV. & B.), Silvestri ('96), p. 119,
pi. 3, figs. 7—9.

L. marginata (VV. & B), Flint ('99), p. 307, pi. 54,
fig. 2.

Most of the examples are small, nearly circular in
outline, and have the keel fairly well developed. Frequent.
*There is a variety present, vvhich comes very near to
fig. 27 on pi. 44, Brady, Parker, and Jones ('88) It is
placed by them under L. marginata, but it appears to me
to be intermediate between L. marginata (W. & B.) and
L. marginata, var. semimarginata, Reuss.

Lagena squamoso-marginata, Parker and Jones (PI. 2,

fig- 4)-
Lagena squamoso-marginata, Parker and Jones ('65),
p. 356, pi. 18, fig. 2.

10 SiDEBOTTOM, Fovaminiferafrom the Island of Delos.

L. sqiianioso-marginata (P. & J.), Brady ('84), p. 481,
pi. 60, fig. 24.

L. squamoso-inarginata (P. & J.), Millett (:0l), p. 622,
pi. 14, fig. 7.

All the specimens have a single keel. The ento-
solenian tube is attached. The mouth opens out at the
lower edge of the keel, and the hexagonal markings are
raised. About thirty found. Rather rare.

*Lagena marginato-perforata, Seguenza (PI. 2, fig. 5).

Lagena marginato-perforata (Seg.), Millett (:0l), p. 621,
PL 14, fig. 4.

This elegant foraminifer is frequent in these gatherings.
The keel varies in its development, in some cases it
scarcely shows, in others it springs from the edge of the
mouth and is continuous right round the test, being
widest at the aboral end. The internal tube is straight
and free. In all the examples the centre of the test is
unornamented. One specimen of trigonal form occurs.

*Lagena marginata (W. & B.), var. inaequilateralis
Wright (PI. 2, fig. 6).

Lagena marginata (W. & B.), var. inaequilateralis^
Wright, ('86) p. 321, pi. 26, fig. 10.

The specimens are quite typical, beautifully clear, and
in all cases the internal tube is attached to the test ; the
mouth is peculiar and agrees with the Irish examples.
Frequent.

Lagena inaequilateralis (Wright), var. semi-marginata,
nov. (PI. 2, fig. 7).

This is an interesting variation of L. marginata var.
inaequilateralis^ Wright. The mouth is the same as in

Manchester Memoirs, Vol. I. (1906), No. 5. ii

Mr. Wright's form, and the test is also inequilateral. The
keel however is confined to the aboral end of the test and
is well developed. Six specimens found. Very rare.

Lagena irregularis, n. sp. (PI. 2, fig. 8).

One face of the test is bent back in the centre at
about an angle of forty five, the opposite side being
highly convex ; periphery bi-carinate. Aperture, a small
slit situated between the two edges of the keels. Entoso-
lenian.

This curious form is difficult both to describe and to
draw quite satisfactorily. It is very minute, and reference
to the figures will best explain its peculiarities. The
keels are well developed at the oral end, whilst at the
aboral they are not so wide, but the bicarinate condition
is more pronounced. The space between the keels, small
as it is, is partially filled with debris, or shell-growth, and
this increases the difficulty of examination. The internal
tube in all cases runs along the centre of the back until
it nearly reaches the other end, where it turns to one side.
The orifice is a small slit, apparently situated between the
edges of the keels as shown in the drawings, but it is very
difficult to make out.

It has been thought that this form might be a
distorted or wild-growing example of L. marginata, van
inaequilateralis, Wright, but the twelve specimens found
are all alike. The mouth of the test is not protruded as
in Mr. Wright's figures of his species, and as in the Delos
specimens of the same form, in addition to which the keel
is double. Rare.

*Lagena lagenoides, Williamson, sp.
Entosolenia marginata, var. lagenoides, Williamson
('58), p. II, pi. I, figs. 25, 26.

12 SiDEBOTTOM, Foraininifevafroiii the Island of Delos.

Lagena lagenoides (Williamson), Reuss ('62), p. 324,
pi. 2, figs. 27, 28.

L. lagenoides (Williamson), Balkwill and Millett ('84),
p. 82, pi. 2, fig. II.

The specimens agree best with Williamson's fig. 26.
Frequent. *A smaller form is also present, in which the
mouth only slightly protrudes, extending nearly the
whole width of the test. Mr. Wright, of Belfast, Ireland,
considers this to be L. nrnata, Williamson, which is a form
of L. lagenoides. Rare.

* Lagena lagenoides, var. tenuistriata, Brady.
(PI. 2, figs. 9, 10.)
Lagena lagenoides, var. tenuistriata, Brady ('84),
p. 479, pl. 60, figs. II, 15, 16.

L. lagenoides, var. tenuistriata (Brady), Balkwill and
Millett ('84), p. Z2, pl. 2, fig. 12.

In the Delos s[)ecimens, the centres of the tests are
sometimes free from striae. The cntosoleiiian tube is
short and straight. Frequent.

* Lagena orbignyana, Seguenza, sp. var. (Pl. 2, fig. 11.)

Entosolenia viarginata (pars.), Williamson ('58), p. 9,
pl. I, figs. 19, 20.

Lagena orbignyana (Seguenza), Brady ('84), p. 484,
pl. 59, figs. I, 18, 24-26.

L. orbignyana (Seguenza), Brady, Parker and Jones
('88), p. 222, pl. 44, fig. 20.

L. orbignyana (Seguenza), Flint ('99), p. 30S, pl. 54,
fig. 4.

The test in this variety is slightly twisted ; and in
the middle, on either face, is an oval ridge of clear shell-
substance. The internal tube is very much curled. In
one or two cases the central keel is split and filled up

Manchester Meuioirs, Vol. I. ( 1 906), No. 5- 1 3

with debris. Most of the specimens are badly fractured
at the edge of the test. Rather rare.

Lagena orbignyana ? var. falcata, nov. (PI. 2, fig. 12.)

The test is compressed, and has two recurved spines,
springing respectively from either side of the shell, near
the orifice. The fine line running on each side of the
delicate keel (or very angular margin) is continuous
except at the oral end of the shell. The mouth is oval
and slightly produced. The internal tube is short and
free. Mr. Millett kindly drew my attention to Dr.
Chaster's Lingiilana Jierdmani ('92, pi. i, fig. 9), a detached
chamber of which would have a strong resemblance to
the Delos specimens, but these latter bear no evidence of
fracture. They could hardly be the initial chambers of
Dr. Chaster's species, as his examples bear no spines
on the initial chamber. The lines above referred to are
also absent in Lingulina het'dinani. The two specimens
found are exactly alike. Very rare.

Lagena bicarinata, Terquem, sp. (PI. 2, figs. 13, 14, 15.)

Fissiirina bicarinata, Terquem ('82), p. 31, pi. 9,
fig. 24.

Lagena bicarinata (Terquem), Balkwill and Millett
('84), p. 82, pi. 2, fig. 4.

L. bicarinata (Terquem), Balkwill and Wright ('85),
p. 342, pi. 12, fig. 30.

L. bicarinata (Terquem), Wright ('86), p. 320, pi. 26,
fig. 8.

L. bicarinata (Terquem), Halkyard ('89), p. 66, pi. 2,
fig. I.

L. bicarinata (Terquem), Millett (:0l), p. 624, pi. 14,

fig- 13-

14 SiDEBOTTOM, Foraminif era from the Isla?id of Delos.

Three varieties occur which have the two keels, and
although very different in other respects, I have brought
them together under the above heading. Fig. 13 has the
two keels well developed, but has no neck, and is of fre-
quent occurrence. Fig. 14 has a well developed neck with
phialine lip surrounding the orifice, also a ridge on either
side encircling the body of the test, and is rare. Fig. 15
has the keels very poorly developed and is frequent. All
the varieties have the internal tube attached to one face
of the test.

Lagena orbignyana, var. clathrata, Brady (PI. 2, fig. 16).

Lagena clathrata, Brady ('84), p. 485, pi. 60, fig. 4.

L. clathrata (Brady), Balk will and Millett ('84), p. ^2,
pi. 2, fig. 14.

L. orbignyana, var. clathrata (Brady), Millett (:0l),
p. 628, pi. 14, fig. 23.

The examples agree with the form figured by Messrs,
Balkwill and Millett in the above reference. Frequent.

* Lagena fimbriata, Brady.

Lagena fimbriata, Brady ('84), p. 486, pi. 60, figs.
26—28.

L. fimbriata (Brady), Balkwill and Millett ('84), p. 82,
pi. 2, fig. 5.

Brady, in the " Challenger " Report, speaks of this as
being a deep-water form, but it has been found both by
Mr. Millett and Mr. Wright in shallow-water off the coast
of Ireland. The two or three specimens found at Delos
agree well with the Irish forms, although in the former
the oval wing surrounding the base is not well developed
and is more compressed. Very rare. Better examples
occur at Palermo.

Manchester Memoirs, Vol. I. (1906), No. 5. 15

*Lagena alveolata, Brady (PI. 2, fig. 17).

Lagena alveolata, Brady ('84), p. 487, pi. 60, figs. 30, 32.

In the above reference, Brady states that L. alveolata
is only found in deep water, and reports its occurrence in
the North Atlantic, 2,750 fms. ; in the South Atlantic,
2,200 fms. ; in the Southern Ocean, 2,600 fms. ; in the
South Pacific ; and in the North Pacific, 2,300 fms.

The depth at which the Delos examples were found
varied from 8 to 14 fms. The tests are very transparent,
and all have a long entosolenian tube attached to one face
of the shell. About 15 were found. Rare.

*Lagena protea, Chaster (PI. 2, fig. 18).

Lagena protea, Chaster ('92), p. 62, pi. i, fig. 14.

There are about fifteen specimens of this protean
form, no two of which are alike. I have one specimen
attached to a piece of shell, and probably the one figured
has likewise been adherent. Rare. Still finer examples
were found in the Palermo material.

BIBLIOGRAPHY.

Balkwill, F. p., and F. W. Millett ('84). " The Foraminifera
of Galway." Journ. Micr. and Nat. Set., vol. 3, pp.
19-28, 78-90, 4 pis., 1884.

Balkwill, F. P., and J. Wright ('85). "Report on some
Recent Foraminifera found off the coast of Dublin and in
the Irish Sea." Trans. R. Irish Acad., vol. 28, Science,
pp. 317-372, 3 pis. and tigs, in text, 1885.

Brady, H. B. ('84). " Report on the Foraminifera collected by
H.M.S. 'Challenger' during the years 1873-76." Zool.
Chall. Exp., vol. 9, 814 pp., 115 pis., 1884.

1 6 SiDEBOTTOM, Foraviinifera froiH the Island of Delos.

Brady, H. B.. W. K. Parker, and T. R. Jones ('88). "On
some Foraminifera from the Abrohlos Bank." Trans.
Zool. Soc, vol. 12, pt. 7, pp. 211-239, pis. 40-46 and
chart, 1888.

Burrows, H. W., and K. Holland ('95). "A Monograph of
the Foraminifera of the Crag." By T. R. Jones and
others. 402 pp., 7 pis. PalcTontographical Society.

Chapman, F. ('94). " The Bargate Beds of Surrey and their
Microscopic Contents. VIII. Foraminifera from the
Bargate Beds." Quart. Journ. GeoL Soc, vol. 50, pp.
693-728, pi. 34, 1894-

Chaster, G. W. ('92). " Report upon the Foraminifera of the
Southport Society of Natural Science District." First Rep.
Southport Soc. Nat. Sci., pp. 54-71, pi. i, 1892.

Egger, J. G. ('93). "Foraminiferen aus Meeresgrundproben,
gelothet von 1874 bis 1876 von S. M. Sch. 'Gazelle.'"
Abhandl. k. Bayer. Akad. IViss., CI. 2, vol. 18, pt. 2, pp.
193-458. pis- 21, 1893.

Flint, J. M. ('99). "Recent Foraminifera, a descriptive cata-
logue of specimens dredged by the U.S. Fish Commission
Steamer 'Albatross.'" Rep. U.S. Nat. Mus., pp. 249-349,
pis. 80, 1897.

Halkvakd, E H. i'89). " Recent Foraminifera of Jersey."
Trans, and Ann. Kept. Manchester Micr. Soc, pp. 55 — 72,
2 pis. J 1889.

Jones, F. W. O. Rymer ('72). "On some Recent Forms of
LageUcX from Deep-sea Soundings in the Java Seas."
Trans. Linn. Soc. Lond., vol. 30, pp. 45-69, pi. 19, 1872.

Millett, F. W. (:0I). "Report on ihe Recent Foraminifera
of the Malay Archipelago, collected by Mr. A. Durrand,
F.R.M.S.,y<?//;7^. R.Micr. Soc, igoijpp. i — 11, 485 — 497,
619 — 628, pis. I, 8, 14.

Orbigny, a. D. d' ('39). " Voyage dans I'Amerique meridionale,
Foraminiferes." 4to. Paris and Strasburg, vol. 5, part 5,
pp. 1-86, 9 pis., 1839.

Manchester Memoirs, Vol. I. ( 1 906), No. 5. 1 7

Parker, W. K., and T. R. Jones ('65). " On some Foramini-
fera from the North Atlantic and Arctic Oceans, includiim
Davis Straits and Baffin's Bay." Phil. Trans.., vol. 155,
PP- 325-44i> pis. 12-19, 1865.

Reuss, a. E. ('50). " Neue Foraminiferen aus den Schichten
des osterreichischen Tertiarbeckens." Denkschr. k. Akad.
IViss. Wien, vol. i, pp. 365-390, pis. 46-51, 1850.

('62). " Die Foraminiferen-Familie der Lagenideen."

Sitzungsb. k. Akad. IV/ss. IVien, vol. 46, part i, pp.
303-342, pis. 1-7, 1862 (1863).

('70)- " 1^^6 Foraminiferen des Septarien-Thones von

Pietzpuhl." Sitziingsb. k. Akad. Wiss. Wien, vol. 62,
part I, pp. 455 493> 1870.

SiLVESTRi, A. ('96). " Foraminiferi Pliocenici della Provincia
di Siena," part r, vol. 12, pp. 1-204, 5 pls-, 1896, and
part 2, vol. 15, pp. 155-381, 6 pis., 1898.

Terquem, O. ('76). " Essai sur le classement des animaux
qui vivent sur la plage et dans les environs de Dunkerque."
Fasc. 2, pp. 55-100, pis. 7-12, Paris, 1876 (also published
in Mem. Soc. Dunkerquoise).

('82). " Les Foraminiferes de I'Eocene des Environs de

Paris. Alem. Soc. G'eol. Frafice, ser. 3, vol. 2, Mem. 3, pp.
1-193. pis. 9-28, 1882.

Williamson, W. C ('48). "On the recent British Species of
the Genus Lagena." A?inals Mag. Nat. Hist., ser. 2, vol.
I, pp. 1-20, pis. I, 2, 1848.

('58). "On the recent Foraminifera of Great Britain."

Ray Society, 7 pis., 1858.

Wright, J. ('86)- "Recent Foraminifera of Down and Antrim."
Froc. Belfast Nat. Field Club, 1884-5, App. 9, pp. 317-326,
pi. 26, 1886.

1 8 SiDBBOTTOM, Foraniinif era from the Island of Delos.

Figs.

1 .

2, 3-

4, 5-
6.

9-12.
13-16.
17-

18-20.
21, 22.

1-3-
4-

5-
6.

8.

9, 10.

II.
12.

13-15-
16.

17-
18.

EXPLANATION OF PLATES..
Plate L

Lagena botelliformis, Brady, var. x 50

,, avipulia-distoma, Ry. Jones, Fig. 2

X 5°. Fig. 3 X 75

„ semisiriata, Williamson x 75

„ laevigata, Reuss x 75
„ laevigata, Reuss, var. acuta.,

Reuss X 75

,, lucida, Williamson x 75

„ fasciata, Egger x 75
„ fasciata, Egger, var. carinata,

nov. X 75

„ staphyllearia, Schwager x 75

„ quadrata, Williamson x 75

Plate IL

Lagena quadtata, Williamson x 75
,, squamoso-marginata, Parker and

Jones X 100

,, margin ato-perforata, Seguenza x 75
„ marginata (W. & B.), var. in-

aequilateralis, Wright x 75
„ inaequilattralis (Wright), var.

setni-carinata, nov. x 75

,, irregularis, n. sp. x 75
,, lagenoides (Will.), var. tetiui-

striata, Brady x 75

„ orbignyana, Seguenza x 75
,, orbignyana} (Seg.) vax. falcata,

nov. X 75

„ bicarinata, Terquem x 75
„ orbignyana (Seg.), var. clathrata,

Brady x 75

,, alveolata, Brady x 75

„ protea, Chaster x 25

Page
2

9
10

10
II

12
12

13
13

14
15
15

Manchester Memoirs, Vol. L.

\a

4a

Plate /.

170!

18o

\\)n

•20o

'21a

22a

n. fiidebettnin, del. nil nnt.

Foraminifera from tlie coast of tlie island of Delos.

Manchester Memoirs, Vol. L.

ia

Plate 11.

5a

6a

Sa.

9re

10a

12«

13a

If. Sldeljottoiii, del. lul iiat.

.\\

Manchester Memoirs, Vol. I. (1906), No. 6-

VI. The Cytological Aspect of Parthenogenesis
in Insects.

By C. Gordon Hewitt, B.Sc,

Deiuoiistrato}- of Zoology in the Umversity of Mauchcsier.
Received and Read ATatrh 13th, igo6.

The cytological aspects of the biological phenomenon
Parthenogenesis, that is the changes which take place in
the maturation and development of the unfertilized q^^%,
have only been studied within comparatively recent years,
though our knowledge of the occurrence of the pheno-
menon dates back to the time of Aristotle.

At the present stage of the inquiry, it appeared to me
that a useful purpose would be served if a short summary
were made of our present knowledge of the cytological
phenomena associated with parthenogenesis in that group
of animals — the Insects, in which the problem was first
studied, and in which examples of the different types of
parthenogenesis are found.

For the sake of brevity and clearness, little reference
will be made to the general parthenogenetic phenomena,
as these are fairly well known and have been summarised
recently by Phillips (78). Since his review was written,
our knowledge of the maturation of the parthenogenetic
ovum has been increased, chiefly by researches made
with a view to investigating the chromosomes, the study
of which is occupying an ever-increasing number of
workers. We shall have more light thrown upon those
great biological problems — Sex, Fertilisation, and
Heredity, about which, the last especially, we know

May 2 1st, igo6.

2 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

very little at present, by the careful study of par-
thenogenesis in all its aspects.

Parthenogenesis occurs in the majority of the orders
of Insects, and in this group we can recognise all the
types of the phenomenon, as they are known from a
study of the general or somatic changes which take place.

To obviate the necessity of explaining at length in
each case the type of parthenogenesis which occurs in
that particular family which we may be considering, it
will be convenient if they are arranged in the following
tabular form, which is a slight modification of Henne-
guy's (45) classification : —

1. TycliopartJienogencsis (Henneguy).

This includes those cases of Parthenogenesis which
are accidental and of exceptional occurrence,
such as are found in the Lepidoptera and
Coleoptera.

2. HomopLirthenogenesis (Henneguy).

Under this term which replaces Hatschek's term
Isoparthenogenesis, all cases of normal
Parthenogenesis are included. They may be
further divided into

{a) Thelyotoky.

Parthenogenesis in which only females are produced.
This occurs in some species of Tenthredinidae.

{b) Arrhenotoky.

Parthenogenesis in which only males are produced.
This occurs in the Hymenoptera Sociales and
some species of Tenthredinidae. (The workers
of some Ants, Bees, and Wasps may produce
parthenogenetic eggs, though this is only
occasional, but in all cases they produce males).

Manchester Memoirs, Vol. 1. (1906), No. 6. 3

{c) Deuterotoky.

Both sexes are produced as in some species of
Tenthredinidae. (Deuterotoky also occurs in
the next division — Heteroparthenogenesis).

3. HeteropartJienogenesis (Henneguy).

Parthenogenesis in which there is an alternation of
generations, a cyclical alternation of sexual
and parthenogenetic forms. This may be either

{a) Regular as in the Cynipid^e and Aphidae, or
ib) Irregular as in some species of Tenthredinidai and
in the Psychidae.

Under a fourth head Pcedoparthenogenesis, Henneguy
places that rare form of parthenogenesis in which young
are produced by the immature insect, which may be either
the larva or the pupa. This phenomenon occurs in a few
Diptera (Cecidomyidae and Chironomidae) and is usually
termed Paedogenesis. I prefer that this term should be
retained, and as this is an extremely specialised form of
parthenogenesis, it should not be classified with the other
forms.

It is impossible at the present stage of enquiry to
make a satisfactory classification of parthenogenesis from
a cytological point of view. Marchal (60) and Loisel
have each brought forward classifications, which include
those cases found in other groups of animals. The results
which we have so far appear to indicate that partheno-
genesis is equivalent to internal agamogenesis.

In this summary only those insects in which the
maturation of the parthenogenetic ovum has been studied
will be considered. The chief feature of these cytological
changes which occur is the formation and fate of the polar
nuclei.

4 H E WITT, Cytologkal A sped of PartJienogenesis in Insects.

HYMENOPTERA.
Apid^.

Apis mellifica. An excellent account of the progress
of our knowledge of the parthenogenesis of this insect and
the theories connected with it is given by Phillips (78).

Blochmann (11) first described the polar bodies of the
drone-egg. Petrunkewitsch (74) has studied the changes
which take place in the maturation of the drone-egg, and
in a later paper (75) he gives an account of his investiga-
tions as to the fate of the polar nuclei.

The nucleus of the ovum approaches the periphery
of the egg. Here it undergoes two successive divisions
{Fig. 8). In this manner four nuclear masses, groups of
chromosomes, are formed. Of these the two outer portions
are the halves of the first polar nucleus, and the two
inner are the second polar nucleus and the female pro-
nucleus. Each nuclear mass contains eight chromosomes,
the original nucleus having contained sixteen. The inner
half of the first polar nucleus and the second polar nucleus
fuse (/^/o-.9),and form a single nucleus (j-) containing sixteen
chromosomes. Petrunkewitsch terms this the " copulation-
nucleus," an unfortunate term in my opinion, and I propose
to call it the Syntelosovie. The outer half of the first polar
nucleus wanders outwards, becomes flattened against
the periphery and degenerates ; the female pronucleus
migrates inwards {Fig. 10).

Petrunkewitsch has followed the syntelosome, and
finds that it gives rise to the germ cells of the male, by
subsequent divisions in the dorsal region of the embryo.
The cells thus formed being the primordial germ cells
forming the spermatogonia. As these arise from the

MancJiester Memoirs, Vol. I. (1906), No. 0. 5

original syntelosome they contain sixteen chromosomes.
Giglio-Tos (38) has pointed out that these observations
of Petrunkewitsch are discordant with those of Meves (64),
who has attacked the problem from the other end by
studying the spermatogenesis of the drone bee. Un-
fortunately Meves' description and figures of the chromo-
somes are not clear enough to permit me to come to a
definite opinion, and I do not think that they can be of use
in either contradicting or supporting Petrunkewitsch's
observations.*

The somatic cells are formed from the female
pronucleus. As this only contains half the somatic
number of chromosomes, Petrunkewitsch suggests that the
normal number is formed by a subsequent mitotic division
without a separation of the halved chromosomes.

Tenthredinid/E.

Many species of saw-flies lay eggs which develop
parthenogenetically. Taschenberg (93) gives a list of the
species in which parthenogenesis is found. Two forms
occur — Homoparthenogenesisand Heteroparthenogenesis,
and there may be thelyotoky, arrhenotoky or deutero-
toky. As a general rule the females are much more
numerous than the males. It has also been found that
the parthenogenetic young are constitutionally weak and
may die before attaining maturity. As either or both
sexes may be produced, this constitutional weakness

* Meves (64) finds that the primary spermatocyte contains 16 chromo-
somes. It gives off a small anucleate bud which he considers to be a
rudimentary spermatocyte. The large secondary spermatocyte then divides,
forming a small spermatid which he describes as abortive ; the large residual
spermatid forming a single spermatozoon. The author does not state whether
there is any reduction in the number of chromosomes in the last division, and
his figures are not clear enough to enable me to judge.

6 H EWITT, Cytological A spect of Parthenogenesis in Insects.

cannot be said to be compensated for by thelyotoky, and,
as Sharp has pointed out, it is in the most abundant
species in this country, the currant saw-fly, Nematus
ribesii, that one finds arrhenotoky occurring.

Doncaster (34) has recently investigated the matura-
tion and early development of the parthenogenetic ova of
several species of Nematus and allied forms.

Nematus ribesii. Doncaster finds that the nuclear
changes in the maturation of the parthenogenetic ovum
of this species are very similar to those observed by
Petrunkevvitsch in the drone-egg. There are two successive
mitotic divisions of the nucleus {Figs. 1-4) which take
place in a line at right angles to the surface of the Qgg.
The outer half of the first polar nucleus becomes flattened
against the Qgg membrane ; the inner half fuses with the
second polar nucleus to form the syntelosome (Figs.c^,6,s).
The female pronucleus travels inwards and becomes
buried in the yolky material. The syntelosome contains
either 14 or 16 chromosomes (the author is not quite
certain) and remains as a group of isolated chromosomes
as far as the blastoderm stage.

Nematus lacteus. (This species is probably arrheno-
tokous, as it closely resembles JV. pavidus^ which produces
males from unfertilised eggs.) The maturation and fate
of the polar bodies of the unfertilised &gg of this species
are very similar to that of A^. ribesii. There is not, how-
ever, a complete fusion of the inner half of the first polar
nucleus with the second polar nucleus. The two groups
of eight chromosomes lie side by side. One of these soon
disappears, the other persisting a little longer. In the
thelyotokous species Poecilosoma liiteolunt, Hemidiroa
rufa (which may produce a few males) and Croesus varus,
there is no conjugation of the inner half of the first polar

Manchester Memoirs, Vol. I. (1906), No. 0. 7

■nucleus with the second polar nucleus i^Figs. 1 1-14). The
process of polar nuclei formation is otherwise similar to
that of the previous species. They may take up positions
farther apart. They move towards the periphery of the
^^% and in P. luteoluin degenerate, or the products of the
first polar nucleus degenerate and the second becomes
resolved into chromosomes. It is also found that the
number of the chromosomes in the yolk nuclei, which are
derived from the egg-nucleus, remain at the maturation
number, as far as can be followed, that is to the forma-
tion of the blastoderm. Doncaster does not, with
Petrunkewitsch, fall back on the theory of Weismann,
that every ovum possesses a power of growth sufficient
to double its nuclear substance, but thinks that there is
no reduction, as in Rhodites rosae. This conclusion is only
provisional, and it cannot be definitely decided until the
oogenesis and spermatogenesis have been worked out.

FORMICID.-E.

Lasius niger. The maturation of the unfertilised ovum
of this arrhenotokous ant has been studied by Henking(44).
He found that three polar nuclei were formed. At one stage
there is a fusion of the inner half of the first polar nucleus
with the second polar nucleus to form a syntelosome, this
stage being very similar to that which he found in the
maturation of the fertile ovum {cf.figs. 327, 259). Later
he observed a dissolution of these nuclei. In a further
stage he observed a scattered mass of chromatin bodies,
which he considered to have been formed from the polar
nuclei, in the peripheral cytoplasm of the &^'g. Their fate
is unknown ; their resemblance to the mass of cells result-
ing from the sub-division of the syntelosome in the bee
is striking.

8 H E WITT, Cytological A sped of Parthenogenesis in Insects.

CHALCIDID/E.

The embryonic development of some species of
parasitic Hymenoptera is very remarkable and extremely
interesting. As the unfertilised eggs of the species
investigated are able to develop parthenogenetically
giving rise to males, they will be considered here.

Encyrtus fuscicollis. This small hymenopteron
deposits its eggs in the eggs of the larva of a lepidopteron
Hypono7ne2ita cognatella, and probably in other cater-
pillars. Bugnion (19) first described the development of
the larval Encyrtus but did not study the embryological
changes. This, however, has been done recently by
Marchal (60) whose account is most clear, and is accom-
panied by excellent figures. The first stage of the ovum
which he was able to obtain was that in which there were
two nuclei. No mention is made of polar bodies as he
was unable to obtain earlier stages. The ovum lies in the
body cavity of the embryonic caterpillar, and grows rapidly.
One of the nuclei, which he calls the paranucleus, grows
very large, the others subdividing until a large number
are formed. These last collect in groups, the protoplasm
surrounding them becomes rounded off and they form
separate morulae. The paranucleus divides up into a large
number of parts, which ultimately with their accompany-
ing cytoplasm form an investing sheath round the morulae.
External to this an enclosing membrane is formed from
the tissues of the host. By continued growth and
development of the morulae, they displace one another
until they are placed end to end in a string. Each of
these morulas forms a separate embryo. They are all
enclosed by two investing membranes, an inner which
Marchal terms the trophoamnion, which also separates
the embryos from each other by transverse septa, and is

■ Manchester Memoirs, Vol. I. ( 1 906), No. 6. 9

derived as shown before from the original paranucleus ;
and an outer layer of polygonal cells originating from the
host. The trophoamnion regulates and controls the food
of the embryos which it derives from the blood of the
host. In this manner the larvae grow until they attain
a certain size, they then enter and live in the blood of the
host, and finally, after devouring the vital parts, pupate in
the skin of the full grown caterpillar.

This process of germinogony as observed in Encyrtiis
and also in Polygnotus niimitits which Marchal also in-
vestigated, has been studied in another Chalcid Litomastix
truncatelhts by Silvestri (85).

Silvestri was fortunate enough to study the maturation
of the ovum. He found that the maturation of the
parthenogenetic ovum is similar to that of the fertile
ovum. Three polar nuclei are formed, which fuse later to
form a single nucleus. In the segmentation of the ovum,
this single ' polar ' nucleus becomes segmented off in an
apical mass of cytoplasm, in which it divides up into a
large number of nuclei. As development proceeds, this
cap of cytoplasm containing dividing nuclei surrounds the
segmenting germ mass; the original ovum is not mono-
embryonic, but gives rise to a large number of embryos.
These embryos are male if the ovum is parthenogenetic
and female if it has been fertilised, as is found in Encyrtus
and Polyg7iotus also.

The fate of the polar bodies is very remarkable in
Litomastix, and if the development of this species be
compared with that of Encyrtus, it seems probable that the
paranucleus of the latter may have arisen from the polar
nuclei, though unfortunately Marchal was unable to study
the maturation stages. He found the stage with two
nuclei, one of which may have been the female pro-
nucleus and the other the 'polar' nucleus.

10 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

Cynipid.e.

Rhodites rosae. Thelyotoky and deuterotoky both
occur in the gall-flies. Henking (44) has studied the
maturation of the unfertilised ovum of Rhodites rosae. In
this insect the males are rare, but they are occasionally
found. Henking found that in the nucleus of the ovarian
egg there were only nine chromosomes, the somatic number
being 18. The nucleus undergoes two divisions; three
polar nuclei and the female pronucleus are formed in this
manner, each containing nine chromosomes. In the
cleavage nucleus, however, 1 8 chromosomes were observed.
The second polar nucleus and inner half of the first polar
nucleus fuse, the outer half of the first polar nucleus going
to the periphery. Later the outer half of the first polar
nucleus wanders inwards, and fuses with the syntelosome.
This takes place about the same time as the female pro-
nucleus begins to divide. As Henking did not examine
any later stages he throws no light on the fate of the
' polar' nucleus.

In the maturation divisions of Rhodites there is no
reduction. The doubling of the chromosomes in the
female pronucleus before undergoing division is also
interesting. Henking considers that there was a pairing
of chromosomes before the first maturation division — the
pairs being resolved at the beginning of the cleavage
divisions.

DIPTERA.
Paedogenesis occurs in this order of insects in some
species of Miastor, a genus of the Cecidomyidae and in
Chironomus.

Miastor. Metschnikoff (63) studied the development
of the unfertilised ovum of this insect. Although he was
unable at that time to investigate the minute nuclear

Manchester Memoirs, Vol. I. (1906), No. 6. H

changes, his observations are of interest. In the early-
stages of the segmentation of the ovum, he describes and
figures ' Polzellen.' He traced these and found that they
underwent further subdivision and finally formed the
genital rudiment. Meinert (62) six years later investigated
the development of the same insect. He considered that
the ' Polzellen' were separated off from the layer of blasto-
derm cells, and disbelieves the idea that they wander
off later to form the genital fundament, although he
apparently did not trace their fate. The ' Polzellen ' may
not be the true polar bodies as is shown in the next
account of Chirononms.

Chironomus. Balbiani (i) followed the 'globules
polaires' which are formed at the posterior end of the ovum
of this insect and traced them to the primitive germ cells.
Ritter (81) investigated the early embryonic development
of Chirojionms. He describes the true polar nuclei, which
apparently have no relation to the ' Polzellen,' they
disintegrate before the latter are formed, and lie in the
anterior region of the 0.%^. He traced the ' Polzellen ' to
the genital rudiments, and so confirmed Balbiani's conclu-
sions. Metschnikoffs conclusions as to the fate of the
' Polzellen ' in Miastor were probably correct.

From Ritter's observations it appears that the
Polzellen are in no way connected with the polar nuclei ;
they probably arise in the same way in Miastor, but the
subject needs reinvestigating.

LEPIDOPTERA.

A few cases of Tychoparthenogenesis, which occurs
in this group, have been investigated cytologically. Males
and females may be produced. The maturation of the

12 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

unfertilised eggs has been studied chiefly by Blochmann
(8) Platner (79) and Henking (44). In Liparis dispar^
and Bombyx niori two polar bodies are formed as in the
fertile q^^. In many cases the embryo is unable to
complete its development. Nussbaum (72) found that
only 2% out of 1,100 unfertilised eggs oS. B. ;;^£>^7 segmented,
but they did not hatch out. Henking obtained an
unfertilised ovum of Lencoma salicis, but he examined it
at too late a stage to observe the formation of the polar
nuclei. Tichomiroff (94-99) has caused the partial develop-
ment of unfertilised ova of B. mori by mechanical and
chemical stimuli. It is difficult to say whether this is a
result of artificial parthenogenesis in every case, as the
eggs might have had already the power of developing by
natural parthenogenesis.

HEMIPTERA.

Aphid.e.

Aphis rosae. Blochmann (8) first investigated the
maturation of the parthenogenetic ovum of Aphis. He
found that a single polar body was given off. His
observations have been confirmed by the more recent work
of Stschelkanovzew (91) and Miss Stevens (86). I have
recently studied the maturation of the parthenogenetic
ovum of this species and my observations confirm those of
Miss Stevens. There is a single mitotic division and one
polar body is given off from the Qgg. It lies at the peri-
phery, but is absorbed later {Fig. 7). The number of
chromosomes is 10 and there is no reduction in the single
maturation division. They occur in five pairs of different
sizes. Stschelkanovzew observed I4 chromosomes in one
equatorial plate, but I think he counted chromosomes

* L. dispar is the destructive ' gipsy moth ' and is also called Ocneria
disfar.

Manchester Memoirs, Vol. I. (1906), No. 6. 13

which had been cut as they are often long and curved.
After the reabsorption of the polar body, the polar nucleus
appears to take no further part in the subsequent cleavage
stages. A simple mass of chromatic substance can be
followed to a late segmentation stage.

Stschelkanovzew believes that there is a passage of
achromatic material from the cytoplasm of the ovum into
the egg-nucleus, where it is changed into chromatic
substance. It is interesting to note that the oocytes which
develop into parthenogenetic ova are different in character
and more numerous in the ovary than the fertile ova.
They are smaller and very poor in yolk, the difference
between the two kinds of ovary being very noticeable,
even before any maturation has taken place. This fact is
of some importance as showing that the egg is already
destined in the ovary oi ApJiis to be either parthenogenetic
or fertile. This may be either a state of affairs which has
been arrived at through very many generations of
parthenogenetic cycles, the phenomenon having been long
acquired, or it may be the effect of nutrition upon the
reproduction of sex.

ORTHOPTERA.
Parthenogenesis occasionally occurs in this group, and
although it has been studied in a general manner, I am
unacquainted with any cytological investigations.

COLEOPTERA.

Osborne first recorded the occurrence of partheno-
genesis in this order. The only investigation which has
been made into the cytological changes is that of Saling
(82) on Tenebrio inolitor. The ovum after some remarkable
nuclear changes, which I am inclined to believe were of a
pathological nature, failed to segment. Very brief reference
is made to the maturation stages.

14 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

GENERAL CONSIDERATIONS.

In studying the work which has been done on partheno-
genesis in insects, certain points of similarity occurring in
the changes undergone by the nucleoplasm and cytoplasm
of the ovum are noticeable. In the majority of forms in
which the maturation of the fertile ovum has been studied,
it is found that true polar bodies are formed ; and
these may be either separated completely from the ovum
or lodged in a depression. In other cases the two polar
nuclei, or sometimes only one, may be contained in a
cytoplasmic projection at the periphery of the ovum.
Except in the case oi Aphis and the tychoparthenogenetic
ova of Lepidoptera which may, perhaps, be left out of con-
sideration on account of their exceptional occurrence, and
the usual failure of the resulting embryos to complete
their development, we do not find any definite polar
bodies formed in the parthenogenetic ova of insects.
Three polar nuclei are formed, but they are not extruded
as polar bodies.* This appears to me to have some
significance as regards the power which the ovum
possesses of developing parthenogenetically.

The two results of fertilisation are Amphimixis and
Embryogenesis. In parthenogenesis the former is ex-
cluded, and only the latter takes place. The question
then arises — what is the factor in the parthenogenetic &%^
which replaces the stimulus to development which is
brought about by the entrance of the spermatozoon ?

The statement of Boveri (12) that 'das Centrosoma
ist das eigentliche Theilungsorgan der Zelle : es vermittelt
die Kern- und Zelltheilung' is certainly disproved by

* In the fertile ova of a few insects it is found that the polar nuclei are
not extruded, but remain at the periphery of the egg

MaticJiester Memoirs, Vol. I. {igo6), No. ^. 15

parthenogenesis, as there is no centrosome brought in by
a spermatozoon, and in many cases the centrosomes do
not appear until the segmentation stages ; we cannot,
therefore, look to the centrosome as the cause of Embryo-
genesis.

The suppression of the second polar body in Aphis,
and as Brauer (18) found in Artemia, its fusion with
the &^^ nucleus, seemed at one time to support Boveri's
(13, p. 73) suggestion that parthenogenesis is the result of
fertilisation by the second polar body, that is, the second
polar nucleus took the place of the sperm nucleus. This
suggestion, however, receives no support from the majority
•of cases of parthenogenesis in insects, in which there is no
fusion of the second polar nucleus with the (t^^ nucleus.

The complete suppression of definite polar body
formation, or at least of the formation of the second polar
body may explain the ability of parthenogenetic eggs to
■develop. There may be a cytoplasmic fertilisation, so to
speak. The evidence we have, seems to me to suggest
that there is a chemico-physical relationship between the
nucleoplasm and cytoplasm, which is necessary for the
initiation of embryogenesis, and which is brought about
by the entrance of the spermatozoon. The non-extrusion
of polar bodies may bring about this relationship. What
the exact nature of this relationship is, or what are the
causes which operate, it is impossible at present to say.
It can be brought about by artificial means as Loeb
(58-59) Delage (23-27) Petrunkewitsch (77) Morgan (66-69)
and others have shown and differences of osmotic pressure,
ferments, metallic ions have all been suggested as possible
causes, but the problem is still unsolved.

As Morgan (70) has recently pointed out little atten-
tion is paid to the role of the cytoplasm. The work of
Delage (29) in Merogony, in which he fertilised portions

1 6 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

of egCTs not containing a nucleus and obtained embryos
from the same disproves the idea that fertilisation is an
affair of the nuclei only.

Nature of Parthenogenesis.

Parthenogenesis may be regarded as internal agamo-
genesis. But it is not by any means of the same nature
throughout the group. The polar bodies are now generally
regarded, I think, as abortive ova. The experiment of
Francotte (36) who was able to fertilise a rather large polar
body was satisfactory proof of the fact. It follows then^
that if we regard the polar nuclei as nuclei of abortive
ova, or better abortive gametes, there are different kinds
of individuals produced in the different types of parthe-
nogenesis which we find occurring.

In Aphis the formation of a second polar body is
suppressed, the resulting individual is then developed from
a cell the nucleus of which may be taken to represent the
fusion of two sister gametes. In Bouibyx mori and
Liparis dispar where two polar bodies are formed the
individual is a monont,* having developed from a single
gamete. In only one case has the complete development
of a parthenogenetic ovum, which does not extrude polar
bodies, been followed, namely, that of the bee. In this
form the second polar nucleus fuses with the inner half
of the first polar nucleus, and from the syntelosome
so formed, the male germ cells develop. These germ cells
develop then, from a nucleus which is really a diont, the
result of the fusion of two (abortive) germ nuclei of the
relation of cousins. The somatic cells of the bee are
derived from the q.^% nucleus. Consequently the drone
bee is a compound structure, as Blackman has also pointed
out. The germ cells have a different origin from the
* I am indebted to Dr. Sharp for this word.

Manchester Memoirs, Vol. I. (1906), No. 6. 17

somatic host which lodges them. This fact is interesting
in the light of Gaskell's (37) recent paper in which he
regards the adult individual as a neural syncytium holding
in its meshes the germ cells, and others, unconnected with
the nervous system. Petrunkewitsch's observatious re-
quire confirmation before they can be finally accepted.

The fate of the polar bodies has not been traced in
any other cases except by Silvestri in Litoniastix where
they develop into an embryonic investment, which I
suggested might be the fate of the polar bodies oi Encyrtiis
and Polygnotus. In the majority of cases they simply
become resolved into a number of chromosomes and
disintegrate.

Determination of Sex.

The desire on the part of some writers to treat the
sex character as being of the same category as somatic
characters, which include secondary sexual characters,
seems to me to be quite unjustifiable. Sex has a much
more fundamental significance than have somatic
characters. In the sexes themselves, the essential differ-
ence between the male and female consists in the fact
that they produce different kinds of reproductive cells.
All other differences — that is, somatic differences, are
secondary and subservient to the act of reproduction and
the development of the individual.

At the present stage of our knowledge, I think that
Castle's (21) attempt to fit in the facts of maturation with
the Mendelian principles of dominance and segregation
is premature. Comparatively few cases of partheno-
genesis in insects have been worked out with a view to
studying the maturation of the ova, but of the few cases
which we already know, a fairly large percentage would

1 8 Hewitt, Cytologicnl Aspect of Parthenogenesis in Insects.

have to be considered by Castle as exceptions to his
view (21, p. 200) that ' a segregation of sex characters
takes place at the formation of the second polar cell,' and
later, ' hence, if the egg which has formed two polar cells
develops without fertilisation, it must develop into a male.'
The two exceptions which he considers are RJiodites
rosae and Bonibyx mori. As has already been stated,
very few males are produced by the parthenogenetic eggs
of R. rosae. There is no reduction in the number of
chromosomes, so that there is no segregation of sex
characters. Castle admits this, and suggests that the
egg retains a capacity to eliminate the dominant female
character, which it does occasional!}', and so males are
produced, as in other parthenogenetic animals, under
appropriate conditions. (The italics are mine.) He finds
it necessary to bring in other conditions than the mere
segregation of male and female characters in the matura-
tion divisions. His treatment of the second exception,
that of B. vwri and L. dispar is still less convincing. In
the parthenogenetic eggs of these moths two polar bodies
are given off, as in the normal fertile &%^, and the small
proportion of these eggs which develop into perfect insects
are of both sexes. In attempting to explain this excep-
tion, he says (p. 205) ' But it is entirely possible that
in the very exceptional ^^^ which develops normally,
a second maturation division has for some reason failed
to take place, or after it has taken place, a reunion has
occurred of the second polar nucleus with the q^^ nucleus,
as sometimes in the &^^ of Arteniia, according to Brauer.
Such a reunion would bring together again the sex
characters segregated in maturation, and would produce
the physiological and morphological equivalent of the
cleavage nucleus of a fertilised egg. A similar result
would follow the complete suppression of the second

Manchester Memoirs, Vol. I. (1906), No. 6. 19

maturation division.' I do not consider the suggestion
drawn for a comparison with Arteviia a satisfactory
exj^lanation. The parthenogenesis of Arteniia is highly
speciaHsed, is the chief mode of reproduction and is
probably of old acquisition, whereas both B. mori and L.
dispar ox\\y occasionally reproduce parthenogenetically and
the nuclear changes in the maturation of their partheno-
genetic eggs do not differ essentially from those of the
fertile &^^. There does not appear to be any tendency
for either a suppression of the second polar nucleus or a
fusion of the latter with the &%^ nucleus.

Petrunkewitsch (76) believes that Brauer's second type
of maturation of the parthenogenetic ovum of Ai'teiuia,.
where there was a fusion of the second polar nucleus
with the egg nucleus, was pathological, as he searched
specially for it in material from the same locality and
failed to find it. Castle thinks that Petrunkewitsch
investigated the winter egg and that the second type of
maturation occurs in the summer egg, although Brauer
himself does not mention the fact. He cannot, however,
bring forward any such reason as this for Henking's and
Platner's failure to observe a similar phenomenon in the
eggs of B. mori and L. dispar.

But now there are some more exceptions which
Castle's theory would not be able to explain, namely the
thelyotokous saw-flies such as Poscilosoma luteohun and
others which Doncaster has investigated. In these, the
number of the chromosomes of the maturation divisions
remains the same and there is no fusion of the nuclei ;
the polar nuclei apparently disintegrate. These exceptions
to Castle's theory are sufficient, I think, to show that it
fails to explain the problem of sex in the parthenogenesis
of insects. They are not insignificant in number but form
a fair proportion of the cases which have been studied

20 Hewitt, Cytological Aspect of PattJienogenesis in Insects.

cytologically up to the present time. I am also unable to
fit in the facts which we find in Aphis with this theory,
Miss Stevens too, appears to find it difficult. In referring
to Castle's theory, Bateson (5, p. 127) says ' While ad-
mitting the likelihood of this suggestion, we feel that for
the present it should be received with caution. In par-
ticular, we doubt the conclusion that both ova and
spermatozoa (after a reduction division) are always bearers
of either the male or the female character. It seems
more likely that special cases will present special pheno-
mena in this respect.'

Doncaster (34) in attempting to explain the phenomena
which he observed in the maturation of the unfertilised
eggs of the saw-flies, combines Castle's hypothesis of the
separation of male and female bearing nuclei with that of
Le Dantec, who considers that maleness and femaleness
are similar to molecular forces which cause an attraction
between bodies bearing them, comparable to bodies
charged with opposite kinds of electricity. In this manner
he explains the fusion of the inner polar nuclei in the
arrhenotokous forms by supposing that the &^^ nucleus is
$ and the three polar nuclei ? , $ and ? respectively,
proceeding outwards, so that there will be an attraction
between the two inner nuclei. In the thelyotokous forms
the &^^ nucleus is $, and the polar nuclei c?, $ and 9
respectively, there will be no fusion then of the inner polar
nuclei. His idea is very ingenious, but requires further
testing, as he admits.

Another recent theory of sex is that of Ziegler (106)
which is based on the two assumptions that sex is a
character which can be transmitted by inheritance and that
the hereditary characters reside in the chromosomes.
The fact that Ziegler states that he knows that ' diese
Erklarung der Entstehung des Geschlechts nicht fiir alle

Manchester Memoirs, Vol. I. ( 1 906), No. 6. 2 1

Tiere zutrefifend sein kann. Die eigentiimlichen Fort-
pflanzLingsverhaltnisse der Honigbiene, der Gallwespen,
der Daphniden, der Rotatorien, des Dinophilus, u.s.w.
lassen sich nicht in dieser Weise aufklaren,' is of itself a
sufficient reason for not accepting it, were it not otherwise
unsatisfactory, as Morgan (71) has shown.

It is not by excepting these 'pecuh'ar' cases of
parthenogenesis, but by studying them, that we shall
have more light thrown upon the problem of sex. The
solution of the problem of the determination of sex is still
far off, and will continue so until we have more evidence
than is available at the present time.

The view that nutrition is an important factor in the
determination of sex has received some support. Siebold
(84) first showed that in the saw-fly {Nematus ventricosus)
more females than males were produced when abundant
nutrition was available. Maupas (61) was able to control
the sexes of Hydatina by feeding. Other evidence is
afforded by Nussbaum on Hydatina. Hoffmann (52) from
his study of plants came to the conclusion that the males
were incompletely developed individuals, formed under
unfavourable conditions. The Lepidoptera have furnished
material for many experiments with a view to finding out
the effect of nutrition on sex. Mrs. Treat (lOO) stated that
males and females were formed according as the
caterpillars were poorly or well-fed, Cuenot (22), however,
was unable to come to this conclusion from his experi-
ments, and Kellog and Bell (53) experimenting on the
same insects failed to obtain evidence in support of it.
The difficulties in investigating the effect of food on
caterpillars are considerable, the chief being that the
female imagines are larger than the males, and con-
sequently require more food during their larval stage, any
diminution then in the quantity of the food will affect the

22 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

female larvae first, and so by their deaths the percentage of
males produced will be increased. Strasburger (87) for
plants. Beard (7) and Lenhossek (57) all support the idea
on theoretical grounds that sex is already present in the
germ.

Absence of fertilisation, that is, development by
parthenogenesis, may produce different sexes in closely
allied forms, as in the saw-flies and in other Hymenoptera.
It is very probable that in the ovum we have only one sex ;
this may differ in allied forms, but it is characteristic of
the species. The production of the opposite sex is
determined by other factors, chief of which is fertilisation.
What other factors influence sex remains to be deter-
mined. I believe that nutrition does in some forms
influence the production of sex, for I fail to see otherwise
how a sexual generation is produced in the Aphides, as
shown by Kyber (55), by the diminution of sap in their
food, caused by the tissues becoming woody, in the summer
months, that is when everything which man can observe
is favourable for the continued production of partheno-
genetic females, except nutrition.

The Chromosomes.

The facts which we have at hand relating to the
number, division, and role of these bodies in cytological
phenomena are almost beyond comprehension. That
the chromosomes should have attracted so much atten-
tion is not surprising, as they are the chief visible factors
of cell division and reproduction which can be dealt with.
The rapid progress which has been made in the study of
these bodies has been due very largel}' to the fact that
they are considered to be the bearers of hereditary
characters, so it has been thought that by a study of

Manchester Memoirs, Vol. l. (1906), No. 0. 23

their changes some explanation might be found for the
problems of heredity and sex.

It may be of some interest to examine the evidence
we have for considering the chromosomes the sole bearers
of hereditary characters.

In 1866, Haeckel ^43, p. 288) wrote ' wir werden den
Kern der Zellen als das hauptsachliche Organ der Verer-
bung, das Plasma als das hauptsachliche Organ der An-
passung betrachten konnen." When the chromosomes
were observed we find Hertwig (47), Strasburger (87, 88),
and Weismann (104) considering them as the bearers of
the ' Vererbungsubstanz.' It was, however, considered by
many to be a proven fact that the chromosomes were the
bearers of the hereditary characters when Boveri (14)
made known his experiments on fertilising fragments of
Echinoderm ova.* He shook up echinoderm eggs in a
tube half filled with sea-water for some time. After this
rough treatment, he poured spermatozoa of another species
into the liquid containing whole eggs and fragments. In
this manner he obtained Plutei, concerning which he made
statements to the effect that there were a number of
small specimens, which he concluded came from frag-
ments of eggs ; as the nuclei of these same plutei were
very small, he concluded they came from enucleated
fragments ; also they bore pure paternal characters.
Therefore he concluded that the cytoplasm of the ovum
does not transmit any maternal characters ; it followed
then, that the nucleus alone was the bearer of the heredi-
tary characters. From this experiment it certainly
appeared as if the nucleus were the only part of the cell
concerned in the transmission of the hereditary characters.

When these experiments were repeated by Vervvorn

* O. and R. Hertwig (48) first attempted fertilising fragments of Echino-
derm ova. But their experiments were not conclusive.

24 Hewitt, Cytological Aspect of PartJienogenesis in Insects.

(lOl), MorfTan (67), and Seeliger (83), and more recently
by Delafje (28, 29, 30), their observations flatly contradicted
those of Boveri. The Plutei which they obtained by
fertilising enucleated portions of eggs with spermatozoa of
a different species, did not bear wholly paternal characters,
but maternal ones also. They found that the nuclei vary in
size ; the nucleus is often below the normal size in embryos,
having small cells coming from nucleated portions of
ova, consequently it is not at all certain that the nuclei
were absent in the fragments which gave rise to Boveri's
plutei. Further, the hybrid plutei which develop from
whole fertilised ova are very variable, their characters are
not necessarily intermediate between those of the parent
species ; some may be intermediate, but there are others
having characters wholly paternal.

These observers performed their experiments much
more carefully than Boveri, who did his very roughly,
and did not demonstrate the absence or presence of the
nuclei in the fragments. Nor did he obtain his enucleated
portions of ova, if there were any [there is no proof that
there were], by cutting them up individually. The results
of Delage and others are much more trustworthy. But
in spite of the fact that Boveri's experiments have been
contradicted, though he has attempted to defend his views
(16), a great amount of faith is placed in them. The
tendency is to take it as a proven fact, and to argue on
the assumption, that the chromosomes are the sole bearers
of hereditary characters, which I believe to be far from
proven.

Working on this assumption. Castle (21), Sutton (92),
Hacker (42), Montgomery (65), and others have sug-
gested hypotheses to account for and explain the nature
of heredity and sex. Each worker has attempted to
make the chromosome data fit in with the theory which

Manchester Memoirs, Vol. I. (1906), No. 0- 25

he champions, with a certain amount of success one must
admit. But it certainly appears to be pushing the matter
beyond reasonable limits when Boveri and other investi-
gators suggest that different sized chromosomes may
bear different or correlated characters.

The cytoplasm of the ovum and the spermatozoon is
left out of the question entirely in questions of heredity
by these investigators. This is an unjustifiable pro-
ceeding. The cytoplasm of the spermatozoon is in a
highly concentrated condition, as also is the nucleu.s, and
v/e have no evidence against the view that the cytoplasm
of both ovum and spermatozoon or of either of them may
bear some hereditary characters as well as the nucleo-
plasm. It must be remembered also, as will be shown
later, that in some ova a large quantity of nucleoplasm is
scattered in the cytoplasm. Until it has been proved
that the cytoplasm takes no part whatever in the trans-
ference of hereditary characters, all hypotheses which
only take the chromosomes into account, must be received
with due caution.

In those parthenogenetic eggs of insects in which
there is a reduction in the number of chromosomes, it is
probable that the normal number is formed later as a
physiological necessity. I believe, with Delage, that the
cell is able to re-establish the normal number of chromo-
somes, and that ' si le nombre des chromosomes est
constant chez les animaux, ce n'est pas, comme on le
croit, parceque ces organites ont une personalite qui les
rend individuellement permanents, c'est parceque ce
nombre est une propriete specifique de la cellule, une
constante de la cellule.' This raises the question as to
whether a fixed number of chromosomes always occurs.
It is usually considered that this is the rule, but certain
observers have found that this is not always the case.

26 YiY^SYYT^Cytological Aspect of Parthenogenesis in Insects.

Von Winiwarter (105) has found that the number is very
variable in the rabbit, varyint^ from 36 to 80, the average
being about 42. The reduced number in the sexual cells
of the rabbit is 12, according to this, then, the somatic
number should be 24. Guignard (41) has shown that in
certain plants the chromosomes may remain at their reduc-
tion number in the somatic cells. Farmer and Shove ('35)
have shown that in the somatic cells of Tradescantia the
chromosomes vary in number from 26 to 33. There are two
varieties of Ascaris, univalens and bivalens, and according
to Brauer two types of parthenogenetically developed
ArteniicB., possessing 84 and 168 chromosomes respectively.
Delage (32, p. 127), who also calls attention to this fact,
found in his experiments on fertilising enucleated portions
of the ova of Strongyiocentrotus, that the cells of the
developing embryo contained 18 chromosomes, although
the spermatozoon only contained nine, the reduced number.
He also found 18 in the cells of the embryos of the same
species which had been made to develop partheno-
genetically. All these facts tend to make one believe
that there is not an absolute permanency in the number
of chromosomes, and favour Delage's view that the
number of chromosomes is ' une propriete cellulaire.'

The suggestion of Stschelkanovzew's (91) raises the
further question, to what extent do we get chromatin
formed by the change of achromatic material which has
entered the nucleus from the cytoplasm ? It is extremely
probable that this takes place to a greater extent than is
recognised. Cameron (20) has found in studying the
development of nerve cells that there is a chromatisation
of achromatic material, which he considers as chromatin
in a nascent condition. The presence, in some cases, as
Alcyoniiim and other Coelenterates, as shown by Hickson,
of a large quantity of chromatic material in the cytoplasm

Manchester Manoirs, Vol. I. (1906), No. i». 27

supports such an idea. But our knowledge of the micro-
chemistry of the cell is still too meagre to arrive at absolute
certainty on this and other points.

The usual preconceived idea of the unchangeable
nature of the nucleus does not rest upon a solid foundation.
As Gruber (40) in the Protozoa, Hickson (50), Hill (51),
and Hargitt in the Coelenterates, Henking in Insects, and
many others in other groups, have all shown that the
nucleus in many cases loses its compact character and
fragments, nor is this a pathological condition, but quite
normal. In some cases the chromatin granules remain
clustered together, in others they become scattered
through the cytoplasm of the ovum, either remaining in
that condition, or collecting later into definite nuclei.
These facts do not support any theory of the individuality
of chromosomes, as put forward by Rabl (80), Baumgartner
(6), Boveri and others. Nor do they furnish evidence for
any hypothesis which considers the chromosomes only as
the bearers of hereditary characters.

The evidence available on the cytology of partheno-
genesis in insects is too small and too diversified at
present to allow us to draw any conclusions on questions
of heredity and sex, about which we know little. It is
certain, however, that the accumulation of observations
and facts on this subject and not of hypotheses and
theories on insufficient facts, will be the only way in which
we shall be able to elucidate those, at present, unsolved
riddles.

28 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

LITERATURE REFERRED TO.

Balbiani, E. G. " Sur la signification des cellules polaires
des Insectes." Compt. Rend., vol. 95, p. 927-929, 1882.

. " Contribution a I'etude de la formation des organes

sexuels chez les Insectes." Rev. Zool. Suisse, vol. 2, p.
525-588, 1885.

•. " Sur les conditions de sexualite chez les Pucerons."

Intermed. Biologistes, vol. i, p. 170-174, 1898.

4. Bataillon, E. " Pression osmotique de I'ceuf, et poly-

embryonie experimentale." Compt. Rend., vol. 130,
p. 1480-1482, 1900.

5. Bateson, W., Miss E. R. Saunders, R. C. Punnett &

C. C. Hurst. " Reports to the Evolution Commictee,
Royal Society," 2, 154 p., 1905.

6. Baumgartner, W. J. "Some evidences for the indi-

viduality of the chromosomes." Biot. Butt., vol. 8, p.
1-23, 3 pis., 1904-

7. Beard, J. " The Determination of Sex in Animal Develop-

ment." Zoot. Jahrb. (Anat.), vol. 16, p. 703-764, pi.
45, 1902.

8. Blochmann, F. "Ueber die Richtungskorper bei

Insekteneiern." Morph. Jahrb., vol. 12, p. 544-5 74^
pis. 26-27, 1887.

9. . "Id." Biol. Centralb., vol. 7, p. 108-111, 1888.

10. . "Ueber die Richtungskorper bei unbefruchtet

sich entwickelnden Insekteneiern." Verh. Ver.
Heidelberg, N.s., vol. 4, p. 239-241, 1888.

11. . "Ueber die Zahl der Richtungskorper bei befruch-

teten und unbefruchteten Bieneneiern." Morph. Ja/irb.^
vol. 15, p. 85-96, pi. 5, 1889.

Manchester Memoirs, Vol. I. ( 1 906), No. O- 29

12. BovERi, T. " ilber den Antheil des Spermatozoon an der
Theilung des Eies." Sitzungsb. Gesell. Morph. Phys.
Miinchen, vol. 3, p. 1 51-167, 1887.

13. . " Die Bildung der Richtungskorper bei Ascaris

inegalocephala und Ascaris hcnil'ricoides." "Zellen-
studien." Jena, part i., 94 p., 4 pis., 1887.

14. . " Ein geschlechtlich eizeugter Oiganismus ohne

miitterliche Eigenschaften." Siizungsl>. Gesell. Morph.
Phys. Miinchen, vol. 5, p. 73-83, 3 figs., 1889.

15. . " Ueber die Befruchtungs- und Entwickelungs-

fahigkeit kernloser Seeigeleier und iiber die Moglichkeit
ihrer Baslardirung." Arch. Etitiv. Aleck., vol. 2, p.
394-444, pis. 24-25, 1895.

16. . " Ueber mehrpolige Mitosen als Mittel zur Analyse

des Zellkernes." Verh. Phys.-med. Gesell. Wiirzburg,
N.S., vol. 35, p. 67-90, 1902.

17' • " Ergebnisse iiber die Konstitution der chromatis-

chen Substanz des Zellkernes." Jena. (Fischer). 130 p.,
75 figs-, 1904-

18. Brauer, a. "Zur Kenntniss der Reifung des partheno-

genetisch sich entwickelnden Eies von Arttmia salina."
Arch. Mikr. Anat., vol. 43, p. 162-222, 4 pis., 1894.

19. BuGNiON, E. " Recherches sur le developpement postem-

bryonnaire, I'anatomie at les mceurs de lEncyrtus
fuscicollis." Recneil Zool. Suisse, vol. 5, p. 435-534,
pi. 20-25, 1891-

20. Cameron, J. " Histogenesis of Nerve Fibres." To appear

in Joutn. Anai. Phys., vol. 40, 1906.

21. Castle, W. E. "The Heredity of Sex." Bull. Mus.

Covip. Zool. Harvard, vol. 40, p. 189-218, 1903.

30 Hewitt, Cy to logical Aspect of Parthenogenesis in Insects.

22. CuENOT, L. " Sur la determination du sexe chez les

animaux." Bull. Sci. France et Belg., vol. 32, p. 462-
S3S> 1899.

23. Delage, Y. " Parthenogenese par I'acide carbonique

obtenu chez les oeufs apres remission des globules
polaires." Arch. Zool. Exper., ser. 4, vol. 2, p. 43-46,
1904.

24. • " L'acide carbonique comme agent de choix de la

parthenogenese experimentale chez les Asteries." Compt.
Rend., vol. 135, p. 570-573, 1902.

25. . " Sur le mode d'action de l'acide carbonique dans la

parthenogenese experimentale." Compt. Retid.., vol. 135,
p. 605-608, 1902.

26. . " Elevage des larves parthenogenetiques d'Asteries

dues a Faction de l'acide carbonique." Compt. Rend.,
vol. 137, p. 449-551. 1903-

27. . " Nouvelles recherches sur la parthenogenese

experimentale chez Asterias glacialis.'''' Arch. Zool.
Exper., ser. 3, vol. 10, p. 213-235, 1902.

28. . " Embryons sans noyau maternal." Compt. Rend.,

vol. 127, p. 528-531, 1898.

29. . "Etudes sur la nierogonie." Ai-ch. Zool. Exper.,

ser. 3, vol. 7, p. 383-417, II figs., 1899.

30. . " Sur interpretation de la fecondation merogonique

et sur une theorie nouvelle de la fecondation normale."
Arch. Zool. Exper., ser. 3, vol. 8, p. 511-527, 1899.

31. . " Sur la maturation cytoplasmique chez les Echino-

derms." Arch. Zool. Exper., ser. 3, vol. 9, p. 285-326,
14 figs., 1901.

32. . " Les theories de la fecondation." Verhandl. des

V. Inter nat. Zool. Congresses zu Berlin, p. 1 21-140, 1901.

Manchester Memoirs, VoL I. (1906), No. 6. 31

33. Delage, Y. " L'Heredite, et les grandes problemes de la

biologic generale." Paris. 1903.

34. DoNCASTER, L. " On the Maturation of the Unfertilised

Egg and the Fate of the Polar Bodies in the Tenthredi-
nidae (Saw-flies)." Quart, Journ. Micr. Sci., vol. 49'
p. 561-589, 2 pis., 1906.

35. Farmer, J. B. and D. Shove. "On the Structure and

Development of the Somatic and Heterotype Chromo-
somes of Tradescantia virginua. Quart. Journ. Micr.
Sci., (n.s.), vol. 48, p. 559-569, 2 pis., 1905.

36. Francotte, p. " La maturation, la fecondation et la seg-

mentation chez les Polyclades." Me/u. Cour. Acad.
Belg., vol. 55, 72 p., 3 pis., 1897.

37. Gaskell, W. H. " On the Origin of the Vertebrates

deduced from the study of Ammocoetes. Part xii. The
Principles of Embryology." /our?i. Atiat. Fhys., vol. 39,
p. 371-401, 1905.

38. GiGLio-Tos, E. " Delia parthenogenesi e della spermato-

genesi nell'ape." Anat. Anz., vol. 26, p. 369-373, 1905.

39. Guerin, p. " Les connaissances actuelles sur la feconda-

tion chez les Phanerogames, Paris." 160 p., 31 figs.,
1904.

40. Gruber, a. " Ueber Kern und Kerntheilung bei den

Protozoen." Zeitschr. iviss. Zool., vol. 40, p. 1 21-153,
2 pis., 1884.

41. Guignard, L. " Le developpement du pollen et la re-

duction chromatique dans le Naias." Arch. Anat. Micr.,
vol. 2, 1899.

42. Hacker, V. " Bastardirung und Geshlechtszellenbildung."

Zoo/, lahrb., Suppl. 7, p. 161-256, 14 figs., pi. 12, 1903.

43. Haeckel, E. " Generelle Morphologie." Vol. i. Berlin,

1866.

3 2 H EWITT, Cytological A sped of PartJienogenesis in Insects.

44. Henking. "Untersuchungen iiber die ersten Entwick-

lungsvorgiinge in den Eiern der Insekten. iii. Specielles
und AUgemeines." Zeilichr. 2viss. ZooL, vol. 54, p. 1-274,
12 figs., pis. 1-12 (430 figs.), 1892.

45. Henneguv, F. " Les Insectes." Paris, 804 p., 1905.

46. . "Les modes de reproduction des Insectes." Bull.

Soc. Philom. Paris, ser. 9, vol. i, p. 41-86.

47. Hertwig, O. " Das Problem der Befruchtung und der

Isotropic des Eies, eine Theorie der Vererbung." Jen.
Zeitschr. Nat., vol. 18, p. 276-318, 1885.

48. Hertwig, O. and R. Hertwig. "Ueber den Befruchtungs-

und Teilungsvorgang des tierischen Eies unter dem
Einfluss iiusserer Agentien." Jeti. Zeitschr. Nat., vol. 20,
p. 120-241, and p. 477-510. P^s- 3-9' 1^87.

49. HiCKSON, S. J. " Fragmentation of the Oosperm nucleus in

certain ova." F/vc. Camh. Phil. Soc, vol. 8, p. 12-17,
1892.

50. . " Early stages in the Development of Distichopora,

with short essay on the fragmentation of the Nucleus."
Quart. Joiirn. Micr. Soc, vol. 35, p. 129-158, i pi., 1894.

51. Hill, M. D. " Notes on the Maturation of the Ovum of

Alcyonium digitatiun." Quart. Journ. Micr. Sci., vol. 49,
P- 493-505. 7 figs-, 1905-

52. Hoffmann, H. " Uber Sexualitat." Bot. Zeit., p. 145-

153, 161-169, 1885.

53. Kellog, V. L. AND R. G. Bell. " Notes on Insect Bio-

nomics." Journ. Exp. ZooL, vol. i, p. 357-367, 1904.

54. Kulagin, W. " Ueber die Frage der geschlechtlichen

Vermehrung bei den Tieren." Zool. Anz., vol. 21,
p. 653-667, 1898.

Manchester Memoirs, Vel. I. (1906), No. C 33

55. Kyber, J. F. " Einige Erfahrungen und Bemerkungen iiber

Blattliiuse." Germar's Mag. Entom.,\Q\. i, p. 1-37, 1815

56. Lameere, a. "Apropos de la maturation de I'oeuf par-

ihenogenetique." Thhe couronnee an concours de
i'enseignejiiejit supcrieur pojtr 1880- 1889. Bruxelles.

57. Lenhossek, M. " Das Problem der geschlechtbestim-

menden Ursachen. Jena." (Fischer). 1903.

58. LoEP., J. " Ueber Methoden und Fehlerquellen der

Versuche iiber kiinstliche Parthenogenese." Arch.
Ent. Alech., vol. 12, 1902.

59. • "Ueber Eireifung, natiirlichen Tod und Verlangerung

des Lebens beim unbefruchteten Seesternei (Asterias
Forbesi) und deren Bedeutung fiir die Theorie der
Befruchtung." Arch.ges. Physiol., vol 93, p. 59-76, 1902.

60. Marchal, p. " Recherches sur la biologie et le

developpement des Hymenopteres parasites. i. La
Polyembryonie specifique ou Germinogonie." Arch.
Zool. Exp. (4), vol. 2, p. 257-336, pis. 9-13, 1904.

61. Maupas, E. " Sur la determinisme de la sexualite chez

rHydaiina scfita.'" Coinpt. Rend., vol. 113, p. 388-390,
1891.

62. Meinert, F. " Om Q^ggets Anloege og Udwikling og

om Embryonets forste Dannelse i Miastor larven."
Naturhist. Tidskr., vol. 8, p. 345-378, i pi., 1872.

63. Metschnikoff, E. " Embryologische Studien an In-

secten." ZeitscJir. wiss. Zool., vol. 16, p. 389-500, pis.
23-32, 1866.

64. Meves, F. "Ueber Richtungskorperbildung im Hoden

von Hymenopteren." Anal. Anz., vol. 24, p. 29-32
8 figs., 1904.

34 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

65. Montgomery, T. H. "The main facts in regard to the

cellular basis of Heredity." Froc. Amer. Phil. Soc,
vol. 43, p. 5-14, 1904-

66. Morgan, T. H. " Experimental studies in Echinoderm

eggs." Anal. Anz., vol. 11, p. 141-152, 4 figs., 1893.

67. . "The fertilisation of non-nucleated fragments of

Echinoderm eggs." Arch. Ent. Mech., vol. 2, p. 268-
281, 3 figs., 1896.

68. . " Action of salt solutions on the unfertilized eggs of

Arbacia and other animals." Arch. Ent. Mech., vol. 8,

P- 448-539> 1899-

69. . " Further studies on the action of salt solutions

and of other agents in the eggs of Arbacia." Arch. Ent.
Mech.y vol. 10, p. 489-524, 1900.

70. • " ' Polarity ' considered as a phenomenon of

gradation of materials." /ourn. Exp. ZooL, vol. 2,
p. 495-506, 1905.

71. . " Ziegler's theory of sex determination and an

alternative point of view." Science, n.s., vol. 22,
p. 839-840.

72. NussBAUM, M. "Zur ParthenogenesederSchmetterlingen."

Arch. Mikr. Anat., vol. 53, p. 444-480, 1899.

73. Paulcke, W. "Zur Frage der parthenogenetischen Ent-

stehung der Drohnen. {Apis Mellifica $ )". Ant. Anz.,
vol. 16, p. 474-476, 2 figs., 1899.

74. Petrunkewitsch, a. " Die Richtungskorper und ihr

Schicksal im befruchteten und unbefruchteten Bienenei."
Zool. Jahrb. {Anat.), vol. 14, p. 573-606, pis. 43-45,
1901.

[75. . " Das Schicksal der Richtungskorper im Drohnenei.

Ein Beitrag zur Kenntniss der natiirlichen Partheno-
genese." ZO0I. Jahrb. {Anat.), vol. 17, p. 481-516, 1902.

Manchester Memoirs, Vol. I. (1906), No. 0. 35

76. Petrunkewitsch, a. " Die Reifung der parthenogene-
tischen Eier von Artemia salina." Anat. Anz., vol. 21,
p. 256-263, 4 figs., 1902.

77. . " Kiinstliche Parthenogenese." Zool.Jahrb., Suppl.

7, p. 77-138, pis. 8-10, 1904.

78. Phillips, E. F. " A Review of Parthenogenesis." Proc.

Amer. Phil. Soc, vol. 42, p. 275-345, 1903.

79. Plainer, G. " Die erste Entwicklung befruchteter und

parthenogenetischer Eier in Liparis dispart' Biol.
Ceiitralbl., vol. 8, p. 521-524, 1889.

80. Rabl, C. " Ueber Zelltheilung." Morph. Jahrh., vo\. \q,

p. 214-330, 5 figs., 7 pis., 1885.

81. Ritter. " Die Entwicklung der Geschlechtsorgane und

des Darmes bei Chironomus." Zeitschr. iviss. Zool.,
vol. 50, p. 408-427, pi. 16, 1890.

82. Saling, T. " Notizen iiber Parthenogenese bei Tenebrio

tnolitorP Zool. Anz., vol. 29, p. 587-590, 2 figs., 1905.

%2y Seeliger, O. " Giebt es geschlechtlich erzeugte Organ-
ismen ohne miitterlich Eigenschaften ? " Arch. Ent.
Meek., vol. I, p. 203-223, 3 figs., 2 pis., 1895.

84. SiEBOLD, C. Tn. E. von. " VVahre Parthenogenese bei

Schmetterlingen und Bienen. Ein Beitrag zur Fort-
pflanzungsgeschichte der Thiere." Leipzig. 144P., ipl.,
1856.

85. SiLVESTRi, F. " Contribuzioni alia cognoscenza biologica

degli Imenotteri parassiti. i. Biologia del Litomastix
triincatellus (Dalm.)." Ann. delia R. Scuola Sup.
d'AgricoIttira di Portici., vol. 6, p. 1-52, 13 figs.,
pis. 1-4, 1906.

36 Hewitt, Cytologicnl Aspect of Par'tJtenogotesis in bisects.

86. Stevens, N. M. " A Study of the Germ cells of Aphis

rosae c^' A. a'liotherae." Journ. Exp. ZooL, vol. 2,
P- 313-334, pis. 1-4, 1905.

87. Strasburger, E. " Neue Untersuchungen iiber den

Befruchtungsvorgang bei den Phaneiogamen, als
Grundlage fur eine Theorie der Zeugung." Jena. 1884.

88. . " Uber Kern- und Zelltheilung im Pflanzenreich,

nebst einem Anhang iiber Befruchtung." Jena. 1888.

89. . " Versuche mit diocischen Pfiangen in Riicksicht

auf Geschlechtsverteilung." Biol. Cen/rall)!., vol. 20, p.
637-665, 6S9-698, 721-731, 753-785-

90. • " rj)ie Stoffiichen Grundlagen der Vererbung im

Organischen Reich." Jena (Fischer), 68 p., 34 figs., 1905.

91. SrscHELKANOvzEW, J. P. " tjber die Eireifung bei ovi-

paren Aphiden." Bioi. Centralbl., vol. 24, p. 104-112,
7 figs., 1904.

92. Sutton, W. S. " The Chromosomes in Heredity." Biol.

Bull., vol. 4, p. 231-248, 1903.

93. Taschenberg, O. "Historische Entwicklung der Lehre

von der Parthenogenesis." Abhaiidl. naturf. Gesell.
Halle, vol. 17, p. 111-197.

94. TiCHOMiROFF, A. " Die kiinstliche Parthenogenesis bei

Insekten." Arch. Anat. Fhys., Phys. Abth., Leipzig,
Suppl., p. 35-36, 1886.

95. . " Sullo svillippo delle uova del bombice del gelso."

Boll. 77ie.iis. di Bachicolt, 1886.

96. . [Parthenogenesis of Bombyx] Russian. Selsskoe

chosaisttvo i Liisobodstivo. c.L., vol. 4, 1887.

97. . " Nochmals iiber Parthenogenesis bei Bombyx

viori.'" Zool. Anz., vol. 11, p. 342-344, 1888.

Manchester Memoirs, Vol. I. {igo6), No. iV 37

98. TiCHOMiROFF, A. "Zur Biologic des Befruchtungspro-
zesses, viii." Kongr. russisch. natiirf. u. Aerzte, vol. 3.
Abstr. Bio/. Centralbl., p. 424-425, i88g.

99. . [Parthenogenesis of Bombyx] Russian. Moscow,

100. Treat, Mary. " Controlling sex in Butterflies." Amer.

Nat., vol. 7, p. 129-132, 1873.

101. Verworn, M. " Die physiologische Bedeutung des

Zellkerns." Pfliigers Archiv. f. ges. Physiol., vol. 51,
1891.

102. ViGUiER, C. " Influence de la temperature sur le de'-

veloppement parthenogenetique." Compt. Rend., vol.
135' P- 60-62, 1902.

103. Weismann, a. " Die Continuitat des Keimplasmas als

Grundlage eine Theorie der Vererbung." Jena
(Fischer), 122 p., 1885.

104. • "Essays upon Heredity." Oxford, vols. i. and ii.

105. Winiwarter, H. von. " Le corpuscle intermediare et le

nombre des chromosomes chez le Lapin." Arch. Biol.,
vol. 16, p. 685-705, pi. 29, 1900.

106. ZiEGLER, H. E. "Die Vererbungslehre in der Biologic."

Jena (Fischer), 74 pp., 9 figs., 2 pis., 1905.

38 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

EXPLANATION OF PLATES.

^. Egg nucleus, /i. Outer half of first polar nucleus, /'i. Inner
half of first polar nucleus. p2. Second polar nucleus.
Pb. Polar body. S. Syntelosome (= p'\ + p2).

Plate I.

I. Formation of polar nuclei in parthenogenetic egg oi Nematus

ribesii (diagrammatic after Doncaster).

First polar spindle.

First polar spindle dividing into the two polar spindles

of the second polar mitosis.
Second polar mitosis.
Egg nucleus {e) and three polar nuclei.
Fig. 5. Fusion of two inner polar nuclei to form the syntelo-
some.
Fig. 6. Resolution of syntelosome into 16 chromosomes.

2. Aphis rosae.
Figs 7. Two mature parthenogenetic ova of Aphis rosae in their
follicles. The single polar body of the upper ovum
lying at the periphery ; the polar body of the lower
ovum has been absorbed: x 1500 (drawn with
camera lucida).

Fig.

I

Fig.

2

Fig.

3-

Fig.

4.

Manchester Memoirs, Vol. L.
1.

Plate I.

>5JCf5> l^\ V^^Pf^i^i:

E> p(

: W

C. lif "f '

f,_

/^V {\;-

..-■v/ )X ■-■"

) ■

S-.^*>v-<:;

^r-

"'~^iJ

oChi.^ '

\^':i'.

%5

■a:-P

'>■•"'•>. :

iSi^i

3.

vi.-,.0-t: ;

Pb

Pi

40 Hewitt, Cytological Aspect of Parthenogenesis in Insects.

Plate II.

I. Formation of polar nuclei in parthenogenetic egg of Apis
(after Petrunkewitsch).

Fig. 8. Second polar mitosis, each of the three polar nuclei
and the egg nucleus contain 8 chromosomes, the
reduced number.

Fig. 9. Fusion of the inner half of the first polar nucleus and
the second polar nucleus to form the syntelosome
( Petrunkevvitsch's " conjugation-nucleus ").

Fig. 10. The degeneration of the first polar nucleus, and
wandering inwards of the egg nucleus. The syntelo-
some contains 16 chromosomes.

2. Formation of polar nuclei in parthenogenetic egg of Poecilosoma
luteolum (diagrammatic after Doncaster).

Ftg. II. First polar spindle.

Fig. 12. Second polar mitosis.

Fig. 13. Egg nucleus and three polar nuclei : inner largest,

outer smallest.
Fig. 14. Degeneration of polar nuclei and inward migration

of egg nucleus {e).

Manchester Memoirs, Vol. L.

Plate II.

11.

b'/ .

^n<

■•■■ '^-;'^f"'

10.

•-£.

13.

*rfl#:

14.

y~--

I ;<*

■«:;>v;^i---£

Manchester Memoirs, Vol. I. ( 1 906), No. 1.

THE WILDE LECTURE.
VIL "Total Solar Eclipses."

By Professor H. H. Turner, D.Sc, F.R.S.

Delivered March 20th, igo6.

To those who have never seen a total eclipse of the
sun, but have seen partial eclipses, the importance
attached to the former may seem somewhat puzzling.
A partial eclipse is an interesting but not a very striking
phenomenon : we watch the queerly-shaped disc of the
sun through smoked glass (or by reflection from water)
and we may have seen the crescent-shaped patches of
light which filter through the trees ; but there is not
much suggestion of an astronomical opportunity differing
essentially in character from that afforded by full day-
light. Those, however, who have been fortunate enough
to see the sun completely hidden by the moon's disc
know that, when the hiding is complete, there flashes
into view a spectacle entirely strange, a wonderful halo
or glory of light of a diffuse character which has been
called the Corona. Close to the sun this light is very
bright — not by any means so bright as a portion of the
ordinary sun, but still brilliant compared with the outer
Corona. Often there are jets of especially brilliant light
of rosy colour, and this colour may be traced, close to
the black central disc of the moon, for some distance
round the circumference. On the Chromosphere, as this
close envelope is called, an important observation was

May 26th, igo6.

2 TURNElR, Total Solar Eclipses.

made in 1868, independently by two observers, Janssen
and Lockyer. They found that its spectrum was made
up of distinct bright hnes — special and very definite
colours, that is to say — and by making use of this know-
ledge it has since been found possible to see this inner
envelope in full daylight without a total eclipse. More
recently a new instrument, the spectro-heliograph, has
been invented which enables us, following out this same
principle, to pJiotogj-aph the Chromosphere, as well as other
portions of the sun's surface, of which we had practically
no knowledge a few years ago. But although many
attempts have been made to detect in similar ways the
Corona or any part of it, they have up to the present
resulted in failure, and a total eclipse remains unique as
an occasion for studying a most important part of the
sun, viz., all that out-lying portion of many times his own
diameter which we call the Corona.

It must not be thought, however, that partial eclipses
are of no value ; they afford us opportunities for very
accurate measures of the relative places of the sun and
moon, and this at a time when we ordinarily cannot well
observe the place of the moon ; for when the moon is
nearly " new," it is during the daytime lost in the sun's
glare, and near sunrise or sunset is so low in the sky that
other troubles arise. Even without accurate measurements
(such as we can now make, but which thousands of years ago
were undreamt of), a mere record that an eclipse took place
is sufficient, if trustworthy, to give us valuable information
as to the motion of the moon, or of the earth in its orbit
round the sun. There have indeed been found difficulties
in interpreting many of these old records, difficulties so
great that of late years it has been doubted whether the
records are trustw^orthy enough for this purpose. Ideas
of accuracy have no doubt changed in the thousands of

iManchester Memoirs, Vol. I. {igo6), No. 7. 3

years since the first crude attempts were made to write
history, and early historians have been suspected of intro-
ducing eclipses into their narratives as scenic effects, much
as a painter might introduce into a picture figures or
clouds which might have been, but were not, actually there.
Within the last year, however, Mr. Cowell, of the Green-
wich Observatory, has thrown a new light upon some of
these old records. He has shown* that by a new and
simple supposition no less than five important records of
eclipses, which it had been previously considered difficult
if not impossible to reconcile with what we know of the
moon's motion from modern observations, could all be
brought into line. His hypothesis is indeed a little start-
ling, being no less than a gradual quickening of the
velocity of the earth in its revolution round the sun, which
we have been accustomed to regard as having been con-
stant for ages, if we disregard variations of a periodical
character. But there is more than one possible vera causa
for such a change of motion. We have long been familiar
with such a change in the case of the moon, and indeed
there has been a controversy of no mean order about this
same " secular acceleration," a controversy which is some-
what of a curiosity,! for it centred round a purely mathe-
matical calculation, about which it might have been
supposed there could be no doubt. But in the case of the
moon there is no difficulty in assigning the causes for this
change, whereas in the case of the newly discovered
change in the earth's motion difficulties do arise. We
may suppose, for instance, that there is a resisting medium
of an extremely tenuous character, and this would produce
the proper effect upon the earth ; but it would be natural

*See Mon. Not. Roy. Astron. Sac, vul. 65, pp. 861, 867; vol. 66.

PP- 3. 5, 35. 36.

tSee, for instance, Mon. Not. Roy. Astron. Soc, vol. 53, p. 198.

4 Turner, To fa/ So/ar Ec/z/>ses.

to suppose that the planet Mercury, which is nearer the
sun and therefore moves more rapidly, would be affected
at least as much as the earth. Mr. Cowell has accordingly
tested* this hypothesis by analysing the observations of
transits of Mercury, and found confirmation of the change
in the earth's motion, but no evidence of change in that of
Mercury. We have learnt, however, in Astronomy to be
patient in expecting the interpretation of results so long
as we can make sure of their reality from observation,,
and we may await therefore with patience, although with
great interest, the unravelling of this new puzzle which
Mr. Cowell's preliminary work has introduced to us.

For the present it is enough to remark that although
this work lies in a region where partial eclipses are not
by any means useless, nevertheless even here the
superiority of total eclipses is manifest. It is one of the
points made by Mr. Cowell that the records are much
safer to interpret when there is some reference to the
appearance of the Corona, and we are able to infer that
the eclipse was total. In modern times, also, observations
of position are certainly rendered more complete on the
occasions of total eclipses, although it is not often that
advantage can be taken of this fact, chiefly on account of
the more pressing need for spectroscopic and other ob-
servations on the nature of the Corona. In 1887, however,
the late Otto Struve organised an extensive programme of
observations of this kind, hoping to measure very exactly
not only the relative places but the relative sizes of the
sun and moon. Unfortunately, cloudy weather prevented
observations of that eclipse, and not only Struve's assistants
but some forty other European Astronomers from all
nations returned home disappointed. In 1898 the
Astronomer Royal carried out in India a programme of

* Jl/ivi. Not. Roy. Astron. Sjc, vol. 66, p. 36.

Manchester Memoirs, Vol. I. (1906), No. 1. 5

observations for position, with the efficient aid of the Royal
Engineers, and last year, one of the prominent members
of this Society, Mr. Stromeyer, made some ingenious
suggestions for work of this kind at the eclipse which was
to take place on August 30th. He had in view not only
the determination of places of the sun and moon, but the
measurement of distances on the earth's surface, if the
observations could be made at more than one place. I
had the pleasure of discussing the possibilities with him,
and at one time I hoped that something might be done at
any rate to make the preliminary trial of his ideas, which
might encourage more complete application of them in the
future. But when the time arrived, and the difficulties of
work in Egypt were fully realised, it became clear that
any such attempt must be dropped unless we could afford
to sacrifice observations demanded by considerations of
continuity.

Total eclipses have, however, one great disadvantage —
their extreme rarity. The spot on the earth from which
the sun is entirely obscured from the moon is a very small
one. If we cover up the sun's disc with, say, a threepenny
bit held at arm's length, we can only cover it up for one eye ;
if we open the other eye the sun is not obscured from it,
as we become painfully aware. And so though one spot
on the earth may be favoured by a total eclipse, another
at a little distance is not. The favoured spot does not
remain stationary on the earth owing to the motion of the
moon between us and the sun, and also to the rotation of
the earth. It travels over the surface ; but even so the
track made is narrow compared with the wide area over
which the partial eclipse is visible. Moreover the track
has a wayward habit of wandering across regions which
are either inaccessible or occupied entirely by ocean ; and
although we could in the latter case enjoy the spectacle

6 Turner, Total Solar Eclipses.

from a ship, unless we can set up instruments firmly on
land the observations made are of little use. In 1904 an
eclipse track crossed the Pacific Ocean, skilfully avoiding
contact at any point with even one of the numerous islands
which seem to have been almost purposely scattered
about to entrap it. In January, 1908, a similar track will
meet with less success, for we learnt a few days ago that
although it will lie for nearly the whole of its length over
deep sea, it does cross two small islands, both uninhabited,
and both devoted to the growing of cocoa-nuts for the
purpose of making Sunlight soap. Some of us are hoping
that Messrs. Lever Bros, may see an opportunity of com-
bining business interests with scientific enterprise, by
sending out an expedition to observe the eclipse of the
sunlight on their property. It is nearly a quarter of a
century since a neighbouring island, Caroline Island, was
used for such a purpose, but meantime other journeys
almost as long, and to places nearly as inaccessible, have
been undertaken. In 1896 I had a most delightful expe-
dition to Japan, in company with the Astronomer Royal
and Major Hills. We saw many most interesting sights
on the way there and back, which lay across America.
For instance, we were present at almost the foundation of
the great Yerkes Observatory, near Chicago, which has
since become famous in astronomical history. We were
escorted to our station by a portion of Her late Majesty's
China Fleet, and dined on the Flagship with the Admiral;
and under the able and hospitable guidance of Sir Ernest
Satow we were able to admire the skill and charm of
that wonderful people who had just completed their war
with China, and have since shown themselves more than
worthy foes of one of the great European powers. But
unfortunately, delightful though the whole expedition was,
the central object of it failed. On the day of the eclipse

UliDickestcr Meiiiuirs, Vol. I. (1906), A'^. 1. 7

there was a dense fog which rendered all our long and
arduous preparations futile. In the same year another
party had what may be called an inverse experience.
After we had lefc England for Japan, and after other
observers had set out for Norway, Sir George Baden
Powell took in his private yacht Mr. Shackleton and the
late Mr. Stone to Nova Zembla. They narrowly escaped
shipwreck, running on a rock not marked on the chart, so
that for some days the ship's deck lay at an angle of 45° ;
when ultimately they got off and effected a landing, the
weather was atrocious and hampered their preparations ;
but the day of the eclipse was fine, and they got excellent
photographs, one of which (taken by Mr. Shackleton) of
what is now called the "flash-spectrum," marks a definite
epoch in such work. Sometimes Fate is cruel enough not
only to inflict hardship, but also to rob the observers of
the compensation of a fine eclipse. Last }'ear the Labrador
expeditions were particularly unfortunate in this way.
Mr. and Mrs. Maunder have told us how on arrival they
found their intended station occupied by some Indians,
who had caught measles, and wished to be near a white
doctor ; and how in order to reach the next clearing, nearly
a mile away, it was necessary (in default of any kind of
labour or horses) to wheel thither all their instruments in
wheel-barrows, during a protracted shower of rain lasting
some days ; and how, after they had braved all these
discomforts and hardships, they were rewarded by a hope-
lessly cloudy day for the eclipse. And we must not for-
get that these expeditions are not only liable to disappoint-
ment and fraught with hardship, but are sometimes
accompanied by positive danger. In 1889 Father Perry,
of Stonyhurst, who, although always a martyr to sea-
sickness, never shirked the longest expedition in the cause
of science, bravely started to occupy a station in a malarious

8 Turner, Total Solar Eclipses.

region near the French convict settlement at Cayenne.
Perhaps if it had been known beforehand how bad the
conditions really were the expedition might have been
abandoned, but this was not realised until too late. He fell
ill with dysentery, and although he successfully completed
the work for which he was sent out, he shortly afterwards
died at sea. He was ill on the actual day of the eclipse —
so ill that when the observations were completed, and he
called for three cheers for the success, he was unable to
join in them himself. " I can't cheer," he said, "but I will
wave my helmet." It is a memory worth preserving, —
that a brave swimmer caught by the cruel waters of
disease should go down waving his helmet for the success
which had cost him his life.

So great a disaster is fortunately rare in scientific
work ; but hardship must often be faced, and disappoint-
ment is always risked, on eclipse expeditions, which have
for their object the best utilisation of the few precious
moments during which the corona is visible. It might be
difficult to justify this enthusiasm from the cold, logical
standpoint of actual knowledge : we do not know that the
Corona is the most important part of the sun ; indeed from
what we know up to the present, it may be comparatively
unimportant. But the other extreme is also possible ; the
study of the Corona may give us an invaluable clue for
interpreting other solar phenomena, and this very uncer-
tainty is in itself attractive. In any case it scarcely needs
explanation that a phenomenon which can only be observed
for a few hours altogether in a whole century, should
attract more than its strictly logical share of attention.
At any one place a total eclipse lasts a few minutes only,
say two minutes on the average. Now there are twelve
"families" of total eclipses which recur in a regular cycle
every i8 years, ii| (or I2i, according to the occurrence

Manchester Memoirs, Vol. I. ( 1 906), No. 7. 9

of leap years) days. The one-third of a day is important,
because it moves the eclipse track on the earth's surface
just one-third round the earth at each recurrence ; so that
for the third recurrence (that is, after 54 years, i month,
and a few days) the track comes back nearly to the same
position. Nearly, but not quite, since all these figures are
approximate only, and not exact. The track is moved a
little north (or a little south) every 54 years, and ultimately
disappears from the earth at the north (or south) pole.
But to replace any family which dies in this way, another
one is born, so that the total number recurring in the 18
years is always about twelve. A new family is to be born
at the North Pole in 1909, which is of special interest to us,
because at its second recurrence (on June 29, 1927) the
track crosses the North of England ; and for the first
time since 1724 we shall be able to see a total eclipse
without leaving England. The track is a little unkind to
Manchester, which is left just outside ; so that it will be
necessary for this Society to make an eclipse expedition
to Liverpool. The duration of totality is also very short —
only 25 seconds — but the occasion can nevertheless be
recommended to everyone as one of remarkable interest,
and there is plenty of time for leisurely preparations.

The observations to be made at a total eclipse vary
greatly in character. We may direct attention to the
parts of the corona lying near the sun, or to the fainter
parts at a distance : we may study its form, and its
possible changes of form ; or its sj^ectrum, and variations
in spectrum ; or the polarisation of the light ; or we may
leave the corona altogether and look for planets close to
the sun, for seeing which a total eclipse affords a unique
opportunity just as for the corona. One modern feature
of all such observations may be emphasized — they are
nearly always photographic. It is easier and safer to

lO Turner, Total Solai^ Eclipses.

take proper photographs and study them at leisure than
to attempt visual observations in the limited time and
under the stress of excitement.

I shall not attempt to give even a brief description
of all these observations ; but shall, with your permission,
select one special point for consideration from a line of
investigation to which my own work at recent eclipses
has been directed. My special object will be to shew
how a quantitative measure helps us in studying natural
phenomena. The point is not a new one — it is as old as
science itself; but every new illustration of it comes with
a certain freshness. A beautiful illustration was given
nearly 30 years ago by Professor Schuster in almost
exactly the same domain as that of which I ain about to
speak. He attacked * the problem of the distribution of
particles round the sun which might, either by scattering
the sun's light, or by themselves becoming incandescent,
give rise to an appearance such as the corona, and wrote
thus :—

" Our problem is an inverse one, and seems at first
sight very hopeless. From the observed polarisation of
light we are to find out what part of it is due to scattering
particles, and, as will be seen, we cannot do this without
finding out at the same time in what way the scattering
particles are distributed round the Sun, and in what way
the light due to other causes varies with the distance
from the Sun. I began the calculation in the hope of
getting a rough idea only of the amount of polarisation
which we might expect. But it appeared that even such
observations as we can make during the short time avail-
able during a total solar eclipse may yield most important
inf(.M-mation as to the constitution of the solar corona.
I shall shew that combined measurements {a) of the

* Moil. Not. Riy. Astroii. Soc, vol. 40, p. 35.

Manchester Manoirs, Vol. I. (1906), No. 7. ii

polarisation at different distances from the Sun, and
(d) of the decrease in intensity of the total light of the
corona with increasing distance from the Sun will be
sufficient to determine all our unknown quantities. Even
if such measurements are incomplete, we may gain a
rough idea of these quantities, and even a solitary observa-
tion like that of Mr. Winter during the eclipse of 1871
will give some results."

In an elegant mathematical investigation, complete in
all details, Professor Schuster then proceeds to demon-
strate these propositions. But the measurements requisite
to utilise his results did not immediately follow, probably
because the difficulties of obtaining them by visual methods
were too great. The gradual introduction of photographic
methods, and especially the invention of the " dry-plate,"
has made them much easier, and at recent eclipses photo-
graphs have been obtained which will probably, when
suitably measured, give the required information. As
yet, however, measures of the polarisation (marked (a)
above) have not been carried out, and the complete prob-
lem formulated by Professor Schuster has not yet been
solved. But extensive measures of class (/'), the total
brightness of the corona at different distances from the
sun, were made on photographs taken in 1893* and i898-|*
(others have been made but are not yet published), and
an approximate law was deduced as follows :

The brightness of the Corona is inversely proportional
to the sixth power of the distance from the sun's centre.

Now, making use of Professor Schuster's paper above
referred to, we can immediately deduce from this that the
particles must be distributed according to the inverse 4^
power of the distance (see Note I.), and I propose now

* Proc. Roy. Soc, vol. 66, p. 403.
t Froc. Roy. Soc, vol. 68, p. 36.

12 Turner, Total Solar Eclipses.

to examine a little more in detail what further information
this gives us about them.

But to prevent misconception, I must recur to one or
two points which have been omitted so far in order to
simplify the statement. When we come to consider the
nature of the Corona, we must remember that it is certainly
a complex structure. The spectroscope gives us evidence
of the existence of gases which show bright lines in the
spectrum. One of these, at least, is a gas unknown to us
on earth, which has been called coronium. But besides
gases, there are in the Corona solid, or perhaps liquid,
particles ; for part of the light which comes to us is
polarised light, and from the character of the polarisation
we can infer the existence of particles of this kind. More-
over we learn something of their approximate size ; many
of them cannot be very much bigger than a wave length
of light, or there would be no such effect. Of course,
there maybe larger particles than these in the Corona, but
these will not polarise the light, and therefore do but
dilute the effect. If we find that the polarisation is
strong, this is evidence that the number of large particles
is small. We see here how important it is to get an
actual measure of the amount of polarisation, and measure-
ments of this kind are in progress. We do know, however,
that the polarisation is strong, and therefore the small
particles numerous, and for the present we are going to
consider them alone, as if they constituted the whole
Corona. We can afterwards make allowance for the
elements neglected.

But the particles are almost certainly not stationary
relatively to the sun ; they are either rising or falling, or
rising and falling : and it is into their state of motion
that I propose to enquire, in the light of the measure-
ment above quoted.

Manchester Memoirs, Vol. I. (1906), No. 1. 13

Let me first take perhaps the simplest case. We
have said that the particles must be small. Now Clerk-
Maxwell pointed out years ago how the radiation of light
would exert a pressure on bodies receiving the light,
which would be quite insensible for large bodies but
might become important for very small ones. This
remark has scarcely received proper attention until
recently, but in the last year or two Professor Poynting*
and others have stated in definite form the amount of
this light pressure, and shown that for bodies smaller
than two wave lengths of light the pressure may be so
great as to counterbalance the sun's gravitational attrac-
tion. If so, the particles may be continually repelled
from the sun, instead of attracted. Repulsion could, of
course, also follow from electrical action of the sun. For
our purpose these two hypotheses can be considered
together. Supposing, then, particles to be repelled from
the sun, how far would their distribution fit in with the
observed law of brightness ? To a certain extent, the
supposition looks promising ; starting from the surface,
where the particles may be densely packed, they are spread
over a larger and larger surface as they travel outwards.
Moreover, since they are travelling with ever-increasing
velocity, their density will further diminish owing to
this fact. Further, the light received by each diminishes
with increasing distance from the sun whether the light
be simply scattered sunlight or the incandescent light of
the particle itself, heated up as it must be by the intense
radiation from the sun : but this is taken account of in
Professor Schuster's paper and the deduction we made
from it, whereby we converted the law of brightness into
a law of density of particles, viz., as the inverse 4^ power

* Phil. Trans. ^ Series A, vol. 202, pp. 525-552; and J'roc. Physical
Society of London, vol. 19 ; also Phil, ^fag., April, 1905.

14 Turner, Total Solar Eclipses.

of the distance. Combining, then, for comparison with
this observed density the two former contributing causes —
greater surface, which goes as the square, and greater
velocity, which is as the square-root at most* — we get only
23^ instead of 4^ : and our su imposition does not fit the
observed facts. Hence, if particles travelling outwards
(under repulsive force) exist in the corona, they can only
dilute the effect we wish to explain, and we must look
for other particles moving in such a way that the index
is even greater than 4^, so that it may be reduced to 4j^
on the average : just as when, in climbing a hill of which
the average slope is known, if we find an easy gradient
for some distance, we know that we shall ultimately find
an unusually steep part somewhere.

Next let us take another supposition of an inverse
kind. Matter cannot be travelling continuously outwards.
Can it be, perhaps, travelling continually inwards from
space? This supposition is worse than the former ; the
condensation as we approach the sun being not even so
great as before. We still have the "concentration" due
to the decrease in surface of concentric spheres, but when
we come to the velocity with which a particle crosses any
sphere, it is now greater near the sun instead of less, as
before. Accordingly our power of the distance is no
longer even so large as 2\, but on the supposition of
simple falling, it would be only 1^. There is, however,
one new consideration which may be taken into account.
Is there anything resisting the falling ? We know, for
instance, that there is in the Corona a gaseous portion
made up of hydrogen, coronium, and perhaps other gases.
Do these check the action of the falling particles? When
the great eruption of Krakatoa took place twenty years
ago, a large amount of dust was flung sky high by the

* See Note I.

Manchester Memoirs, Vol. I. (1906), No. X 15

volcano, which took years to fall. Many of us remember
the beautiful sunset glows of the years near 1884, which
were probably due to the existence of this dust high up
in our atmosphere, and Professor Stokes showed how the
observed rate of fall of these particles was just such as
might be inferred from observations of the way in which
they scattered light. But if I understand his formula
rightly, the rate of fall would be nearly uniform at all
heights. It would not at any rate diminish as the particles
approach the earth at a rate so marked as the cube of the
distance, which is the kind of change we require. We
must, of course, be cautious in arguing from our own
atmosphere to that of the sun, but in default of positive
knowledge to the contrary, I put this supposition aside as
not helpful for the illustration of our present problem.
[It seems, however, just possible that the resistance of the
gaseous atmosphere, combined with a resultant force out-
zvards, might satisfy the conditions. This consideration
occurred to me too late for full examination.]

Let us, then, turn to some other supposition. Light
pressure or electrical repulsion must not overcome gravity,
but they may considerably reduce it. One result of this
reduction is that a much smaller velocity is necessary to
carry a particle to a given height from the sun. If gravity
were not so reduced, a velocity of 382 miles per second
would be necessary to eject a particle completely from
the sun so that it never returns, and a velocity not much
less is required to eject it to heights such as are repre-
sented in the corona. For instance, to reach a height of
one solar diameter the velocity must be 270 miles per
second. There is nothing impossible or even unlikely
in the existence of such velocities at the sun's surface
a priori, though presently we shall see some reason for
doubting their existence. But for the moment let us

l6 Turner, Total Solar Eclipses.

leave on one side the consideration of the absolute magni-
tude of these velocities, only remarking that a diminution
of the sun's attraction is for our purpose equivalent to a
diminution in the velocity of projection. We can substi-
tute one for the other and leave the path of the particle
practically unaltered ; so that we lose no generality by
asking how particles ejected with different velocities and
in different directions would be distributed, keeping their
size and the sun's attraction the same. For simplicity
let us restrict the problem further still and consider the
variation in magnitude and direction separately. Let us
first suppose a velocity given in magnitude but varying
in direction. Take, for instance, a velocity just sufficient
to project a particle vertically to the height of one radius
of the sun ; then the same velocity, with horizontal pro-
jection, would send the particle skimming round the sun
completely as a very close satellite ; while intermediate
directions would cause it to describe trajectories of
different heights ; all, however, less than one radius. If
particles be shot out from a point on the surface in all
directions with this velocity, there will certainly be more
of them at any moment near the surface than far away.
Beyond one radius there are none at all, and just within
that limit there are only the few projected nearly vertically.
Lower down we get a greater number from directions
more and more inclined to the vertical, and so as we
approach the sun the density increases. Close to the sun,
indeed, it becomes infinite ; for the particles which come
skimming along the surface occupy that region for a
finite time, whereas all other particles are at any particular
distance for an infinitesimal time only. Working out
the law of density mathematically, we find that it is not
unlike that required, excluding the little shell close to
the surface which may be regarded as part of the chromo-

Manchester Memoirs, Vol. I. (1906), No. 1. 17

sphere (see Note III.) But the early promise of this
supposition is not followed up. Firstly, it would only give
us a corona one radius deep and the observed corona is
much bigger than this ; accordingly we must increase
our possible velocity so that particles may rise several
radii from the sun, and on coming to this more general
case we find quite a different law of distribution. Let
us take, for example, a velocity which would carry a
particle to the height of three radii vertically. If particles
were spouted from a point on the sun's surface in all
directions with this velocity, the densiiiy at the surface of
the sun comes out infinite as before ; at a little distance
it is finite but decreasing very slowly. The decrease
ultimately stops, and then becomes an increase ; and we
get anotJier shell zvith infinite density, due to the fact that
in the neighbourhood of a point on the other side of the
sun from the eruptive centre there is a great accumulation
of particles in a part of their orbits where they remain
for some time at nearly the same solar level (see Note IV.)
Outside this shell the density falls off at the rate we
desire, but fails altogether at a height of three radii.
This supposition accordingly is totally at variance with
the observed facts and we are led to the conclusion that
variations in direction of projection apart from variations
in amount do not give us an adequate explanation of the
corona.

We must fall back on variations in magnitude of
velocity. But the work already done has not been thrown
away ; it has simplified the problem. We have seen that
with a given magnitude of velocity there is a certain shell
where the density becomes infinite compared with that
inside and outside, and hence in adding together the
results of velocities differing in magnitude, we need only
take account of this particular shell in each case, and we

l8 Turner, Total Solar Eclipses.

arrive at a conception of the corona as made up of a series
of shells one inside the other, corresponding to different
velocities. But this does not accord at all well with its
appearance ; it would give a stratified corona, whereas the
observed corona is distinctly radial in character.

For reasons which would be tedious to give at length,
the evidence seems to indicate that variations in direction
of velocity must be very small, and that we may assume
the direction to be approximately vertical. The corona
near the sun is formed from loiv vertical jets ; that at a
distance, from jets which reach a great height ; and since
the former is so much denser, there must accordingly be
many more velocities of small size than of large. The
law of degradation of brightness is accordingly now
become a law telling us how many more small velocities
there are than large ones, and we accordingly return to
what was originally said about the magnitude of these
velocities.

It may not really be the velocities themselves which
vary, but the force of the sun's attraction, as diminished
by light pressure. When the light pressure and attraction
nearly balance one another, a very slight change in the
adjustment wiil double the difference, and it is obvious
that we have here great possibilities for variation in mag-
nitude. When we introduce this variation into our
equations (see Note V.), we find a notable effect, some-
what of the kind required, on the distance to which the
particles are ejected. If we assume a series of attractive
forces, in the ratios lo, 9, 8, 7, 6, 5, then a velocity which
would, under force 10, eject a particle to the height of i
radius from the surface (or 2 from the centre — for com-
parison with our law it is better to measure from the
centre), would, with the forces 9, 8, 7, 6, eject it to 2'2, 27,
3"5, 6'0 radii from the centre ; while under force 5, the

Maitchestcr Memoirs, Vol. I. (1906), No. f. 19

particle would leave the sun altoi^ether. The scattering
of the particles is thus very rapid at distances greater
than two radii from the centre. But the same does not
hold within this distance. If we extend the series of
forces in the other direction by the numbers ii, 12, 13,
14, 15, the corresponding distances from the centre are
rSj, 171, 163, r56, I'SO : in other words, the inner
corona would be too nearly uniform, unless we have some
other source of variation. This we must seek in the
distribution of the sizes of the particles. We must suppose
that those which rise to considerable heights, because
light pressure nearly balances the sun's attraction, are
comparatively few, while there would be many for which
the balance was less complete.

We have, therefore, to a certain extent, only replaced
one difficulty by another — or rather, only one enquiry by
another. Instead of looking for a cause for a certain
distribution of density, we now seek the reason of a
certain distribution of size. But I think that we have
advanced a step, although we may not have completely
solved the problem. That there should be this variation
in size near the point where light pressure nearly balances
attraction is of the nature of a vera causa.

At the risk of appearing to argue in a circle, I now
call attention to the effect of this supposition on our
estimate of the absolute magnitudes of the velocities. If
we are right in regarding the variations of magnitude as
taking place in the force (and not in the velocity of
projection), then the average force is small, and the
velocities will also be small. This affords us a loophole
of escape from a possible difficulty. If particles were
-ejected with velocities of this magnitude, we might
reasonably expect the corona to change its form, at any
rate in detail, within a very short time. In one hour, for

20 Turner, Total Solar Eclipses.

instance, a particle would travel over a million miles —
more than a solar diameter ; and though a jet may retain
the same general appearance, though composed of entirely-
new matter, we might hope to detect differences of detail
in it. Good snapshots of Niagara, for instance, differ in
details, however quickly they may succeed each other.
Now pictures of the corona at an interval of one hour do
not differ in this way, and though we must not lay too
much stress on the evidence as yet, it affords a pre-
sumption in favour of lower velocities. Hence it is a
point gained that light-pressure enables us to look for
lower velocities of ejection. For instance, instead of the
382 miles per second necessary to eject a particle com-
pletely from the sun, one-tenth of this velocity only is-
required if the force be 100 times smaller, so that we may
now contemplate velocities of ten or twenty miles per
second, which would with such force raise a particle to a
height of several radii, as taking part in the phenomenon.
Now it is interesting to find that we have indications of
velocities of this size from a quite independent research,
namely, that on the Solar granules. The surface of the
sun in a telescope presents a mottled appearance. Forty
years ago, Mr. James Nasmyth, in a letter to a member
of this society — of which within a few months he became
a corresponding member — claimed that he had discovered
as an explanation of this mottled appearance, that there
were scattered over the sun's surface a number of objects
resembling " willow-leaves." Thereupon ensued a curious-
scientific controversy. Other observers saw the pheno-
menon indeed which Mr. Nasmyth intended to describe,,
but found fault with his description. Some preferred the
name "rice-grains" for the objects which they admitted
were there ; others spoke of bits of straw. Their various
ideas may be illustrated on the screen. Attention was

Manchester Memoirs, Vol. I. (1906), No. T. 21

after some years distracted from this controversy by the
work of the spectroscope, but within the last year our
interest has been reawakened in it by some very successful
photographs* taken by Mr. Hansky, of St. Petersburg,
who has not only photographed these granules, but
demonstrated their continuous existence as separate bodies
of some kind, moving about among one another. But to
identify them as individuals it is necessary to take photo-
graphs at intervals of less than one minute, for the move-
ments even in a few seconds are large. By measuring
the photographs, Mr. Hansky estimates that the velocities
of the granules are something like 10 to 20 miles per
second. If we may assume that there are in a vertical
direction velocities similar to these horizontal ones, we
get just the sort of velocity which would suit the pheno-
mena of the corona on the lines above indicated. I do
not lay, however, too much stress on this illustration, if
only for the reason that Mr. Hansky's discovery is too
recent. But I would add one word further of what may
be frankly called speculation in the direction of very low
velocities.

If for any reason it were to seem probable that veloci-
ties even smaller than these might play a part in the
phenomenon of the corona, velocities, let us say, of one
mile per second, then we must not forget to take into
account the possible effect of the sun's rotation. Sup-
posing, for a moment, that the particles owed their ejec-
tion to this rotation alone, we can see that they would be
chiefly flung off from the equator and the corona would
have the appearance of a comparatively flat extension,
which we see at a time of minimum sunspots. The
extension in other directions seen when sunspots are
numerous might then be due to the introduction of some

*Mitt. der Nikolai-Hauptstermvai te zu Pitl/cowo, vol. i, no. 6.

22 Turner, 7\nal Solar Eclipses.

disturbing cause in addition to the sun's rotation. But
for considerations of this kind to have any value, the
velocities of ejection must be of the same order as that
of a particle rotating at the sun's equator, which is about
one mile per second. And for this to be effective in
carrying it out to several radii from the sun, we must
suppose the attractive force diminished by light pressure
to the ioo,OOOth part of itself — a supposition which has
not much to recommend it.

To sum up, then, the result of this discussion : —
The observed falling off in light of the Corona must
be referred to the fact that light pressure nearly, but not
quite, balances gravity for the particles forming the
corona ; that the difference accordingly varies in magni-
tude ; that there are many more particles for which this
difference is large than for which it is small ; and that,
without any further considerable diversity in conditions,
particles ejected vertically from the sun's surface with
velocities similar to those observed in the granules might
then distribute themselves according to the observed law.
It must be remembered that we have not taken account
at all of the incandescence of the particles, which may
vary according to almost any law of distance. We have
started from the fact that the corona shows strong polar-
isation— in other words, that scattered light must play,
at any rate, a considerable part in the appearance — and
limited the enquiry to this part of the received light.

The discussion is therefore partial only, and reaches
no final conclusion. But it is, perhaps, the more character-
istic of eclipse work. We are only slowly spelling out,
with long intervals between each letter, the lessons to be
learnt from the Corona ; and even those we seem to have
mastered we must be ready to modify in the light of new
facts.

MancJiester Memoirs, Vol. I. (1906), No. 1. 23

NOTE I.

In Table II. of Professor Schuster's paper (yMon. Not..,
vol. 40, p. 55) the brightness of the stronger component at
different distances from the sun's centre is calculated for
different laws of distribution of particles. To get the whole
light /, -f /j we must combine Tables II. and III., forming

Let us form this quantity for 6 = 30°, i.e.., 2 radii from the
centre : and 6 = 90° or i radius (the sun's limb). The ratio of
the distances is thus 2 ; and the brightnesses observed are in the
ratio 2'' or 64. We have to find what law of density of particles
gives us this law of brightness. Writing down the brightnesses
for different laws of density from Professor Schuster's paper, we
have

Value of I^ + Ii

Law of density r°

At distance i, I„ + 1^ = yo

2, /„ + /i=r22
Ratio of brightnesses 2 "46
Ratio as power of 2 i'3
Power of 2 for density o
Difference +i"3

Thus if the density falls off as the wth power of the distance,
the brightness falls off about as the {;« + i7)th power, when tn
is near the value we require. And since we have observed

;;/ + I • 7 = 6

we find )n = 4"3.

But it is perhaps scarcely advisable to choose as one of the
points the actual limb of the sun, since the chromospheric
phenomena complicate matters here ; and indeed the measure-
ments of brightness were not made quite so close to the sun as
this. Let us rather take the pair of points for which (^ = 52°"5
and 0= 22'''5.

r -

r-*

r'"

i"54

I'20

I -02

0-131

0-0234

0 0046

iiS

51

222

3-6

57

77

2

4

6

+ 1-6

+ 1-7

+ 17

r -

r-^

r

•5969

•2753

•1428

•0573

•0058

•0007

10-4

47'4

204

3'2

5 3

7-3

2

4

6

I"2

I "3

I "3

24 Turner, Total Solar Eclipses.

The distances from the sun's centre are, in terms of the
sun's radius, the cosecants of these angles, or 1-26 and 2"6t, the
ratio of which is 2*07 (log = 0 "3 16). The above table is then
modified as follows : —

Value of lo + Ii-

Law of density r„

At distance T '26, /0 + /1 =20704
„ „ 2-6i,/„ + /i = -9191
Ratio of brightnesses =2-25
Ratio as power of 2*07 = I'l
Power of 2*07 for density = 0
Difference i"i

The difference 17 has been reduced to i "3 : and the law of
density approaches the inverse fifth power instead of the inverse
4th. For our purpose we may take the index as - 4'5.

NOTE II.

The velocity is given by the equation

2 C" 2

^\a r

and thus although v increases with r it is not expressible as a
power of r. But we can represent it approximately by a power
of r within a limited region. Let J^ be the Sun's radius. The
velocity is zero when r=2«; and if this occurs at the Sun's
surface 2a = R ; otherwise 2a<^R, and r is always greater than
2a. Let us compare the velocities at distances r and 2r. The
ratio v'Ji^ is (r- a)/(r- 2a) which tends to unity when r is
large compared with a. It is less than 2 unless r is less than
3a. Thus 7'Jz\ is less than 2- except within a possible thin
shell close to the Sun. This shell does not exist at all if R'^^a,
that is, if the square of the velocity of projection be greater than
fji/R: and its maximum thickness is R/2, when the particles
start from rest at the surface.

Manchester Memoirs^ Vol. I. (1906), No. 7- 25

NOTE III.
The velocity in a trajectory, with sun's centre as focus, is

given by the equation

\r a

At the sun's surface r = J^, 7j= Fsay

If this be just sufficient to carry a particle to height Ji above the

surface, when the direction of projection is vertical, then a = Ji

and ti = Ji VI

For any other direction of projection, making an angle 0 with

the vertical, let the direction of motion make an angle f with

the focal distance at any point. The velocity outwards from the

sun's centre, i.e., along the focal distance, is z^cosi^. We must

express this in terms of r and the initial quantities Fand 6.

The factor v is given above. As regards 0 we have, if / be the

perpendicular on the tangent from the sun's centre,

. ^ p h VR%\x\B

smri)= -= — = ,

r vr vr

h being the well-known constant.

Let us now consider the number of particles projected in
the direction 0. We may take them as proportional to 2n-sin0 . dQ.
Within the spherical shell r to r + dr there will be a number of
these proportional to dr/vcos(p ; and

V-

= —i{2rR-r-R-?,\n"d)

Hence for the density within the shell, whose volume is
47rrVr, we shall have

K^ sinddd

Vr- [R'co^'d - {r - Rff

26 Turner, Total Solar Eclipses.

where K is a constant depending on the number of particles
ejected in unit time. To get the total density we must integrate
this expression between limits. The lower limit is clearly 0 = o.
For the upper, we must include all directions of projection
providing particles which reach the shell ; some do not. The
direction for which a particle just reaches the shell is given by

r- R
cos Q = = cosf(, say

and for larger values of d than this the denominator of the
expression to be integrated becomes imaginary.
For the integration, put

R cos Q = {r - R) sec -^
- R sin Bdd = (r - R) sin \p sec'^\Ld\p
[R-cos-0 - (r - R-]i = {r- R) tan yh

and we get, since R cos a = {r — R),

K

A /

777,-log,, tan \ - +
VRr => V4 2

-los,ntan

VRr

where R cos a = r - R^ and A" is a new constant.

This becomes infinite at the sun's surface, i.e., when r=R:
which is otherwise obvious since a number of particles remain
at this distance for an indefinite time as close satellites. For
distances greater than this we can quickly calculate

log]„tan( - + -\ = L

say, as in Table I. If L varies at any point as r~", then

logZ + n\ogr= const.,

and hence differences of logZ, divided by differences of logr,
will give n, as in the last column of Table I. The average value
of n is about 3 up to distance i"8, where il rapidly increases.

Manchester Memoirs, Vol. I. {igo6), No.H. 27

Hence the density of particles, which varies as Ljr, or as i/r"+^
falls off about as the inverse fourth power of the distance up to
r= I '8, and then much more rapidly.

TABf.E I.

logiotan

rlR.

a.

c^:)

logZ..

Differ-

log rlR.

Differ-

Ratio.

ences.

ences.

IT

84°-2

i'2954

0-II23

■1153

-0414

•0378

3'i

\'2

7«-5

0-9932

9-9970

•0857

•0792

•0347

2 "5

I "3

72*6

0-8152

9'9ii3

-0788

■II39

•0322

2-5

I "4

66-4

0-6800

9-8325

-0752

•1461

•0300

2-5

1-5

6o*o

0-5719

97573

•0779

-1761

•0280

2-8

1-6

53"i

0-4780

9-6794

•0893

-2041

•0263

3'4

17

45'6

0-3892

9-590I

•1124

-2304

-0249

4-5

rS

36-9

0-3004

9'4777

•1712

•2553

■0235

7-3

1-9

25-8

0-2025

9-3065

oc

■2788

-0222

oc

2'0

O'O

O'OOOO

oc

•3010

NOTE IV.

If we do not take the special case where a = R, we arrive at
the integral

K C sin9/zfi

where

and is positive if

negative if

^J \{2a.

VrJ \{2uR- A''}cos'y -f Z\

Z={a — Rf - {r - ci)'

a - R^ r - a or r <^ 2a - R ;
r-^za-R.
We thus have two cases to deal with in the integration.

28 T U R N E K , 7"tf /rt/ Solar Eclipses.

Case I. r '^la - J?, Z= - ^- say.

This case bears the closer resemblance to that of Note 11.
The limit for 0 is determined by the vanishing of the denomi-
nator. Putting 2aR - FP' ^p"^, the denominator vanishes when

pcosd = ^.

Let a denote this value of 6. Then the integral becomes
A' r s'mddB

Vrj {p-cos~0 - q-)^

which is the same as that considered in Note II., with/ written
for i? and q for {r - R). The value is thus

^log,tan(^+^

Since q^ = {r- of -{a- R)- and vanishes when r=2a- R, the
density becomes infinite at this distance, as it did in Note II.
for r=R. Outside this distance the density falls off in a
manner somewhat similar to that already tabulated for Note II.
It is scarcely necessary to give tables, which would have to be
made for different values of a, since the study of Case II. shews
us that the original supposition will not fit the facts. But if
tables are required, perhaps the quickest way of getting one for
any value of a would be to utilise Table I., keeping the values
of a the same and calculating the corresponding values of r
from the equation

p cos a = q

R{2a - R) cos^a = (^ - R){r - 2a - R)

or

(r — a)" = a^ - R{2a - R) sin'-a.

For instance \{ a= 2R, we have

r/^=2 + (4-3sin2a)i
= 2 +.(l + 3 C0S-a)5.

Manchester Memoirs, Vol. I. (1906), No. 1. 29

Hence we should get a table as follows, knowing that the
first column of Table I. is 1 +cosa.

The ratio in the last column, increased by unity, indicates
the inverse power of the distance at which tiie density falls otf,
and we see that the power is much higher than before.

Table II.

Cos a.

(l+SCOS^a).

r//v =
sq. root + 2.

log.;-/ A'.

Difterences.

Difference
ofZ.

Hatio.

01

1-03

3-02

•480

•006

■I153

'9

0'2

fI2

3-06

•4S6

•oio

•0S57

S-6

o"3

1-27

3"i3

•496

•012

•0788

6-6

0-4

1-48

2,-22

•50S

•013

•0752

5-S

0-5

175

yz^

•521

•016

■0779

4-9

0-6

2-o8

3-44

■537

■016

•0893

5-6

07

2-47

3-57

■553

•016

"1124

7-0

08

292

371

•569

•017

•1712

100

0-9

ZAZ

3-85

•586

■016

oc

a

vo

4"oo

4"oo

■602

We now come to consider

Case II. When 2a - R^r'^R the integral becomes

where

jY^^ f sin Md

s'^^{a-Rf-{r-a)'^.

30 Turner, Total Solar Eclipses.

The limits are now

e = o to « = '^,

2

since all directions of projection are represented in the shell of
radius i: Put

p cos Q = s tan -i^
— p sin QdQ = s sec-^\pd\p
(/>- cos-0 + /-)i = i' sec -d/.

For limits, when 0 = o, J tan ip =/>. Let this value of ^p be
denoted by \.
When

TT
0 = -5 -i, = O.
2

Hence

-TIT- lo^„tan I - + -
f^rp °'' \4 2

a similar expression to those obtained previously and it is most
easily tabulated by taking cot \ as the argument. We have to
find r from the equation

j'-tan-A. =/"
or

(r - a)-- = {a- R)- - {2a R - R^)co\:^\.

Let us take the same particular value of a as in Case L, viz.,
a = 2R. Then

rjR = 2 + ( I - 3C0t"/\)j.

Hence tan'-A. is greater than 3, or \ lies between 60° and 90°.
The former table only contains 5 values of \ (or «) between
tliese limits, but they will suffice to illustrate the solution to be

Manchester Memoirs, Vol. /. (1906), No. 7- 31

expected, since we may give either sign 'o the radical, or what
is the same thing, the solution for {a - r) is the same as that
for {r- a).

Table III.

cotx.

I -3001^ X.

;-/A'.

log;/ A".

Differences.

Diff. log L.

Ratio.

•102

■969

2-985

•4749

■0073

-■"^^"j?.

-,58

•204

•875

2935

-4676

■0143

-•0857

-6-0

•313

•706

2-840

'4533

-0294

-•0788

-2-7

•437

•427

2-654

•4239

■1228

--0752

-06

•577

•000

2-000

-3010

-1720

+ -0752

+ 0-4

•437

•427

I '346

-1290

•0645

4- -0788

+ 1-2

'3^2,

•706

i-i6o

-0645

•0371

+ -0857

+ 23

•204

•875

1-065

-0274

-0210

+ ■1153

+ 5-5

•102

•969

1-015

-C064

Hence the density falls at first, being infinite at the surface ;
but when r=2R it ceases to decrease and begins to increase,
becoming infinite again when r ■= 2a - H or r= 3 A".

NOTE V.

The variation in height attained for different velocities
follows at once from the equation

V = p

For vertical projection, the maximum height, given by v = o, is
r=2a. Suppose this is one radius {R) from the surface, or

32 Turner, Total Solar Eclipses.

2R from the centre, so that a^ R: and that in this case yn = 10.
Then the velocity Kat the surface {r = R) is given by

2 I

For any other force /.t and the same initial velocity io/7? we
have

R ro

a ~ ^ A*'

Putting ju = 9, 8, &c., in succession, we get the results in the
text.

Manchester Memoirs, Vol. I. (1906), No. 8-

VIII. On the Range of Progressive Waves of Finite
Amplitude in Deep Water.

By R. F. Gwyther, M.A.

(Communicated by Professor H. Lamb, LL.D., D.Sc, F.R.S.)

Received and Read February 2jt]i, igo6.

The waves considered in this paper are those first,
and very fully, discussed by Stokes,* in his paper on the
" Theory of Oscillatory Waves," and in the " Supplement "
to that paper.

The object of the investigation is to confirm the
hypothesis that this class of wave is mathematically
capable of propagation with uniform velocity and without
change of form, and to determine the limit of the range
of such waves in height, and as far as possible, the change
in shape of the profile oi the wave as the ratio of the
height to the wave length is increased.

The method which I employ is that of tracing the
paths of the fluid particles. It is not supposed that the
knowledge of these paths is of interest ; but there seems
no more complete method of testing whether the motion
assumed can actually take place than one which shews
how the motions of the particles are to be coordinated
in order to adapt themselves to the circumstances of the
case. Since this investigation shews that the motions of
the individual particles can, within a certain range, be
traced to any desirable degree of accuracy, it follows
that this is a form of wave of finite amplitude which
represents an actually possible phenomenon.

* Mathematical and Physical Papers, vol I., p. 196, and Sitpplement, p. 314.

J!i[ay 31st, igo6.

2 GWYTHER, Range of Progressive Waves in Deep Water.

In order to obtain the expansions employed, I make
use of a solution of Lagrange's Equations of Fluid Motion
contained in a paper* read before this Society, of which
I reproduce a portion for the convenience of readers.

The expressions which I find for the motion of the
fluid particles bring out prominently the general character
of the motion, namely that, as Stokes discovered, the
fluid particles have a progressive motion in the direction
of the wave propagation which diminishes rapidly as the
depth of the particles increases. This is necessary in
order that the fluid particles may reconcile themselves to
the conditions of the permanent progression, and, failing
this, the propagation cannot be continued. Since this
progressive motion increases with the height of the waves,
it seems probable that this circumstance plays a consider-
able part in practically determining the limit of their
height.

The subject of the shape of the wave profile for waves
of different heights is of greater interest. In this section
of the paper it is shewn that if the co-ordinates of the
wave surface are expanded in a series of trigonometrical
terms, the coefficients of the successive terms diminish
very rapidly, the fall from the first to the second
coefficient being very remarkable. The results also
shew that the successive terms in the several coefficients do
not exhibit any marked tendency to converge. Hence,
although it is not desirable to take in many of the trigo-
nometrical terms, it may be desirable to have the co-
efficients of the terms retained worked out more fully-
than is here done.

The last, and perhaps the more interesting, part of
the paper deals with the mathematical limit of the height
of the wave and the shape to which the profile in this

* Manchester Memoirs, Vol. xliv. (1900), No. 10.

Manchester Memoirs, Vol. l. (1906), No. 8. 3

case tends. An attempt is made to shew (and I hope the
argument is sound) that all progressive waves in deep
water are of the class which possesses a horizontal series
of poles above the surface of the water, and that there is
no limit to the closeness to the water surface wh ich this
series of poles may assume. As this process is continued
the point of inflexion of the wave surface continually
approaches the crest, and in the final stage coincides with
it, and the wave profile shews a finite angle at the crest.

In this critical case, the wave-profile is the same as
that investigated by Mitchell * in his paper on " The
Highest Wave in Water."

The Solution of Lagrange's Equations.

The functional solution of Lagrange's equations for
irrotational motion of a fluid in two dimensions used in
this paper is found as follows.

Since the equations of condition may be written in
the form

■^(x + iy) ~. ix + iy)
^ M ^/ •^

= 0,

x-\-iy must be a function of x—iy and /, and in the case
of steady motion which I am here considering, a function
of x—iy only.
Writing

2^{x + iy) = (l>\x-iy),

we get

^'{x + ty)^fx + iy) = ^{x + iy)f'{x - iy)

= a real quantity,
* F/ii7. Mag., November, 1893.

4 GwYTHER, Range of Progressive Waves in Deep Water.

and therefore

^(.v + iy) = u + ib,

where u is real and b is independent of t.

If we suppose now that this relation can be reversed,

we may write

x-\-iy=f{u^ib) . .' . (i).

This is, of course, exactly comparable with the
Eulerian relation usually written

x^iy=f{<p + i-^).

Proceeding again with this relation, we have

d . (ill

-{x + ty)=^f{n-,tb)-.

This must be a function of x—iy, and therefore of u — ib,
and it can only take the form cj f\ii — ib), where c is an
absolute constant.

We therefore must have

f\u + ib)f' {u - ib)du = cdt
and

Jf{u + ib)f'{u-ib)dii = ct + a . . (2).

This relation gives, in any case, ti as a function of a and
b, and these symbols in the Lagrangian equations are now
introduced as constants of integration.

I now proceed to show that continuing to use the
Lagrangian method, the expression for the pressure takes
the same form as that derived from the Eulerian method.
This is solely a matter of form, since the result could not
be otherwise.

The equations with which we have to deal are

bail, ^^ -^^ -^ ' ) dci. ca -^ ha)

I'
We have

etc.

X- +y^ = cu ,

• ex -ly lu
X^r + Vtt- =Ct~

ca ca da

Manchester Memoirs, Vol. l. (1906J, No. 8. 5

• Ix • cy ?u ,

-^T^ +J'?7 = ^^7 5 SO that
CO -^ to do

c {p 1 • "> , ^ ^ '"i/A ^

ca(p *' - J ^//V W
with a precisely similar equation.

The condition that the motion shall be irrotational
requires that c shall be independent of a and b, which is
here satisfied since c is an absolute constant.
The pressure is given by

P 1 •

-xy + hcu = constant

P

or

~-gv+—fr, — . .,. .,, r^^ = constant . . (3).

p ^- 2f{u + i0)f{u-w) ' '

It follows from this that if x + (}'=/{(p -{- t\p) is a
solution of the Eulerian equations which can be made to
satisfy the conditions of a steady motion, then

X + iy =/ {u + ib)
where

J f {u + ib)f\u - ib)dii = c/ + a ,

is the solution of the Lagrangiari equations under the
same circumstances.

Application to Stokes' Problem.

The mode of treatment of waves of finite amplitude
in deep water employed by Sir George Stokes in his
" Supplement to a paper on the Theory of Oscillatory
Waves " * leaves the question in the form to which the
method of the last chapter is applicable.

Confining myself to the steady motion, from which
the progressive wave is to be obtained in the usual way, I
write, following Stokes,

X + iy =/(7i + ib) = u + ib - /v^^'»*("+'*) . . (4),

* Mathematical and Physical Papers, vol. i., p. 314.

6 GWYTHER, Range of PiOgressive Waves in Deep Water.

where w is an integer, and /^ is a quantity taken as
defining the wave length and is a constant.
From this it follows that

f\u + ib) = I + 2//„^-"'V"'" = I + :S^,«'"'" . (5),
where

H^^h^fi-""' . . • • (6).

Since the surface is, in the steady motion, the path of
a fluid particle, it is one of the lines for which /? = constant,
and we may most conveniently select this to be the line
from which b originates, so that <^ = 0 defines the free
surface of the fluid.

The condition necessary to secure the constancy of
pressure at the free surface is, from (3), that

2gy - c^lf\u + ib)f\it - ib)

shall be independent of u, when /; = 0.
This requires that

2g^-h,f:o's,nku + ^(^(i + 2/^,/'"'")-\i + 2;//„^-""'"j-^

71

shall be independent of ji.

The determination of the resulting relations to any
desired degree of approximation, after the manner of
Stokes, is merely a matter of labour. I have carried the
approximation to the 6th order, in order to show rather
more of the character of the final solution of the Lagran-
gian equations. Stating the results only, I obtain

^e = H^/lf (7)

Manchester Memoirs, Vol. I. (1906J, No. 8- 7

On comparing with Stokes' results {loc. cit., p. 318),
and allowing for the change in the form of the coefficients
which I emplo}-, the results will be found to agree as far
the 4th order.

If 2a stands for the height of the wave from trough to
crest, we get from (7) and (4)

The next stage in the process is to apply the condi-
tion (2) to express the necessary relation between ii, a, b.,
and /.

This requires

or

/[{ I + 2y¥;-} + 2 {iT; + 2/r„y7;,+i}cos ku

+ 2[H,,+ I,//„JI„+.,}cos 2ku + ...'\du^ct ^a.
This becomes

{ I + ^H;)ku + f {/^i + 2Zf„/7„+i}sin ku

+ %{H--v^H„H,,^.)%\x\ 2ku+ ... =k{ci + a).

For convenience, I shall divide by the coefficient of

ku, and write this

^ + ^jSin '^ + -£'^sin 2i^ + ... =^ . . (8),
where

(p = ku,

H = k{a + a)j{i +I.ff„%
E=2{H^ + 2^,^.+,}/(i . { I + ^H,}]),
E.^=2{H.^ + ^H„H'„+.,]I{2 . { I + Si7„2}), etc.

These expressions contain, though not in a very
convenient form, the connections which we are seeking,
and the solution is contained in (8) and the relations

kx = <h + 'L — ^sin«0,

H

ky = kb -^ -- cos n<h.

71

8 GWYTHER, Range of Progressive Waves in Deep Water.

The Reversal of the Series.

The next stage in the process is to shew how to
express ^, and finally x and y, explicitly in terms of ju,
and therefore in terms of t. The whole of the results thus
obtained as far as (20), are generally applicable in cases
of wave motion, and are independent of the special values
of the constants which are introduced at the next stage.

It will be noted that the relation (8) is of the same
character, though of course more complex, as that con-
necting the mean and the eccentric anomaly, and that it
is capable of reversal by the method which Bessel applied
to the astronomical problem. The same method may
also be applied to express x and y in terms of /.

The relation (8) is

^ + iS'jSin(/» + -£'2sin 2^ + ... =;U,

hence when 0 = ^Tr we have ^ = r-K, and we may properly

assume

0 = y^ + ^jSinju + ^„sin 2/i+ ... (9).

Differentiating this last equation with respect to ju,
multiplying throughout by cosr/^ (where r is any positive
integer), and then integrating with respect to ju between
the limits 0 and tt of ju, we remain with only the term
arising from sin r\i in the series in (7).

Accordingly we obtain

0

—COS rfidfi,
dfx

and since, as we have noted, 0 has the value rir when fx
has the value rw, it follows that we may change the
independent variable from fi to ^, and retain the limits
unchanged.

MnncJicster Memoirs, Vol. I. (1906), No. 8. 9

Hence

"^rA,. = / cos r^^(^
0

— / COS r(0 + iijSin0 + ^jSin 2^ 4- ...)r/<^ . (10).

The series for kx—<^ can also be similarly replaced by
a series in terms of sines of multiples of ^,
As before we are entitled to assume

kx - <^ = J/j sin n + A[„ sin 2^t + . . . (11).

Treating this equation as we have just treated (9), we
obtain

"^rM^ = / — i^-'v - <f) cos rfxdi.1
2 J du.

0 '

= / — (-^.v - (h) COS rud(h
J df

== / ^I/„t COS m(p cos r{(p + ^j sin (/< + ^., sin 2f... )d(p
/ ' . (12).

Finally the form properly to be assumed for k{y — b')

is

k{y - />)= - A^j cos iLi - iV„ cos 2^1- ... (13).

and, as before,

— {ky) sin rudu
dp.

= / — {ky) sin rj^idij)
J ^0

= / 277,,, sin w0 sin r(0 + ^j sin(^ + ^., sin 20 + ...)^0...
/ ' ■ • (14).

10 G\NYTiiER, Range of Progressive Waves in Deep Water.

The Expansion of the Coefficients in the
Series.

Although the formal expressions for x,y, and ti (or <p)
have now been obtained in terms of /.t and therefore of /,
it is still necessary to express these coefficients in ascend-
ing powers of H^ and /^, , before the series add anything
to our knowledge of the motion of the fluid in Stokes'
problem.

It is unlikely that this step can be performed by any
method without a considerable amount of labour. In
order to retain, in all cases, and as conveniently as may
be, expansions which proceed in sines and cosines of
multiples of 0, I adopt the method of expanding

cos /-(£", sin ^ + ^o sin 20 + ...),

and

smr{E^s\n f + E.,sm 2(p + ...),

by means of series in which the coefficients are Bessel's
Functions.

The relations which I continually employ are

cos (/\ sin ■^) = Jo{X) + 2E.7„,„(\) cos 2m\p
and

sin (X sin \p) = 22J^,,„_j(X) sin {2m - i)\p.

I shall not give the details of the work, but write
down the full expansions to the 6th degree of approxima-
tion. The only simplification which I make is to replace
Jo(A) by unity in cases where X' occurs to a sufficiently
high power to justify my doing so.

The expansions are

cos r[EiS\n f + E.2sln2<j) + ... +^Gsin 6^}
= JXrEy.,{rE.;)J,{rR) - 2J.lrEyirE.).

Manchester JMevioirs, Vol. i. {igo6), No. ^. ii

- 2J^{rE^){JlrE;) + J.XrE,) + J,{rE,)}cos <p
+ {2JJrE^)[JXrE.^ + J.^rE.^] - J^irEy^rE.^)

- J ^(rE.) J ^{?Ei)} cos 2(j>
+ 2J,{rE,){J,{rE.^ - J,{rE,)}cos s<p

+ 2\J,{rE;) + JXrEyirE.:) ^ J .{rE^J.irE,)

-Ji{rE^)]\ cos 40
+ 2[JJ,rEyi,rE,)+J^{rE.:)[JlrE;) + J^{rE.)]\ COS 50

+ 2 \J,{rE;) + JlrE.) + J.^rEyirE.^ + JlrE^^rE;^

+ J,{rEi)J^{rE,) + J,{rE,)J,(rE,)] cos 60 . . (15),
and

sinrljE*! sin 0 + ^2 sin 20 + ... + ^esin 60}

= 2{JSrEy,{^E^ + jpEyirE.;)} sin 0

+ 2{ J.(^^2)K(^^i) - Mr^i) - MrEy)J,{rE,)]

+ J.lrEyj^rEi)]%\n 20

+ 2{ j;(r^J - Jl^rEyirE.^) + .U.Ey^{rE.)] sin 30

+ 2{J^{rEXJ.^rE^) - J,{rEy{rE.)] + J,lrEy^{rE,) }sin 40

+ 2{ J^l^^i) + -/i(^^5) + J,{rE,).h{rE,)

+ JlrEypE:)]smsf

+ 2{ J-:(r^6) + ./3(^^2) + J,{rE,)[.h{rE,) + .7i(r^i),/i(r^3)]

+ J2(''^i)«A(''-£"4)} sin6(/i . . (16)

The coefficients A^, y?,... of the expansion of 0 in (8)
can now be written down. They are
A,= - 2J,{E„){J,{E,) + J.XE;) + J,{E^,)}
-2{J,{Ey,{E.) + J^Ey,{E,)},

2A,= 2{/,(2^,')[/„(2E.,) + Jl2E.^] - Jl2Eyi2E.^

-J,{2E,)J,{2E,)]

- 2{ Ji(2^,)[./„{2^l) - JI2E,) - J,{2Eyi2E,)]

+ J,{2Ey,{2E,)]

2,A.= 2J^{2,Ey,U^Eyj,{2>E:)\

- 2[Jl2>E,) + Jo{2,Ey.{zE,) - J,{zEyizE.y,

12 G\\'YTU.EY<., Range of Progressive Waves in Deep Water.

- 2\Jf,{sEi) + JiiS^i) + ^1(5^1)^2^5^2)

+ '>^.(5^x)'>^i(5^3)}
6^6 = 2 ; J6(6^i) + ^2(6^2) + J^{6E,)J^{6E,)

+ J,{6E,),M6E,) + J,{6E,)J,(6E,) + J,{6E,)J,{6E,))

- 2\Jr{6Ee) + M6E,) + J,{6h\)[J,{6Ei) + J,(6E,)J,{6E,)]
+ J,{6E,)J,{6E,)\ (17)

The coefficients Jl/^ and /V,. in (11) and (13) can best
be found in conjunction, thus

~{M,+ K) = J :^/J,SOs[{r-m)f + r{E,s\n<p+ ...)\di>...{iS)

and

Trr C

—{M,.-N,^= / 2://,„cos;(r + ;«)0 + r(£'isin^ + ...)j^0--(i9)

and the values of each integral can readily be written
down by means of (15) and (16).
1 thus obtain

JA + iVi= 2H,JXE,)JXE.^

^2HJ,{E,)[i-J,{E.;)\

+ 2B,{ME\) + J,{E.^],
M, - N, = 2H,\J,{E,) - J,{E,)J,{E,) - JXE,)J,{E.^]

+ 2B,\J,{E\)J,{E.^ - J,{E,) - J,{E.^] ,

2{A/, 4- A.^ = - 2B, [J,{2E.^[J,{2E,) + J,{2E,) + J,{2 E,)]

+ J,{2E,)JX2E.;} + J,{2E\)J,{2E,)}

^2H.Jl2E^JX2E.^

,2llM2E,)\l-Jl2E.^\
+ 2H,\J.l2K,) + J,{2E.^\

MancJiestcr Memoirs, Vol. I. (1906), No. 8- 13

2{M, - TVg = 2H,\Ji2h\\J{2K.^ + Jl2E.^ - .7,(2 E,)\

-J,{2E,)-J„{2E,)J,{2E,)\
+ 2HJyJ,{2E,) + J,{2E,) - J,{2E,)

+ J,{2E,)J,{2E,}-J,{2E,)J,(2E,)] ,

3(Af, + N,) = 2//,\J,{sE,) - J„{zE,).A{:,E,)

-2l/.J,{3E,)\i^.U2,E,)\
+ 2H^IJ,3K^

+ ^1(3 A\)Ji(3^-.) - ^i(3^.V.(3^i)} ,

4(.l/, + ;V4) = 2 H,; j;(4^,)[.A(4^.) + ^.(4 ^.) - ^1(4^4)]

-J,{^E,)-JU]^,)JME,)]

+ 2l/,\J,{4E,) - JME,).U4K,) - JME,)J,UE,)\
-2h,J,{4E,)[i+JME,)]
+ 2HJXAEy)
+ 2HaJi(4^i)

a{M, - ^\) = 2H,\J,{4E,)[J,UE,) - X(4^,)]
+ J,(4E.;)[J,{4E,) + J,(4E,)]

- JME,) - JI4E,) - Jl4E,)JME.:))

s{^f, + ^\)= 2H^\JisE,)~.UsE.)+J.lsE.:) + JlsEyisE.:)

-J.isEyisEd)

+ 2lIJ,J,{sEy^{5E.^ -J(sE,) - J^sEj]

+ 211,,

14 Gw'YTHER, Rang-e e>/ Progressive Waves in Deep Water.

6( J/e + .V,) = 2 H^\Ji6E,)[J,{6E,) - J,(6^J]
+ U6E„)[JleE,) + J,{6E:)-\

- J,[6E,) - J,{6E,) - J„_{6E,)J,{6E,)}
+ 2 H,{J,\6E,) + ^,(6^,) - ^4(6^4) + -^(6 E,)J,{6 E.^)

-J,{6E.;)J,{6E,)]

+ 2lQJ,{6E,)J,{6E.) - J.16E,) - JA6E.:)]
+ 2I/,{J,{uE,)-J,{6E;)}
-2H,J,{6E,)

M,-]S\ = 0 . (20).

The Numerical Values of the Coefficients in
THE Case of the Free Surface.

From the conditions (7) which we have found for the
free surface we may find, from (8), the corresponding
values of the E's (which I shall write in small type when
b = 0) in terms of /r^. I shall also omit the suffix from hi,
since this is now the only one of the set retained.

These values are found to be

e^= 2h-¥ 2//+ lo/f,

^6 = i^-^^« . . (21)

Also, we shall have

J^rey) = I - rVr -Ur^- -V^^ -(iir-r' + ^h\
JXre^ =1 -r'h'- ^/z«,

MaticJiester Memoirs^ Vol. I. ( 1 906), No. 8.

iS

Jlre,

J.,{re^
J-lre,

J. ire.
All of t

^K-I^''Ht-'H

.^6

'tr^'-S")'

5 '

=^(-(^3)-).

= 6-'''

=f:(--(F-)-)^

=77/^';

6r

(22).

hese have been carried to the 6th order

The result of the substitution in the formulae of the
last section gives

1 6 G\NYTll^R, Range of Progressive Waves zn Deep Water.

A.^

-\h'

^llh\

M =

-\h^

+ fl^^

N =

\h'

9 7 7,6 .

-3 6^^ ;

A,=

1 i5
-3 6^ '

M* =

1 i5
-2 4^ '

N.=

2 4'* J

A,=

2 4 0-^^'

M,=

T2F'^ '

N,=

1 JA

^,=j/,=.y,=^« = j/6=iv;=o . (23).

Indicating by the suffix 0 that the symbols refer to
the free surface, we have now obtained these values
correct to the proposed degree of approximation :

Jat^ = ^„ = ^„ - (3// 4- ih' + -jV/5) sin ^<, - {\h' - HM^) sin 2^,

- gV/i^ sin 3/z„ - ^io//'' sin 4^„ . . (24).

kx, = ix, - {h + 1// + ^th') sin ^, - (1 /^* - \yf) sin 2;.,

- fV/= sin 3^„ - ^V^/' sin 4^„ . . (25).

ky^ = - (/^ + yf + ?lh') cos ^„ - (1/^* - U^') cos 2;x„

- ^Vi^ cos 3^,, - T^^/z'' cos 4^/„ . . (26).
where, from (8),

^„ = k{ct + «)/( I + /r + 4/^' + '^^-h'),

We also have the relation, from the surface-conditions
in (7), that

and we can readily verify that the surface-condition, in the
form that 2gy^ — ai^ shall be constant, is satisfied to the
degree of our approximation.

The noteworthy features which the results just
obtained present to us are their striking simplicity as

MancJiester Memoirs, Vol. I. ( 1 906), No. 8. 1 7

compared with the cumbrous expressions from which they
have been derived, and also the rapid fall in the leading-
coefficients in the implicit equation to the wave-form
contained in (25) and (26) after the first trigonometrical
term, and the close approximation of that wave-form to
a trochoid.

The equations also indicate that for a closer know-
ledge of the form of finite waves, we need more terms in
the coefficients of the earlier trigonometrical terms rather
than a greater number of terms. On this account no
diagrams of the wave profile have been drawn.

The Wave-Profile.

Returning to equations (25) and (26), which give the
implicit equations of the wave-profile, we will now write

?-•

and

so that

^=i^/^'- + 4/^* + ^/A
\ 4

Then the equations may be written
x = ^ (ct + a)

Kg

-—(h + ^//' + —Jf ) sin ^{ct + a)

2iv\ 2 4 / Aq

27r\3 18 y Ao

A 15 ,, . 67r,

27r 72 A/ '

27r 00 A„

1 8 G\\YTll¥.K,Ra)ige of Progressive Waves in Deep Wafer.

y= -—(h + ~h^ + — /^= COS-- (ct + a)
■^ 27r\ 2 4 y A/ '

27rV6 36 / Xo ^ '

X T „ 67r, .

- — • — /r COS ^ let + a)

27r 24 A,/ '

- — • • — M^cos-r-ici + a). . . (27)

27r 120 A/ ' ^ '

Hence the horizontal distance apart of two particles
indicated by (a,o) and (a + \,o) at any instant is A, so that
in the motion the wave-length (not only at the surface
but throughout the motion) is A. At a depth d the particles
will be indicated by (a,l?) and (a + X/^,d'), where the value of
Aj is readily expressed from (8)

The motion of the particles is given in the next
section, but the general character of the motion is
illustrated by replacing A„ by A^ in the equations just
written and remembering that the coefficients will die
away exponentially. It will be noticed that A = A^ with
this notation.

We have for the height of the wave

27rV 2 24 y

which on reversal becomes

/l=z!!^- 3/^' Y _ I / 2 Tray
A 2V A / 24V A y '

and this might appear a convenient form for substitution
in the results just obtained. It can be shewn, however,
that this reversal is only justifiable for small values of /i,
and in fact for values of /i less than that value which will
be determined later as the limit of the range of /^.

The general solution of Lagrange's equations for the
motion of a particle within the fluid mass is not of great
importance, but, as the necessary calculations are pre-

MancJiester Memoirs, Vol. I. (1906), No. 8.

19

pared, in order to complete the solution, I shall state the
results.

With the object of making use of the simplifications
already found, I consider the actual motion as obtained
by means of a correction on a hypothetical motion dying
away from the surface motion by the exponential law
resulting from the substitution in the surface motion of
H(J.e., he'^'') in place of h, the correcting terms being of
the type H\Jfi-H% etc.

For example, from (7)

and I consider this as

2m + ^* + -^-W + i^H"' + ".^H'){Ii' - H'')

and similarly with the other quantities.

By this method the chief terms (using the word
" chief" not in a physical sense, but in reference only to
the analytical method) in the values of A, M, N, are
deduced from values derived from those of (23) by
substituting // in place of /i throughout.

The correcting terms are readily calculated from (17)
and (20), and are

A=-

-2,H\k^-IP),

M = ..

+ \H\h^-H%

N = ..

. + -m\h''-W),

A.,= ..

- H\K^ - H'-) - ^f-H\h^ - m) - -y-H%/i'- - H'^Y,

M = ..

. + \U\K^ - m) + s/Z^/z' - H"-) + \%H\]{- - m)\

N,= ..

+ hH\p - R') + ^-H\/i' - 7/2) + f^H\/? - try,

A=...

-lU\/f-H%

M = ..

+ j\HVr-ir),

^3=-

+ ^\H\/i-^-H%

A,= ..

-i^w^h-^-m),

M,= ..

+ J^H%/r-H-'),

N,= ..

. + ^H\k'' - W).

20 GWVTHER, Range of Progressive Waves in Deep Water.

From these we obtain, writing Jie~^'' for H, as the
equations corresponding to (24), ('25), and (26) : —

ku = ^i-\ ihe-"" + zh'e-'" + ir'iy-'""' + ^^^-''*'')1 sin ^

- \h\e-"^" - il-^-*') + ii'i^^-e-""' + if e--^^" - ^tf-«^'')|sin 2/i

+ sV^-tHs^"''' - 6^-"') sin 3;z - ^l^h\2oe-'"' - i y^""*") sin 4/i
At = ;x - J/^^-^* + |;r^-^^'' + h\^e- ■«•* + >/^^-^^*)! sin ^t

- {hWe--"' - le-'"} + y^«(f 1^--^* + ^e-'"' - W-^-"^*)] sin 2^

- ^\/i%6e-''"' - e-'"') sin 3^ - -^h^/i%ioe-''' - e'"''') sin 4^
^_y = Z'^ - \/ie-"' + ^/fe -"■' + /I'ip-''" + -V-^-«*")l cos fx

- {/I'iU-"-'" - y'''") + /iM'^e--'" + U-'" - -^''e-''')} cos 2^

- ^V^^(2^-^*" - e-''") cos 3,u - ^l^/iX^e-''" - 2e-""') cos 4ju.

(28).

From these we may evaluate 2gy — ai, to determine
the law of variation of the pressure downwards, and so
measure the error in Gerstner's assumption that the
pressure is uniform along each stream line. The
expression is, however, very cumbrous, as might be
expected.

The Approximate Determination of the Extreme
Value Permissible for h.

When the values of //,, Z/^,... from (7) are substituted
in f\u + ib), the successive terms in the expansion have
the general appearance of those of a divergent series. If,
however, attention is confined to the leading terms in the
coefficients in the series, it may be noticed that there is
a marked resemblance, in point of divergence, with those
of the Binomial expansion of (1-3//^"")"^ In order to
bring into prominence and to test the extent of this

Manchester Memoirs, Vol. I. ( 1 906), No. 8. 21

resemblance, I write /'(?^ + /^), when (^^0, in the form
i\>{ikii), and work out the value of {0(//-«)}~^ on the
assumption of convergence.

I thus find —
{'^{iku)\--'-
= I - 3/^d'^" - 3(/?^ + %V^'')t;''*" + {Vi - ^k'y^''

+ {h' - i/iy''"' + -V-// v^" + -V-/^''t?«''•" + . . . (29)

In this the reduction of the magnitude of the later
coefficients is very noteworthy. And, although a con-
tinuation of the series in the coefficients is desirable,
it would appear that under proper conditions the right-
hand side is rapidly convergent and that the series can
be properly reversed.

To examine the necessary conditions more fully, it
becomes manifest that the expression on the right-hand
side of (29) will have a factor of the form (i - ie"") where
^ is not widely different from 3//, and to which we may
approximate more closely.

Thus, when we write

we may determine a, /3 and y approximately so as to
satisfy the condition

,_3^_3(^^_+3r3 + =0

from (29), with some amount of accuracy. The results
will only give an approximation to the actual values,
by the nature of the method used.

The equations obtained are :

I - a + ^\a' + sVa* + ^^n^ + ^-f ^a" = 0,

r=-(.VaV + Tfl8«'') • • • (30).

22 GwYTllEK, Range of Progressive Waves in Deep Water.
These give approximately •

a= 1-054,

/3= - -068,

7= - •016 . . . (31),

and these lead to

^^=■3514 -'023^"- -0054' . . (32).

This value increases with increasing fractional values
of 4, and reaches the value '323 when 4 reaches the value
unity.

Since the hypothesis of convergency would certainly
fail when Z, exceeds unity, we may take this to give
approximately the upper limit of the range of values
permissible for //.

With these numbers we also find from (/j

— = i + -i23r+-0374^+-oi9;^ , . {t,^)-

and

ka = '35i<C+ •0414^ + -oigii;'^

where 2a is the height of the wave from trough to crest,
and

2a

Y= -1124 + -oi 34"'+ -0064', . (34).

if X is the distance from crest to crest of the wave.
This gives for the greatest value of this ratio

20
^-=•131

approximately, a value not very greatly in defect of that,
viz., "142, found by Mr. Mitchell for the pointed wave.
With the same numerical values I find

= (i - 'Ce'^'fli - (-018^- -0234'- -0054^'*"

- ('oi 84" - •0094"')^''*" - (-0114^ - •0034^)^^'^"

- -0064 V''^" - •0034^6'^'*" .... (35)

Manchester Memoirs^ Vol. I. (1906), No. 8. 23

If we introduce X, in place of Ji in the equations (27)
to the wave profile, and omit terms in which the coefficient
is less than 00 1, we get

x=-\ct'^a)

27r A„

- — (-0054:* --on l") sin ^{ct + a),
27r Ag

\ -> ''TT

}= - —(■35141'+ "0424^+ •oi84'^)cos-^^(^/ + a)
2 7r A„

- A(-oo34'* - -006^) COS 4!'(r/ + a),
27r \

where —"= i + •1234"+ •0454;*+ •0284''' . . (36).

\

From these it will be seen that although it is needless
to obtain further terms in the trigonometrical series, the
continuation of the coefficients of the first two trigono-
metrical terms would be necessary in order to draw
diagrams, except for moderate values of X,.

The Connection with Mr. Mitchell's Highest
Wave in Water.

The results of the last section suggest a change in the
method of treating the equation of surface condition in
order to determine the coefficients in the expansion of
the velocity more conveniently, and also to justify more
fully the convergency of the expressions obtained. It is

24 GWYTHER, Range of Progressive Waves in Deep Water.

also proposed to find a connection with the process
employed by Mr. Mitchell.

For this purpose, I differentiate with respect to //, the
equation of surface condition (3), and obtain

^\^{iku)-if{-iku^^^e^\^{iku)<^{-iku)\^ . . (37).

In the first case, I write

and obtain

fiiku) = [\P(iku)y\

miku)\i-[^-tku)\^

2 du

(38).

On rationalising the left-hand side of this equation,
the right-hand side will be found also to have become
rational. Thus

■i^'Jku) -\p(- ikii)

jiu

., .d

"§,' —

du

\^{iku)-^{- iku)\^
i\){iku) \p( - iku)

(39).

It would appear that a continuation of the terms in
(29) could be most readily obtained from the condition
in this form.

The stage most consonant with the final form of the
result will be reached by writing

^{lh^) = {l- 4V^")-V^1

(40).

and <p{ - iku) = {i - (^e- '"'")-' jV\

where U and V are rapidly converging series of the types

MancJiesier Memoirs, Vol. I. (1906), No. 8. 25

i + c/'" +<::/'■'■■" +...

and I + Cj(?"''" + C/"''^" + . . .

In this case the velocity {qc^"') is given by

and

,./._(i-4>-"")' r

In this case the point of inflexion in the surface
approaches the crest continually as Z, approaches, but
is finitely less than, unity,* and the character of the
singularity at the crest when ^— i is clearly indicated by
the formula.

We may now obtain the expression employed by
Mr. Mitchell. For

UV

= r{(i-^.-)*(i-C.-)j ^

2kK %\X\ku

may be written

w(.-y\V_/.-^.--\Vi

<lVi-{<-'V Vi -&■'■" 7 /
_^,|(,-y">!(.-^.-")i ^

\ 2 dii- '

* To a first approximation the position of the point of inflexion is given
by cos kit = ^.

26 GWYTHER, Ranoe of Progressive Waves in Deep IVater.

This becomes, in the special case when 4 = 1, identical,
except for the difference in units, with the equation
which Mr. Mitchell has employed to obtain the circum-
stances of the Highest Wave in Water.

We may, however, proceed to rationalise the ex-
pressions as in the previous case, and thus obtain

This may be used when 4 is less than unity to determine
the coefficients to the expansions of [/ and V, and also,
when 4=1, after division by (i - <r^''*")*(i - 4^-"")*, it may
be conveniently made use of for the same purpose, since
the series on both sides would proceed in cosines of
multiples of the angle ku/2.

By the theory now proposed, the form of the solution
IS the same whether for rounded or for the critical pointed
wave. In both cases the solution shows poles, which in
the case of critical pointed waves lie on the surface of the
fluid, but in the case of the rounded waves are situated
above the surface at a less or greater distance according
to the height of the waves.

It has not been shown that the equation of surface
condition has a solution of this special form, but merely
that the terms of the solution in series found by another
method submit to a great simplification when treated in a
special way.

This being the case, a conclusion in regard to waves
under other circumstances would not be justified, but it

Manchester Mcvwtrs, Vul. I. (1906), No. 8- 27

would seem not unlikely that the solutions in the case
of other irrotational waves are susceptible of similar
simplification.

The Tractical Limit of the Height of Waves.

In this paper I have not considered either the origin-
ation of the wave, nor its propagation into still water, but
only the circumstances under which uniform propagation
is possible ; and a differential motion of drift is one
such circumstance. The magnitude of this drift of the
fluid which diminishes from the surface downward is
worth a closer consideration, for it appears probable that
the practical limit of the height of deep water waves is
effectively determined by this magnitude and not by the
theoretical limit to the extreme shape of the wave.

When the steady motion with which we have started
is reduced to a progressive wave motion, the measure of
the rate of this drift at the water surface is, by (27).

or, on simplification,

[/e+^h'+r^-hy . . . (44).

If we give to // the extreme value "323 which we have
approximately assigned to it, this will turn out to be
about 'iSc, or nearly one-sixth of the velocity of the
wave. If we give h the value -25, the rate of drift falls to
•07^ about, that is to less than one-half of its value in the
theoretically extreme case.

After allowing for the fact that the series used in (44)

28 GWYTHER, Range nf Progressive Waves in Deep Water.

have not been proved to remain convergent for the range
of values given to h, it will still remain that the amount
of drift necessary increases very rapidly with the height
of the wave, and that this will be an important factor, if
not the predominant factor, in limiting the height of such
waves in nature.

The same conclusion will follow from the considera-
tion of the horizontal momentum of the fluid per
wave-length which will, to a first approximation, vary as
the square of the height of the wave, and, including the
further approximate terms, at a more rapid rate than this.

Manchester Memoirs, Vol. I. ( 1 906), No. 0,

IX. Observations on a Captive Mole {Talpa europcea).
By Lionel E. Adams, B.A.

Received March 22nd, igo6. Read April lo/h, /god.

The following extracts from my field note book may
be of interest, as they touch upon some of the many mole
problems yet unsolved : —

At noon, on December 15th, 1904, I found a mole on
a path in Reigate Park. The little beast was wandering
aimlessly about without attempting to burrow. I brought
him home in one of my leather gloves — not in my pocket
for reasons which previous experience had furnished —
and placed him in an empty sugar case with six inches of
earth banked up at one end. During the afternoon I
noticed him shivering as if with cold, and I provided him
with a large handful of hay rolled up to resemble a mole's
nest, placing it on the top of the earth in his box. He
soon came upon it, went inside and lay perfectly still ; I
believe he went to sleep. I fed him with worms at
intervals during the evening, and next morning found
him active and hearty, hurrying about in search of food.
That he had survived the night with only a few worms
left was doubtless due to the fact that he was enabled to
keep warm in the hay. I do not know whether it has
ever occurred to anyone to provide a nest for a captive
mole, but this is doubtless necessary to sustain the animal's
heat, especially when there is no great depth of soil for
burrowing ; and the constant failures to keep a mole alive
in captivity are due to the want of this quite as much as
to the alleged lack of sufficient food. Now, I kept this

May ^isf, jgo6.

2 Adams, Observations on a Captive Mole.

mole for eight days, and turned him out at last as hearty
and vigorous as when I first caught him, and this without
leaving nearly as much food as I have done on previous
occasions when the captives succumbed during the night.

I noticed one day, when the weather was somewhat
colder, that he made the nest more compact by pulling in
the hay from the inside. On the fourth day he became
very restless, so I changed all the earth in the box as
well as the hay, thinking that the dead worms and mice
with which I had supplied him had fouled his restricted
quarters, though I could detect no offensive odour attend-
ing him ; but he continued restless till I gave him his
liberty on the eighth day of his captivity, placing him in
a neighbouring field where he made many mounds, and I
finally lost sight of him when the field was ploughed.

Writing in the Ulster Echo, July 31st, 1903, Mr. M.
Woodward, the author of an article on the mole in
Pearsons Magazine for July of that year, quoting Mr. A.
Runciman, says : " [The mole] having caught his worm
. . . seizes it by the tail, off which he bites a small piece ;
then he turns the worm round. This is accomplished
with his paws, the sides of which, while their palms turn
towards the front, grip the worm firmly. Then the mole,
having turned the worm, draws it into his mouth with a
series of short quick jerks, at the same time moving his
paws slightly forward. And the effect of this movement
is to cause the earth to squirt out at the tail end, the
tip having been cut off purposely to give the earth free
vent. Thus the mole secures a clean meal, without any
distasteful clay. Evidently he knows that the tail is the
proper part of the worm to bite off, and that he must
begin feeding at the nose to effect his purpose."

Now, during the time that my captive was under close
observation — indeed I spent most of my time in his

Manchester Memoirs, Vol. I. (1906), No. 1). 3

ompany — I watched to see him adopt this method of
feeding, but in vain. I never once saw him bite off the
tail of a worm or turn it round. My mole after seizing a
worm with its mouth would hold it down with his paws
and feel the way with his snout to one end (as often the
head as the tail), and would eat the worm from that end
to the other in exactly the manner that Mr. Runciman
has described — by a series of short quick jerks. I have,
however, noticed now and then that the mole would brush
away the external earth from the worm with his snout
and paws before commencing to devour it. That the
mole is particular about the earthy contents of a worm is
doubtful, as I have as often as not found the stomachs of
moles full of earthy matter. Some moles, nevertheless,
may be more fastidious and scour their worms, but I have
never seen this done by any of the many moles I have fed
and watched.

On one occasion a large lobworm had burrowed nearly
out of sight, when the mole came upon it and seized it,
but instead of tugging at it furiously, as I had expected,
and thereby breaking it, he held it taut, and presently
yielding to the gentle tension, it was secured whole. This
knowledge of the fragile nature of a worm, to say nothing
of the ultimate consequences of haste and fracture, is
remarkable, and the self-restraint on the part of so
impetuous a creature is still more so.

On one occasion, when I had fed him till he could
eat no more, he took a worm, bit it with quick bites along
its whole length, and then crammed it into the earth, left
it, and turned about to find another. I gave him another,
a large lobworm, which he treated in precisely the same
manner, thrusting it into the same hole, which he straight-
way covered up by scraping the earth over it with his
fore paws. On two other occasions I watched him bury

4 Adams, Observations on a Captive Mole.

worms, and once a dead mouse, in the same way. I
had never previously seen moles bury worms, and it is
interesting, as it tends to confirm the old stories of moles
storing up worms for food.*

Worms are undoubtedly the favourite food, though
my captive ate slugs without hesitation [Auialia sozverbyi,
Arion hortensis, and Agnoliviax agrestis).

I offered him also freshly killed bank voles and long-
tailed field mice, but he would only tackle them when
worms failed, and then only when slit open. The heart,
lungs, and intestines were the only parts touched.
Respecting the mole's food, the Rev. A. Woodrufife-
Peacock writes : " I have known young birds held by the
feet by moles in the shallow runs, a blackbird and young
pheasants .... Within the last four winters my brother
and I found a hooded crow picked clean by a mole in the
middle of a meadow. The frozen snow and tracks to its
earth proved it to be a mole which had fed on the dead
bird."

1 he more I watch moles, the more convinced I am
that they are practically blind. When a worm is placed
near a mole the latter immediately shows signs of excite-
ment, being aware of its proximity by smell or hearing,
but it is only after poking about with its snout that it
strikes it haphazard. The mole never goes straight for
the worm, and when a half-eaten worm is dropt and has
crawled a little way, the same hunt for it is repeated.

I kept the captive in a greenhouse, and one evening
he escaped by climbing up the rough corner of his box

* Shortly after writing the above, I met a farmer on whose land I study
moles, and he informed me that a few days previously (March 30) while
digging out a fortress on one of his meadows he came upon a heap of dead
worms in the nest cavity close beside the nest. He described it as " three
spadefuls." This is most interesting, and is a peifectly reliable confirmation
of the stories of stored worms.

Manchester Memoirs, Vol. I. ( 1 906), No. 0. 5

and played havoc among the borders agahist the walls,
raising mounds and uprooting plants. I took a candle to
aid his recapture, and placed it on the ground just where
he was about to emerge. Upon coming to the surface he
paid no attention to the candle in front of him, and
blundered into the candlestick, but directly I stirred to
seize him he was off again. I took the candle and
watched him burrow and emerge in another place ; I
waved the candle in front of him, but he took no notice.
This experiment with a lighted candle I have repeatedly
tried with captive moles at different times, and invariably
with the same result, and I cannot avoid the conclusion
that moles are practically blind, for surely no animal
would be indifferent to such a startling and unfamiliar
phenomenon as a light waving suddenly close in front of
its eyes, unless, indeed, the light was imperceptible.
When I had caught and replaced him I fastened some wire
netting over his box. The next morning I found him
actually hanging to the wire netting with his front paws,
poking his nose through the meshes, and travelling from
one hole to another with his body hanging perpen-
dicularly. 1 regretted the light was not good enough to
obtain a photograph of him in this unique position.

It is astonishing how soon moles become tame or
rather indifferent to being handled. When first caught
the}' squeak and bite viciously, but within an hour I have
taken them up by the skin of the back without their
protesting. I have frequently stroked and tickled my
captives while they were engaged in eating, and I have
held them up by the tail while they lapped water. I
taught my last captive to come blundering along for food
when I scratched upon the earth or the side of his box.
This sudden tameness or indifference I have observed
with long-tailed field mice, short-tailed field mice, and

6 Adams, Observations on a Captive Mole.

young bank voles which I have come upon in hedge
banks ; by approaching them carefully I have been able
to tickle them with a stick, and even to take them in my
hand.

The folklore of the mole is well worth collecting.
The Rev. A. Talbot, of Church Eaton, near Stafford, once
told my friend, Mr. C. Oldham, and myself that three or
four years ago an old mole catcher in the neighbouring
village of High Onn assured him that an infallible cure
for the toothache was to catch a mole, cut off its fore-feet,
and then release the animal, which would carry away the
pain ! So long as the two feet were carried on the person,
immunity from toothache was assured. The old man
had a mole's fore-feet screwed up in paper in his pocket ;
these Mr. Talbot secured after his death. Mr. Oldham
sends me the following extract from the report of a
lecture by Dr. C. B. Plowright on Mediaeval Medicine.
"Among amulets in use in 1903 and 1904, a teething
amulet formed of a piece of violet root {Iris florentina)
and suspended by a tape to the babe, and a pair of
mole's hands."

Pennant in his " History of Quadrupeds," vol. 2, p. 484,
speaking of moles, says : " Pabna cJuisti and white
hellebore, made into a paste and laid in the holes, destroys
them."

Since my first paper on the present subject was
published * I have been in correspondence with Mr.
W. Evans, of Edinburgh, who has kindly presented me
with a copy of his " Mammalian Fauna of the Edinburgh
District" (Edinburgh, 1892), containing a most interesting
article on the Mole, with a plan of a fortress by the

*"A Contribution to our Knowledge of the yioXo. {Talpa europira)"
vol. 47, part II., of the Memoirs and Proceedings of the Manchester Literary
and Philosophical Society, Session 1902-1903.

Manchester Memoirs, Vol. I. (1906), No. t). 7

author, who, of course, forestalls me in exploding the old
legend of the fortresses being universally constructed as
in the time-honoured figure.

When I was writing my former paper, 1 read all the
literature on the subject that I could hear of, and I
regret that neither of the above publications came under
my notice. The only author that I was then aware of
who cast a doubt on the legend was the hero of our boy-
hood. Captain Mayne Reid, in his " Naturalist in Siluria"
(London, 1889), a delightful book of real observations,
which deserves to be better known.

Manchester Memoirs, Vol. I. (1906), No. 10.

X. A New Fern from the Coal Measures :

Tjibicaulis Sutclijfii spec. nov.

By Marie C. Stopes, D.Sc, Ph.D.,

Assistant Lecturer and Demonstrator of Botany in tlie
University of Manchester.

Received and read May Stii, igo6.

I. Introductory.

The " Age of Ferns " is the misleading name which
has been sometimes apph'ed to the period during which
the Coal Measures were deposited. This name, which no
longer represents the facts known to us, is now threatened
with the fate of most of man's surmises, and the " Age of
Pteridosperms " seems to be replacing it. Since the pub-
lication of the paper by Oliver and Scott (104) on Lageno-
stoma, the seed of Lyginodendron , recent work on similar
lines has rapidly accumulated numbers of important facts,
and, as Scott (:05) demonstrated in the Wilde Lecture,
we must conclude that many of the paleozoic " ferns "
were not true pteridophytes. The great phylogenetic
importance of these discoveries has led to an acces-
sion of interest in the plants of this period, which
are of such assistance to botanists in filling the gaps
in their knov/ledge of the " Natural System " aimed at
since the time of Linnaeus. The discovery of seeds on
fern-like plants has also increased our curiosity regarding
the actual ferns of the same period ; but strange to say,
in this ' Age of Ferns ' the number of such plants with
their structure preserved so that anatomical study is
possible, is exceedingly limited. Though scattered frag-
ments are of frequent occurrence, ferns are all too rare

June 2gth, igo6.

2 StopEvS, a New Fern from the Coal Measures.

which show both axis and branches or petioles associated
so that a fairly complete idea of the plant is obtainable.

The discovery, therefore, in the Bullion mine at Shore,
of a new fern with its main axis surrounded by numbers
of petioles and roots, and of such a size that a series of
sections could be cut from it, was an exceedingly welcome
one. I am extremely indebted to W. H. Sutcliffe, Esq.,
F.G.S., of Shore, the owner of the mine, for his great
generosity in placing in my hands the whole series of
sections of this unique specimen for description in the
interest of science. I have named it Tubicaulis Sutcliffii,
in honour of the man who has done so much on many
occasions for the cause of Pala^obotany.

The plant is not only new to Britain, but so far as I
am aware is the first of its genus to be found in the Coal
Measures. The single other species of this rare genus
has been found but once, and that in 1815, in the Middle
Permian of the continent.

2. General Structure.

The specimen was about 4^ inches in length, and,
including all the petioles preserved together, about 2x4^
inches in diameter. From this single block Mr. Lomax,
of Bolton, succeeded in cutting about 60 sections in trans-
verse and longitudinal directions, leaving a block uncut of
about an inch in thickness. It seems to have been the
upper end of the plant ; and most of the transverse
sections show the central main axis relatively small,
surrounded by a large number of petioles of varying size,
those just coming off being smaller than the main axis,
and those further removed much exceeding it. Between
these larger branches are innumerable small ones, of which
the bulk are adventitious roots (see Fig. 2). Towards
the upper end of the plant the differentiation of the tissues

MancJiester Memoirs, Vol. I. (1906), No. 10. 3

becomes much less perfect, and the number of the petioles
less, till finally in the last few sections the tissues of the
main axis and neighbouring petioles appear to be extremely
immature, and must lie near the growing point.

The general structure and appearance of the plant is
that of a herbaceous fern. The distinctive character of
the numerous petioles, which come off from the main
axis as small branches and increase in size as they leave
it, is only found in ferns that live in the ground. The
general arrangement is like that of an Aspidiiim or
Athyrium among living ferns, or an Osmtindites among
fossils. The rapid increase in size on the part of the
petioles till they greatly exceed the main axis in diameter
separates it further from most ferns, though this is also
characteristic of some species oi Athyrium. The form of
the meristele, however, differs from any of these, as its
simple horseshoe curve is orientated with the convex side
toward the main axis, a feature of great rarity, which
occurs only in Asterochlaena dulbia, possibly in Anacho-
ropteris, and in Tubicaulis Solenites.

The petioles apparently come off from the axis
equally on all sides, and bend slowly away from it. The
main axis itself appears to have been nearly vertical, but
to have bent slightly. Large numbers of roots coming
off from the axis and petioles branch there, and run in a
most irregular fashion. One is led to believe that the
plant was probably growing underground, the main axis
being a nearly vertically running rhizome. The enormous
number and the size of the petioles in comparison with that
of the stem indicate that in the air the axis could not
well have been self supporting, while their close arrange-
ment and very short internodes preclude the idea of a
climbing plant. The leaf bases apparently persisted for
a long time, as there are many in the lowest section

4 Stopes, a Nezv Fern from the Coal Measures.

(which is furthest from the growing region), most of which
must have been attached some distance further back
along the rhizome beyond the fragment now in our
hands.

The internal anatomy is simple for so large a plant,
there being but a single solid stele in the main axis, and
a simple band of vascular tissue in the petiole, in neither
are there definite sclerised bands, though the outer cortex
is somewhat thickened.

The preservation throughout is of that curious type
well known to palsobotanists as " roof preservation " (see
p. 17), which affords local patches of tissue exquisitely
fossilised with every detail retained, and other parts
entirely wanting. From this cause, the work of recon-
structing the plant has been extremely difficult, for in one
section the axis bundle or whatever is being studied may
be well preserved, while in the neighbouring sections of
the series no trace of it may be seen. This partly accounts
for the many instances in which it has been impossible to
speak quite definitely about a point under discussion.
Such difficulties are always encountered to a certain
extent in fossil work, but in " roof nodules " of this type
they are more serious than usual.

3. The Main Axis.

Although it is a little difficult to determine the exact
limit of the cortex, the diameter of the main axis appears
to be just short of one centimetre. Its external form is
approximately cylindrical, broken at frequent intervals
by the off-coming petioles. The external surface was
apparently smooth, but the petioles are so closely arranged
that but little of it would be exposed.

The stele is a single nearly circular one, of the
simplest monostelic type. The xy/on mass is absolutely

Manchester Memoirs, Vol. I. (1906), No. 10. 5

uniform and solid, and about 2 mm. across (see Figs. 3
and 4). The position of the protoxylems is not absolutely
determined, partly owing to the preservation, partly to the
fact that they appear not to have been very markedly
differentiated from the rest of the tracheides. In trans-
verse sections certain rather indefinite groups of smaller
elements are found near to or at the edge of the stele,
and from the uniform appearance of the central portions
one may safely say that the bulk of the wood, if not all,
was centripetally developed (see Fig. 3). In some cases
such groups appear oblique, in transverse sections, and
they show simple scalariform markings. A number of
longitudinal sections were specially prepared, but in none
did the preservation justify the trouble. Spiral and annular
markings appear to be absent (or never preserved), not
only in the axis, but also in both petioles and roots. This
is not uncommon in fossil ferns, and may be a sign of
slow growth ; the small scalariform tracheides may well
be the actual protoxylem elements. The tracheides of the
main mass are on the whole very uniform, about 015 mm.
in diameter, their corners fitting closely in together.
The wood mass appears to be absolutely solid, as in none
of the sections can I find any suggestion of soft inner
tissue (as in Zygopteris, Racliiopteris corrugata, etc.) or of
xylem parenchyma. The form of the tracheides is fern-
like, many being simply scalariform, with one or two rows
of elongated pits ; the larger ones, however, show series
of small round pits. The larger tracheides of this type
show 6 — 8 rows of these pits {cf. Fig. 7), and are not
paralleled, so far as I am aware, in any recent fern, though
frequent in the Botryopterideae.

The phloem is not preserved in most of the sections,
but obviously it would have lain in the space surrounding
the xylem. In two sections in which the axis has locally

6 Stopes, a Neiv Fern from the Coal Measures.

undergone "carbonisation " prior to its fossilisation,a zone
of smallish thin walled cells is to be seen just outside the
xylem. Of these, one or two rows have much the form of
the rather large celled metaphloem, frequent in living
ferns {sqq Fig. 5), and it is likely that they are the phloem
in this case. Pericycle and endodermis are not preserved
in any determinable form, but the presumption is natural
that they surrounded the vascular elements as in the
normal fern type.

The cortex is in general less well preserved than that
of the petioles ; there is a large space between the stele
and those layers of the cortex which are preserved, which
may represent a more delicate inner zone. The cortical
layers present are composed of closely -fitting rounded
cells with rather thickened walls and no air spaces. I
have not been able to observe the epidermis.

There seems to be nothing peculiar or even character-
istic about the outer zone of the cortex which might lead
to the correlation of this plant with any other impressions
or fragments, as neither zones of thickening, hairs nor
emergencies of any sort are to be found on it.

Towards the upper end of the specimen the tissues
become much less clearly differentiated, and finally, though
still at least an inch from the actual apex, no xylem is
recognisable. The other tissues are less thick-walled and
the cells contain more small fragments which may repre-
sent cell contents, while many of them contain blackened
contents and others numbers of small roundish bodies
which may be starch grains (see p. 10 and compare i^z^. 9).
These appearances, combined with the smaller size and
number of the petioles surrounding the main axis (these
also appearing undifferentiated), lead to the conclusion
that the apical growing point is approaching. The series
of sections stops here because the preservation of the cen-

Manchester Memoirs, Vol. I. (1906), No. 10. 7

tral parts is so poor that there seemed no prospect of
being able to see the actual growing point. There is a
small block remaining, however, in which we can feel
confident that the apex existed whether it is now preserved
or not. The sections of this region reveal the fact that
the parts attained a considerable size prior to the
differentiation of the xylem, but they threw no light on
the question of the position of the protoxylem or of the
direction of differentiation of the metaxylem.

4. Branching.

The main axis, or rhizome, runs the length of the
specimen without branching directly, or giving off any
structure of the same order as itself Surrounding the
main axis are the numerous petioles which it gives off in
rapid succession. In each transverse section there are
several petioles closely associated with it, though it is
impossible to be sure whether or not they have entirely
separated from the main cortex, as the outer layers of
both axis and petioles are wanting, of these, three are
generally undoubtedly within the main axis cortex. At
the time when the two cortices are certainly distinct the
diameter of the petiole is slightly less than that of the
main axis, and about one quarter that of the largest
petiole preserved.

The pJiyllotaxy appears to be exceedingly near to, if
not actually f, and the spiral is a close one. The reasons
that there is any doubt about the matter are, (i) the
rather uncertain preservation, (2) the fact that both axis
and petioles bend, though but slightly, independently and
at different angles. The plant with which it is natural to
compare this, viz., Tubicaulis Solenites, is described by
Stenzel ('89) as having a spiral of ]^ further out, and ^t
or y\ near the axis, but in the case of my specimen, to

8 Stopes, a Neiv Fern from the Coal Measures.

choose any such complicated fraction seems more than'
the conditions justify. In living plants it frequently
happens that there is a slight twisting of the line of
insertion which makes it difficult to determine the phyllo-
taxy from a small piece of the plant. In the case of a
fossil, therefore, in which has been detected a slight'
bending of the axis, and of which there is at best but a
few internodes from a single specimen, 1 prefer to choose
the simplest common fraction which comes out nearly
right, and therefore provisionally state the phyllotaxy to
be |. (See Text-fig. i.)

Fig. I. — Plan of the petioles in a transverse section, shewing the
phyllotaxy to be |. {Slide T 13.)

The branching of the meristele from the main axis
is very simple, a group of tracheides (apparently including
one or two of the supposed protoxylem elements) separates
from the outer part of the main mass, and bends away
from it (see Figs. 3 and 4). In transverse section this
group appears oval or bean-shaped, gradually taking the
form of a thick slightly bent arc as it leaves the stele.
The convex side of the meristele is from the first directed
towards the main axis. As it passes out the crescent form
becomes more accentuated, and the ends begin to bend
slightly round to form the C of the free petiole. In the

Manchester Memoirs, Vol. /. (1906), No. 10. 9

larger petioles the stele becomes very narrow, the C being
somewhat laterally elongated and the ends more incurved,
its orientation as regards the main axis, however, remains
constant.

In many cases, as the simple group of tracheides is
leaving the stele to pass out to the petiole, a smaller strand
separates from the main axis and bends away from the
stele to one side or other of the petiole strand. It runs out
through the cortex and forms what I take to be an
adventitious root (see Figs. 3 and 4). It is soon lost'
among the crowds of other rootlets which bend and
branch among the bases of the petioles.

Of anything corresponding to a true " axillary shoot",
such as is found in Zygopteris and HyniefiopJiylliim, there
appears to be no trace.

How soon the main petioles branched to secondary
ones it is not very easy to say, certainly it could not have
been very soon after leaving the main axis, as I can find
no case of the petiole stele branching to give off pinnules,
and in two cases only can I find branch steles within the
big petiole cortex. In one case this is in a fragment of
what must have been a very large petiole, lying in the
matrix remote from the main axis near the upper part of
the plant. This fragment is very well preserved, though
it is not repeated in the neighbouring two sections of the
series, so that it gives no information as regards the
method of branching. (See Slide T. 19.) The cortex is
in part beautifully preserved, and lying in it a few cell
rows from the outside to the left of the main stele is the
small branch bundle. It consists of a rather scattered
xylem group with some delicate soft tissue preserved, the
phloem apparently lies almost entirely to the outside, and
the whole is surrounded bj' a definite sheath of delicate
small cells, most of which are not well preserved, but

lO Stopes, a New Fern from the Coal Measures.

some appear very clear and like endodermis. The second
case is in the large petiole ao (see Fig. i and Slide T. 5) in
the cortex of which lie two branches both far out. The
neighbouring slides do not repeat them, so that little is
gained, beyond the fact that when the petioles get some
distance from the axis they do branch. Petiole ao {Fig. i)
is the largest and furthest out of the series.

The petioles give off a number of adventitious roots,
though none shew the stelar branching with any degree
of clearness. Among these numerous roots several small
branches similar to them but of rather different type, are
to be seen. These have rather more massive wood than
the ordinary root, and have three or four protoxylem
groups instead oi the typical diarch stele. They may
be small ramifications of the petioles, but there is no
direct evidence that they are so. One only shews a well
marked crescent stele {Slide T. 2), and this lies near a big
petiole of which it may be a ramification, they are in the
proportion of 2 mm. to 28 mm. in diameter, and their
steles are orientated back to back, but there is no direct
proof that the small one is a branch of this petiole in
particular ; though it is evidence that some of them must
branch. Among the two or three small branchlets about
the size of roots, but of uncertain nature, are two in which
the steles have divided, each into a smaller and a larger
portion (see Slide T. 25) but the series does not reveal
what happens to the branches. We have, therefore, no
clear case of branching among the petioles ; a point in
which this plant differs from the description of T. solenites,
where, according to Stenzel the petioles branched re-
peatedly giving off small pinnules while they were still
exceedingly near the main axis. He remarks on this as
being unusual (p. 45), in the words that " ungestielte,
gefiederte Blatter mochten bei lebenden Farnen kaum

Manchester Meinoiis, Vo/. I. (1906), No. 10. 11

vorkommen." Petioles so large as those of T. Sutclijfii
would be likely to remain unbranched for some distance,
and there is sufficient indication that in those further re-
moved from the axis branching took place, to conclude
that ultimately they may have ramified freely.

5. The Petioles.

The separation of the petioles from the main axis has
just been described, as have the slight changes in shape
undergone by the meristele during its passage out from
the main stele. The individual petioles are approximately
circular or oval in outline, with apparently no expanded
portion or side wing at the base of attachment ; their
rather irregular outline at times appears to be the result
of fossilisation. In some cases both stele and cortex are
wanting, and there is only a large clear calcite mass about
the shape and size of the petiole to represent it.

The meristele of a medium-sized normal petiole in
most cases shews only the xylem in a good condition ;
between it and the preserved cortex is generally a gap
larger on one side than the other, which probably repre-
sents the phloem, but is also partly due to the tearing
away of the stele from the cortex, which is so common in
ferns. In transverse sections \hQ. protoxyleni groups stand
out conspicuously, and are chiefly on the outer edge of
the wood and on its convex side (see Fig. 8). In longi-
tudinal sections, although many tracheides are of the
simple scalariformtype,others arepitted like those described
for the main axis (see p. 5). In longitudinal sections a
few cells are suggestive of spiral protoxylems, though it
is doubtful whether they are so in reality ; there are
also a few cells which look like sieve tubes, but no trace
of actual sieve plates is to be found. Several sections
shew a zone composed of 3 or 4 rows of small thin-walled

12 Stopes, a Neiv Fern froui the Coal Measures.

cells just outside the xylem ; these are probably the
phloem^ and they appear to surround the xylem on both
sides. In some places, between them and the cortex is a
darker streak of crushed cells, which probably represent
the endoderniis and pericycle in their normal position.

Se\'eral of the small branchlets, the nature of which
is uncertain but which are possibly fine ramifications of
the rachis (see p. lo), shew an extremely simple stele, the
xylem, which alone is preserved, being a single solid mass,
sometimes with 3 groups of proto.xylem on the same side,
like a very simple case of the form once called R. triden-
tata. In its simplest terms, this type of stele appears in
the small branches of several ferns, and seems therefore
to be of little value as a criterion of specific affinity.

In the small petioles {i.e. those which have recently
left the main axis) the cortex appears to be a simple tissue
of compact, probably rather thickened cells. In the larger
ones, however, there seems to be a distinct differentia-
tion into two zones, an inner one composed of thin-walled
tissue which in tangential section has almost the appear-
ance of" spongy parenchyma," and an outer one of rather
thick-walled cells i^Pl. 2, Fig. 2). Within the soft cells of
the inner zone in many of the petioles are to be seen
numbers of small circular or oval bodies, which have the
appearance of actual cell contents (see PI. 3, Fig. 9). In
some parts they are exceedingly crowded, and bear a
strong superficial resemblance to starch grains. Their
likeness to fungal spores, as preserved so frequently in
fossils, is also marked, but I incline to the view that they
are starch rather than fungi, on the following grounds :
Each little separate body is apparently unconnected with
its neighbours by any thread or filament, as it would be
if it were a fungal sporangium ; there are further no
obvious fungal hyphae which one would expect in con-

Manchester Memoirs, Vol. I. (1906), No. 10. 13

siderable numbers were the enormous masses of these
grains of fungal nature ; the tissues in which they are best
found are well preserved, and shew no sign of destruction
as might have been expected from masses of invading
fungi ; the grains are only present in the soft tissue, and
not in sclerenchyma and stele also, as one might have
expected had they been fungi ; their size, shape, and
general structure are exceedingl}' like recent storage starch
in ferns ; they are more numerous and conspicuous just
in the positions one would expect were they starch in the
rhizome and petioles of a recent fern, and appear in great
numbers toward the apical part. Further, they are too
deep seated to be chlorophyll grains, and also other indi-
cations tell us that the specimen was the underground
portion of the plant. Prof. Oliver has observed some-
what similar structures in some new sections of R. corru-
gata Will., which are still undescribed. Williamson ('88)
figured very similar bodies in the cortices of several
plants, but inclined to the belief that they were algae, or
intrusive unicellular organisms of some sort.

Some of the few minute branches which may possibly
be the axes of pinnules (see p. 10) have apparently three
zones of tissue in the cortex. Outside the space surround-
ing the stele is a zone of two or three rows of slightly
thickened cells ; outside them some soft tissue which is
rather broken down, and which appears thin-walled, but
varies in the specimens, and may only be a slightly less
thickened zone continuous with the outermost zone of
thickened cells.

The eptdennis is seldom well preserved, and when
present consists of small regular cells, in no way special.

As in the case of the main axis there appears to be no
characteristic external point which might lead to the con-
nection of this plant with other fragments.

14 Stopes, a Nexv Fern from the Coal Measures.

6. The Roots.

The many adventitious roots in the specimen originate
from both the main axis and the petiole bases, in particular
there is one or more [Fig. 4, il) coming off with each
petiole stele as it leaves the main axis. Large numbers
of the roots cut in several directions are present in each
transverse section. They surround the axis and petioles,
but each one is entirely free ; there is no inclusion of the
roots in the stem cortex, as is so common in tree ferns
and the Psaroneae. They are of varying size, the larger
ones being about 2 to r5 mm. in total diameter.

In some cases it is possible to see the dichotomous
branching in transverse section, when the stele has equally
divided and the cortex not yet completely separated.

The Stele of an undividing root is of the simple diarch
type so characteristic of ferns, and is composed of about
10 metaxylem tracheides, with a group of very small
protoxylem elements at either end (see Fig. 6). These
protoxylem groups are well marked in transverse section,
but true spiral or annular elements have not been seen in
longitudinal view. The larger tracheides are clearly pitted
with multiseriate pits (see Fig. 7). As a rule there is no
trace of the soft stelar elements, a space between the stele
and cortex indicating their position.

The cortex is apparently uniform, composed of
closely packed, rather thick-walled cells, which leave no
air spaces. The outer layer does not seem to differ from
the others, and neither root-hairs, nor exodermis has been
found in any of the sections examined.

The roots as a whole shew no indications of having
been specialised as air or water roots, and seem to be
identical with the ordinary earth root of the modern fern
rhizome.

Manchester Memoirs^ Vol. l. (1906), No. 10- 15

7. Foliage,

No trace of foliage can be identified in any of the
preparations. As the tissues preserved are those near the
main axis, and the size of the petioles suggests a large
leaf, there is a strong probability that no laminae would
have been developed so close to the axis.

That the foliage was fern-like in appearance, the
general anatomy can leave but little doubt. As was
pointed out by Stenzel ('89, p. 4), in his description of the
allied species, petioles of such great size indicate that the
fronds must have been big, and were probably complex.
It is very possible that some of the large frond impressions
known generally as "tree fern" leaves, may belong to
these plants. In T. Sutcliffii but two of the numerous
petioles are seen to be branching to give off a pinnule
rachis, but this in no way militates against the view that
they were ultimately much divided.

Unfortunately, as before mentioned, p. 6, there is no
characteristic feature on the cortex of either stem or
petiole that might indicate which of the many unassociated
impressions belong to this plant ; hence it is unlikely that
the connection of foliage impressions and structural
material will be established in the way in which Sphenop-
teris and R. aspera were recognised as being the same. The
probability is that until a happy chance reveals the secret
in other structure-specimens, the foliage will remain un-
known.

The likeness in general structure between the small
petioles of Anachoropteris rotundata (R. gleicJie Will.) and
T. Sutcliffii makes it possible that some of the smaller
ones known to us from our coal measures as A. rotundata
may belong to Tubicaidis.

Petioles of ^. rotundata of small size are now frequently

1 6 Stopes, a Nezv Fern from the Coal Measures.

found in the lower coal measures, and any day they may
turn up with foliage attached. It must not be assumed,
however, that the plant has any connection with Tubicaulis
Sutcliffii beyond the similarity in type of its meristele.
As was pointed out long since by Williamson ('74, p. ^7^^,
similarity in structure of petioles alone is a very poor
indication of true affinity, and is sometimes exceedingly
misleading.

8. Fructification.

Unfortunately there is no fructification in organic
connection with the plant. There are, however, several
isolated sporangia scattered among the petioles and roots.
The mere association of sporangia in such a position
would carry but little conviction with it in the case of an
ordinary bullion, which is packed with fragments of many
things in extreme confusion ; but in the sixty sections
which this nodule has yielded, no plant structures are
preserved but the one under discussion, a single minute
fragment of cortex which looks as though it might belong
to another plant, and these sporangia. It is generally the
case that the " roof nodules " contain but one specimen
(see p. 18), and, as a consequence, the importance of the
association of the sporangia with Tubicaulis is considerable.
Though there must still remain some doubt about the
question, the probability is that they are indeed the
fructification of Tubicaulis Sutcliffii.

Each sporangium is small, being from 0"2-oi5 mm. in
diameter, and most of them are circular in outline. As is
to be expected from their small size, each appears but
once in the series of sections, so that it is impossible to
determine whether the well-marked annulus is multiseriate
or not. As is shewn in Figs. 10, 11, and 12, the
annulus is well defined, and only on one side of the

Manchester Memoirs, Vol. I. (1906), No. 10. 17

sporangium, in which character they agree with Botry-
opteris rather than Zygopteris ; the annulus appears to be
of but one row of cells, but from the specimen it is
impossible to be sure. Only the outer cells of the
sporangium wall are sufficiently well preserved for recog-
nition, though the debris on their inner sides indicates
the possibility that they had the soft inner lining which
occurs so often in fossil ferns. No spores are preserved.
One of the sporangia is bigger than the others and
somewhat crushed, this shows the annulus very well
(see Ft'g. 12), and may be more mature than the others.
In one slide (T. 15) several sporangia are lying close
together, but they are entirely free from one another.
It is unlikely that sporangia of this type formed anything
approaching a synangial sorus. These sporangia are
extremely like those of a modern leptosporangiate fern,
and there appears no reason to suspect them of being male
fructifications, or in any way different from those of a
typical fern.

9. Geological Horizon and Preservation.

The mine in which the nodule containing Tubicaulis
Sutclijfii was found, is one which works the lower coal
measure seam known variously as the " Bullion," " Upper
Foot," or " Halifax Hard Bed," and is identified in many
localities in Lancashire. The actual nodule was not one
of the plant nodules found in the seam of coal itself, but
was one of those in the " roof" which usually contain no
plant remains, but are packed with goniatites of all
sizes, many visibly sticking on the outside, and must
undoubtedly have been formed under marine conditions.
Sometimes such nodules contain plant remains, in which
case there is generally but one plant in the nodule
embedded among the goniatites and not mixed with

1 8 Stopes, a Neiv Fern from the Coal Measures.

other plant debris. Each probably represents a drifted
plant, which had become water-logged and sank among
the shells on the sea bottom. This minimises chance
association among plants so preserved in one nodule,
for the vissicitudes of drifting would tend to separate any
parts which were not organically connected. Hence the
association of sporangia with the vegetative organs in the
present instance is of much more value than it would be
in the case of most fossils which have been preserved
together with the mass of debris of the forest floor. In
sections, the matrix of the nodule contains numbers of

Fi<r. 2. Small portion of the central axis sliewing the "shattering"
of the tracheides. (Slide T. 4.)

goniatites of varying size, some extremely small and only
revealed by the microscope ; they are embedded in what
appears to be a fine mud which is nearly opaque in
section, but which does not penetrate the tissues of the
plant. Where, as is frequently the case, parts of petiole
or axis are wanting, they are replaced by a clear, cream-
coloured calcite frequently shewing crystalline structure.
The preservation of the plant tissues is in parts excellent,
but these good regions are tryingly local, and neighbour-
ing sections may show a mass of calcite in place of the
continuation of the tissues. Local " carbonisation " also
occurs, which, though generally blurring the tissues, has

Manchester Memoirs, Vol. I. (1906), No. 10. 19

in this case preserved the phloem which is not otherwise
recognisable (see p. 5, Fig. 5). Many of the tracheides in
the best preserved portions show curious "shattering,"
the cause of which is difficult to surmise, and which is
more marked in this case than in any other I have
observed. The tracheid walls are not crushed, but have
broken up into fragments as though they had been very
brittle. {Text-fig. 2.)

The roof nodules, lying as they do immediately above
the coal, must be slightly more recent in point of actual
time than those bullions in the seam itself, though this
difference is geologically so minute as to be of no moment.
A more important difference than that of time seems to
be indicated between the plants forming the " roof " and
" seam " nodules, and that is the oecological conditions
under which they grew. There are indications that in
the " roof" nodules we are not dealing with a swamp
flora, but one which had inhabited higher ground, so that
it is likely that Tuhicaulis Sutcliffii grew on dry land.

10. Affinities.

Among fossil plants the only one which shews a close
likeness to this new fern is Tubicaiilis Solenites, a plant
which has been but once found, and that in 1815, near
Chemnitz. It received its present name from Cotta ('32),
and more recently was fully described by Stenzel ('89).
It is not of Coal Measure age, being attributed to the
" Rothliegende " (a division of the Middle Permian), but
the anatomical details of axis, petioles, and roots are so
strikingly alike in the two plants that generic identity is
certain.

Owing to the kindness of the authorities of the
Museum flir Naturkunde, Berlin, I have been able to
handle parts of the original specimen of this unique plant,

20 Stopes, a New Fern from the Coal Measures.

while to Count Solms-Laubach I am indebted for the
gratification of seeing a microscopic section shewing
several of its petioles. From the examination of these
specimens, as well as from the facts stated in Stenzel's
description, I gather that though they are so alike, there
is enough difference between them to indicate that the
plants are specifically distinct.

As has been already stated, the preservation of
Tubicaulis Sutcliffii is not propitious, and that of T. Sole-
nites is also rather unsatisfactory, but in a totally different
way, so that minute comparison of the two plants is very
difficult. The tissues agree in a general way, and in the
shape of the meristele and axis, and the size of the
tracheides, the agreement is exact. T. Soletiites evidently
had pitted tracheides like those of T. Sutcliffii, though
I did not see them in the sections. Stenzel describes-
them (p. 7) as having about four rows of short cross-pits,
though he had no very satisfactory longitudinal sections
to deal with. The differences between the two plants,,
which taken together appear to warrant specific separation,
are the following : — The relative differences in the sizes
of petioles and axes, T. Solenites with a main stele 6 mm.
in diameter, has smaller petioles than T. Sutcliffii with a
main stele 2 mm. in diameter. The petioles of T. Solenites
seem to have branched much earlier than those of
T. Siitcliff'ii (see Stenzel, //. i, figs. 2 — 8), giving off
many pinnules very close to the main axis ; this was
pointed out by Stenzel as a peculiarity of this fossil plant,
and unlike recent ferns. The phyllotaxy in T. Solenites
is stated to be '^ and \\, while that of T. Sutcliffii is |.
Further there is a considerable difference of horizon and
locality, which presupposes a difference of species, as but
few organisms run absolutely unchanged from the Lower
Coal Measures to the Middle Permian.

]\Ianchcster Memoirs^ Vol. I. (1906), No. 10. 21

The new species appears to be closely allied to the
original species of the genus, and according to Stenzel's
description of this limited genus, it is the only other
known plant belonging to it. The name Tiibicaulis has
been given by a number of authors to different plants in
the course of time, but none of these appears to have real
connection with it, and were all discarded by him in his
revision of the group in 1889. Since then Scott (:00,
p. 296) suggested that Williamson's Rachiopteris corrugata
would have been included in the genus by Stenzel, but
now that more is known of the plant it is clearly seen to
belong to another group.

The genus Tubicaulis is, therefore, a very small one,
consisting of but two species, of each of which but one
specimen is known. Stenzel placed it next Asterochlaena,
to which it undoubtedly shows some affinity, but in
Stenzel's monograph all the known forms of this series are
not included, and it appears to be more like Renault's
Grainniatopteris Rigolloti. In Renault's description of
this species ('91, p. 16 and//tr/^ x.,figs. 11 and 12) there
appear to be many points in which it is like Tubicaulis,
particularly in its simple solid circular stele, and in the
simple band-shaped petiole steles, which, however, remain
almost straight and do not curve to form a C as in Tubi-
caulis. Botryopteris ramosa (R. ra^nosa of Williamson
'91") has also a similar solid main axis with simple masses
of wood coming off to the petioles, but there any close
agreement ceases ; the arrangement and relative size of
its petioles being very different from those of Tubicaulis.
Asterochlaena in spite of its corrugated and irregular main
stele, approaches more nearly to the type of Tubicaulis,
and in the genus there are some varieties which may form
a scries. In A. dubia the simple C of the meristele bends
its convex side towards the main axis as a Tubicaulis, in

22 StopeS, a Neiv Fern f rout the Coal Measures.

A. krigisica the meristele is apparently unbent, while in
A. laxa and A. raviosa the concave side is toward the
main axis in the normal manner of ferns. In all cases
the main axis is surrounded by a mass of petioles which
come off in rapid succession, enlarging somewhat as they
get further from it, but they never, grow to so great a size
as to exceed the main axis, a character which seems to
belong to Tiibicaiilis alone.

Another group which may also be near these plants
is the AnacJio7'opteris series, in which the meristele is a
single horseshoe shaped curve. Unfortunately the main
axis is only known in one species, A. Decaisnii ; this was
described by Renault ('69) as being very like Zygopieris,
that is, with a single five-rayed stele with the central
vascular elements of smaller size than the rest and mixed
with parenchyma. The meristele first comes off from this
in the form of an elipse of hollow wood (see Renault,
plate 10, fig. 4), which soon becomes an open curve.
A. rotunda ta and A. pulclira have similar meristeles,
which though detached are described by Corda ('67,
p. 86-7) as being disposed with the concave side away
from the main axis ; this he gathered from the shape of
the petiole cortex which seems to indicate the fact. It
this is correct, then AnacJioropteris is like Tubicaulis in
this unusual feature as well as in the form of its meristele,
though the latter is slightly more incurved and compli-
cated in Anachoropteris than in Tubicaulis where it is
absolutely simple.

All these plants belong to the large family of
Botryopterideae, and within this group, somewhere near
the base of the series among its simpler members, it will
be safe to place Tubicaulis, near to Graniviatopteris and
AsterocJdaena. The sporangia associated with Tubicaulis
in no way militate against this view. To go into further

MuncJiester Memoirs, Vol. I. (1906), No. 10- 25

detail, however, and attempt a more ambitious classifica-
tion, would be extremely unwise in view of the fact that
next to nothing is known of the foliage and fructification
of most members of the family.

A plant was described in 1849 as a " Tubicaulis von
Ilia," and this Stenzel considers to be an Osiniindites, a
view which appears to be fully justified. He goes further,
however, and considers that " im Sinne Cotta's " it should
be associated with the Tubicaulis group, and he therefore
places Osmwidites after Bathypteris in the series, including
Zygopteris and many other Botryopterideae. The detail
of its woody structure, however, which does not agree
with that of the Botryopterideae, as well as the enormous
difference in time between the palaeozoic group and the
tertiary Osniundites, incline one to leave the two series
for the present less closely associated than they are in
Stenzel's table.

When we turn to recent ferns, there appears to be
none which shows any close affinity with Ttibicaulis. It
is possible that through Zygopteris one may trace a family
connection with the Hymenophyllaceae, but there is no
direct likeness between them be}-ond a certain simplicity
in the anatomy of the main stele, and it is doubtful whether
that carries much weight. The small annulate sporangia
associated with the fossil suggest a Leptosporangiate form,
but without the foliage and sori one is at a disadvantage
in discussing affinity.

Tracheides with multiseriate pits such as have been
noted in Tubicaulis are found in most members of the
group of the Botryopterideae, and they may well be of
more interest and value than has hitherto been allowed.
So far as I am aware they are not found among the living
ferns, in which the scalariform elements are so essentially
characteristic. On the other hand, well marked pits with

24 Stopes, a New Fern from the Coal Measures.

circular borders are equally a gymnospermic feature.
Though we may find here and there instances to the
contrary, taken as a whole there are few things so
characteristic of the asexual pteridophyte and the asexual
gymnosperm respectively, as the markings on their wood
elements. Were the distinction between these elements
with multiseriate pits and those with scalarlform marking
so unimportant as many suppose, then one would expect
to find the former type among the ferns of to-day, but
this, so far as I can ascertain, is not the case. Now the
Botryopterideae, though they have scalariform elements,
have many elements with series of pits recalling a L'ygino-
dendron and other primitive gymnosperms, and it appears
that in this, and perhaps in this alone, they are not abso-
lutety true ferns, but have taken one step towards gymno-
sperm-anatomy. Other considerations lead one to suppose
that they may be the group from which the Pteridosperms
sprang, and it is not surprising that while perhaps remain-
ing true ferns in every other sense of tht word, some of
them have assumed one feature in advance of the general
run of ferns of to-day. As one of the openers of the
discussion at the Linnean Society on the " Origin of
Gymnosperms" this March (1906), Mr. Arber stated that
there was every reason to believe that true ferns did exist in
the FalcBozoic, and that these would belong to a class from
which both Leptosporangiate and Eusporangiate groups
arose. To this old stock he gave the name " Primofilices,"
and stated that the Botryopterideae constitute one of its
families, adding that the origin of the modern ferns from
the Primofilices is clear. The point at present under
consideration is whether the Botryopterideae are really
members of the group Primofilices in the true sense of the
word, which presupposes that its members shall be at least
as primitive as those living to-day. Many of the Botry-

Manchester Memoirs, Vol. I. (1906), No. 10. 25

opterideae, however, appear to shew in the structure of their
wood a character which places them above the modern
fern, and, unless we are to assume that the modern ferns
have lost the habit of forming such wood, and confined
themselves to the simple primary wood of scalariform
elements, then we cannot consider the Botryopterideae as
a whole as Primofilicinean, though, of course, some of its
members may have remained less advanced. The group of
Botryopterideae appears to contain members which are
" Pro-Pteridosperms " rather than Primofilices pure and
simple, and with them was Tubicaulis, which I therefore
conclude has no direct phylogenetic connection with any
living fern.

There is one interesting group of " ferns " to-day with
which they m.ay have some remote affinity, and that is
the Ophioglossaceae. This small family is by some con-
sidered to be the remnant of a group from which Lyco-
pods and ferns sprang, but it has also certain anatomical
characters which are perhaps recently acquired. In
Helniinthostachys bordered slightly oval pits are described
by Farmer and Freeman ('99, figs. 19 and 20) and in
some material kindly supplied by Prof. Oliver, I have seen
much more strikingly gymnospermic pits with perfectly
circular borders. No direct connection with Tubicaulis is
suggested, but it is possible that this group is in some
points a nearer parallel to the Botryopterideae than are
most of the recent ferns.

The fact that of the genus there are but two species,
and that each has been found but once, with an interval
of nearly 100 years, possibly indicates that Tubicaulis
was one of the less successful types, which may very
well have left no direct descendant among more modern
plants.

26 Stopes, a Neiv Fern from the Coal Measures.

II. Specific Description.

The genus as described by Stenzel ('89) is as
follows : —

"Tubicaulis Cotta, Dendrol. S. 15.

"Truncus petiolorum basibus persistentibus obtectus
herbaceus, fasciculo vasculari centrali simplice, terete,
singulos fascicules per corticem crassum in folia emittente.
Petioli basi tenues, ascendentes valde incrassati, fascicu-
lum vascularem simpHcem, fasciaeformem, canaliculatum,
cavitate extrorsum spectante includentes.

"Rohrenstein Breith. — Endogenites Sprengel p.p. — Seleno-
chlaena Corda."

As before stated (p. 21), he included but one species
in this genus, the description of which follows.

" Tubicaulis Solenites, Cotta.

" T. caule digitum crasso erecto herbaceo ; cortice
crassiore quam fasciculo vasculari simplice terete ; petiolis
basi tenuioribus, ascendentibus valde incrassatis, com-
pressis, ultra pollicem latis, fasciculo centrali canaliculato,
cavitate extrorsum spectante, cortice interno molli, externo
e cellulis minoribus firmioribus composito."

To this I would add : Petiolis ramosis juxta caul em ;
phyllotaxis ^-^ et 7/^.

"Rohrenstein, Breithaupt in Oken's Isis, Jahrg. I. (1820),
Heft, v., S. 440, Taf. 4. " Ueber eine Art von Palmen versteine-
rung : Rohrenstein." — Schippan, H. A., Quer- und Langen-
durchschnittsriss einer in Sachsen gefundenen versteinerten
Palme. Freyberg, 1824. 4^. 6 Seiten und i Tafel.

" Endogenites Solenites p.p. Sprengel, comentatio de Psaro-
lithis, 1828, p. 32.

"Tubicaulis Solenites Cotta, Dendrolithen. 1832, S. 21, 22,
Taf. 2, Fig. 1—3.

Manchester Memoirs, Vol. I. (1906), No. 10. 27

" Selenochlaena Reichii Corda, Beitr. z. Flora d. Vorwelt.
1845, S. 81.— Unger, gen. et spec. pi. foss. 1850, p. 200 —
Goppert, Flora d. perm. Form. 1865, S. 44. — Schimper, traite d.
Paleont. J. 1869, p. 697."

Tubicaulis Sutcliffii sp. nov.
Similis T. Soleniti ; caule quam Solenitis minore,
petiolis majoribus ; phyllotaxis 4 ; petiolis juxta callem
sine ramis.

12. Summary.

Titbicaulis Sutcliffii xs: a new species discovered at Shore
in the roof of the Bullion seam, Lower Coal Measures.

The single specimen consists of an axis with a solid
monostele, surrounded by numerous petioles, each with a
simple C-shaped meristele with the convex side toward the
axis.

The petioles are small on leaving the axis, but rapidly
increase till they greatly exceed the axis in diameter.

The phyllotaxy is -| ; no "axillary shoot" has been
observed.

Many adventitious roots arise from petioles and axis,
which branch freely and generally dichotomise.

No foliage is associated in any way, though compound
leaves are suggested.

Several small annulate sporangia are associated, and
as the nodule is a ''roof" one this strongly indicates true
connection.

The plant is closely allied to T. Solenitcs Corda, but is
specifically distinct.

It appears to be one of the simpler Botryopterideae,
and to have no direct affinity with any living fern.

28 Stopes, a New Fern from the Coal Measures.

In conclusion I must acknowledge my indebtedness
to several people in the course of this work. To W, H.
Sutcliffe, Esq., as before stated, for the generous way in
which he placed the specimen in my hands ; to the
Authorities of the Berlin Museum and Count Solms
Laubach for the loan of their valuable type specimens of
the allied species T. Solenites ; to Prof Weiss, in whose
department the work was done, for his kindly and helpful
interest ; to Prof Oliver, for permission to examine the
specimens of Botryopterideae at University College, and
other kindness ; and to Mr. D. M. Watson for photo-
graphing a number of the sections. I may add that
owing to Mr. Sutclifife's generosity, the whole of the
specimens have been deposited in the Manchester
Museum, with the following exceptions : — Sections T 13,
T 4, L 14, have been presented to Prof F. W. Oliver,
section T 16 to Mr. James Lomax, sections T 9 and L 12
have been retained by Mr. Sutcliffe, and T 12, T 22,
T 26, and LE 2 by myself

Manchester Memoirs, Vol. I. {igo6),No. 10- 29

13. Literature.

CoRDA ('67). "Beitrage zur Flora der Vorwelt." Berlin, 1867.

CoTTA ('32). " Die Dendrolithen." Dresden and Leipzig, 1832.

Farmer and Freeman ('99). " On the structure and affinities
of Helmmthostachys zeylanicaT Ami. Bat., vol. 13,
pp. 4 2i-444> 1899.

Oliver and Scott (:04). "On the Structure of the Palaeozoic
'St^d Lagenostoma Lof/iaxi." P/iiL Trans., ser. B,vo\. igj,
p. 193-247, pis. 4-10, 1904.

Pettko ('49). "Tubicaulis von Ilia." JVa^. wiss. Abhandl.
Haidinger, vol. 3, p. 3-9, 1849.

Renault ('91). " Note sur la famille des Botryopteridees."
Bull. Soc. hist. nat. Autun, vol. 4, p. 3-27, 1891.

('69). " Etude sur la tige de Zygopteris." Ann. Sci. Nat.

Bot. (5), vol. 12, p. 161-190, pis. 3-14, 1869.

Stenzel ('89). " Die Gattung Tubicaulis Cotta." Bibl. Botanica,
p. 1-50, 1889.

Scott {:oo). "Studies in Fossil Botany." London, 1900.

(-OS)- "The Early History of Seed-bearing Plants, as

recorded in the Carboniferous Flora." Manchester
Memoirs, vol. 49, no. 12, pp. 1-32, 3 pis., 1905.

Williamson ('74). " On the Organisation of the Fossil Plants
of the Coal-measures." Mem. 6, Phil. Trans., p. 676-703,
1874.

('88). " On some Anomalous Cells developed within the

interior of the Vascular and Cellular Tissues of the
Fossil Plants of the Coal-Measures." Ann. Bot., vol. 2,
pp. 1-9, 1888.

('91)- " On the Organisation of the Fossil Plants of the

Coal-measures." Mem. 18, Fhil. Trans., ser. B, vol. 182^
P- 255-265, 1891.

30 Stopes, a Nezv Fern from the Coal Measures.

DESCRIPTION OF PLATES.

Figs. I and 2 are from photographs taken by Mr. James
Lomax, of Bolton, and Figs. 3, 4, 5, 6, 7, 8, and 10 were specially
photographed by Mr. D. M. S. Watson, to whom I am much
indebted.

Plate I.

Fig. I. Transverse section across the whole plant shewing
the main axis a surrounded by numerous rootlets and petioles.
i^Slide T 7.). X 175 diam.

e. Petiole of which but a small fragment is preserved, though
the shape is indicated by the clear calcite mass {see
p. 11).

ao. The largest petiole, found in many sections of the series,
and shewing small branch bundles in the cortex (see
p. 10). The narrow band like meristele is seen with its
convex side toward the axis.

Q.

^

^

^

s

32 Stopes, a New Fern/roju the Coal Measures.

Plate 2.

Fig. 2. Several roots cut in different directions, root a magnified
in Fig. 6. b part of outer cortex of a petiole.

Fig. 3. Central axis, showing branches of stele going to petioles
and roots. x 14.

a. Solid central meristele ; its star shape is due to the

local breakdown of Tracheides.
h. Strand just leaving for a petiole.

c. Meristele now enclosed in petiole cortex.

d. Root bundle coming out at the side of the petiole.

e. Cortex of main axis.

Fig. 4. Similar to Fig. 3. The root d cut in transverse section.

]rig. 5. Small portion of the main axis, shewing a strand b
beginning to separate from the main mass of the wood w.
Surrounding the wood and within the dark band lie the
smaller, thin walled phloem cells, //^. {Slide Z i.)

Fig. 6. Root a of Fig. 2 enlarged. Shews the diarch stele
with large metaxylem elements {cf. Fig. 7).

r^>rfy^\j^]

*jj^
^

i

i^* '- '•-'

T »^^

''^'

M

ft ,

■«s

34 Stopes, a Nezv Fern from the Coal Measures.

Plate 3.

Fig. 7. Longitudinal section of a root shewing the multiseriate
pits of the woody elements, x 140. [Slide L 6.)

Fig. 8. Small portion of a petiole shewing a piece of the stele
and the cortex. x 44.

7V. Wood with the protoxylems, px., on the convex side.
s. Crushed cells, probably phloem and sheath.
c. Cortex.

Fig. 9. Drawing of cells from the cortex containing circular
bodies, possibly starch. x 400.

Fig. 10. Photograph of a sporangium. x 184.

Fig. II. Drawing of a sporangium shewing the circular form and
well marked annulus. x 230. {Slide L i.)

Fig. 12. Drawing of cells from the annulus of a larger sporan-
gium. X 290. {Slide L 9.)

I

Manchester Memoirs, Vol. L. {No. 10.)

7-

Piaie III.

''^f^ii^

Manchester Memoirs, Vol. I. (1906), No. 11.

XI. On the Difference between Physiological and
Statistical Laws of Heredity.

By A. D. Darbishire, M.A.

Demonstrator of Zoology in the Royal College of Science, London.
Received and read March 6th, igo6.

Contents.

1. Introductory. p. i.

2. Statistical Laws.

{a) Pearson's Law, p. 4.

{b) Galton's Law. p. 4.

{c) The difference between the two. p. 5.

3. Physiological Laws.

(a) The Law of Diminishing Individual Contribution.

p. 8.

(b) Mendel's Law. p. 11.
{c) The difference between the two. p. 14.

4. The Difference between statistical and physiological

Laws.
{a) The former " descriptive " : the latter " explan-
atory." p. 15.

{b) Mendel's Law true of units : Pearson's of masses.

p. 19.

{c) Examples of the confusion between physiological
and statistical Laws. p. 23.

{d) Description of a method of dealing with the
material of a breeding experiment in such a
way that the data obtained may be used to
test the validity both of Mendel's and of
Pearson's Law. p. 28.

((?) Why do white sheep eat more than black ones ?

P- 31-

July 6th, igo6.

2 Darbishire, Laws of Heredity.

I.

There are those who maintain that it is not the part
of the biologist to argue, to discuss, and to explain ; and
who assert that he is transgressing his proper limits when
he ceases to confine himself to the description of observa-
tions and experiments, and to drawing from them certain
obvious conclusions.

I do not hold this opinion : because I am convinced
that if as much (I do not say more) trouble were taken to
understand the meaning of a term as is spent in
establishing the authenticity of a fact, the progress of our
knowledge of fundamental natural processes — heredity,
variation, the determination of sex, to name a few — would
be more rapid than it is at present. For it seems to me
to be evident that nothing short of a firm but unprejudiced
grasp, in the mind of the investigator, of the relation
between the facts themselves and past, present and
possible attempts to account for them can enable him to
advance toward a closer knowledge of these phenomena.

I think that the reader will admit the truth of this
contention, if he is not one of those who still cling to the
Baconian delusion that all that is necessary for the
elucidation of the problems of nature is the bringing to
light of as many facts as possible by as many workers as
possible ; whereas as a matter of fact it is obvious that
that which hinders the progress of natural knowledge is
not the slowness with which facts are brought to light,
but the paucity of investigators capable of dealing with
them properly.

The incapacitating fault among biologists which is at
once the commonest and the most serious is the uncon-
scious ease with which they fall into the error of using a
term without having previously ascertained its meaning.
And so long as biologists turn a deaf ear to speculation

Manchester Memoirs, Vol. I. (1906), No. 11. 3

this disease will flourish. That which is necessary, there-
fore, to make progress both surer and swifter is a greater
aptitude to formulate a clear idea of the meaning of the
terms which are employed — a habit of mind which is not
likely to be common so long as the consensus of biological
opinion regards with less favour the attempt to discover the
essence of a newly suggested hypothesis, than the attempt
to describe the course of the vas deferens in a newly
discovered worm.

In the study of heredity in particular the most
extraordinary confusion has resulted from the fact that
not only has the same term been used to mean different
things by different writers, but very often has had many
significations in the writings of a single author. This
state of affairs is due, in my opinion, to the absence
of any patient and laborious attempt to thresh out the
meanings of the terms continually on the lips of those
who take part in the discussion of this subject ; and
this absence is due in its turn to the callousness, if not
disfavour, with which such an attempt is likely to be
regarded. Nevertheless I propose to make it.

Space forbids me to discuss the question of the
advisableness of using the term " law " at all as summaris-
ing vital phenomena, more than to say that the fact that
I use it 166 times in this paper demands some apology.

I use it because, besides possessing the advantage of
brevity, it is of all terms in biology the vaguest ; signify-
ing as occasion demands either a theory, or a resumd, or
a hypothesis, or a formula, or a generalisation — to name
a few of the more or less legitimate senses in which it is
used : and because it shelters, under its wide roof. Laws
whose authors aim at explanation, and those whose
authors are satisfied with description. And I spell it with

4 Darbishire, Laivs of Heredity.

a capital L because that is the conventional way of
writing terms the discussion of whose meaning is post-
poned.

2 (a). Pearson^ s Law.

No one has any excuse for not knowing what the
Law of Ancestral Inheritance is ; the essential features of
it are outlined by Pearson in the following words : —

" Taking our stand then on the observed fact that a
knowledge neither of parents nor of the whole ancestry
will enable us to predict with certainty in a variety of
important cases the character of the individual offspring
we ask : What is the correct method of dealing with
the problem of heredity in such cases ? The causes
A, B, C, D, E, ... which we have as yet succeeded in
isolating and defining are not always followed by the
effect X, but by any one of the effects U, V, IV, X, Y.
We are, therefore, not dealing with causation but corre-
lation, and there is therefore only one method of procedure
possible ; we must collect statistics of the frequency
with which U, V, W, X, Y, Z respectively follow on
A, B, C, D, E .... From these statistics we know the
most probable result of the causes A, B, C, D, E and the
frequency of each deviation from this most probable
result. The recognition that in the existing state of our
knowledge the true method of approaching the problem
of heredity is from the statistical side, and that the most
that we can hope at present to do is to give the probable
character of the offspring of a given ancestry, is one of
the great services of Francis Galton to biometry.*

2 {b). Galton's Law.
Galton formulated his Law as follows : " The two

• Pearson, :03a, p. 215.

Manchester Memoirs, Vol. I. {igo6), No. 11. 5

parents contribute between them on the average one-half,
or (o'5) of the total heritage of the offspring ; the four
grandparents, one-quarter, or (o"5)" ; the eight great-grand-
parents, one-eighth, or (o'5)^ and so on. Thus the sum of
the ancestral contributions is expressed by the series
{(o'5)+(o"5/-f(o'5)^ &c.}, which, being equal to i, accounts
for the whole heritage." *

2 {c). The Difference between Pearson's and Galton's Laiv.

It will be seen how profoundly Galton's differs from
Pearson's Law. Yet the belief that the two are much the
same is not rare, and the statement that the latter is
merely an extension of the former is often made. A clear
appreciation of the difference between the two is necessary
to anyone who wishes to be conversant with modern
theories of heredity.

One feature the two have in common ; both of them
are true only of masses, and do not pretend to apply to
individuals. This is so obvious to the careful thinker that
Pearson only refers to it in a footnote if yet it is often
forgotten. The difference between the two lies in this :
Pearson's Law measures the degree of correlation between
a character or characters in a given generation, and
some similar (or dissimilar) character or characters in
the preceding generation. Galton's Law states the
amount which a given generation contributes^ to the
generation which it produces. It definitely states that
on the average a half of the filial generation are like the
parental, a quarter like the grandparental,and an eighth like
the great-grandparental, and so on. From the knowledge
that the parents of a given generation of cats are tabbies,

• Galton, '97, p. 402.
t Pearson, :04, p. 161.
+ See Appendix B.

6 Darbispiire, Laws of Heredity.

and that half of its grandparents are tabbies, a quarter
whites, and a quarter blacks, you are enabled to predict
by Galton's Law the proportions in which these three
kinds of cats will occur in that generation. Pearson's Law
does not enable you to do this : it is of an entirely different
kind. For Pearson's Law to be true it is not necessary

FATHERS.

Black.

Dark
Grey.

Grey.

Pale
Grev.

White. Totals.

Purple.

Bluish
1^ Purple.

K Purplish
P Blue.

I— I

() Blue.

Pale
Blue.

Totals.

3

II

2

16

8

27

19

8

62

6

1

30

39

32

3

no

12

27

33

18

90

4

10

8

22

17

80

91

83

29

300

The degree of correlation in the above Table is that between the
number of dice exhibiting 4 or more than four points (or "pips") upper-
most, in first throws and the number exhibiting those faces uppermost in
second throws in a series of 300 double tiirows. The two throws are
correlated by leaving half the first throw on the table, so that the second
throw has half the dice lying exactly as they fell in the first throw. For
details of this process the reader is referred to Weldon :o6, p. 100.

The degree of correlation in my imaginary Table is "54 ; for calculating
which I am indebted to Mr. Udny Yule. The "coefficient of parental
heredity," therefore, in this case is identical with that for the inheritance of
deafness, of which Pearson's Law is true, recently worked out by Schuster
(:o6, p. 478). Yet, in my imaginary case, none of the children could be
mistaken for any of the parents.

Manchester Memoirs, Vol. I. (1906), No. II. 7

that any of the children should be like any of the parents ;
all that is necessary is that a particular kind of parent
should be associated with {i.e. should produce) as often as
not a particular kind of child. On the opposite page is an
imaginary Correlation Table, in which Pearson's Law is
borne out, yet in which none of the children are like any
of the parents.

The fact that the relation between a given generation
and those that precede it, is described by a series of figures
which in the case of Galton's Law is '5, '25, '125, "0625 etc.,
and which in the case of Pearson's, for eye colour in man,
for example, is "4947, ■3166, "1879,* has led some to believe
that the figures mean the same thing (which, of course,
they do not), and has thus constituted a trap for the un-
wary. Castle has done good service to progress in the
study of heredity by falling into it.f

I hope I have made clear what the difference
between the meanings of the two series is ; for to under-
stand this is to understand the difference between the
two Laws.

This difference is sometimes expressed in the state-
ment that Pearson's Law is more comprehensive and less
biological :|: than Galton's : and inasmuch as it embraces
sets of facts which are not described by Galton's formula,
the first of these statements is true : and inasmuch as the
relation between successive generations which it measures
is the same as the relation between two series of throws
of dice (in which reproduction is unknown), of which
every throw of the second series consists of half the dice
lying exactly as they fell in the corresponding throw of
the first series, the second is true a!so.

* Pearson, '.O'^a, p. 221.
t Castle, .-03, p. 224.

JFruwirth (:05, p. 147) goes so far as to say that " das Ahnenerbengesetz
ist kein biologisches Gesetz, . . . ."

8 Darbishire, Laws of Heredity.

3 id). The Law of DiminisJiing Individual Contribution.

In my paper on the supposed antagonism of biometric
to Mendelian theories of heredity, I showed that a set of
facts (summarized in the Table on page 6 of that essay),
appearing at first to be a complete refutation of Mendel's
Law, could easily be shown to be equally in accord with
both Mendelian and Galtonian theories.* Mendel's Law
describes the individual phenomena in this Q3.se perfectly :
Galton's Law describes the mass result composed of these
very individuals mating at random perfectly. The latter
describes the proportions, the former accounts for them.
The Galtonian deals with individuals from the point of
view from which the physicist deals with atoms ; the
Mendelian deals with them from that of Clerk Maxwell's
demon.

Now just at the same time that I announced my
discovery that the proportions of the albinos in this case
were not evidence against the truth of Mendel's law,"f-
Castle made the same discovery.:|: But he argued from
his discovery, not (as I did) that the two theories were
compatible but that Galton's was wrong ; that is to say,
he must have thought that the two theories were mutually
exclusive ; which indeed he did : but not in the same way
that I did. For whilst the way in which I made that
error was by lifting Mendel's Law from the level of a
would-be explanatory to that of a purely descriptive
Law, he made it by lowering Galton's from the level of
a purely descriptive to that of a would-be explanatory
one. And the reason that I discovered my mistake
before he did was that it is easier to see that Mendel's
Law is something more than a purely descriptive one, than
it is to see that Galton's is not a would-be explanatory

* Darbishire, :05a, p. 6.
t Darbishire :05a, p. 9- + Castle :05, p. 17 £/ seq

Manchester Memoirs, Vol. I. (1906), No. II. 9

one. And the reason again of this is that Galton's Law is
confused with another one which resembles it in one
respect, but differs from it in being would-be explanatory.
The remarkable thing about this Law is that whilst it is
characteristic of most Laws to be enunciated and receive
a name first and then become widely believed in after-
wards, the reverse is the case with this one ; for it is
believed in by all biologists who are not Mendelians, by
all breeders of animals or plants, and by all persons not
belonging to these classes who think about heredity at all.

But it has not yet received a name. I propose to call
it the Laiu of Diminishing' Individual Contribution.

According to it : the germ plasm of an individual con-
tains contributions from all of its progenitors : the amount
of the contribution being large in proportion as the progenitor
is near, i.e., large in the case of the parents, smaller in the
case of the grandparents, and so forth.

It is a very good type of biological Law : it has the
advantage of simplicity : it is also, except in a few cases,
untrue.

I will now give 3 cases to shew how widespread belief
in this Law is.

The first that I give is that of the result of crossing a
yellow and white pink-eyed Japanese waltzing mouse with
a pink-eyed white mouse — that is, an albino. The result
is, usually, a black-eyed grey mouse.* And to anyone not
familiar with it, the result is most astounding : it is quite
the opposite of what one would expect. Expect from
what ? From one's — possibly unconscious — belief in the
Law of Diminishing Individual Contribution.

Another case. Now that we know that a blue Anda-
lusian fowl is a heterozygous form produced by mating a
black and a white ; and that Andalusians when mated

* Darbishire, :04a, p. 7.

lO Darbishire, Laivs of Heredity.

together produce 25 per cent. Blacks, 50 per cent. Anda-
lusians, and 25 per cent. Whites, we no longer try to
get a pure strain of Andalusians by throwing away the
blacks and whites and by continuing to breed from the
Andalusians for many generations, because we know that
we can always get Andalusians and nothing else by
mating blacks and whites*. What is the conception of
heredity which underlay the old-fashioned attempt to
breed pure Andalusians by weeding out the blacks
and whites, but the Law of Diminishing Individual
Contribution ?

Again Coutagnef, in discussing the possibility of
the hybrid nature of some dark-lipped individuals of
Helix hortensis, which occurred in a collection of that
species and Helix nevioralis living in one locality,
concluded from the fact that these supposed hybrids were
unhanded, whereas the great majority of the H. nemoralis
in that locality were banded, that they were not hybrids.
To translate his own words "If the H. nemoralis were the
parents of the 113 black-lipped individuals there is every
reason to believe {tout porte a pr^sunier) that this char-
acter of banding would appear at least in some cases
in these 113 individuals." Through Lang's;]: work we
know now that in a cross between a banded H. nemoralis
and an unhanded H. hortensis the unbandedness is
dominant. So that now we should not expect " this
character of banding " to appear in any of the indi-
viduals : and Coutagne's argument falls to the ground.

But what is '' tout porte a pnfsuiner" but the expecta-
tion based on a firm belief in the Law of Diminishing
Individual Contribution?

* Punnett, 105, p. 28.
t Coutagnc, :95, p. 72.
+ Lang :04 (p. 497) and :06. See also Darbishire .O^f', p. 196.

Manchester Memoirs, Vol. I. (1906), No. II. ii

These three cases shew how widespread is behef in
this Law ; and they also shew that in these three cases at
least it is not valid.

The difference between the expectation based on this
Law* and the accurate knowledge of what actually takes
place (which it is the business of Mendelian investigation
to supply), is the same as the difference between common
sense and science, and the same as the difference between
that which stands to reason and that which rests on
evidence.

3 {U). Menders Law.

I do not propose to discuss here the statement that
the time has not yet come when we are justified in
speaking of Mendel's Laiv, nor to enquire into the meaning
of this statement : the question I propose to answer is,
" What is the essential feature of that which is called
Mendelism by those who believe in it, and Mendelianism
by those who do not ? "

I divide definitions of the Law into two primary
categories : —

(i.) Suitable for those who desire to establish the
invalidity of the Law.

(ii.) Suitable for those who wish to discover whether
Mendelism has helped us or is likely to help us
to attain to a more intimate knowledge of
heredity.

There is no difficulty in finding a definition of the
first class : a very satisfactory one is one which binds the
Mendelian down to the Law exactly as enunciated, and
the description of the phenomena exactly as given by
Mendel himself If this is carefully done, no difficulty

* Huxley must have been thinking of some such Law as this when he
made the remarkable statement that science was organised common sense.

12 Darbishire, Lazvs of Heredity.

will be encountered in establishing the invalidity of
Mendel's Law as facts accumulate.

To discover a definition of the second class is not so
easy. To my mind, there are two perfectly distinct things
included under the one term Mendelism. One is belief
in the existence of character-units in the germ, and in the
thesis that these units are pure in respect of the characters
which they represent. The other is the hiethod by which
the extent, separateness, and transmission of these units
is discovered. The first may be called the Mendelian
theory, the second the Mendelian method — which is
the application of the experimental method to the study
of heredity. I think that both these things are implied
when the term Mendelism is used ; and whether they are
or not — and it does not in the least matter — I believe
that the Mendelian method will do as great service, in
accounting for the phenomena of heredity, as that parti-
cular theory which Mendelians happen to be employing
at the moment.

For I think that it must be evident to anyone who
has followed closely the Mendelian work of the last few
years that, while the method which workers of that school
have employed has remained the same, the actual theories,
by testing the validity of which they have sought to attain
their end, have been from time to time considerably
modified. This procedure, of course, makes it difficult for
those who wish to criticise or base statistical calculations
on the theory itself. Pearson comments on it in these words:
" The original Mendelian theory has been replaced by
what are termed " Mendelian Principles." In this aspect
of investigation the fundamental principles propounded
by Mendel are given up, and for each individual case a
pure gamete formula of one kind or another is suggested
as describing the facts. This formula is then emphasized,

MancJiester Memoirs, Vol. I. (1906), No. 11. 13

modified, or discarded, according as it fits well, badly, or
not at all with the growing mass of experimental data.
It is quite clear that it is impossible while this process is
going on to term anything whatever Mendelian as far as
theory is concerned."*

But should we be right in refusing to commend the
efforts of a well-digger, if, in sinking his well, he alternately
used a spade, a pickaxe, and dynamite, according as he
had to deal with gravel, sandstone, or granite, provided
that he found, or even that he thought he would find,
water at last ?

The aims of the Mendelian and the well-sinker are
the same — to discover something ; and they each employ
a definite method, but the tools they use are continually
being changed. That is why I think that the method is
at least as essential a part of Mendelism as the theory.
And that is why I think that there is no more connection
between Pearson's generalized theory of alternative
inheritance (with special reference to Mendel's Laws), and
Mendelism, than there is between the second law of
thermo-dynamics and the Maxwellian demon's knowledge
of atoms plus the method by which he has acquired it.

There is a definite relation between a generalized
theory of alternative inheritance and that particular
doctrine on which it is based : it is the same as the
relation between the second law of thermo-dynamics and
the theory held — ex hypothesi — by the demon as to the
nature of the atom.

But there can be no relation between any generalized
theory of inheritance and Mendelismf unless that term

* Pearson, :03^, p. 53.

t I do not, of course, intend to imply that Pearson tries to establish
any relation between a generalized theory of inheritance and Mendelism :
I know his was a generalized theory of alternative inheritance based on the
theory of the pure gamete. All I wish to insist on is that the theory which

14 Darbishire, Lazvs of Heredity,

signifies the Mendelian theory only ; and, even so, this
relation cannot be permanent unless the MendeHan is
pledged not to change his theory in the smallest degree.
I hold that, in the first place, Mendelism has, as 1 have
shewn, a wider signification, namely, that it embraces the
method as well ; and secondly, that no Mendelian can be
expected to take the pledge demanded, if, by doing, so he
believed, as he probably would, that he would' be pre-
vented from attaining his end. It is idle to accuse
him of inconsistency. What should we think of the
consistency of a well-digger who died of thirst because
he would stick to his spade although only a few feet of
granite separated him from water?

And what right have we to expect that the demon
should pledge himself not to alter his theory of the nature
of atoms if he hopes that by being free to alter it he will
attain a knowledge of them that will enable him to live
in a warm compartment without having to do any work
for it? What right has a physicist to expect a demon
not to alter his theory, on the ground that such an altera-
tion makes it exceedingly difficult for him (the physicist)
to use the theory as a basis for statistical calculation ?

3 {c). The difference betweeri Mendel's Law and the Law
of Contribution.

Now that we come to discuss the difference between
Mendel's Law and the Law of Diminishing Individual
Contribution we must clearly understand that by Mendel's
Law we mean the theory associated with that name, and
not the method.

Mendelians happen to be testing at the moment is, to my mind, not the
essential thing in Mendelism. If the commonly accepted explanation of the
proportion iDD : 2DR : iRR were shewn to be false, would experiments,
called Mendelian, now in progress be prosecuted with less zeal? By no
means. Such a discovery would even be an incentive to more strenuous
search.

Manchester Memoirs, Vol. I. (1906), No. 11. 15

These two Laws resemble each in being physiological,
in that they attempt to picture the way in which characters
are represented in the germ cell. But they differ pro-
foundly in the picture which they draw. The difference is
so obvious that it is hardly necessary to speak about it. A
remark on the difference between the predictions of the
two Laws as to the nature of the offspring of extracted
recessives will suffice. Suppose that two hybrid mice
with grey coats and black eyes were to produce an
(extracted) albino — which, if the Law of Contribution
were true, they could not do : the Mendelian prediction
about the offspring of a pair of such albinos is that they
will be all albinos : the expectation based on the Law of
Contribution is that a quarter of the coat of each indi-
vidual child will be grey — supposing the proportions for
individuals in which each progenitor contributes, according
to that Law, to be the same as that demanded for popula-
tions by Galton's. The Mendelian prediction is right.* In
fact, the Law of Contribution is so utterly invalid that every
case of alternative inheritance is a contradiction of it. It
may apply to some cases of blended inheritance. But
the reason that I have formulated it, and given it a
name, is not that it may perhaps apply to one or two
cases ; but because unless it is definitely enunciated
it will not be reckoned as having any claims to recognition ;
and because, the sooner it is widely recognized, the easier
will it be to put an end to its confusion with Galton's
Law.

4 {a). Statistical Laws^ " descriptive " .• Physiological
Lazvs, "■ explanatory y

The remark might be made about the Law of Contri-
bution that it is Galton's Law made applicable to the

* Darbishire :04", p. 23.

1 6 DarbiSHIRE, Laws of Heredity.

individual : and this in a sense is true : but there is a pro-
found difference between the two ; for, whilst the Law of
Contribution is an attempt to picture the way in which
characters are represented in the germ cells of individuals,
Galton's Law is merely a statement that the characters of
the ancestry of a population reappe'ar in certain definite
proportions in that population. It is only concerned
with that which is above the horizontal line AB \x\. the
figure on the next page. Moreover it is only true of the
aggregate of adults, and not of the individuals which
compose that aggregate. The Law of Contribution, on
the other hand, deals with that which is below as well as
with that which is above the horizontal line, and it is true
of the individuals.

Now, although Galton's Law is true of the mass but
not of the component individuals, the Law of Contribution
is true of the individuals and of the mass as well : because
according to it all the individuals are contributed to, in
the same degree.

The widespread belief in the existence of a Law which
is true both of mass and individual is the result of the
interaction of two outstanding characters of the human
mind, (i.) an inability to distinguish between truths about
masses and truths about individuals — and (ii.) a passion for
explaining things ; for possessing a formula for familiar
phenomena ; in the case of heredity for seeing below the
line in the diagram opposite.

Man has observed that on the whole Galton's Law is
true of masses — though he has not expressed himself in
these words. Feature No. (i.) had led him to think that the
truth is of individuals, and has thus satisfied No. (ii.) and
supplied him with a brief formula, a simple picture of the
way in which characters are inherited — a picture of what
is below the line AB in the diagram.

Manchester Memoirs, Vol. I. (1906), No. 11 17

cO

(mV

^ — f."

;:::::«»

::.:

r.::T6«-

::.:.

:r::-:x>*

A.
V

\

\

%.

.....

\\\

..„...\o*

::~::6^

\ t

\

r7)

:s 'JJ

i2

¥ ^

l8 Darbishire, Laws of Heredity.

Even if this is not a true history of the origin of the Law
of Contribution, the point about that Law which I wish to
enforce is that it is a physiological theory of heredity. It
is an attempt to account for the phenomena of heredity
by picturing the way in vvhich the characters of an
organism are represented in the germ cell which produces
it. And being a physiological Law, it is profoundly
different from a statistical one like Gallon's, which does
not pretend to account for anything, but is a generalization
about the relation between the aggregates of adults of
successive generations.

In a word : a physiological Law deals with the indi-
viduals of each generation on both sides of the line AB in
the diagram : a statistical one, with the generation as a
whole on the upper side only. Another way of marking the
difference between physiological and statistical Laws of
heredity is to say that the former are explanatory while
the latter are descriptive. To which it will be immediately
objected that no Law ever explains anything. I am
perfectly aware of this, and of the fact, moreover, that no
theory does ; in fact that we cannot explain anything at
all in nature ; that all that we can do is to describe. But at
the same time it cannot be denied that there is all the
difference between attempting to account fora phenomenon
and contenting oneself with describing it : and one is, I
hold, perfectly logical in making this attempt to explain,
although one knows that however intimate a knowledge
of causation one has acquired one has done no more
than describe phenomena. One great difference between
these two things, the logical attempt to explain, and
satisfaction with mere description is that the method of the
former is experiment and that of the latter is observation.
Another difference is that it is only by attempting to
account for things that we have been enabled to get what

Manchester Memoirs, Vol. I. (1906), No. ||. 19

knowledge we have of the causation of natural phenomena
and so to obtain what control we have of the operation
of natural processes. Just because we know that ex-
planation is after all only description, it does not follow
that we should abandon the attempt to account for things.
We are now in a position to classify Laws of heredity
under these two headings.

Phvsiolog ical. Statistical.

(a) Mendel's Law. (a) Galton's Law.

(/;) The Law of Contribution. {i>) Pearson's Law.

Now inasmuch as of physiological Laws d has been
shewn to be invalid, and of statistical ones a has been
shewn to be less comprehensive than d, the discussion of
the mutual relation of physiological to statistical Lav^s of
heredity resolves itself into a discussion of the relation of
Mendel's Law to Pearson's Law of Ancestral Inheritance.

4 (/;). Mendel's Law True of Units : Pearson's of
Masses.

Discussing a little while ago with a friend the relation
of Mendelism to biometry, I suggested as the briefest
possible statement of the difference between the two that
Mendelism treated of units, and biometry of masses of
units : my friend replied. ' You speak of the animals and
plants with which you deal as breeding true to such and
such a character. What do you mean by this statement?
Are the offspring absolutely identical with their parents
in respect of the character under consideration? If they
are not, how like are they ? and how is the degree of this
similarity to be measured, except by biometric methods?'

I saw that there was truth in what he said ; but I
could not see the relation which the idea in his mind
bore to my original idea, to which I still adhered,

20 Darbishire, Laws of Heredity.

that Mendelism dealt with the units while biometry
was concerned with masses. Now I see that my
inability to do so 'was due to the fact that biometry
meant to my friend that the only way to measure the
resemblance between parents and children is the method
of the correlation table ; in which he was quite right :
while biometry called up in my mind the Law of Ancestral
Inheritance, and especially the manner in which data
designed to establish that Law are collected and dealt
with ; such, for example, as in the case of one of the last
series of data from which such correlation coefficients
have been worked out — that of greyhounds. I had in my
mind the collective treatment in one correlation table of
such different characters as Black, Brindled, Fawn, White,
and Red ; * while he was thinking of the only method of
measuring the intensity of inheritance within a single such
character — say, Red. Now this I believe brings us to the
heart of the matter. When I say that the Mendelian deals
with units and the biometrician with masses, I mean, not
that the former deals with a few and the latter with many,
but that the former first settles what character he is
going to treat as a unit and then only deals with it
in large numbers when he is sure that the component
units of this number are identical ; their sameness having
reference to such properties as can be discovered by
mating them with their like and with their unlike. It
is just as necessary for the Mendelian to have a large
record of matings, as the biometrician, to establish his
generalizations. But though the Mendelian might allow
that the only method of measuring the similarity between
parents and offspring within a group (such as the unit
character Red) was that of the correlation table, he would
vehemently maintain that the biometric method over-

* Rartington, etc. :05, p. 264.

Manchester Memoirs, Vol. I. (1906), No. 11. 21

stepped its limits when it included in such a table more
than one such category. What Bateson means when he
says that Mendel saw by sure penetration that masses
must be avoided is that the biometric method oversteps
its limits when it does this.

The answer that I should now give to my friend is
this : ' I fully admit that the only method of measuring
the degree of resemblance between a generation and the
one which produced it, within such a unit, is the biometric
method : I fully agree with the biometrician, when he
says that all green peas are not alike in respect of their
greenness because all green peas are green, and that the
biometric method is the only one to measure their dis-
similarity ; but I stoutly maintain that when he puts
green and yellow peas together into a correlation table he
has started on a path which will not lead to a more inti-
mate knowledge of heredity.'

Biometry furnishes the only means of actually
measuring the intensity of heredity within a unit :
Mendelism furnishes the only means by which a fuller
knowledge of the properties of these units may be
acquired.*

It is sometimes easy to determine the extent of these
units as in the case of discontinuous characters such as
purpleness and whiteness of flower in Pisum : f it is often
difficult as in the case of characters varying continuously
such as the weight of beans.:|:

Careful consideration of the table on p. 6 of my paper §
on the supposed antagonism of Mendelian to biometric
theories shews that the accuracy with which Mendelian

* Exactly the same idea is expressed by Lotsy (:o6, p. 143).
+ Fruwirth :06, p. 141.
X Johannsen :03.
§ Darbishire :05«.

22 Darbishire, Laws of Heredity.

or Galtonian Law describes the facts of heredity depends
on the composition of the unit tested. There is nothing
a p nor 2 iWoQical in treating the sharply defined category
of dark-eye-and-coloured-coat, in mice, as a unit. Suppose
this is done : Galton's Law fits the facts beautifully, while
Mendel's is triumphantly refuted (by shewing that the
amount of albino ancestry of a hybrid affects the per-
centage of albinos produced by such hybrids mated
inter se) : to which the Mendelian would make the follow-
ing answer : " Are you sure that your unit ' dark-eye and
coloured-coat ' is incapable of resolution into still simpler
units. Are you sure that you are not regarding as a
simple thing that which is really compound, just as the
' fixed alkalies ' were regarded as elements until Davy
showed them to be compounds.*^ I can prove that
you are ; for by testing the gametic constitution of the
dark-eyed and coloured-coated forms, I can shew that
they are sharply distinguished into heterozygous and
homozygous forms. Now I claim to have discovered
what should really be treated as a unit ; namely that
character which, like the elements of the chemist,
cannot be split up into simpler characters. 1 do
not pretend that my formulas of heredity describe the
numerical results obtained by jumbling a lot of my
elemental units together, any more than you pretend that
the Law of Ancestral Heredity describes the phenomena
exhibited by my units when dealt with separately."

* Davy 'o8.

+ This illustration may at first sight appear not to be strictly parallel.
A chemist reading it would think that what was going to be shewn was
that the unit "dark-eye-and-coloured-coat" was resolvable into "dark-eye"
and "coloured-coat," whereas, of course, what is really going to be shewn
is that the units "dark-eye and coloured-coat " are of two quite distinct
kinds. The true parallel to my case is the idea ol the element before
Davy's time. Amongst the things that were classed as elements, some
really were undecomposable (cf. the homozygous forms), while others — the
alkalies — were decomposable (cf. the heterozygous).

Manchester Memoirs, Vol. I. (1906), No. II. 23

The point that I wish to bring home to the reader is
that the statement that Mendelism deals with units, and
biometry with masses, is not merely a brief summarising
statement which pleases the mind, but that it has an
actual meaning in relation to the facts : that is to say,
that the limits of these units are not set by the imagina-
tion, but are discovered by experiment.

4 (f). Examples of the confusion between physiological
and statistical Laws.

Having spoken this much on the difference between
physiological and statistical Laws of heredity I propose to
consider a few cases where failure to perceive this difference
has led to confusion. As I have already referred to
Castle's case I will finish with it before I proceed to others:
he says* " The foregoing results show very clearly that
albinism conforms in the mode of its inheritance to
Mendel's law of heredity. The fact, however, must not
be overlooked that a somewhat different explanation of
its inheritance^ has recently been given, based on Galton's
"law of ancestral heredity." I shall not at this time
enter into a detailed discussion of Galton's hypothesis,
which was an entirely rational one in the form in zviiich it
was originally proposed'^ and quite in harmony with the
phenomena oi ga?netogenesis^ as then interpreted. I have
showrx elsewheref by a specific test in the case of mice,
based on the observations of von Guaita, that Galton's
law fails to account for the observed facts^ concerning the
inheritance of albinism, but that Mendel's law does this
perfectly. Nevertheless Darbishire, likewise dealing with
albinism in mice, though admitting that certain of his
results are not in disagreement with Mendel's law, is

* Castle :05, p. 16. t Castle, -.03, p. 231.

24 Darbishire, Lmvs of Heredity.

inclined rather to interpret the phenomena on some such
hypothesis as that of Galtoii!'^

The nume'als refer to the words in italics preceding
them.

1. This shews that Castle followed me in confusing Galton's

Law with the Law of Contribution.

2. Here we see that Castle has started on a right track :

he has perceived that Galton's Law in the sense in
which I used it, meaning the Law of Contribution,
is not the same as that Law as first enunciated.

3. And yet he thinks that in its original form it was in

harmony with the phenomena of gametogenesis as
then interpreted, whereas it seems to me that the
chief characterist c of a statistical Law is that it is
independent of any theory of gametogenesis what-
soever.

4. Of course it does : because it does not attempt to. What

he means is that the Law of Contribution attempts and
fails : and this is quite true.

5. Here again as in (2) we see light breaking in on the

confusion between Galton's Law and the Law of Con-
tribution. Castle sees that tlie theory of heredity I
had in mind is not quite the same as Galton's : I
have shewn (p. 15) exactly how it differs from it.

I will now refer to a case in which the confusion
between Galton's Law and that of Contribution is com-
plete.

In 1904 I wrote:* "/ do not propose to discuss here
the difference^ between Mendelian principles and the
statistical conception of inheritance^ but to consider one
part of the hypothesis put forward by Mendel, which is at
variance with Galton's theory? I refer to the phenomenon
of segregation. We have seen what Mendel says. But
this is flatly contradicted by the Galtonian generalization*^
according to which the greater number of generations a

* Darbishire '.o^b^ p. 9.

Manchester Memoirs, Vel. I. (1906), No. 11.

25

given hybrid is from the first hybrid , the fewer

pure recessive and dominant forms is it likely to produce
when mated with another hybrid of its own generation."
r. For the very good reason that I did not then understand

it : as I do now.
If " Law of Contribution " is put in the place of the terms
2, 3, and 4, this passage is quite true. As it stands,
it is nonsense.

Perhaps, after all, the most complete example of this
confusion is to be found in Castle's writings : it occurs
in his paper on Galton's and Mendel's Laws : he gives
a table * to shew the difference between the Mendelian
and Galtonian prediction of the number of albinos pro-
duced by crossing Japanese waltzing mice with albino,
together with the actual numbers that occurred in von
Guaita's well known experiment.f

I give the top 2 lines of the table.

Generation.

Total Young.

Mendel's Law.

Calculated No.

of Whites.

Observed

No. of
Whites.

Galton's Law.

Calculated No.

of Whites.

II.

28

0

0

14

Substitute Law of Contribution for Galton's Law, and
the idea conveyed by the Table is sensible and true.

I will refer to one more passage in which Galton's
Law is used in the sense of the Law of Contribution :
" Nach Gallons Theorie muss jede Gamete, welche von
einem Individuum produziert wird, imstande sein, alle
Merkmale der Sippe, welcher es angehort, auf die Nach-
kommen zu ubertragen ; es ist unmoglich, dass gewisse
Gameten fUr immer von der Ubertragung gewisser

* Castle :03, p. 231.
t Guaita '98 :oo.

26 DarbisHIRE, Lazvs of Hereaity.

Merkmale ausgeschlossen werden."* Here " Galtons
Theorie " is made to refer not merely to the individual
but to the gamete borne by it, and the expression as here
used means nothing more nor less than the Law of
Contribution. It is true that Galton himself tenta-
tively suggested,! when he formulated his Law, that it
might become applicable to the individual. :J: But his
Law as it stands is a statistical Law true of masses of
units ; and when a physiological theory of heredity, as in
the above quotation, is spoken of as " Galtons Theorie" it
is high time that a new term is invented to describe it : I
have proposed the " Law of Contribution."

Nothing could be more fatal to profitableness of dis-
cussion than that two such profoundly different things as
Galton'.s Law and the Law of Contribution should go by
the same name.

So long as physiological are not clearly distinguished
from statistical Laws of heredity, biologists will continue
to slide from meaning a physiological to meaning a
statistical one : and the transition will be unconscious
because the term by which they denote these two different
things is the same — namely Galton's Law. Progress in
the study of heredity will be slow as long as this confusion
prevails. For so long as it prevails we shall continue
to hear the insensate statement that ancestry makes a
difference. Of course it makes a difference — in the mass;
which it is the business of the biometrician to measure
and of the Mendelian to account for. Anyone who pro-
claims that his results prove that ancestry njakes a
difference, without making it clear whether he has in mind
a physiological or a statistical theory, is drawing a con-
clusion which is meaningless. For his conclusion to have

* Lotsy :o6, p. 152.
t See Appendix A.
:|; Galton '97, p. 403.

Manchester Me JHoirs, Vol. I. (1906), No. H. 27

a meaning he must make this clear. If he is referring to
the former, he is declaring for the Law of Contribution ;
if to the latter, for the Law of Ancestral Inheritance.

When the Mendelian says that ancestry does not
make a difference, he is not denying the validity of the
Law of Ancestral Inheritance but the Law of Diminishing
Individual Contribution. At least I think this is the
correct attitude.

And I cannot bring myself to agree with Bateson
when he says that facts once describeable by Mendel's
Law are permanently removed from the operation [sic] of
the Law of Ancestral Inheritance, unless all that he means
by this statement is that when we have gained this deeper
knowledge of certain hereditary phenomena their further
treatment by the method of the correlation table will
not increase our knowledge of them. I should like to
think that this is all he means : but his writings prevent
me : for he imputes to upholders of Pearson's Law belief
in the Law of Contribution :* yet on the next page he
shews that he has not confused the two, by saying that
the Law of Ancestral Heredity ''does not directly attempt
to give any account of the distribution of the heritage aJiiong
the gametes of any one individual." I do not know
whether Bateson still holds that Mendel's Law is
antagonistic to the Law of Ancestral Inheritance as well
as to the Law of Contribution. If he does, I do not
understand on what grounds. Pearson has investigated
the relation between the two and concludes " that in the
theory of the pure gamete there is nothing in essential
opposition to the broad features of linear regression, skew
distribution, the geometric law of ancestral correlation, etc.
of the biometric description of inheritance in populations.''^

* Bateson, :02, p. 21, second half,
t Pearson, '.O'^b, p. 86.

28 Darbishire, Laws of Heredity.

And no flaws in the argument of my paper on the
supposed antagonism of Mendelian to biometric theories
have been pointed out to me ; in fact, Correns* and
Giard"f* have expressed their agreement with it.

I feel most strongly that so long as we confuse
physiological with statistical Laws of heredity we are
wandering in the dark : we cannot know in what
direction our studies are leading us, whether we are
establishing correlations among the leaves or are digging
among the roots. It may or may not be that what we
learn by the former method is all that we shall ever
know, and that we shall find nothing by our digging ; but
be this as it may, I hold that it is essential to progress
in discovery, no less than to clearness of thought, that
we should know which of the two we are doing.

4 {d). Description of a method of dealing %vith the material
of a breeding experiment in such a way that the
data obtained may be used to test the validity both
of Mendel's and Pearson's Laiv.

When I had finished my last paper on my hybridiza-
tion experiment with mice,| I was still of the opinion
that Mendelian and biometric Laws of heredity were
mutually exclusive, and that if I could discover which of
the two was true, I should be making a forward step in
our knowledge of heredity. I therefore devised an ex-
periment which was destined to settle this question ; and
wasted a year in carrying it out. As soon as I discovered
the true relation of the two Laws I devised a method
of dealing with my experiment, of such a kind that the
results could be utilized by the Mendelian or the bio-

* Correns, :05, p. 43.
t Giard, :0S, p. 22.
Ij: Darbishire, :04^.

Manchester Memoirs, Vol. I. (1906), A*?. 11. 29

metrician to test his own particular Law ; for the stringency
with which the mice were selected in the previous part of
the experiment rendered the results useless for anyone
who wished to test the Law of Ancestral Inheritance by
them.

What was wanted was some device to ensure the
random mating of the mice, and, at the same time to
ensure the possibility of tracing all the ancestors and all
the offspring, in fact, all the relations of every degree of
every individual mouse ; the second condition had been
fulfilled in the previous part of my experiment ; but the
first had not, because the different kinds of mice had been
rigidly selected.

The method by which I mated the mice at random
was very simple. I wrote the catalogue-name of each
mouse on a counter ; then I put the counters representing
female mice into one hat and those representing males
into another : all that remained to be done was to draw
out at random a counter from the ' female ' hat and
similarly one from the ' male ' hat, and to mate the actual
mice represented by these counters. As I have said, this
method enables one to test the Law of Ancestral Inheri-
tance and Mendel's Law. As far as the first is concerned
it is the most perfect conceivable ; but for the second it is
clumsy and involves unnecessary labour : because what is
aimed at in a Mendelian experiment is the discovery of
the properties of character-units, as far as they can be
discovered by determining the specific results of their
union with similar and dissimilar character-units. Now,
some particular combination of characters may turn up
very seldom by the method of random union ; and if one
wishes to discover the result of such a combination one
has to wait until the drawings from the hat give it. One is
in the position of an observer, and if one wishes to

30 Darbishire, Lmvs of Heredity.

attain the knowledge of the result of such a combination
quickly and in large numbers, the random mating and
the counters must be discarded and one must deal experi-
mentally with the material, isolating the individuals the
properties of whose character - units one wishes to
determine.

I had planned at the beginning of last year (1905),
to do the same experiment with peas by mating at
random peas with green round seeds (Eclipse) with peas
with yellow wrinkled seeds (British Queen), and with
each other ; and had already sown the seed ; when it
occurred to me that I need not have done so. We know*
the result of crossing a yellow wrinkled with a
green round pea, and of their mating inter se : so that all
that is necessary is to start with a hat containing equal
numbers of yellow and green counters representing pistil
parents, and a hat with similar contents representing
pollen parents, and to mate the contents at random, the
result of each of the 3 possible unions, ^x^.^xj'.j'Xjr,
being known by previous experiment. And the result of
the matings of the various kinds of offspring can be
predicted from the knowledge, which we have, of their
gametic constitutions. Thus, for example, in F,^ a yellow
resulting from the union yellow x yellowj- will produce
only yellow when mated with green ; but a second yellow
(indistinguishable by outwardly observable features from
the first) produced by the union yellow x green will
produce half yellows and half greens when mated with
green ; while a green of what ancestry soever will always
produce green when mated with green.

Ex Jiypotliesi Mendeliano it is possible to predict the
result of these unions for however many generations

* Hurst :04.
f These colours refer to the gametes.

Manchester Memoirs, Vol. I. (1906), No. \\. 31

through which the experiment is continued, because in its
simple form that Law states that the result, for example,
o^ DRxDR will always be the same whether the mating
of hybrids takes place in F^ or /'V The Mendelian
hypothesis in this simple form may or may not be right ;
and I for one think that it is not. But this does not
damage my argument. My point is that you can deter-
mine the properties of the hybrids in different generations
— supposing that they are not the same in all ; and having
acquired this knowledge you can then return to the
counters and see whether the result of mating your
material at random can be described by the Law of
Ancestral Inheritance. What I want to make clear is
that the knowledge of heredity acquired by the Mendelian
is deeper ; is nearer the phenomena theinselves, than is
that acquired by the biometrician ; and is such that the
latter if he is inclined can use it as material with which
to test the Law of Ancestral Liheritance, without the
labour of conducting a breeding experiment.

Having devised my method, therefore, I discovered
that it was unnecessary to use it. So f abandoned it ; in
the case both of the mice and of the peas. I am now
investigating the properties of the various kinds of
individuals in various generations in both cases, accumulat-
ing information (of a physiological nature) which will be
available to the biometrician for use in testing the Law
of Ancestral Inheritance.

4 {e). Why do zvhite sheep eat more than black ones ?

I was asked the other day this well-known riddle :
and as I had forgotten the answer I was told it : " Because
there are more of them."

The supplying of the answer never provokes a laugh,
yet the relation between it and the question is full of

32 Darbishire, Lazvs of Heredity.

interest. Let us discuss it. When you ask the riddle
you do not say that you are not referring to individual
white and black sheep, but the man of whom the riddle
is asked invariably thinks that you are : in attempting to
answer it, the ideas that rush through his mind may
either take the form of seeking for some pun on the
words or perhaps for some humorous quotation in which
they appear ; and so forth : or, what usually happens, he
thinks that as a matter of fact a white individual does eat
more than a black, and (if he is a biologist) he may be
trying to think of some physiological explanation of the
fact, in connection possibly with the well-established
relation between pigmentation and the getting rid of
waste products.

In the answer he is told that the amount eaten by
the sum-total of white sheep as compared with that eaten
by the sum-total of black sheep is the subject under
discussion ; and not any peculiarities of ingestion, diges-
tion, or egestion associated with whiteness as compared
with blackness.

If the antithesis between truths about masses, and
truths about individuals which constitutes the point in this
riddle were more widely and more clearly perceived than
it is to-day, there would no longer be that confusion in
the minds of most biologists which prevents them
seeing the profound difference that exists between a
physiological Law like Mendel's, which is true of units, and
a statistical one like the Law of Ancestral Inheritance,
which is true of masses. All intending students of heredity
should be asked this riddle; and if they cannot detect the
fallacy in it they should be declared unfit for their intended
task.

The similarity between the impression called into
existence in the mind by asking the question and Men-

Manchester Meinoi}-s, Vol. I. (1906), No. 11. 33

delism ; and that between the idea conveyed by the
answer and the Law of Ancestral Inheritance does not lie
only in the fact that while the question and Mendelism
deal with individuals, the answer and the Law of Ancestral
Inheritance refer to masses. The idea implied in the
question is like Mendelism, because it suggests what
Mendelism effects, the discovery of a hitherto unsuspected
order in familiar phenomena ; while the truth conveyed
in the answer is like the biometric treatment of heredity,
because it is the accurate statement of a relationship that
you already know to exist. Everyone knows that the
sum-total of children are more or less like the sum-total
of their parents ; the biometrician accurately measures
the degree of this resemblance.

The answer you expect is physiological. The answer
you get is statistical.

34 Darbishike, Lazvs of Heredity

APPENDIX A to p. 26.

It is interesting to enquire what Galton himself said when he
formulated his Law, on the subject of its applicabihty to individual
cases. He said ('97, p. 403) "It should be noted that nothing
in this statistical law contradicts the generally accepted view that
the chief, if not the sole, line of descent runs from germ to germ
and not from person to person. The person may be accepted
on the whole as a fair representative of the germ, and, being so,
the statistical laws which apply to the persons would apply to
the germs also, though with less precision in individual cases.
Now this law is strictly consonant with the observed binary sub-
divisions of the germ cells, and the concomitant extrusion and
loss of one-half of the several contributions from each of the
two parents to the germ-cell of the offspring." Mark his words,
" . . . . though with less precision in individual cases " — the
itaUcs are mine. If one were referring to Gallon's Law (in the
form in which it is true of masses only) one would say, " . . . .
without applying at all to individual cases"; and if to the Law
of Contribution, " . . . . with absolute precision to individual
cases." But I may be interpreting this wrongly, for the " less "
may refer not to the difference between population and individual,
but to the difference between person and germ. And, in fact, I
think the following quotation from the previous page ('97, P- 402)
justifies us in concluding that Galton conceived his Law as being
true solely of masses without being true of the component
individuals. " The neglect of individual prepotencies is justified
in a law that avowedly relates to average results ;...." At
any rate it simplifies matters very much to consider that Galton's
Law as he formulated it is true of masses only, and not of their
component units ; for if we do not, we have to keep three laws
distinct in our minds.

I. Galton's Law as he formulated it : true of masses, but
also, though with less precision, of individuals. Statistical
and Physiological.

Manchester Memoirs^ Vol. I. (1906), No. 11. 35

2. Galton's Law : true of masses only. Statistical.

3. The Law of Contribution : true of units. Physiological.

APPENDIX B to p, 5.

There is nothing, of course, in the word 'contribute' to
definitely signify that the thing which is contributed is the same
as that which contributes : in fact, in the everyday usage of the
term this is hardly ever the case. But it is reasonable to hold
that Galton's Law is the generalisation that like contributes like
and not unlike : and it is certain that Galton himself meant this,
as the last words of his illustration of particulate inheritance
readily shew " . . . . each piece of the new structure is derived
from a corresponding piece of some older one, as a lintel derived
from a lintel, a column from a column, a piece of wall from a
piece of wall." ('89, p. 8).

APPENDIX C to A {[>) pp. 19— 23.

There is an apparent paradox, in the ideas just expressed,
about which I think it is necessary to say a few words, in case
the reader should detect it himself and think that it had not
occurred to me.

I have said that biometry deals with masses and Mendelism
with units ; but I have also said that the biometrician exceeds
his proper limits when he goes beyond the boundary of a unit,
while the Mendelian is concerned with the mutual properties of
numerous units : in other words the sphere of the biometrician
is within the unit while that of the Mendelian is outside it.

36 Darbishire, Laws of Heredity.

The fundamental idea on which the Law of Ancestral In-
heritance is based is that set forth in the quotation from Pearson
on p. 4 of this essay ; it is that a knowledge of the characters of
the parents does not enable us to predict the character of the
offspring in individual cases.

The fundamental idea in Mendelian theory is that the
ascertainable gametic characters of the parents do enable us to
predict the character of the offspring in individual cases.

How are these two diametrically opposite ideas about
heredity to be reconciled? The answer which most naturally
suggests itself is that the biometrician happens to have dealt
with cases about which it was impossible to predict in individual
cases ; while the Mendelian happens to have dealt with cases in
which prediction was possible. This answer presupposes the
existence of two sets of phenomena in heredity, those about which
it is possible to predict, and those about which it is not. Now
let us grant for the moment that the Mendelian theory (which I
think by no means proven yet) that the characters of an organism
consist of a number of separate character-units is true. What
relation, if any, do the two sets of hereditary phenomena — the
predicable and the non-predicable bear to these units? Just
this. The non-predicable phenomenon is the incomplete
correlation between the degree in which any character x is
exhibited by a parent, in a single case, and the degree in which
that same character is exhibited in its child. The predicable
phenomenon is the result of the union of x with x., or of x with y.

Chemistry furnishes a parallel. The chemist cannot predict
the rate at which any given atom in a litre of oxygen is travelling ;
he can only deal with 'statistics of average conduct'; but he can
predict the result of passing an electric spark in a vessel con-
taining oxygen and hydrogen. Yet he who deals with the
properties of the elements may be said to deal with units, and
he who deals with the component atoms — and one can only
deal with them in large numbers — may be said to deal with
masses.

It is true that the biometrician possesses the only means of

Manchester Memoirs, Vol. I. (1906), No. 1|. 37

measuring the "intensity of heredity" in a non-predicable case,
but it seems to me that to extend the appHcation of these means
to predicabJe cases is fallacious. If the true function of the
biometrician is to give us statistics of average conduct where we
cannot predict individual conduct it seems to me that to deal by
the biometric method with cases where we can is not only un-
profitable, but likely to lead men to think that where there are
two methods dealing with the same material of which the one
can predict while the other cannot, the latter is fallacious.
Whereas if the biometrician confined himself to the non-
predicable and the Mendelian to the predicable, the general
conclusion would be that each had his proper sphere— which
indeed, in my belief, he has. I do not set forth these views in
any spirit of dogmatic certainty ; and nothing could please me
less than that they should go unchallenged by anyone who
believes me to be mistaken.

LITERATURE REFERRED TO IN THE TEXT.

'08. Davy, Humphry. " On some new Phenomena of chemical
Changes produced by Electricity, particularly the De-
composition of the fixed Alkalies, and the Exhibition
of the new substances which constitute their bases ; and
on the general Nature of alkaline Bodies." Phil. Trans.,
vol. 98, p. I.

'89. Galton, Francis. "Natural Inheritance," Macmillan& Co.

'95. CouTAGNE, G. "Recherches sur le Polymorphisme des
MoUusques de France." Lyon.

'97. Galton, Francis. "The average Contribution of each
several Ancestor to the total Heritage of the Offspring."
Proc. Roy. Soc, vol. 61, pp. 401-413.

'98. GuAiTA, G. VON. " Versuche mit Kreuzungen von verschie-
denen Rassen der Hausmaus." Ber. d. naturf. Gesell. Frei-
burg, vol. 10, p. 317.

:00. . 2** Mittheilung, etc., ibid. vol. 11, p. 131.

38 Darbisiiire, Laivs of Heredity.

:02. Bateson, W. " Mendel's Principles of Heredity." Univer-
sity Press, Cambridge.

:03. Castle, W. E. " The Laws of Heredity of Galton and
Mendel, and some laws governing Race improvement by
selection." Proc. Amer. Acad. Arts and Sci.y vol. 39, no. 8.

■.03. JoHANNSEN, W. " Ucber Erblichkeit in Populationen und
in reinen Linien." Jena. Verlag von Gustav Fischer.

:03«. Pearson, Karl. "The Law of Ancestral Heredity."
Biometriha, vol. 2, p. 211.

:03^. . "Mathematical Contributions to the Theory of

Evolution. Xn. On a Generalized Theory of Alternative
Inheritance, with special Reference to Mendel's Laws."
Phil. Trans., sen A., vol. 203, pp. 53-86.

:04a. Darblshire, a. D. " On the Result of crossing Japanese
Waltzing with Albino Mice." Biomefrika, vol. 3, p. i.

■.Qt^b. . "On the Bearing of Mendelian Principles of

Heredity on current Theories of the Origin of Species."
Manchester Afemoirs, vol. 48, no. 24, 19 pp.

:04. Hurst, C. C. "Experiments in the Heredity of Peas."
Journ. R. Hort. Soc, vol. 28, pp. 483-494.

:04. Lang, Arnold. " Ueber Vorversuche zu Untersuchungen
iiber die Varietatenbildungen von Helix hortensis Miiller
und Helix nemoralis L." Festschrift zum siebzigsten
Geburtstage von Ernst Haeckel, p. 439. Jena.

:04. Pearson, Karl. " On the Laws of Inheritance in Man."
II. " On the Inheritance of the mental and moral charac-
ters in man, and its comparison with the inheritance of
the physical characters." Biometrika, vol. 3, p. 131.

:05. Barrington, Amy; Lee, Alice; and Pearson, Karl.
" On the Inheritance of Coat-Colour in the Greyhound."
Biometrika, vol. 3, p. 245.

:05. Castle, W. E. " Heredity of Coat Characters in Guinea-
Pigs and Rabbits." Pub. by the Carnegie Institution of
Washington.

:05. Correns, C. " tJber Vererbungsgesetze." Berlin. Verlag
von Gebriider Borntraeger.

Manchester Memoirs, Vol. I. (1906), No. W. 39

:05«. Darbishire, A. D. " On the Supposed Antagonism of
Mendelian to Biometric Theories of Heredity." Man-
chester Memoirs, vol. 49, no. 6, 19 pp.

:0^b. . " Professor Lang's Breeding Experiments with

Helix hortensis and H. nemoralis ; an Abstract and
Review." Jourti. of Concholo}:,y, vol. 11, p. 193.

:05. Fruwirth, C. " Die Ziichtung der landwirthschafdichen
Kulturpflanzen," vol. i. "Allgemeine Ziichtungslehre,"
2'" Aufl.

•.05. GiARD, A. " L'Evolution des Sciences biologiques."
Congres de l' Association Fran^aise pour f Avanctment des
Sciences. Cherbourg.

:05. PuNNETT, R. C. " Mendelism." Macmillan and Bowes,
Cambridge.

:06. Fruwirth, C. " Die Ziichtung der landwirthschaftlichen
Kulturpflanzen," vol. 3.

:06. Lang. Arnold. " Uber die Mendelschen Gesetze, Art —
und Varietatenbildung, Mutation und Variation, insbeson-
dere bei unsern Hain — und Gartenschnecken." Verhandl.
der Schweiz. Naturf. Gesell.

:06. LoTSY, J. P. "Vorlesungen iiber Deszendenztheorien mit
besonderer Beriicksichtigung der botanischen Seite der
Frage " [especially Lectures 8, 9, 10 and iij. Jena.
Gustav Fischer.

:06. Schuster, E. " On Hereditary Deafness : A Discussion
of the Data collected by Dr. E. A. Fay in America."
Biometrtka, vol. 4, p. 465.

:o6. Weldon, W. F. R. " Inheritance in Animals and Plants,"
pp. S1-109, in "Lectures on the Method of Science."
Edited by T. B. Strong, Clarendon Press, Oxford.

40 Darbishire, Laws of Heredity.

Further List of Articles, d^c. (dealing with Heredity from the

Statistical and Physiological Aspects), ivhich have appeared

during and since igo4 and not included in the list at the end

of my paper 'Si\b, in the preceding list.

103- Yule, G. Udny. " Professor Johannsen's Experiments in

Heredity: A Review." Netv Fhytologist, vol. 2, No. 10.

[Tills paper sliould liave been in tlie list at the end of my

first paper to this Society. Darbishire, '.O^b.^
:04. Allen, G. M. "The Heredity of Coat Color in Mice."

Froc. Amer. Acad. Arts and Sci., vol. 40, no. 2.
:04. Bateson, W., Saunders, E. R., and Punnett, R. C.

" Reports to the Evolution Committee of the Royal

Society," Report 2.
:0^a. Bateson, W. " Presidential Address to Section D." Brit.

Assoc. Reports, Cambridge, 1904, p. 574.
:Q\b. " Albinism in Sicily : A Correction." Biometrika,

vol. 3, p. 471-
;04. BiFFEN, R. H. "An 'Intermediate' Hybrid in Wheat."

Brit. Assoc. Reports, Cambridge, 1904, p. 593.
:04<2. Correns, C. " Experimentelle Untersuchungen iiber

die Gynodioecie." Ber. Deutsch. Bot. GeselL, vol. 22,

part 8, p. 506.
'.Q^b. . " Ein typisch spaltender Bastard zwischen einer

einjahrigen und einer zweijahrigen Sippe des Hyoscyamus

niger." Ber. Deutsch. Bot. GeselL, vol. 22, part 8, p. 517.
:04. CuENOT, L. " Un paradoxe hereditaire chez les Souris.

Reunion biolngique de Nancy, p 1050, seance du 13 Juin.
:04. Darbishire, A. D. " On the Result of crossing Japanese

Waltzing with Albino Mice." Brit. Assoc. Reports, Cam-
bridge, 1904, p. 591.
:04. DoNCASTER, L. " On the Inheritance of Tortoiseshell and

Related Colours in Cats." Froc. Cambridge Fhilos. Soc,

vol. 13, pt. I.
:04. Durham, F. M. " On the Presence of Tyrosinases in the

Skins of some Pigmented Vertebrates." Froc. Roy. Soc,

vol. 74, p. 310.

Manchester ATemoirs. Vol. I. (1906), No. II. 41

•.0/[a. Hacker, V. " Uber die neueren Ergebnisse der Bastaid-

lehre, ihre zellengeschichtliche Bedeutung und ihre Bedeu-

tung fiir die praktische Tierzucht. Arch. f. Rassen- und

Gesellschajts-Biologie, Jahrg. i, part 3.
'.Q^b. . " Bastardierung und Geschlechtszellenbildung."

Zool. [ahrb., Suppl. 7, Festschr. f. Weismann.
:04a. Hurst, C. C. " Mendel's Discoveries in Heredity." Trans.

Leicester Lit. and Phil. Soc, vol. 8, pt, 2. [Contains a

useful list of references.]
:Onb. . " Experiments on Heredity in Rabbits." Brit. Assoc.

Reports, Cambridge, 1904, p. 592.
:04. Le Dantec, Felix. " Les Influences ancestrales." Ernest

Flammarion. Paris.
:04. Lock, R. H. " Experiments on the Behaviour of Differen-
tiating Colour-characters in Maize." Brit. Assoc. Reports,

Cambridge, 1904, p. 593.
:04. MoENKHousE, W. T. J. "The Development of Hybrids,"

Amer. Journ. of Anat., vol. 3.
:04. NooRDUijN, C. L. W. " Over Erfelijkheid en Verandering

der Kleuren. Album der Natmir, August.
:04. Pearson, Karl. " Note on Mr. Punnett's Section on the

Inheritance of Meristic Characters." Biometrika, vol. 3,

P- Z^i-
:04. Petrunkewitsch, A. " Gedanken iiber Vererbung." Frei-

burgi. Br. Speyer & Kaerner.
:04. PuNNETT, R. C. " Merism and Sex in ' Spinax Niger.' "

Biometrika, vol. 3, p. 313.
:04. Raynor, G. H., the Rev., and Doncaster, L. "Note on

Experiments on Heredity and Sex-determination mAbraxas

grossiilariata.'" Brit. Assoc. Reports, Cambridge, 1904,

P- 594-
:04. Rosenberg, C. " Uber die Tetradenteilung eines Drosera-

Bastardes." Ber. Deutsch. Bot. GeselL, vol. 22, part i,

p. 47.
:04. Saunders, E. R. " Heredity in Stocks." Brit. Assoc.

Reports, Cambridge, 1904, p. 590.

42 Darbishire, Laws of Heredity.

:04. Staples-Browne, R. " Experiments on Heredity in Web-
footed Pigeon." Brit. Assoc. Reports.^ Cambridge, 1904,

P- 595-
:04. TscHERMAK, E. " Weitcre Kreuzungsstudien an Erbsen,

Levlcojen und Bolinen. Zeiischr. f. d. landw. Versuchs-

wesen in Oest, 1^04.
:04. WoLTERSTORFF, W. " Tritoii Blasii und die Mendel'schen

Regeln." Comptes rendiis du 6^ Congres ititernational de

Zoologie, Session de Berne.
:04. . " Triton blasii de Plsle^ ein Kreuzungsprodukt

zwischen Triton marmoratus und Tr. cristatus.'''' Zooi.

Aiiz., vol. 28, Sept.
:05. Bateson, W. and Punnett, R. C. "A Suggestion as to

the nature of " walnut " comb in Fowls." Froc. Cambridge

Philos. Soc, vol. 3, pt. 3, p. 165.
:05. Bateson, W., and Gregory, R. P. " On the Inheritance of

Heterostylism in Primula." Proc. Roy. Soc, B. vol. 76,

p. 581.
:05. Bateson, W., Saunders, E. R., and Punnett, R. C.

" Further Experiments on Inheritance in Sweet Peas and

Stocks : Preliminary Account." Proe. Roy. Soc, B. vol. 77,

p. 236.
:05. Castle, W. E. " Recent Discoveries in Heredity and their

Bearing on Animal Breeding." Pop. Sci. Monthly, July.
:05. CoRRENS, C. An edition of " Gregor Mendel's Briefe an

Carl Nageli." Abh. d. math. phys. Kl. d. Kon. Sachs.

Geseil. d. IVissensch., vol. 29, no. 3.
:05. Cu^not, L. " Les Races pures et leurs Combinaisons

chez les Souris " (4'"'' Note). Arch, de Zool. exp. et. gen.

[4], vol. 3, Notes et Revue, no. 7, p. cxxiii.
:05- DoNCASTER, L. "On the Inheritance of Coat-colour in

Rats." Proc. Cambridge Philos. Soc, vol. 13, pt. 4, p. 215.
:05. Gai.ippe, V. " L'Heredite des Stigmates de Degenerescence

et les Families souveraines." Paris, Masson et C'^
:05. Hatschek, B. " Hypothese der organischen Vererbung."

Leipzig, W. Engelmann.

Manchester Memoirs^ Vol. I. (1906), No. 11. 43

:05. Hertwig, Oscar. " Ergebnisse und Probleme der

Zeugungs- und Vererbungslehre." Jena. Gustav Fischer.
:05. Hurst, C. C. " Experimental Studies on Heredity in

Rabbits." Journ. Linn, Soc. — Zoology, vol. 29, p. 283.
:05. Farabee, W. C. " Inheritance of Digital Malformations

in Man." Papers of the Peabody Mus. of Amer. Archceol.

and Etlmol.., Harvard University, vol. 3, no. 3.
:05. Farmer, J. B., and Moork, J. E. S. "On the Maiotic

Phase (Reduction Divisions) in Animals and Plants."

Quart. Journ. Micr. Sci., vol. 48, pt. 2.
:05. Lock, R. H. "Studies in Plant Breeding in the Tropics,"

Annals Roy. Bot. Gard., Peradeniya, vol. 2, pt. 3, p. 357.
:05. MacDougal, D. T. (assisted by Vail, A. M., Shull, G. H.,

and Small, J. K.). " Mutants and Hybrids of the Oeno-
theras." Pub. by the Carnegie Inst, of Washington.
:05a. Morgan, T. H. " The Assumed Purity of the Germ Cells

ill Mendelian Results." Science, N.S., vol. 22, pp. 877 — 879.
:05^. • " Ziegler's Theory of Sex Determination, and an

alternative Point of View." Science, N.S., vol. 23, no. 573,

PP- 839—841.
:05. NooRDUijN, C. L. W. " Die Farben-und Gestalts-Kanarien."

Magdeburg, Creutz'sche Verlagsbuchhandlung.
:05. Schuster, E. H. J. " Results of crossing Grey (House)

Mice with Albinos." Biometrika, vol. 4, p. i.
:05. Strasburger, Eduard. " Die stofifiichen Grundlagen der

Vererbung in organischen Reich." Gustav Fischer, Jena.
:05. Yule, J. Udny. " On the Influence of Bias and of

Personal Equation in Statistics of Ill-defined Qualities : An

Experimental Study." Proc. Roy. Soc, A., vol. 77, p. 337.
:05. ZiEGLER, H. E. "Die Vererbungslehre in der Biologic."

Gustav Fischer, Jena.
:06- Barrington, A., and Pearson, K. "On the Inheritance

of Coat-Colour in Cattle. Part I. Shorthorn Crosses and

Pure Shorthorns." Biometrika, vol. 4, p. 427.
:o6. Bateson, W. " Albinism in Sicily." Biometrika, vol. 4,

p. 231.

44 Darbishire, Lmvs of Heredity.

:06. Davies, C. J. " Horns in Polled Cattle : a possible
Explanation." The Estate Magazine., for March, p. 99
[a popular article].

:06. Galton, Francis, and Schuster, Edgar. " Noteworthy
Families (Modern Science)." London. John Murray.

:06. Heider, Karl. " Vererbung und Chromosomen." Gustav
Fischer. Jena.

:o6. Hewitt, C. G. " The Cytological Aspect of Partheno-
genesis in Insects." Manchester- Memoirs, vol. 50, no. 6,
40 pp., 2 pis. [Gives a very full list of papers bearing
on the subject of parthenogenetic inheritance.]

:06fl. Hurst, C. C. "Notes on the Proceedings of the Inter-
national Conference on Plant Breeding and Hybridisation,
\<)02 " Journ. R. Hort. Soc., vol. 29, pt. 4.

■.06b. . " On the Inheritance of Coat Colour in Horses.''

Proc. Roy. Soc, B., vol. 77, p. 388.

:06. Pearson, K., and others. " Cooperative Investigations on
Plants," " III. On Inheritance in the Shirley Poppy."
Second Memoir. Biometrika, vol. 4, p. 394.

:06. Reid, G. Archdall. " On Mendel's Laws, and the Muta-
tion theory of Evolution," being Appendix B to the 2nd
Ed. of his " Principles of Heredity." London : Chapman
and Hall.

:06. Saleeby, C. W. "Heredity." London. T. C. & E. C. Jack.

:06. ScHiMKEWiTSCH, W. " Die Mutationslehre und die
Zukunft der Menschheit." Biol. Centralbl. vol. 26.

:o6. Staples-Browne, R. "Note on Heredity in Pigeons."
Proc. Zool. Soc, 1905, vol. 2.

:06. Wilson, E. B. " Mendelian Inheritance and the Purity of
the Gametes." Science, N.S., vol. 23, no. 577, pp. 112-113.

:06. Woods, F. A. " Mental and Moral Heredity in Royalty."
New York. Henry Holt & Co.

:06. Yule, G. Udny. " On a Property which holds good for
all Groupings of a Normal Distribution of Frequency for
Two Variables, with Applications to the Study of
Contingency-Tables for the Inheritance of Unmeasured
Qualities." Proc. Roy. Soc, A., vol. 77, p. 324.

Manchester Memoirs, Vol. I. (1906), No. \%.

XII. Notes. — On an allotropic form of Arsenic

and

On the Estimation of Arsenic when in Minute
Quantities.

By William Thomson, F.R.S.E., F.I.C.

Read Alarch ijth, igo6- Received for ptibli cation June 2bth, J gob.

I wish to call attention to a peculiar allotropic form
of arsenic which is produced when that substance is
sublimed and rapidly condensed in vacuo or in presence
of an inert gas such as hydrogen or carbon dioxide.
When in small quantities it condenses as a white film
which rapidly becomes black. This blackening is brought
about almost instantaneously by the light from a burning
magnesium wire, or by bright daylight, but it takes place
rather rapidly by ordinary gaslight, and the whiteness is
not preserved when the film is kept in the dark. It
continues white for a longer time when kept at the
ordinary temperature of the atmosphere than when heated
to, say, lOO'^C. It becomes black in about 20 minutes
when cold and in the dark, and in about five minutes in
diffused light.

A platinum wire covered with a thin layer of glass
upon which ordinary arsenic was laid was enclosed in a
glass bulb. A vacuum was made in the bulb by a
Topler vacuum pump till no more air could be extracted,
and left 24 hours over phosphoric anhydride. The pump
was again worked and proved that no leakage of air into
the bulb had taken place. A current of electricity was
then passed through the wire to heat it to redness, and so

August 1 4th, igo6.

2 Thomson, Notes on Arsenic.

evaporate some of the arsenic, which condensed on the
top of the bulb as a white film, and rapidly became black
as it did when sublimed in an atmosphere of hydrogen
or of carbonic anhydride.

It occurred to me that this blackening action in the
dark may be due to some obscure radiations which are
everywhere present, and some experiments were made
by enclosing these films in thick lead tubes, but they
appeared to blacken when so enclosed just as rapidly as
when enclosed in an ordinary wooden pencil case. Never-
theless these white films may prove of interest to the
physicist, as some means may be found for preserving
their whiteness, and so making it possible to determine
the presence of certain radiations as yet unknown.

On placing tubes containing these mirrors in liquid
air they remained white for 5 hours, during which time
they were immersed. The test mirror which was momen-
tarily withdrawn from the liquid air from time to time
became blackened, whilst the others which were immersed
and in the dark, remained white.

I find that this form of arsenic has been studied by
several observers,* but more especially by Hugo Erdmann
and Max von Unruh who find that it is more volatile than
the black form (which our observations confirm). They
say — it is soluble in carbon disulphide, and whilst in solu-
tion it is not affected by light, but may be recovered as a
yellow deposit on evaporation of the solvent, and it is then
very sensitive to all kinds of light. On being left in solution

*Schuller, Math. ti. naturit. Ber. aus Ungarn, 1S89, vol. 6, p. 94.

Retgers, Zeit. anorg. Chem., vol. 4, p. 403 — 409, and vol. 6, p.
397—320.

McLeod, Chemical Ne^vs, 1894, vol. 70, p. 139.

Linck, Zeit, Kryst. Min., 1896, vol. 26, p. 280.

Erdmann & Max von Unruh, Zeit. anorg. Chem., 1902, vol. 32, p.
437—452.

Manchester Memoirs, Vol. I. (1906), No. \%. 3

in a stoppered bottle a reddish form of arsenic is pro-
duced which gradually falls from solution.

Improvement in the cooling method for co7uiensing the
arsenic mirrors in arsenic determinations.

The results obtained by the method which I formerly
adopted of running a stream of cold water over a piece of
tissue paper covering the drawn out portion of the tube,
leave something to be desired, especially in the estimation
of exceedingly minute quantities of arsenic, for the water
at the edge of the paper next the flame does not give an
exceedingly sharp line of cooling surface, and we have
improved on this by taking a piece of block tin tube
8 inches long, having an internal bore of je^hs inch.
This tube is flattened so that the internal measurements
become \ inch x |th inch, and it is bent downwards in a
curve about f inch from the end. Two holes are made
in the edges of the flattened tube opposite each other
2 inches from the bent end, and a thin silver-foil plate
soldered over one of these holes. This plate is perforated
by drilling in it a hole of a diameter sufficient to allow
the drawn out portion of the largest sized glass tube to
fit it. The tube is passed through the silver plate to the
point at which the mirror is to be formed, and the water
turned on through the tin tube, which is connected with
the main by a thin rubber tube. A gentle flow of water is
kept running through this tube, and it is found that if
the length of the bent end of the tin tube is properly
adjusted the water does not escape at the silver plate or
the hole in the tin tube opposite, even when the tubes fit
comparatively loosely. By this device o-(/ooth of a grain
per gallon of arsenic trioxide when working on 50 c.c. of
the solution can be readily seen as a distinct narrow

4 Thomson, Notes on Arsenic.

metallic ring, and the larger mirrors are more evenly
deposited.

Influence of nitrous compounds on the estimation of
minute quantities of arsenic.

I observed in making tests by the electrolytic as
compared with the Marsh-Berzelius method that in some
cases considerably larger mirrors were obtained by the
latter than by the former, and on investigation it turned
out that this difference was due to the fact that all the
nitrous compounds used for the destruction of the organic
matter had not been removed. To be certain of the
results by the electrolytic process it becomes necessary
that the sulphuric acid should be evaporated to its fuming
point, diluted with twice its volume of water and again
evaporated to the fuming point, and this should be again
repeated to ensure the elimination of all of the nitrous
compounds, so that the proper size and depth of arsenic
mirrors may be obtained by the electrolytic process in the
time allowed for the test. With the Marsh-Berzelius
method the presence of some nitrous compounds does
not so materially affect the result.

With a view of studying the influence of nitric acid
in the electrolytic apparatus, a dilute solution equivalent
to half a c.c. of the ordinary strong acid of commerce
(0'48 grm. HNO3) was added separately to an electrolytic
apparatus — the first with a lead, the second with
a zinc, and the third with a graphite kathode. This
acid was gradually converted into ammonia in each
apparatus, the amount of which was then determined.
A current of 3 amperes was employed, and each apparatus
was worked in series. After 45 minutes the ammonia
in the solutions in the various kathode chambers was
determined, and the following results obtained : —

Manchester Memoirs, Vol. I. (1906), No, 13. 5

HNO3 reduced to
NH3 in 45 minutes.

*Lead kathode ... ... ... ... 0*22 grm.

Zinc ,, ... ... ... ... o"2i ,,

Graphite „ ... ... ... ... o'lo ,,

This shews that the rate of decomposition was about
the same with lead and zinc kathodes, whilst the graphite
kathode only exerted about one half the reducing action.

Influence of \ c.c. HNO on the detection of arsenic in the
form of AS2O5, by the electrolytic apparatus when
using kathodes of different substances.

A solution was prepared containing O'O0O,O0O,83 grms.
of AsjOg in each cubic centimetre, i c.c. is equivalent to
yjjVgth of a grain of As^Oe per gallon when using 50 c.c.
of the liquid; 10 c.c. of this solution were mixed with
\ c.c. of HNO3 and introduced into the apparatus, using
kathodes of different substances. The kathodes used
were graphite, lead, cadmium, zinc, and iron.

After the current (3 amperes) had run for 45 minutes
the arsenic mirrors obtained were approximately equiva-
lent to the following c.c. of the standard solution : —

(The mirrors obtained when HNOa is present are very
irregular, and only admit of an approximate estimation.
In each case the number of c.c. is given as the nearest
whole number, and the figures represent averages from a
number of experiments.)

c.c. of standard
solution.

Graphite ... ... ... ... ... i"o

60

tLead
Cadmium..
Zinc
Iron

2"0
4-0
2"0

• This lead kathode was made and kindly given to me by Mr. J. E.
Hackford, of Nottingham. The lead u.sed having been purified from traces
of arsenic and antimony by him by fusing with sodium, and afterwards stirring
up with pure fused sodium chloride while the lead is still at a high temperature.

t Another lead kathode made by ourselves gave practically the same
result, although the results shewed some difference.

Thomson, Notes on Arsenic.

It is remarkable that the arsenic acid was reduced by
each of the different kathodes to the condition of arseniu-
retted hydrogen in presence of nitric acid or some nitrous
compounds, some of which remained intact after the
experiment, and that lead was the most efficient.

Influence nf different kathodes on arsenious acid.

For this series of experiments a standard solution
was prepared containing 0000,000,715,4 grm. of AS4O6
per c.c. I c.c. is equivalent to yoVoth of a grain per gallon
when working on 50 c.c. of the solution, and 10 c.c.'s of
the standard solution were used for each experiment, which
was carried out during 45 minutes, in each case with a
current of 3 amperes.

The mirrors obtained during 45 minutes were equiva-
lent to the following c.c. of the 10 c.c. of the standard
solution taken — the mirrors obtained with a pure zinc
kathode during 45 minutes being taken as the standard : —

Graphite

Lead

Cadmium

Zinc

Iron

c.c. of standard
solution.

9
10

3
10

Influence of different kathodes on arsenic acid solution.

Each c.c. of the standard solution used contained
0"000,ooo,83 grm. AsoOj (=TTro7th of a grain of arsenic
trioxide per gallon when using 50 cc).

The experiments were made as before with 10 c.c. of
standard solution, and the time and electrical current as
above mentioned. The resulting mirrors obtained were
as follows in equivalents of c.c. of the standard arsenic
trioxide solution :

Manchester Memoirs, Vol. I. (1906), No. 1*^. 7

C.C. of standard
solution.

Graphite ... ... ... ... ... i

Lead ... ... ... ... ... 9*0

Cadmium... ... ... .. ... 2*5

*Zinc ... ... ... ... ... 9*o

Iron ... ... ... ... ... 5"o

The iron electrode was made by wrapping closely
round a glass tube pure fine iron wire, such as is used
for making standard solutions.

Part of the arsenic seems to combine with the iron
and to form such a combination that it is not eliminated
as AsHg when the current is passed. When iron which
has previously been used in an arsenic test is used
immediately afterwards with acid free from arsenic, it
gives a perfect blank after working for 45 minutes, but
on being left till the surface oxidises, and then repeating
the test with pure acid, a mirror of arsenic is obtained.

Insensitive zinc in the Marsh- Berzelius apparatus.

When zinc contains small quantities of iron, etc.»
it becomes insensitive when used in the Marsh-Berzelius
apparatus, i.e., it gives off hydrogen free from arseniu-
retted hydrogen, and even when minute quantities
of arsenic are added it still fails to produce a mirror.
Chapman and Law {^Analyst, January, 1906, vol. 31,
p. 358), say that it is due to the hydrogen being given off
at a lower electrical supertension, owing to the metal,
existing as an impurity, being capable of liberating
hydrogen at such lower supertension. They say, however,
if 2 grammes of cadmium sulphate be added to the
solution, that the insensitive zinc becomes sensitive. It
seems remarkable that a metal such as cadmium which
liberates hydrogen at a lower supertension than zinc

* In working with As.iO^ the 9 parts would be represented by the
standard mirrors as 10 for that time of the experiment.

8 Thomson, Notes on Arsenic.

should have the effect of doing away with the influence
of the iron ; in other words that the deposition of
cadmium in a spongy condition on the zinc (which is the
metal dissolved by the acid) should have the effect of
raising the supertension at which the hydrogen is evolved,
whilst the iron still remains and must presumably act as
it did before the addition to it of a spongy covering. I
have repeated his experiment with a sample of electro-
lytic zinc prepared by Brunner, Mond & Co., which
contains a small proportion of metallic iron. It retains
minute quantities of arsenic added to the sulphuric acid
used in the Marsh-Berzelius apparatus. We added
2 grammes of cadmium sulphate to this zinc, as recom-
mended by Chapman and Law, and tried experiments by
introducing minute quantities of arsenic into the apparatus,
with and without cadmium sulphate, but we failed to find
that the cadmium sulphate made any difference. The
cadmium was thrown down on the zinc, and the evolution
of hydrogen much reduced thereby, but in our hands it
remained as insensitive as it was before, although the
strength of the sulphuric acid was increased so as to give
about the same flow of hydrogen.

Chapman and Law mention, as I previously found,
that magnesium is insensitive. I endeavoured to obtain
magnesium free from arsenic and antimony, but failed.

1 took, however, the sample which contained the smallest
quantity of arsenic, and made the test by dissolving com-
pletely I "4 grammes of magnesium. The experiment was
repeated with the addition to it of 0'2 gramme cadmium
sulphate. In the latter case a slightly larger mirror was
obtained than in the former. 4 c.c. of the standard
arsenic trioxide solution was then added with 1-4 grammes
of magnesium, and a mirror obtained equivalent to from

2 to 2^ c.c. On repeating the experiment with the addi-

Manchester Memoirs, Vol. I. (1906), No. VI. 9

tion of 0'2 gramme cadmium sulphate, a mirror equivalent
to 4 c.c. was obtained.

I dissolved 0'5 gramme of nickel in 100 grammes of
pure zinc, and tried this alone ; it gave no mirror. I then
added 4 c.c. of standard As^Og solution, and it gave a
mirror equal to that obtained when pure zinc is employed
in the manner previously recommended by me.* On
repeating this with the addition of 0'2 gramme of cadmium
sulphate a mirror of about the same size was obtained.
Even I c.c. of the standard solution (which is equivalent
to 0'000,ooo,7i4 gramme of AS4O6) gave the full mirror.
Thus the zinc nickel alloy was quite as sensitive as pure
zinc, although it dissolved much more readily in the
sulphuric acid, and the addition of 0'2 gramme of cad-
mium sulphate made no difference in the size of the
mirrors obtained.

We then made alloys of zinc with 0*5 gramme of
cobalt, iron (in the form of wire), and copper respectively,
and of these iron shewed the most marked retention of
arsenic, for from 20 c.c. of standard arsenic solution a
mirror equivalent to about 3 c.c. was obtained, whereas
cobalt and copper gave mirrors corresponding with about
20 c.c. in each case.

The addition of 0'2 gramme of cadmium sulphate
increased the sensitivity of the iron alloy, a mirror equi-
valent to about 10 c.c. being obtained from 20 c.c. of the
standard arsenic solution.

It seems to me that the action of cadmium sulphate
in rendering zinc and other metals more sensitive in the
determination of arsenic in the Marsh-Berzelius apparatus
must be further studied.

I have pleasure in stating that I am indebted to the
care and ability which has been shewn by my assistant,
Mr. Edwin Hopkinson, M.Sc. (Vict.), in carrying out these
experiments.

* British Food Journal, vol. 4, nos. 44 and 45.

Manchester Memoirs, Vol. I (1906), No. 13.

XIII. Notes on the Palaearctic Species of Coal-Tits.

By Francis Nicholson, F.Z.S., &c.

Received and read April loth, igo6.

During a visit to London in 1903, I obtained eleven
specimens of the Cyprian Coal -Tit {Parus Cypriotes of
Dresser), eight males and three females, which I have
presented to the Manchester Museum. On comparing
these birds with the series of Coal-Tits in the British
Museum, I made the following notes, which may be of
some interest to ornithologists.

In 1894, an excellent series of papers on the Coal-Tit
{Partis ater) and its allies was published by the late
Mr. J. P. Prazak in the ScJiwalbe^ the Journal of the
Ornithological Club of Vienna, and these essays were
afterwards published by him in a collected form in a
pamphlet of 44 pages, entitled, " Einige Bemerkungen
Uber die Tannenmeise {Pants ater, L.) und ihr nahe-
stehende Formen." He recognised the following races,
and gave their full synonymy and geographical distribu-
tion.

1. Parus ater.

2. Parus ater britannicus.

3. Parus ater Cypriotes.

4. Parus ater michalowskii.

5. Parus ater phaeonotus.

6. Parus ater aemodius.

7. Parus ater rufipectus.

8. Parus ater pekinensis.

August 15th, I god.

2 Nicholson, Palcearctic Species of Coal-Tits.

I do not intend in the present short article to reproduce
the whole of the synonymy so elaborately detailed by
Mr. Prazak, but the following notes occur to me.

Mr. Prazak, in speaking of the typical Coal-Tit of
Europe, which is Panes ater of Linnaeus (founded on the
Scandinavian species), includes Great Britain and Ireland
as within its range. In proof of the last-named locality,
he quotes Thompson's " Birds of Ireland," and Seebohm's
paper on Irish birds in the Ibis for 1890 (p. 400) ; but in
these instances the name of Parus ater, Linn., for the
Irish bird was given by Thompson in ignorance that the
British Coal-Tit was different from the typical continental
form, and by Seebohm doubtless from conservative notions
of nomenclature. The Coal-Tit of Ireland is the same as
that of England and Scotland, viz.. Partis britanniciis, and
is not Panis ater, which is the species of Scandinavia and
the continent of Europe. Parus britaiinicus is the resident
species of the British Islands, although the true P. ater is
said to occur occasionally in England. I have never
myself seen an English specimen of P. ater, and there is
not one individual in the collection of the British Museum
or the Manchester Museum, but that occasional specimens
are to be met with cannot be doubted. Considering the
hordes of tiny Goldcrests that annually migrate to the
eastern coasts of Britain, there is nothing wonderful in
the appearance of an occasional Coal-Tit from Scandi-
navia. Our British bird has an olive-brown back in its
full winter plumage, but as this plumage gets worn
during the breeding season, it becomes more grey, and it
then resembles in some slight degree the grey-backed
P. ater, but this latter species is blue-grey both in summer
and winter, and the comparison of a series shows that the
two birds are really quite distinct

I agree with Dr. Bowdler Sharpe's observations on

Manchester Memoirs, Vol. I. (1906), No. 13. 3

the two species (" Handb. Brit. Birds," vol. i, p. I'i^y),
where he remarks that, although in worn plumage during
the nesting season, P. britanniciis may lose some of its
olive tint through the wearing away of the edges of the
feathers, yet there is never any real difficulty in
distinguishing our native bird from the true P. ater, which
can only be considered an occasional visitor to the British
Islands.

Mr. Prazak has given the full synonymy oi Parus ater,
and assigns to it the following geographical range : —
France, Portugal*, Spain, Italy, Sicily, Sardiniaf, Switzer-
land, Belgium, Holland, Germany, Denmark, Norway,
Sweden, Austria, Hungary, Bosnia, Herzegovina, Servia,
Montenegro, Bulgaria, Greece, Macedonia, Asia Minor,
Palestine, Poland, Baltic Provinces, Russia generally,
throughout Siberia to Amur Land, Kamtchatka, Ussuri
Land, Askold Island, Japan, and the Liu Kiu Islands.^

He likewise includes the island of Formosa, and
hereby recalls a ludicrous episode in the history of
ornithology. In Horsfield and Moore's " Catalogue of
the Birds in the Museum of the East India Co." (vol. i,
P- 'h7Z)i 3- specimen of Panes ater from Formosa is
recorded as having been presented by Mr. John Gould.
This specimen has now passed into the collection of the
British Museum, and is undoubtedly only an example of
the ordinary Parus britannicus. It has been prepared by
Mr. Gould's own hands, and is evidently a bird procured
by him, or by Mr. William Briggs, at ' Formosa,' Sir
George Young's beautiful place on the Thames, near
Cookham. In middle age, so I am informed by Dr.
Bowdler Sharpe, who used frequently to meet him,

* Now separated as Partis vieirce, mihi.

t Now separated as Partis sardiis, Kleinschmidt.

X Now separated as Pat-tis insularis, Hellmayr.

4 Nicholson, PalcBurctic Species of Coal-Tits.

Mr. Gould was a constant visitor to ' Formosa,' when
Mr. De Vitre lived there. The head-gardener, Briggs,
was a first-class naturalist, and was always on the look-
out for specimens for Mr. Gould, when he was writing his
" Birds of Great Britain." This is no doubt the way in
which the specimen oi P. ater from ' Formosa ' came into
Mr. Gould's possession, and was afterwards given by him
to the Indian Museum.

It will be noticed further on that the Coal-Tit of
Japan has been separated recently by Mr. Hellmayr
under the name oi Partis insularis (vide infra, p. 8).

In Mr. Dresser's " Manual of PaUtarctic Birds "
(vol. I, 1902, pp. 164 — 16"/), Parus ater is recognised, with
2 sub-species, P. britannicus and P. Cypriotes. P. inicha-
loivskii is united to P. phaeonotus, and P. rufipectus is
made a sub-species of the latter bird, whereas, in my
opinion, it belongs to the long-crested section of the Coal-
Tits, being closely allied to P. pekinensis, which Mr.
Dresser unites with P. ater.

Dr. Bianchi published in 1902 a very useful " Hand-
list" of the Paridae in the Animaire of the Petersburg
Museum, and this was the foundation of Dr. Bowdler
Sharpe's synopsis of the family, in his" Handlist of Birds"
(vol. 4, 1903).

In 1903 appeared part 18 of " Das Tierreich," in which
Dr. Hellmayr monographed the Paridae, with all that
care which he bestows on every one of his undertakings.
It is a very fine piece of work, and leaves little to criticise^
although the system of nomenclature may not commend
itself to all of us.

Mr. Hellmayr places the following species and sub-
species in his sub-genus Periparns.

1. Parus rubidiventris Blyth.

2. Parus rufomichalis Blyth, with two sub-species.

MiDicJiester Memoirs, Vol. I. (1906), No. 13. 5

a. P. I'lifoiuichaiis rufomicJialis Blyth.

b. P. rufonucJialis beavani (Jerdon).

3. Parus melanoloplms Vigors.

4. Parus ater Linn, with ii sub-species.

a. P. ater britamiicus Sharpe & Dresser.

b. P. ater Cypriotes Dresser.

c. P. ater ater Linn.

d. P. ater instdaris Hellmayr.

e. P. ater pekinensis David.

f. P. ater inichalozvskii Bogd.

g. P. ater atlas Meade- Waldo.

h. P. ater pJiaconotus W. Blanford.
i. P. ater aemodius Hodgson.
k. P. ater riifipectus Severtzoff.
/. P. ater ledotici Malh.

Mr. Hellmayr places P. riifipectus, Severtz, and P.
aemodius, Hodgs., in the same section, on account of the
ochre-yellow or pale cinnamon colour of the under surface,
distinguishing the latter by its smaller size and long crest
In P. aemodius the crest-feathers are certainly abnormally
developed for a member of the genus Parus.

P. rufipectus, Severtz. Of this species I examined
several specimens on my last visit to the British Museum,
from the Seebohm and Menzbier collections. In the
same year (1873), Severtzoff gave two names to this
species, Panes piceae and P. ater var. rufipectus. The
former of these, being a nomen nudum, has been dis-
allowed by subsequent authorities, and the name of
rufipectus insisted upon. This method is perfectly correct,
but in one respect it is to be regretted, as rufipectus does
not invariably convey the impression of the colour of the
bird's breast, which has often nothing especially rufous
about it. A specimen from Thian Shan shews a faint

6 Nicholson, Palcearctic Species of Coal- Tits.

tinge of buff below, of much the same shade as in P.
pekinensis and P. insularis, and from this I consider that
P. rufipectiis is more closely allied to P. pekinensis, both
as regards colour and development of crest. The Thian
Shan example is, in fact, a large edition of P. pekinensis.
There are also examples from Transcaspia in the British
Museum, some of them as pale below as the Thian Shan
bird, while others are as cinnamon as P. aemodius.

P. ater of Europe appears to me to be the lightest in
colour of all the pale-breasted group of which it is the
type. It certainly has a whiter breast than most of the
others.

P. pekinensis is, according to Mr. Hellmayr, an
inhabitant of Southern Siberia, east of the Yenesei River
and North China. Its distinguishing character is its very
evident top-knot of long feathers. This feature is certainly
developed in specimens from the far East to a greater
extent than in typical P. ater, but the latter species is
by no means devoid of a crest. Some examples of P.
pekinensis, however, have nearly as long a crest as in
P. aemodius, and it seems to me that the white spots on
the wing-coverts are more conspicuous than in P. ater.

Of P . pekinensis I have examined several specimens
in the British Museum. One from the Ussuri River (Lat.
48° N. : Dybowski) has quite long crest-feathers, and has
a warm ochraceous-buff tint on the sides of the body, the
lower back and rump being also washed with a light shade
of ochraceous-buff. I have also examined, in the British
Museum, one of the typical specimens given to the late
Robert Swinhoe by Abbe David, and bequeathed to
the Museum by Mr. Seebohm. It is from Pekin, and is
a somewhat remarkable bird, not only on account of the
tuft of long feathers on the crown, but from its rufescent
under surface, wherein the breast and flanks are of a

Manchester Memoirs, Vol. l. (1906), No. 13. 7

distinct fawn-colour, approaching that of P. rufipectus,
Severtz. The date of the specimen is December 14, 1867,
so that it is in full winter plumage. I have elsewhere
remarked on the difference in the colour of the under parts
shewn by P. rufipectus, which has also a long crest similar
to that of P. pekinensis, and has a rufous breast when in
full feather, which fades to a sort of creamy fawn-colour
in worn plumage.

I find that the same difference exists in P. pekmensis.
A specimen from Kuatun, presented to the British Museum
by Mr. C. B. Rickett, and obtained in May, is in very worn
plumage, and is therefore very much paler below than the
Pekin bird ; it has an evanescent tinge of fawn-colour,
and also an extraordinary crest of elongated plumes. No
one examining either of these specimens of P. pekinensis
could doubt the distinctness of the -species from P. ater,
and I can only suppose that Mr. Dresser had not examined
these birds in the British Museum when he determined to
unite the two species in his recently published " Manual
of Palsearctic Birds."

Another specimen in the British Museum which I
also consider to be P. pekinensis is a female bird from
Chemulpo in Corea (C. W. Campbell : Seebohm Coll.),
but those from the Gulf of the Amur, Ussuri Land, and
Kamtchatka seem to be true P. ater, as do the birds from
Krasnoyarsk. Siberian birds are slightly more fawn-
coloured below, and shew some approach to P. rifipectus.

A specimen from the Gulf of the Amur River,
collected by the brothers Doerries, and bequeathed by
Seebohm to the British Museum, cannot be separated
from P. ater, as it has no more crest than that species.
The white bars on the wing, are, however, somewhat
broader, and there is a distinct tinge of buff on the under-
parts, especially on the flanks, though the rump is scarcely

8 Nicholson, PalcEarctic Species of Coal-Tits.

tinged. A longer series of specimens from Eastern
Siberia may prove that the Coal-Tit of the far East may
be separable from the European species.

In the British Museum are birds from Irkutsk and
Krasnoyarsk which I refer to P. ater, not to P. pekinensis,
and I believe it to be a mistake to record the latter
species from the Valley of the Yenesei.

P, insidaris, Hellmayr, from Japan, is, as might have
been expected, very closely allied to P. pckinensis, but
there is a much more decided tinge of fawn-colour on the
flanks, and it has the same conspicuous white spots on the
wing-coverts. There is also a fulvescent tinge on the rump.
Of the brown- or olive-backed section it is also difficult to
write down the distinctive characters, but Mr. Hellmayr
has given the characters which I enumerate below.

P. Cypriotes, Dresser, is the darkest race of this whole
section, the back being not so much olive-brown as dusky-
grey, with an olive-brown wash. The smoky-brown tinge
of the sides of the body is also much darker and less
inclined to buff than in any of the other forms, and the
black on the throat extends further on to the chest.

P. p/iaeonotiis, Blanford, is a somewhat larger and
browner bird, with a very little admixture of grey on the
back. The sides of the body are of a pale fawn tint.

P. atlas, Meade-Waldo, from Morocco, is a representa-
tive of P. phaeonotus, but with the sides of the body
darker and more smoky-brown.

P. inoltchanowii is described by Professor Menzbier
from the Crimea as a very distinct form, allied to
P. phaeonotus.

P. inichaloivskii, Bogd. has a very stout bill, and has
pale fawn-coloured flanks, but somewhat deeper in tint
than P. phaeonotus. It is very closely allied to the latter
species, but is a little darker brown.

MancJiestcr Memoir's, Vol. I. (1906), No. 13. 9

P. britannicus, Sharpe and Dresser, is, when specin:iens
in fresh plumage are compared, one of the most distinct
of all the races of P. aier, remarkable for its olive-toned
back, and the pronounced fawn-buff colour of the flanks.
It has a very weak bill compared with that of its allies.

In 1905 was published part 3 of Dr. Hartert's monu-
mental work " Die Vogel der palaarktischen Fauna,"
and he recognises the following forms of Paj'its ater and
its allies.

Pants aier ater, Linn. — Europe to 65' N. Lat,
apparently through N. Siberia to Kamtchatka. In
Europe to the mountains of Spain, Italy, and Sicily.

Pariis ater britannicus, Sharpe and Dresser. — British
Isles.

Panis ater sardus, Kleinschmidt. — Sardinia.

Parus ater atlas, Meade- Waldo. — Atlas Mountains,
Morocco.

Parus ater pekinensis, David. — N. China and Man-
churia, probably west to the Yenesei Valley.

Parus ater aeviodius, Hodgs. — E. Himalayas, east-
wards to mountains of Kansu and Shensi in W. China.

Partis ater insularis, Hellm. — Japanese Islands (Yezo,
Hondo, Liu Kiu Islands).

Parus ater rufipectus, Severtz. — From E. Thian-Shan
Mountains to Issik-Kul.

Parus ater Cypriotes, Dresser. — Cyprus.

Parus ater ledouci, Malh. — N. Algeria and N. Tunis.

Parus ater moltchaiiowii, Menzb. — Southern Crimea.

Parus ater derjugiui, Sarudny and Loudon. — N.
Armenia (Lasistan).

Parus ater inichaloiusku, Bogd. — Caucasus Mountains
to Lenkoran.

Parus ater pJiaconotus, Blanf. — Persia and S. Trans-
caspia.

lO Nicholson, PalcEarctic Species of Coal-Tits.

Partes mfonucJialis rnfo>iuchalis, Blyth. — Turkestan ;
Himalayas, from Gilgit to Gurwhal.

Parns rufonuchalis beavatii, Jerd. — Nepal, Sikhim,
eastward to W. China.

The conclusions arrived at by Dr. Hartert are, as will
be seen by the above summary, confirmatory of those of
Dr. Hellmayr. His work bears evidence of great care
and judgment, and in future studies of the Palaearctic Coal-
Tits, Hartert's essay must receive ample consideration.

Partis ater sardus, of Kleinschmidt, was described in
the OrnitJiologiscJie MonatsbericJit for 1903 (p. 180).

Another species recently described is Partes scJiwederi
{^Partis ater sc/izvederi, Loudon and Tschusi, Orn. Ja/ub.
vol. 16, p. 140, 1904). Hab. Livonia, Baltic Provinces.
Dr. Hartert (/.r., p. 356) does not consider this form
separable from true P. ater.

The following is a list of the species of the sub-genus
Peripartis, or, as I prefer still to call them, Pariis ater.,
and its immediate allies. This list is founded on the
recent works of Hellmayr and Hartert.

Parus rubidiventer.

Fariis rubidivetitris, Blyth, y. A. S. Beng., vol. 16, p. 445 (1S47).
Lophophanes rttbidiventris, Jerd, "B. Ind.," vol. 2, p. 274 (1863) ;

Gates, "Faun. Brit. Ind., Birds," vol. i, p. 58 (1889);

Hartert, "Vog. Pal. Fauna," p. 362 (1905).
Parus {Feriparus) riidiventris, Hellmayr, " Tierreich, Paridae,"

p. 74(1902).
Periparus riibidiventris, Bianchi, Anti. Mus. Zool. Acad. St.

Fetersb., vol. 7, p. 245 (1902).
Feriparus rubidiventer, Sharpe, " Handl. B.," vol. 4, p. 326

(1903)-
No white tips to the wing-coverts ; under surface of
body rusty-red on the centre of the breast and abdomen.
Hab. Himalaya Mountains, from Kumaon to Nepal.

Manchester Memoirs, Vol. I. (1906), No. 13- il

Parus rufinuchalis.

Parus riifomichaiis, Blyth, J. A. S. Befig., vol. 18, p. 810

(1849).
Lophophanes rufonuchalis, Jerd., " B. Ind.," vol. 2, p. 273(1863).
Lophophanes rufinuchalis., Gates, " Faun. Brit. Ind., Birds," vol. i,

p. 58 (1889).
Parus ( Periparus) rufonuchaiis, Hellmayr, "Tierreich, Paridae,"

p. 75 (1902).
Periparus rufinuchalis, Bianchi, An7i. AIus. St. Pitersb., vol. 7,

p. 246 (1902) ; Sharpe, " Handl. B.," vol. 4, p. 326 (1903).
Parus rufonuchalis rufomichalis, Hartert, " Vi'ig. Pal. Fauna,"

vol. I, p. 361 (1905).

No white tips to the wing-coverts ; breast and abdomen
grey ; throat and chest black. Size larger.

Hab. Russian Turkestan ; Afghanistan ; Mountains of
Gilgit, N. Kashmir.

Parus beavani.
Lophophanes beavani, Jerd., " B. Ind.," vol. 2, p. 275 (1863, ex
Blyth MSS.) ; Gates, " Faun. Brit. Ind., Birds," vol. r,
p. 59 (1889).
Periparus btavani, Bianchi, Ann. Mus. St. Petersb., vol. 7,
p. 245 (1902) ; Sharpe, " Handl. B.," vol. 4, p. 326 (1903),
Parus {Pe?-iparus) rufonuchalis beavani, Hellmayr, "Tierreich,
Paridae," p. 75 (1903).
Differs from P. rufinuchalis in its smaller size, and in
having only the throat black.

Hab. E. Himalayas, Nepal to Sikhim ; to W. China
(Kansu, Kokonor).

Parus melanolophus.

Parus melanolophus. Vigors, P.Z.S., vol. i (1830), p. 23 ;

Hartert, " Vog. Pal. Faun.," part 3, p. 362 (1905).
Lophophanes melanolophus, Hume, " Nests and Eggs, Ind. B.,"

p. 403(1874); Gates, "Faun. Brit. Ind., Birds," vol. i,

p. 57 (1889).

12 Nicholson, Palcearaic Species of Coal-Tits.

Parus {Peripariis) 7/iehi?iolophiis, Hellm., "Tierreich, Paridae,'

p. 76 (1902).
Periparus 7nelanolophiis, Bianchi, t.c, p. 245; Sharpe, " Handl.
B.," vol. 4, p. 326 (1903).
With distinct white spots on the median and greater
wing-coverts ; under surface of body grey.

Hab. N.-VV. Himalayas, from Kumaon to Murree and
Kashmir (Gilgit), Mountains of Afghanistan.

Parus ater.

Parus ater, Linn., " Syst. Nat.," vol. i, p. 341 (1766); Prazak,

t.c, p. 4 (1894); Dresser, "Man. Pal. Birds," vol. i,

p. 164 (1902).
Parus {^Periparus) ater ater, Hellmayr, "Tierreich, Paridae,"

p. 78 (1902).
Periparus ater, Bianchi, t.c, p. 245 ; Sharpe, " Handl. B.," vol. 4,

p. 325 (1903).
Parus ater ater, Hartert, " Vug. Pal. Fauna," part 3, p. 356 (1905).

With white spots on the median and greater wing-
coverts ; under surface of body whitish, with the sides
distinctly washed with rust-colour ; back ashy-blue ; no
distinct crest or top-knot ; rump with a slight wash of
olive-yellowish.

Hab. The whole of Europe and Siberia to the Gulf of
the Amur and Kamtchatka.

In addition to its blue-grey back, the sandy-buff sides
of the body are, in P. ater, decidedly duller in colour, and
incline to smoky-brown. Scandinavian specimens are
very clear blue-grey, but show a faint tinge of olive in the
winter dress. A similar faint shade of olive is to be found
on a few birds from the Vosges Mountains, obtained in
October and November, but otherwise birds from Eastern
France, Holland, and Belgium appear to be identical with
those from Norway. Birds from Asia Minor are rather
paler blue.

Manchester Memoirs, Vol. I. (1906), No. 13. 13

Parus insularis.

Parus ater (nee Linn.) Seebohm, " B. Japan, Enip.," p. 82

(1890), et auct.
Funis ater insuiaris, Hellm., Orn. Jahrb., vol. 13, p. 36 (1903);

Hartert, "Vug. Pal. Fauna," part 3, p. 359 (1905V
Parus {Periparus) ater insularis, Hellm., " Tierreich, Paridae,"

P- 75 (1903)-
Periparus insularis, Sharpe, " Handl. B.," vol. 4, p. 325 (1903).

Mr. Hellmayr characterises this species as follows : —
Similar to P. ater, and with the same bluish-grey back,
but of a lighter and purer blue shade ; rump as in P.
britannicus, light olive-yellowish grey ; under surface of
body light olive-reddish-yellow, similar to that of P.
britannicus.

P. insularis is of a very pure ashy-blue on the back.
It only differs from P. pekinensis in its pale fulvescent
under surface, which is rather lighter than in the latter
bird. The crest is well developed, but does not form so
distinct a tuft as in P. pekinensis. Some specimens are,
however, hardly distinguishable.

Parus pekinensis.

Parus pekinensis, David, Ibis, 1870, p. 755.

Parus ater pekinensis, Frazak, AfT. Orn. Ver. Wien, vol. 18, p. 32

(1894) ; Hartert, "Vog. Pal. Fauna," part 3, p. 358 (1905).
Parus {Periparus) ater pekinensis, Hellmayr, "Tierreich, Paridae,"

p. 78 (1903).
Periparus pekinerisis, Bianchi, t.c., p. 245; Sharpe, " Handl. B.,"

vol. 4, p. 325 (1903).

Similar to P. ater, with a blue-grey back, and large
spots on the median and greater coverts, but distinguished
by a very distinct top-knot or crest on the crown.

Hab. Corea. China (Pekin to Foh-kien).

14 Nicholson, Palcearctic Species of Coal- Tits.

Mr. Hellmayr's diagnosis is as follows : Similar to/-*, ater,
but distinguished by an apparently well-developed crest
on the hinder head. Nape-patch generally mixed with
spots of blackish. Upper surface ashy-bluish, sometimes
washed with olive-colour on the lower back. Rump
rusty-yellowish ; otherwise resembling P. ater in tint.
Under surface whitish, the sides of the body washed with
pale rusty-yellowish colour.

Mr. Hellmayr gives the range of P. pekinensis as
Southern Siberia, east of the Yenesei, to China. As
already mentioned in my note on P. ater, I cannot agree
that the birds from the Yenesei in the Seebohm collection
belong to P. pekinensis, but I consider that they are
typical P. ater.

In the British Museum are specimens of P. pekinensis
from Pekin {David: SeeboJim Coll.) ; Kuatun {Rickett
Coll.) ; Nikolaiesk {Seebohm Coll.) ; N. Ussuri Land
{Seebohm Coll.) ; Chemulpo, Korea {Campbell), Kam-
tchatka {Seebohm Coll.).

Parus rufipectus.

Farus ater var. rufipectus, Severtz., " Turkestanskie Jevotnie,"

pp. 66, 134 (1873).
Parus ater rufipectus, Prazak, MT. Orn. Ver. IVien, vcl. 13,

p. 175 (1894); Haitert, "Yog. Pal. Fauna," part 3, p. 359

(1905)-
Parus rjtfipectus. Dresser, " Man. Pal. Birds," vol. i, p. 166

(1902).
Parus {Periparus) ater rufipectus, Hellmayr, "Tierreich, Paridae,"

p. 80 (1903).
Periparus rufipectus, Bianchi, i.e., p. 245 ; Sharpe, " Handl. B.,"

vol 4, p. 325 (1903).

This species, as I have said before, seems to me to be
a large race of P. pekinensis. It shows some affinity to

Manchester Memoirs, Vol. I. (1906), No. \V*. 15

P. aeinodius, but has not such a strongly developed crest,
and it is decidedly larger.

It ranges from Eastern Turkestan to the Tian Shan
Mountains. Some specimens in the National collection
are apparently from Transcaucasia, but I could not deci-
pher the Russian labels.

When in full plumage, P. nifipechis is easily recognised
by its rufous-buff under-surface, the sides of the body
being of the same colour as the breast : the wing-spots
are also tinged with fawn-colour, and are not pure white
as in P. ater. The back is dark blue-grey, and the fulvous
shade on the rump is not particularly pronounced.

Parus britannicus.

Patus brifannicus, Sharpe and Dresser, Ann. and Mag. Nat.

Hist., {4), vol. S, p. 437 (1871); Dresser, "Man. Pal.

Birds," vol. i, p. 165 (1902).
Parus ater britan7iicus, Prazak, MT. Orn. Ver. IVien, vol. 18.

p. 141 (1894); Hartert, "Vog. Pal. Fauna," part 3, p. 357

(1905)-
Parus (Periparus) ater btitannicus, Hellmayr, "Tierreich,

Paridae," p. 77 (1902).
Periparus britannicus, Bianchi, t.c, p. 245 ; Sharpe, " Handl. B.,"

vol. 4, p. 325 (1903).

Similar to P. ater, and with white spots on the median
and greater coverts, but having the back olive-brown, in-
stead of blue-grey ; only the throat black ; sides of the
body very clear rusty-yellow : bill decidedly more slender.

Hab. British Islands.

Parus britannicus, when a series is compared, need
never be confounded with true P. ater, for the difference
in colour of the back is very perceptible. Freshly moulted
birds in August and September are strongly olivaceous
on the back, which in P. ater is clear blue-grey at that

1 6 Nicholson, Palcearctic Species of Coal-Tits.

time of year. On some specimens the white nape-patch
shows a slight yellowish tinge, which may be a sign of a
freshly moulted young bird.

The olive-brown colour of the back is always very
pronounced from August and September to January ; it
becomes a little less obvious as the breeding-season
approaches, and the colour of the back becomes slightly
more grey as the plumage becomes worn, but it is never
so blue as in the continental bird.

Parus vieirae, n.sp. (PI.)

There is in the British Museum a specimen, from
Portugal, which appears to belong to an undescribed form
of Coal-Tit, and for which I propose the name of Parus
vieirae.

Similis P. britannico, sed regione uropygiali et corpore
subtus pallide cinnamomeisdistinguendus. Long. tot. 4"0,
culm. 04, alae 2-2, cauda i'6, tarsi 07.

Hab. Coimbra, Portugal (Dr. L. Vieira).

The dull cinnamon-rufous colour of the underparts
which gives a rufous appearance to the bird is quite
different from the fulvescent tint found in P. ater and
P. brita7iniais. Dr. Sharpe tells me that he has shown
the type to Dr. Bianchi and Dr. Hartert, and that they
both confirm my idea of its distinctness.

Parus sardus.

Parus sarins, Kleinschmidt, Orti. MB., vol. 11, p. 186(1903).
Parus ater sardus, Hartert, "Vog. Pal. Fauna," part 3, p. 358
(1905)-
This is said by Pastor Kleinschmidt to be easily
recognisable from true P. ater by its bright rust- coloured
sides. The English form, he adds, is not so bright in
colour on the flanks, and has likewise a duller colouring

Manchester Meinoifs, Vol. I. (1906), No. 13. 17

on the back. The British typical form is in every respect
much duller and darker in colour than the Sardinian bird.
Hab. Sardinia.

Parus Cypriotes.

Parus Cypriotes, Dresser, F.Z.S., 1867, p. 563, id. "Man. Pal.

Birds," vol. i, p. 165 (1902).
Pants ater Cypriotes, Prazdk, MT. Orn. Ver. JVie/i, vol. iS,

p. 142 (1894); Hartert, "Viig. Pal. Fauna," part 3, p. 359

(1905)-
Parus (Peripari(s) ater Cypriotes, Hellm., t.c, p. 77 (1903).

Periparus Cypriotes, Bianchi, t.c, p. 244; Sharpe, " Handl. B."

vol. 4, p. 325 (1903).

With white spots on the wing-coverts as in P. ater,
but with an olive-brown back, deeper in colour than in
P. britannicus ; lower neck black, as well as the throat :
rest of under-surface of body cream-coloured, the sides
and under tail-coverts washed with brownish.

Hab. Island of Cyprus.

Parus ledouci.

Panes ledouci, Malh., Mem. Soc. H. N. Moselle, p. 45 (1842):

Dresser, "Man. Pal. Birds," vol. i, p. 166 (1902).
Parus ater ledouci {M.:i\h.), Prazak, t.c, p. 20 ; Hartert, " Vog. Pal.

Fauna," part 3, p. 360 (1905).
Parus {Periparus) ater ledouci, Hellmayr, "Tierreich, Paridae,"

p. 80 (1903).
Periparus ledouci, Bianchi, t.c, p. 244; Sharpe, "Handl. B.,"

vol. 4, p. 324 (1903).

This species is confined to N.E. Africa, and is easily
recognised by its yellow face and breast.
• Its home is in Algeria and N. Tunis.

Parus atlas.

Parus atlas, Meade-Waldo, Bull. Brit. Orn. Club, 1901, p. 27;
id. Ibis, 1903, p. 207, pi. 6 ; Dresser, "Man. Pal, Birds,"
vol. I, app. p. 885 (1903).

l8 Nicholson, Palcearctic Species of Coal-Tits.

Farus {Feriparus) aier atlas, Hellmayr, " Tierreich, Paridae,"

p. 79 (1902).
Feriparus atlas, Sharpe, " Handl. B.," vol. 4, p. 325 (1903).
Fartis ater atlas, Hartert, "Vog. Pal. Fauna," part 3, p. 358
(1905)-

This species was discovered by Mr. Meade-Waldo in
the Atlas Mountains of Morocco in 1901, where, he says,
it abounds throughout the moister woods, and ascends
as high as the limit of trees or scrub.

It is placed by the describer, and by Mr. Hellmayr, as
most nearly allied to P. viicJialozvskii, but distinguished
by having the black of the fore-neck descending over the
sides of the chest. Specimens in freshly moulted plumage
also show white spots on the fore-neck. The sides of the
body are dark smoky-buff instead of pale buff.

Dr. Hartert (/.<:.) places the species between P. sardus
and P. pekinensis.

Parus phaeonotus.

Farus phaeonoius, Blanford, Ibis, 1873, p. 88 (1873).

Farus phaeonotus, Dresser, "Man. Pal. Birds," vol. i, p. 166

(1902).
Farus {Feriparus) ater phaeotiotus, Hellmayr, "Tierreich,

Paridae," p. 79 (1902).
Feriparus phaeonotus, Bianchi, t.c, p. 244 ; Sharpe, " Handl.

B.," vol. 4, p. 325 (1903).
Farus ater phaeonotus, Prazak, AIT. Orn. Ver. Wien, vol. 18,

p. 158(1894); Hartert, "Vog. Pal. Fauna," part 3, p. 361

(1905)-

This species is described by Mr. Hellmayr as
approaching P. brltannicus in colour, but exceeding it in
size. The upper surface is olive-brown, the rump and
upper tail coverts of the same colour as the back. Nape-
spot pure white, sometimes very large, sometimes only
slightly developed. The white cheek-spot extends low

Manchester Memoirs, Vol. I. (1906), No. 13- 19

down on the sides of the neck. Median and greater wing-
coverts with large white spots at the ends, which, on the
central and inner greater coverts, are surrounded by a
small edging of rusty yellow, and, as well as the primary-
coverts, show an olive-brownish outer edge. Primaries
whitish, the inner ones, as well as the secondaries, olive-
greenish, the innermost secondaries more edged with
brown. Tail feathers externally edged with olive-grey or
brownish. The throat-patch often extends over part of
the chest. Middle of under surface of body white ; the
sides of the body, abdomen, and under tail-coverts
washed with fulvous rust-colour, the latter less strongly.
Hab. Persia and Southern Transcaspia.

Parus moltchanowi.

Parus moltchanoivi, Menzbier, Bull. Brit. Orn. Ciub, vol. 13,
p. 49, (1903); Dresser, "Man. Pal. Birds," vol. i, app.,
p. 885 (1903).

Pants ater moltchanowi, Hartei t, " Vog. Pal. Fauna," part 3,
p. 360 (1905).

In this species the back is said to be of the same
colour as in P. ater, but of a somewhat lighter grey, the
under-surface of the body with scarcely any tint on the
sides. Wings and tail as long as in P. viicJuilozvskii, the bill
being much longer and slenderer (cf Hartert, t.c. p. 360).

Hab. Mountains of Southern Crimea,

Parus derjugini.

Periparns ater var. derjugini, Sarudny and Loudon, Orti. MB.,

vol. II, p. 129 (1903).
Parus ater derjugini, Hartert, "Vog. Pal. Fauna," part 3, p. 360

(1905)-
This form is said to differ from typical P. ater, in its
longer and heavier bill, and there is a mixture of greyish-
brown in the colour of the back.

20 Nicholson, Palcearctic Species of Coal- Tits.

It is found in the Tschorock District of the Caucasus.

Dr. Hartert says that it has a much longer bill than
P. moltchanowi, and differs from that species, in that the
back is not so pure ashy-grey, but is plainly olivaceous in
tint, as in P. britanniciis, while the sides of the body and
the under tail-coverts are not so whitish, but clearly
tinged with reddish-brown colour.

From P. viicJialowskii it differs in the much more
slender bill and the colouring of the upper surface, which
appears more greyish.

Parus michalowskii.

Parus michalowskii^ Bogd., Tr. Soc. Kazan Univ., vol. 8, no. 4,

p. 87 (1879).
Parus ater michalowskii, Prazak, MT. Orn. Ver. IVien, vol. 18,

p. 143 (1894); Hartert, "Vog. Pal. Fauna," part 3, p. 360

(1905).
Parus {Peripariii) ater michaloivskii, Hellmayr, "Tierreich, Pari-

dae," p. 78.
Periparus michalowskii, Bianchi, i.e., p. 244 ; Sharpe, " Handl.

B.," vol. 4, p. 325 (1903).

This is one of the brown-backed section of Coal-Tits,
and is very closely allied to P. pJiaeonotus, Blanford, with
which Mr. Dresser has united it.

Mr. Hellmayr gives the characters as follows : — Very
similar to P. britannicus, but with a somewhat brownish
tone on the back ; the rump without any olive-grey tint.
The under surface, however is much lighter, and with much
less of a bright wash on the sides. Bill decidedly stouter
than in the allied forms, P. ater and P. phaeonotus.

The habitat is stated by Dr. Hartert to be the
Southern Caucasus, in the river system of the Kura,
Sekarsk to Kedabeg- and Lenkoran.

Manchester Memoirs^ Vol. I. (1906), No. 13. 21

Pariis aemodius.

Parus aemodius, Hodg3.,y. A. S. Beitg., vol. 13, p. 943 (1844).
Parus ater aemodius, Prazak. t.c, p. 174 (1894); Hartert, " Vog.

Pal. Fauna," part 3, p. 358 (1905).
Lophoi)hants aemodius, Oates, " Faun. Brit. Ind., Birds," vol. i,

p. 58 (1889).
Periparus aemodius, Bianchi, i.e., p. 109 (1902); Sharpe,

" Handl. B.," vol. 4, p. 326 (1903).
Parus {Peripar2is) ater aemodius, Hellmayr, "Tierreich, Paridae,"

p. 79 (1902).

This species is distinguished by Mr. Hellmayr from
the other races of P. ater, on account of its ochre-yellow
under surface. I should have called the breast of P.
aemodius pale ' cinnamon-buff,' but I find that it agrees
most nearly with the 'ochraceous-buff' of Ridgway's
'Nomenclature of colours.' In any case, this rufescent
tint of the under surface generally serves to distinguish
P. aemodius and P. rufipectus from P. ater and its allies,
though P. pckinensis often shows a rufous tint on the
under surface. In P. ater and the other forms there is
sometimes a tint of pale buff or cinnamon, but their
general aspect below is greyish white.

Manchester Memoirs, Vol. L . , JVo. 1 3 .

Plate

West,NeJvman' imp

1. PARUS VIEIR^

2. „ BRITANNIC US .

Manchester Memoirs, Vol. I. (1906), No. 14.

XIV. The species of Ctenopteryx, a genus of
Dibranchiate Cephalopoda.

By Dr. J, H. AsnwORTH and Dr. W. E. HOYLE,

Read May Sth, igo6. Received for pziblication Jjily 2.^th, igob.

The genus under discussion was created in 1890 by
Dr. Appellor for a small Cephalopod from Messina, only
15 mm. in length, and characterised as follows : —

Ctenopteryx Appellof ('90, p. 3).
The fins consist of a series of muscular threads, united
down to the base by a very thin, transparent membrane,
so that each fin has a comb-like appearance. The mantle
connective consists of a tract of cartilage on either side
of the base of the funnel, wider behind, and with a very
narrow groove down its middle, corresponding to which is
a cartilaginous ridge on the inner side of the mantle.
Only two pairs of siphonal adductors are present, of which
the upper pair are visible from without ; an external
muscle from the funnel to the head, as seen in Ommas-
trephes, is absent. The ocular opening is drawn out
forwards into a pointed sinus, so that it assumes a pear-
shaped appearance. There is no clearly defined funnel
groove ; the funnel has a valve.

The following forms have been referred to this genus
by various authors : —

Ctenopteryx fimbriatiis, Appellof (type).

Ctenopteryx cyprinoides, Joubin.

Calliteuthis nevroptera, Jatta.

Sepioteuthis sicula, Riippell.
We propose to discuss these identifications seriatim.

August 14th, igo6.

2 ASHWORTH & HOYLE, Species of Ctenopteryx.

Ctenopteryx cyprinoides, Joubin ('94, :oi).

This species was based on three specimens taken
fronn the stomach of a dolphin, which was captured by the
Prince of Monaco's yacht " Princesse AHce," off the island
of Corsica. Pfeffer (:oo) has expressed the opinion that
these two names belong to one and the same species,
though without giving any reasons for his belief. Jatta
('03), however, held the view that they were probably
distinct, judging only from the published descriptions.

A short time ago one of us was fortunate enough to
procure a well-preserved specimen* from Messina, much
larger than that described by Appellof, and comparable
in size with the Monaco examples, as the following table
of dimensions will show: —

Dimensions in Millimetres.

Length, total

Length, total, excluding tentacles ...

End of body to mantle margin (dorsal)

End of body to eye

Breadth of body

Breadth of head

Eye to edge of umbrella (laterally)...

Length of fin

Breadth of fin

Breadth across fins ...

Length of first arm ...
Length of second arm
Length of third arm
Length of fourth arm
Length of tentacle ...

* I would here express my thanks to the Carnegie Trust for the
Universities of Scotland for a grant which enaiiled me to purchase this
specimen. — J. H. A.

t Measured at the opening of the mantle.

Our
specimen.

Monaco
specimen,

129

100

112

54

51

59

21

32t

22

23

12

R. 48, L

•51

—

12

22

38

—

Right.

"uft.

25

25

Zl

29

27

36

29

20

31

28

28

29

57

68

150

Manchester Memoirs, Vol. I. (igo6), No. 14.

In order to settle the question of the identity of this
species with that of Appellof, we applied to Dr. Richard,
director of the Musee oceanographique at Monaco, for
permission to examine Joubin's types, and with a generous
courtesy, which we most gratefully acknowledge, he was
good enough to entrust them to us for comparison. The
following is the result at which we have arrived.

The characters upon which Joubin relies for the
distinction of the two species are : —

C. fimbriatus.

1. The pectinate fin stops
about one quarter of the
length of the mantle behind
its anterior margin.

2. The fin has 24 teeth.

3. The mantle projects
backwards into a notch left
between the two fins.

4. The tentacles at most
reach a length equal to that
of the mantle.

5. The head is sunk into
the mantle, and its breadth
appears to be about equal
to the diameter of the mantle
opening.

C. cyptinoides.

1. The pectinate fin ex-
tends forward as far as the
margin of the mantle.

2. The fin has 25 or 26
teeth, and possibly one
more, which cannot be
made out with certainty
owing to the unsatisfactory
condition of the specimens.

3. The fins are con-
tinuous round the hinder
end of the mantle so that
no notch is formed.

4. The tentacles are at
least three times as long as
the mantle.

5. The head is less deeply
sunk into the mantle, and
its breadth is less than the
opening of the latter.

4 AsHWORTH & HOYLE, Species of Ctenopteryx.

In regard to these several points our example presents
the following appearances : —

1. The pectinated fin extends practically the whole
length of the mantle. The ceasing of the fin behind the
anterior margin of the mantle in Appellof's type we
regard as due to its immaturity : a similar condition is
seen in several other young examples which have come
into our hands.

2. The pectinated fin has 23 teeth on the left and 24
on the right side. On the specimen from Monaco which
appeared to be best preserved we only succeeded in
counting 24 teeth on each side, but it is by no means
easy to be sure of the number owing to the unsatisfactory
state of preservation.

3. The fin is very narrow posteriorly, but is continuous
round the hinder end of the mantle. Here, again, we
attribute the deficiency in Appellof's type to immaturity.
In seven young examples examined by one of us, ranging
from about 4 mm. to 9 mm. in length, the fins are clearly
separate from each other, though in one or two cases
there is only a short interval between them.

4. The tentacles are somewhat longer than the mantle :
their excessive length in the types of C. cyprinoides we
attribute to the maceration which they have undergone.
This elongation of the tentacles is of very frequent occur-
rence in cephalopods taken from the stomachs of cetaceans.

5. This character we do not regard as of any
systematic importance. The extent to which the head
is drawn into the mantle varies greatly in different
examples of the same species and depends on the state
of contraction of the retractor muscles at the time of
death. The diameter of the head in relation to the size
of the opening of the mantle is also dependent on the
state of contraction of the latter.

Manchester Memoirs, Vol. l. (1906), No. 14. 5

Taking these facts into consideration we cannot see
that there is any ground for regarding the Monaco
specimens as being specifically distinct from the one from
Messina, examined by us, and this again differs from
Appellors type only in characters which are dependent on
difference in age. We, therefore, regard C. cyprinoides as
a synonym of C. fimbj'iatiis.

It must be remembered, however, that the specimens
examined by Joubin had undergone severe maceration by
the gastric juices of the dolphin from whose stomach
they were taken and that he only had for comparison
with them the figures and description of a much smaller
example. Hence he was led to attach more importance
to the apparent differences than they deserved.

Calliteuthis nevroptera, Jatta ('96, p. 118, pi. 31,
figs I — 10).

Unfortunately the type specimen of Dr. Jatta's species
appears to have been dissected and the parts not pre-
served. The principal differences between his published
description and our specimen are as follows : —

The fin in Jatta's specimen extends along two-thirds
of the mantle, while in ours it is present along practically
the whole length of the mantle. This difference is probably
accounted for by the fact that ours is larger and more
mature. The tentacular arms are scarcely clubbed in our
specimen, the dilatation being of the slightest. The mar-
gin of the ocular opening is not perfectly round, as in
Jatta's example, but bears a notch at its antero-ventral
border ; this is exactly the same on both right and left
sides. The transparent elliptical disc behind the eye,
referred to by Jatta, is not pronounced in our specimen.
There is on the left side, behind the eye, a paler area,
somewhat reniform in shape, its concave face turned to the

6 ASHWORTH & HOYLE, Species of Ctenopteryx.

eye, which was probably more transparent in Hfe, and
crossed by three round pigment spots. On the right side
this area is larger, not so regular in shape, and less clearly
defined. On the lower side of the head are two elongate
" olfactory organs," apparently not quite so triangular as
those of Jatta's specimen. The buccal membrane is well
developed. The characters of the body in general, the
front edge of the mantle, the nuchal cartilage, and the
arms seem to agree with those of Jatta's specimen.

The differences do not appear to us of generic value,
except as regards the presence or absence of a sinus in
the anterior margin of the ocular opening. A difference
in this respect has usually been regarded as marking a
generic distinction, but we do not think it of sufficient
importance to outweigh the many points of resemblance,
especially as it seems to vary in extent in different
examples (compare Appellof's description and figure with
the description of our specimen above), and may prove
to be dependent on the age of the specimen.

Jatta himself subsequently (:03) came to the con-
clusion that his species should be referred to the genus
Ctenopteryx.

It is more difficult to form an opinion as to the
specific identity of these two forms. Dr. Pfeffer (:00) has
united them ; for ourselves we would merely say that we
have a strong suspicion that they will ultimately prove to
be the same, but pending further evidence it is advisable
to retain Dr. Jatta's name.

Sepioteuthis sicula, Riippell : Verany ('51, p. 75, pi. 27).

Dr. Pfeffer has regarded this form as identical with
Appellof's species, but the characteristic features ot
Ctenopteryx are not indicated by Verany in his drawing,
though a phrase in the description of the colour might be

Manchester Memoirs, Vol. I. (1906), No. 14. 7

interpreted as referring to the remarkable pectinate form
of the fins.* The dimensions given (" trois decimetres "),
unless a misprint, indicate a much larger form than any
Cteiiopteryx hitherto recorded. We are, on the whole,
disposed to agree with Jatta in regarding Ruppell's species
as a doubtful form, the fate of which can only be decided
when further material is available.

Summary.

We conclude that Ctenopteryx cyprinoides, Joubin ('94),
is to be regarded as a synonym of C. fimbriatus, Appellof
('90). C. {CalHteuthis) nevroptera., Jatta ('96), is also
probably to be referred to C. fimbriatus, but pending
further evidence Jatta's specific name is retained. Sepio-
teuthis sicula, Riippell (Verany '51), which has also
been considered to be identical with Appellof's species,
is a doubtful form, the position of which cannot at present
be determined with certainty.

Bibliography.

Appellof, A. ('90). " Teuthologische Beitrage : i. Chteno-
pteryx n.g. ; Veranya sicula Krohn ; Calliteuthis Verrill."
Bergens Mus. Aarsber. for 1889, 34 p , i pi, 1890.

Jatta, G. ('96). "I cefalopodi viventi nel Golfo di Napoli."
Fauna 11. Flora des Golfes von Neapel, part 23. Berlin,
1896.

(:04). "A proposito di alcuni Cefalopodi del Mediter-

raneo." Boll. Soc. Nat. Napoli (i) vol. 17, p. 193-207,
1904.

* " Apres le mort le corps de ce cephalopode prend un teint blanchatre
par Taction de ralcool, les fibres musculaires horizontales des nageoires
deviennent opaques et les nageoires semblent conime rayees horizontalement
de blanc.

"Les peu d'individus de cette espece, qui ont ete jxis n'ont jamais
depasse trois decimetres de longueur, non compris les bras tentaculaires "
(p. 76).

8 AsHWORTH & HOYLE, Species of Ctenopteryx.

JouBiN, L. ('94). "Note sur les cephalopodes recueillis dans
I'estomac d'un dauphin de la Mediterranee." Bull. Soc.
Zool. France, vol. 19, p. 60-68, cuts, 1894.

(:0l)- " Cephalopodes provenant des campagnes de la

Princesse Alice (i 891-1897)." Result, camp. set. Albert
de Monaco, fasc. 17, 135 p., 15 pis., 1900 [1901].

Pfeffer, G. (:00). " Synopsis der oegopsiden Cephalopoden."
/]////. Naturhist. Mus. Hamburg (2 Beiheft z. Jahrb.
Hamburg. Wissensch. Anstalten), vol. 17, p. 147-198, 1900.

Veranv, J. B. ('51). " MoUusques mediterraneens observes,
decrits, figures et chromolithographies d'apres le vivant,
pt. I, Cephalopodes de la Mediterranee." Genes, 185 1.

PROCEEDINGS

OF

THE MANCHESTER LITERARY AND
PHILOSOPHICAL SOCIETY.

Ordinary Meeting, October 3rd, 1905. '

The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table. The following were among the recent
accessions to the Society's lAbra.ry :--'■' Aidum of Philippine
Types,'' by D. Folkmar (obi. 8vo., Manila, 1904), presented by
the American Museum of Natural History; '•'■Flora Capensis,''
vol. 4, sect. I, pt. I, by Sir W. T. Thiselton-Dyer (8vo., London,
1905), purchased ; " CEuvres Completes,''' de C. Huygens,
tome X. (4to., La Haye, 1905), presented by the Societe
Hollandaise des Sciences ; " History of the Literary and Philo-
sophical Society of Nezvcastk-upon-Tyne," by R. S. Watson (4to.,
London, 1897), presented by the author ; " The British Tuni-
cata," vol. I, by J. Alder and A. Hancock (8vo., London, 1905),
purchased ; " Piat>porlen van de Cominissie in Nederlatidsch Indie,
voor Oudheidkundig Onderzoek op Java en Aladoera" 1901, 1902,
1903 (8vo., Batavia, 1905), i)resented by the Bataviaasch
Genootschap van Kunsten en Wetenschappen ; " Contributions
to Practical Medicine" by Sir J. Sawyer, 4th ed. (8vo., Bir-
mingham, 1904), presented by the author; '■^Reports of the
Sleeping Sickness Commission" Nos. 5, 6 (8vo,, London, 1905"^
presented by the Royal Society.

ii Proceedings. [October jrd, igo^.

Mr. Thomas Thorp, F.R.A.S., exhibited a lantern slide
photograph of the total solar eclipse, taken at Burgos, in Spain,
on August 30th, 1905. The inner corona and several streamers,
extending nearly a degree from the sun, were shown very clearly,
as also a number of prominences.

Mr. C. G. Hewitt, B.Sc, read a paper entitled " Note on
the Buccal Pits of Peripatus"

In the discussion which followed Professor S. J. Hickson,
F.R.S., pointed out that the peculiar position which this small
creature occupied as a connecting link between the worms and
insects, and the probability that it would soon become extinct,
gave a special importance to any investigation which resulted in
additional knowledge of the anatomy of Peripatus.

General Meeting, October 17th, 1905.

The President, Sir William H. Bailey, in the Chair.

Mr. C. Gordon Hewitt, B.Sc, Assistant Lecturer and
Demonstrator in Zoology in the Manchester University, was
elected an ordinary member of the Society.

Ordinary Meeting, October 17th, 1905.

The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. T. Thorp, F.R.A.S., made a short communication on
the " shadow bands " seen only during a total eclipse, which were
well observed at Burgos, in Old Castile, on August 30th, 1905.

The President then delivered his Inaugural Address.

The address is published in full in the " Memoirs."

October J I St, iQO^.'] PROCEEDINGS. iii

General Meeting, October 31st, 1905.

Professor W. Boyd Dawkins, D.Sc, F.R.S., Vice-President,
in the Chair.

Mr. Herbert J. Woodall, Associate of the Royal College
of Science of London, Market Place, Stockport ; Miss Mary
McNicoL, B.Sc, Research Scholar in the Manchester University ;
Miss Ethel G. Willis, B.Sc, Science Mistress, Manchester
High School for Girls ; Miss Edith Mary Saxelby, B.Sc,
Research Scholar in the Manchester University, were elected
ordinary members of the Society.

Ordinary Meeting, October 31st, 1905.

Professor W. Boyd Dawkins, D.Sc, F.R.S., Vice-President,
in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. F. J. Faraday, F.L.S., drew attention to a paper
entitled " On a Biological Aspect of Cancer," read by him in
1899, and printed in volume 43 of the Society's " Memoirs."

Several of the conclusions recently arrived at by the Cancer
Research Committee were therein foreshadowed, e.g., that cancer
is not a microbic disease, but is due to an arrest of development
and differentiation among the somatic cells, growth being re-
stricted to mere gemmation. The cure consequently resolved
itself into finding out the nature of the change in nutrition,
nerve-stimulus, or environment, which might be the cause of
this change in the normal life-history of the somatic-cells. In
illustration deficient oxygenization of the blood might be hypo-
hetically regarded from analogy as a possible cause.

Dr. Marie C. Stopes gave an account of some recent
researches into the nutrition of the egg cell in certain plants.
The special group of plants on which the author worked was

iv Proceedings. [October jist, igo^.

that including tlie pine trees, Ginkgo, and the Cycads, viz. : —
the Gymnosperms. Though the egg cells in this group are in
many ways different from those of the flowering plants, the
results have some bearing on the question of nutrition of egg
cells in general, as well as some points of general technique.

The egg cells in all the plants under consideration are
surrounded by a well marked layer of cells, called the "jacket
cells." In the past, it has been stated by Arnoldi, Ikeno,
Coulter, and Chamberlain, and others that the nuclei of these cells
enter the egg cell either bodily or in part, and so provide
nourishment for the growing egg. The " proteid vacuoles " in
the egg of Finns and the proteid granules in Ginkgo and Cycas
had been traced directly or indirectly to these nuclei.

The author shows that the entry of these nuclei does not
usually take place, and that the structure of the wall of the egg,
in which a fine membrane closes all pits, makes it impossible
under normal conditions ; and that it is not necessary to look
only to the nuclei of these jacket cells for the supply of proteid,
as the whole surrounding tissue in Ginkgo and Cycas is packed
with absolutely similar proteid grains to those in the egg. It was
also found, during research in an Alpine laboratory, where the
author examined living material every three hours during the
day, that starch grains occur in the egg cell as well as in the
surrounding tissue. The author lays stress on the fact that
carbohydrate (/.(?., starch) always travels in soluble form as sugar ;
that proteid also travels in some soluble, simpler form ; and that
it is unnatural to expect a sudden change in the mode of entry
to the egg, as would be the case if the nuclei of surrounding
cells entered it as such.

The jacket cells act as a secreting layer and dissolve the
food stored in the endosperm, which then passes into the egg in
the normal way in solution.

In conclusion, the author states that the past unnatural
views have probably resulted from too close attention to material
treated by elaborate technical methods, and that her work on
treated material was always checked by work on living material.

October 3 1 St, ipoj.] PROCEEDINGS. v

Much of the work was done in conjunction with Prof. Fujii,
of Tokio, with whom the author is pubUshing a joint paper on
the subject in Germany.

Dr. Herbert Ramsden exhibited and described a model,
devised by himself, to illustrate the propagation of sound waves.

It consists of a series of magnetised needles, suspended
vertically so as to vibrate in the same plane with their like poles
downwards, and is designed to show (since the needles were
constructed and regulated to have equal times of oscillation)
most of the phenomena of the longitudinal transmission of
waves.

General Meeting, November 14th, 1905.

The President, Sir William H. Bailey, in the Chair.

Mr. George C. Simpson, M.Sc, Lecturer in Meteorology in
the University of Manchester, of Daltoii Hall, Victoria Park ;
Mr. Charles H. Burgess, M.Sc, Demonstrator of Chemistry
in the University of Manchester, Shaftesbury House, Cheadle
Huhne; Mr. Alfred Holt, M.A., Research Fellow of the
University of Manchester ; and Mr. William Edwin Grim-
SHAW, B.A., Physics Master, Manchester Grammar Scnool,
4.6, Broadway Street, Oldham, were elected Ordinary Members
of the Society.

Ordinary Meeting, November 14th, 1905,
The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

The following were among the recent accessions to the
Society's Library : — " The Collected Mathematical Works " of
G. W. Hill, vol. I (4to., Washington, 1905), presented by the
author through the Carnegie Institution ; " The Financial History

vi Proceedings. {^November i^th, igos-

of Baltimore,^'' by J. H. Hollander (8vo., Baltimore, 1899),
presented by the Johns Hopkins University ; " Le Opere "
di Galileo Galilei, vol. 16 (4to., Firenze, 1905), presented by the
Italian Embassy at London ; " Catalogue of Gteek Coins in the
Hunterian Collection" vol. 3, by G. Macdonald (4to., Glasgow,
1905), presented by the Trustees of the Hunterian Coin Cata-
logue Fund ; " Mexico, its Social Evolution,^'' 3 vols. (Fol.,
Mexico, 1900, 1902), presented by the Minister of Commerce
and Industry of the Mexican Government.

The thanks of the members were also voted to Mr. Alfred
Brothers, F.R.A.S., for his presentation to the Society of a portrait
of the late Rev. William Gaskell, who was a Vice-President
during the years 1869-75 and 1882-83.

Professor H. B. Dixon, M.A., F.R.S., who was present at
the meetings of the British Association in South Africa,
exhibited a number of slides, made from photographs taken by
him of some of the places of interest he visited, and of the
various types of natives.

Mr. C. L. Barnes, M.A., read some extracts from the
Classical writers, which showed in how little esteem seaweed
was held by the ancients, it being regarded by them as the
most useless of things. He then pointed out, by enumerating
some of the uses to which the seaweed is now put, that the
moderns had eft'ectually removed this reproach from it.

Mr. C. H. Burgess, M.Sc, read a paper entitled "Some
Convection Effects in a Heated Tube." The author
performed the experiment described in his paper.

Ordinary Meeting, November 28th, 1905.

The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

November 28th, ipoj.] PFOCEEDINGS. vii

Mr. H. MoRRis-AiREY, M.Sc, read a paper entitled "On
the Variation of the Electrical Resistance of Osmium
with Temperature." The range over which the experiments
were conducted extended from the temperature of liquid air up
to dull red heat. The results show that the behaviour of osmium,
like that of the ordinary metals, can be represented by a
parabolic expression.

December 12th, 1905.

The Ordinary Meeting fixed for this date was not called,
but a Conversazione was held in the Society's House on
the invitation of the President and Council. 260 cards of
invitation were issued to members and other persons not
connected with the Society, and a full descriptive catalogue of
the very interesting exhibits brought together on the occasion
was printed and distributed to those present.

Ordinary Meeting, January i6th, 1906.

The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. H. Stansfield, B.Sc, described the behaviour of
liquid films formed from a solution of saponin in water.
Although saponin films have very little mobility, they are
capable of becoming extremely thin. The limiting thickness of
a black saponin film is comparable with that of the thinnest
soap film. In the process of thinning, the saponin films exhibit
a grey stage ; and there are two characteristic abrupt changes
in thickness, the first from the white of the first order to the
grey, and the second from the grey to the black.

viii Procekdings. S^Januaty i6t/i, igo6.

In the apparatus exhibited platinum wires were introduced
into the fihii in order that, after it had become black, it could
be conveniently thickened again so as to show the grey stage by
applying an electromotive force of a few volts. The motion
of material that is observed in the film is in the same direction
as the electromotive force.

Mr. John Allan read a paper, communicated by Professor
E. Knecht, Ph.D., F.C.S., entitled, "On Battack Printing
in Java, with Notes on the Malay Kris, and the
Bornean Sumpitan and Upas Poison."

Specimens of the objects described were exhibited.

Mr. J. W. Jenkinson, M.A., of Exeter College, Oxford, read
a paper, communicated by Dr. F. W. Gamble, entitled
" Remarks on the Germinal Layers of Vertebrates
and on the Significance of Germinal Layers in
general."

General Meeting, January 30th, 1906.

The President, Sir Willl-^m H. Bailey, in the Chair.

Mr. Stanley Dunkerley, D.Sc, Professor of Engineering
in the University of Manchester, was elected an ordinary member
of the Society.

Ordinary Meeting, January 30th, igo6.
The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table. The following were among the recent
accessions to the Society's Library : — " Bibliographical Index of
North Americafi Fungi" by W. (}. Farlow, vol. i, pt. i. (8vo.,

January joih, I po6^^ Proceedings. ix

Washington, 1905), presented by the Carnegie Institution at
Washington ; " Mexicati and Central American Antiquities"
by E. Seller and others (8vo., Washington, 1904), presented by
the Bureau of American Ethnology ; '^ A Monograph of the
British Desmidiacecz^^ vol. 2, by W. West and G. S. West (8vo.,
London, 1905), " The British Freshwater Rhizopoda and
Heliozoa," vol. i, by J. Cash and T. Hopkinson (8vo., London,
1905), purchased from the Ray Society; "■ Les Quantites eie-
mentaires d" Elect ricite : Ions, Electrons, Corpiiscules" par H,
Abraham et P. Langevin, fasc. i, 2 (8vo., Paris, 1905), pre-
sented by the Societe Frangaise de Physique ; " Description
geologiqiie de rile d' Ambon . . .", par R. D. M. Verbeek (8vo.,
Batavia, 1905), ''Atlas" (fol., Batavia, 1905), presented by the
Dutch Colonial Department.

Mr. C. L. Barnes, M.A., shewed a group of stereoscopic
charts of the stars, issued with Mr. T. E. Heath's " Our Stellar
Universe." The charts are reduced from maps on which are
represented all the stars whose parallaxes have been measured,
or are conjecturable with some approach to accuracy. Since
the actual parallaxes are seldom more than small fractions of a
second of arc, it has been found necessary to magnify them
about 19,000 times for purposes of convenience. There are
twenty-six maps in all, two of which shew the polar regions on a
polar projection, eight the equatorial regions on Mercator's pro-
jection, and the remainder the intermediate declinations and
latitudes, north and south, on a conical projection. The object
aimed at is to shew the solidity of space, and to represent the
stars, not as on the surface of a sphere, but at different distances,
such as they actually occupy.

At this point the Chair was taken by Mr. Francis Nichol-
son, F.Z.S.

Mr. R. L. Taylor, F.C.S., F.LC, read the following paper
entitled, " On the Origin of the Salt in the Sea."

This paper is a contribution to the controversy on this
subject, which was started more than thirty years ago by
Dr. Sterry Hunt.

X Proceedings. ^January 30th, igo6.

The most recent and reliable estimate of the total amount
of sea-water which exists gives it at 340 millions of cubic miles.
A cubic mile of sea-water contains, at the rate of about 4'3
ounces per gallon, iio million tons of common salt, and, in
addition, 3"6 of potassium chloride, 12 of magnesium chloride,
8 of magnesium sulphate, and 5*5 of calcium sulphate — all in
millions of tons.

Sterry Hunt's hypothesis* was, shortly, as follows : — When
the earth was in an intensely heated gaseous condition, all the
elements which compose it would be in the free state, and would
only unite gradually as ihe temperature fell. When the temper-
ature reached i2oo°C. the crust at any rate would have become
solid, and would inevitably consist mainly of silicates such as
we are familiar with as forming the principal constituents of
eruptive and igneous rocks. Practically all the chlorine which now
exists as chlorides would then be in the atmosphere in the form
of hydrochloric acid ; the carbon which exists now as coal, as
well as that in the carbonates of lime and magnesium which
now form such an important part of the earth's crust, would also
be in the atmosphere as carbonic acid gas, and in all probability
the sulphur which now exists in the various natural sulphates
and sulphides would be in the atmosphere as well, either in the
form of sulphur dioxide or trioxide. The whole of the water
would also be in the atmosphere in the form of vapour,
so that this primeval atmosphere must have exerted an
enormous pressure — possibly 300 or 400 times its present
pressure. As the temperature fell, the water vapour would
at last condense to liquid water. The condensation would
begin at 37o^C (the critical temperature of water), and the water
would immediately dissolve in it some, at any rate, of the acid
gases from the atmosphere. The primitive ocean would there-
fore be a highly heated dilute solution of hydrochloric and
sulphuric acids. The chemist will readily understand how
rapidly this would attack many of the natural silicates, the acids
finally becoming completely neutralised, and the ocean becoming

*Che)incaI Neivs, vol. 15, p. 314.

January joth, I po6.'\ PROCEEDINGS. xi

a solution of chlorides and sulphates, in which calcium, and
perhaps magnesium, would probably preponderate as bases.
Much silica would also enter into solution, to be afterwards
deposited in the crystalline form. Insoluble or slightly soluble
sulphates such as those of barium and calcium, would also
sooner or later separate out. The atmosphere would still contain
a large amount of carbon dioxide, and the gradual decomposition
of the exposed parts of the earth's crust by the moisture and
carbon dioxide would result in the dissolving out of the alkalies
as carbonates, which would be carried into the sea. The
immense amount of calcium which existed in the primitive ocean
has since been removed by the agency of organisms of various
kinds, and now exists as calcium carbonate, in the form of lime-
stone chalk, coral, &c. The place of the calcium in the water
would thus gradually be taken by the sodium carried down by
the rivers.

David Forbes objected strongly to Sterry Hunt's views, and
an animated and finally somewhat acrid controversy took place.
Forbes contended that, in the cooling globe, the chlorine would
unite with sodium in preference to hydrogen, and thus common
salt would be deposited all over the globe long before the water
became liquid, Forbes said that there would be a layer of salt
lo feet thick all over the globe, which would be immediately
dissolved when the water condensed, so that, according to him,
the sea would be salt from the very beginning. (Forbes' cal-
culation was lamentably wrong ; Joly has since calculated that
there is enough salt in the sea to form a layer 112 feet thick
over the whole globe.)

Probably no modern chemist would accept Forbes' view.
The temperature at which common salt would solidify would be
quite high enough for silica and silicates, in conjunction with
water vapour, of which there would be an immense amount, to
decompose it, and this of course would result in the liberation
of hydrochloric acid. Further, authorities are pretty well agreed
that there is nearly as much potassium as there is sodium in the
earth's crust, and as the heat of combination of chlorine with

xii Proceedings. \^ January joth, igo6.

potassium is higher \};\zx\. with sodium, if the chlorine united with an
alkaU metal at all it would be with potassium preferably, with
the result that potassium chloride would have been the principal
constituent of sea-water.

Hunt's hypothesis gives a reasonable explanation of the
facts as they exist at the present time. It accounts, on the one
hand, for the immense quantities of common salt and of calcium
and magnesium carbonates which are known to exist, and on
the other hand it explains the tremendous amount of silica
which is found in the free state as sand, sandstone, and other
forms. These, basic and acid bodies, are the complements of
each other, and they formerly existed in combination as more or
less complicated silicates.

Under any hypothesis as to the condition of the primeval
earth and ocean, one of the most difficult things to explain
is the fact that while the metals sodium and potassmm
probably exist in approximately equal quantities in the earth's
crust, the salts of the former predominate so greatly in the
waters of the ocean at the present time. There are several facts,
however, which go a long way towards explaining this anomaly.
First, there is the formation of marine deposits contairiing
potash, of which the mineral Glauconite is a good example.
This mineral is a silicate of iron and potassium, and appears to
be a product of the decomposition of certain marine organisms.
It is forming in various places at the present time, and is widely
distributed amongst the sedimentary rocks. There is no doubt
that the formation of substances such as Glauconite will account
for the abstraction of a considerable amount of potash from the
waters of the ocean.

Then there is another fact (which Mendelejeff appeared to
think of great importance) namely, that soil of any kind,
especially when mixed with vegetable remains, r^/a/^^j compounds
of potassium much more readily than those of sodium, and this
probably will account for some more of the deficiency of
potassium salts in the ocean. There is also another consideration
to which I myself should attach considerable importance, and

January joth, I go6.] PROCEEDINGS. xiii

that is the fact that siUcates in which soda is the principal alkah
are, on the whole, more readily decomposed, both by hydro-
chloric acid and by the ordinary weathering processes of the
atmosphere, than the corresponding silicates containing potas-
sium. I have a list of some of the commoner silicates containing
sodium either as the principal or one of the principal alkalies,
and including such minerals as Albite, Natrolite, Nephelite,
Sodalite, Pectolite, &c., and out of 15 such minerals no less than
eight are decomposed by dilute hydrochloric acid, — one or two
only with difficulty, but most of them easily. The commoner
silicates containing potassium do not appear to be decomposed
anything like so easily. I should contend, therefore, that at the
first attack of the dilute acid of the primeval ocean upon the
silicates of the crust, those containing sodium would yield more
readily than those containing potassium, so that there would be
a greater amount of sodium chloride than of potassium chloride
in the sea from the very first. There is strong evidence that
such a preferential action has taken place in the well-established
fact that, whereas the average amount of soda in the igneous and
eruptive rocks is higher than that of the potash in the proportion
of about four to three, in the sedimentary rocks, which have
been derived from the igneous and eruptive rocks, the propor-
tion of soda to potash is only about three to five. Plainly the
soda has been much more completely dissolved out than the
potash.

The paper gave rise to an interesting discussion, in which
several members took part.

Ordinary Meeting, February 13th, 1906.
The President, Sir William H. Bailey, in the Chair.

Dr. R. S. HuTTON and Mr. C. S. Allott, M.Inst.C.E., were
nominated auditors of the Society's accounts for the session
1905-1906.

xiv Proceedings. {February ijth, igo6.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. Charles Bailey, M.Sc, F.L.S., presented to the
Society's Library a quarto MS. volume, found at a bookseller's,
containing records of the minutes of the meetings of the Man-
chester Botanists' Association. The first entry is dated January
13th, t86i, and the last April, 1889.

The thanks of the meeting were voted to Mr. Bailey for his
interesting gift.

Mr. H. SiDEBOTTOM read a paper entitled, " Report on
the Recent Foraminifera from the Coast of the Island
of Delos." Part III. Lageninae. Drawings of the most
interesting forms obtained were exhibited and described.

General Meeting, February 27th, 1906.

Professor Horace Lamb, LL.D , D.Sc, F R.S., in the Chair.

Mr. Joseph Burton, A.R.Coll.Sc. Dublin, of the Tile Works,
Clifton Junction, was elected an ordinary member of the Society.

Ordinary Meeting, February 27th, 1906.

Professor Horace Lamb, LL.D., D.Sc, F.R.S., in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. C. E. Stromeyer, M.Inst.C.E., gave notice that on
March 27th he would move the following resolution : —

February 2yth, rpo6.] Proceedings. xv

" That before every Annual Meeting the out-going Council
" shall prepare ballot papers containing the names of
" Members whom they have nominated for election into
" the Council and for the various Offices. These ballot
" papers to contain the names of five more Members
" than there are vacancies to be filled, due regard being
"taken that the various Literary and Philosophical
" interests of Manchester are adequately represented on
"the ballot papers. The President and Vice-Presidents
"shall, as far as possible, be selected for nomination
"from the out-going Members of Council.

"A notice to be printed on the ballot papers
" pointing out that, while other names may be substituted
" for those nominated by the Council, the total number
" voted for must not exceed the number of vacancies,
"otherwise, the ballot paper will be considered void."

Mr. Francis Nicholson, F.Z.S., presented to the Society's
Library the following works : —

"Annals of Electricity, Magnetism, and Chemistry." Con-
ducted by William Sturgeon. Vols, i — 9. 1837 — 1842.

"The Annals of Philosophical Discovery and Monthly
Reporter of the Progress of Practical Science." Con-
ducted by WilHam Sturgeon, i vol. 1843.

Mr. R. F. GwYTHER, M.A., read a paper entitled, "On the
Range of Progressive Waves of Finite Amplitude in
Deep "Water." The paper was communicated by Professor
H. Lamb, LL.D., D.Sc, F.R.S.

At this point the Chair was taken by Mr, Francis Nichol-
son, F.Z.S.

Mr. A. D. Darbishire, M.A., B.Sc, read a paper entitled,
" On the Difference between Physiological and Statis-
tical Theories of Heredity."

xvi Proceedings. \March ijth, iqo6.

Ordinary Meeting, March i3tli, 1906.

Professor W. Boyd Dawkins, D.Sc, F. R.S., in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. Francis Nicholson, F.Z.S., exhibited a specimen of
the Pine Marten, marfes abietum^ which was captured in
Queen's County, Ireland, last April. He mentioned that
Martens, whilst they do not appear to be very suspicious of
traps as some wild animals are, yet range so far and in such
varied directions, seldom following the same run, that they are
only taken by accident. The specimen exhibited was so caught,
in a trap set for rabbits, and singularly enough by the nail only
of one of its toes. As regards its distribution, the Marten has
met with the same fate in Ireland as in England and Scotland.
At one time it was common throughout the island, but with the
advance of civilisation, — railways, cultivation and the deforesting
of woods, it has been driven to the wildest parts of the country,
but not necessarily to the North and West.

Mr. Nicholson afterwards presented the specimen to the
Manchester Museum at the University.

Mr. C. Gordon Hewitt, B.Sc, read a paper entitled
" The Cytological Aspect of Parthenogenesis in
Insects."

Mr. W. Thomson, F.R.S.E., F.I.C., read a paper entitled
" Notes on Arsenic and on its Estimation in Minute
Quantities."

March 27th, ipo6.] Proceedings. xvii

General Meeting, March 27th, 1906.

The President, Sir William H. Bailey, in the Chair.

The following resolution, proposed by Mr. C. E. Stromeyer,
M.Inst.C.E., and seconded by Mr. E. F. Lange, was carried : —

" That before every Annual Meeting the out-going Council
" shall prepare ballot papers containing the names of
" Members whom they have nominated for election into
" the Council and for the various Offices. These ballot

"nanprc: fn rnntain fhp nnmes of two new Mpnihers.

Special Meeting, March 2oih, igo6.
The President, Sir William H. Bailey, in the Chair.

The Wilde Lecture, on "Total Solar Eclipses," was
deHvered by H. H. Turner, Esq., D.Sc, F.R.S., Professor of
Astronomy in the University of Oxford.

Ordinary Meeting, March 27th, 1906.
The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. C. E. Stromeyer, M.Inst.C.E., read a paper entitled,
"On Recent Mysterious Fractures of Steel Plates."

The author briefly referred to the well-known failures of
steel plates of the boilers of the Imperial Russian Yacht "Livadia,"
and to Mr. Maginnis's experiences with two brittle marine boilers,

xvi Proceedings. {^March ijth, igo6.

Ordinary Meeting, March 13th, 1906.

Professor W. Boyd Dawkins, D.Sc, F. R.S., in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. Francis Nichcjlson, F.Z.S., exhibited a specimen of
the Pine Marten, marfes abietiun, which was captured in
Oueen's Countv. Trelanrl. 1ac;t- Anvil Ho .-.-.o.^fi^.^^-^ ^u-*-

Manchester Museum at the University.

Mr. C. Gordon Hewitt, B.Sc, read a paper entitled
" The Cytological Aspect of Parthenogenesis in
Insects."

Mr. W. Thomson, F.R.S.E., F.I.C., read a paper entitled
" Notes on Arsenic and on its Estimation in Minute
Quantities."

March 27th, ipo6.] Proceedings. xvii

General Meeting, March 27th, 1906.

The President, Sir William H. Bailey, in the Chair.

The following resolution, proposed by Mr. C. E. Stromeyer,
M.Inst.C.E., and seconded by Mr. E. F. Lange, was carried : —

" That before every Annual Meeting the out-going Council
" shall prepare ballot papers containing the names of
" Members whom they have nominated for election into
" the Council and for the various Offices. These ballot
"papers to contain the names of two new Members,
"due regard being taken that the various Literary and
" Philosophical interests of Manchester are adequately
"represented on the ballot papers. The President and
" Vice-Presidents shall, as far as possible, be selected
" for nomination from the out-going Members of
" Council.

" A notice to be printed on the ballot papers
" pointing out that, while other names may be substituted
" for those nominated by the Council, the total number
" voted for must not exceed the number of vacancies,
" otherwise, the ballot paper will be considered void."

Ordinary Meeting, March 27th, 1906.
The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table.

Mr. C. E. Stromeyer, M.Inst.C.E., read a paper entitled,
"On Recent Mysterious Fractures of Steel Plates."

The author briefly referred to the well-known failures of
steel plates of the boilers of the Imperial Russian Yacht "Livadia,"
and to Mr. Maginnis's experiences with two brittle marine boilers.

xviii Proceedings. {^March 2'jtJi, igo6.

and then dealt with recent cases which have come under his
notice. He has collected samples of plates of exploded boilers
from America, Austria, and Russia, and plates which have shewn
themselves brittle in the workshop. In one case the brittleness
is clearly due to an excess of phosphorus, but in all the other
cases there is no discernible cause, and the author suggested
that certain qualities of steel have the property of slowly deterio-
rating. Several of the fractured plates were exhibited, and
sections were prepared shewing the microscopic structure of
the different materials. The chemical compositions and the
mechanical tests were also given.

At the end of the paper the author mentioned that he was
about to carry out experiments on 20 samples of steel which he
had collected, and asked for suggestions as to supplementary
tests which would discriminate between reliable and treacherous
steels.

An interesting discussion followed the reading of the paper,
in which the President, Mr. E. F. Lange, Mr. M. Longridge, and
Mr. E. L. Rhead took part.

Dr. W, E. HoYLE, F.R.S.E., communicated a paper entitled,
"Notes on a Captive Mole," by Lionel E. Adams, B.A.,
which was postponed to the next meeting on April loth.

General Meeting, April loth, 1906.

Mr. Francis Nicholson, F.Z.S., Vice-President, in the Chair.

The Lord Mayor of Manchester (J. H. Thewlis, Esq.),
was elected an ordinary member of the Society.

April lotJi, ipo6.] Proceedings. xix

Ordinary Meeting, April loth, 1906.

Mr. Francis Nicholson, F.Z.S., Vice-President, in the Chair.

The thanks of the members were voted to the donors of
the books upon the table. The following were among the
recent accessions to the Society's Library: — '■'•Report oj the
Committee np07i Mechanical Coal-cuttino,^' pt. 2 (8vo., Newcastle-
upon-Tyne, 1905), presented by the North of England Institute
of Mining and Mechanical Engineers; '' Haida Texts and
Myths, Skidegate Dialect" recorded by J. R. Swanton (8vo.,
Washington, 1905), presented by the Bureau of American
Ethnology; " Codex Diplomaticus Lusatiae Superioris iii. . . . ",
Hft. I, von. Dr. R. Jecht, (8vo., Gorlitz, 1905), " Die mittel-
alterliche Baiihmst Batitzens" von F. Rauda (410., Gorlitz,
1905), presented by the Oberlausitzische Gesellschaft der
Wissenschaften zu Gorlitz ; " Reports of the Commission . . .
for the Invistigation of Mediterranean Fever" pts. 1-4 (Svo.,
London, 1905-06), presented by the Royal Society of London;
" On the Flinders Petrie Papyri" by J- P. Mahaffy and
J. G. Smyly (Cunningham Memoirs, No. 11), (4to., Dublin,
1905), presented by the Royal Irish Academy; ^^ Idea dell'
Universo ovvero Interpretazione delta Naiura e sue Consegicenze
teoricke e pratichef 6\ G. C. Paoli, vol. i (Svo., Milano, 1906),
presented by the author ; " The Collected Mathematical Works
of G. W. Htll" vol. 2, (4to., Washington, 1906), presented by the
author through the Carnegie Institution.

Mr. Charles Oldham read a paper entitled " Notes on
a Captive Mole," by Lionel E. Adams, B.A.

At this point the Chair was occupied by Dr. W. E. Hoyle,
F.R.S.E., while Mr. Francis Nicholson, F.Z.S., read a paper
entitled " Notes on the Palaearctic Species of Coal-Tits."

XX Proceedings. [April 24.1/1, igo6.

Annual General Meeting, April 24th, 1906.
Professor W. Boyd Dawkins, D.Sc, F.R.S., in the Chair.

The Annual Report of the Council and the Statement of
Accounts were presented, and it was resolved: — "That the
Annual Report, together with the Statement of Accounts, be
adopted, and that they be printed in the Society's Proceedings.''^

Mr. J. Boyd and Mr. Charles Leigh were appointed
Scrutineers of the balloting papers.

The following members were elected officers of the Society
and members of the Council for the ensuing year : —

President : Sir William H. Bailey.

Vice-Frestdents : H. B. Dixon, M.A., F.R.S. ; Horace
Lamb, LL.D., D.Sc, F.R.S. ; W. Boyd Dawkins, D.Sc, F.R.S.;
Francis Nicholson, F.Z.S.

Secretaries : Francis Jones, M.Sc, F.R.S. E., F.C.S. ; F. W.
Gamble, D Sc

Treasurer : Arthur McDougall, B.Sc.

Librarian: W. E. Hoyle, M.A., D.Sc, F.R.S.E.

Other Members of the Council: W. Thomson, F.R.S.E.,
F.LC. ; Thomas Thorp, F.R.A.S. ; Charles Bailey, M.Sc,
F.L.S. ; R. L. Taylor, F.C.S., F.LC; Charles Oldham,
Ernest F. Lange, F.C.S.

Ordinary Meeting, April 24th, 1906.

Professor W. Boyd Dawkins, D.Sc, F.R.S., in the Chair.

The thanks of the members were voted to the donors of the
books upon the table. The following were among the recent

April 24.th, igo6.] Proceedings. xxi

accessions to the Society's Library : — " Biblioiheca Chemka : a
Catalogue of the Alchef?iical, Chemical, and Pharmaceutical
Books in the Collection of the late Jaines Young, of Kelly
and Durris . . .", by John Ferguson, 2 vols. (4to., Glasgow,
1906), presented by the Trustees of Dr. J. Young; '■'■The
Physical Geography, Geology, Mitieralogy and Paleontology of
Essex County, Massachusetts, by J. H. Sears (4to., Salem., Mass.,
1905), presented by the Essex Institute; '■'■ Catalogue of lyy 2
Stars, chiefly comprised within the Zotie 8f — go" N.P.D.,for the
Epoch igoo . . . made at the Ratcliffe Observatory, Oxford . . .
under the direction of A. A. Rambaut" (4to., Oxford, 1906),
presented by the Radcliffe Trustees ; " Neiv Reduction of
Groombridgi s Circumpolar Catalogue for the Epoch iSiO'O,"
by F. W. Dyson and \V. G. Thackeray (4to., Edinburgh, 1905),
'■^ Telei:;raph!C Determinations of Longitude made in . . . 1888 —
igo2," (4to., Edinburgh, 1906), presented by the Royal
Observatory, Greenwich.

Ordinary Meeting, May 8th, 1906.
The President, Sir William H. Bailey, in the Chair.

The thanks of the members were voted to the donors of the
books upon the table. The following were among the recent
donations to the Society's Library : — " Mans Responsibility," by
T. G. Carson (8vo., New York and London, 1905), presented by
the author; ''Collected Mathematical Works of G. JV. Hill,"
vol. 3, (4to., Washington, 1906), presented by the author through
the Carnegie Institution ; and " Zelandia illustrata ..." 4^
vervolg, [by] M. Fokker (Svo., Middelburg, 1905),

Mr. Francis Nicholson, F.Z.S., presented to the Society's
Library a volume of scientific memoirs which had been succes-
sively in the possession of Dr. Dalton, F.R.S., Eaton Hodgkin-

xxii Proceedings; [Maj/ 8th, igo6.

son, F.R.S., and Sir William Fairbairn, Bart., F.R.S., who were
Presidents of this Society from 1817 — 1844, 1848 — 1850, and
1855 — 1859 respectively. The dates of the memoirs range
from 1822 — 1830.

The thanks of the meeting were accorded to Mr. Nicholson
for his interesting gift.

Miss M. C. Stopes, Ph.D., D.Sc, read a paper, entitled
"A New Fern from the Coal Measures Tubkanlis
Suicliffii, spec. nov.

Dr. W. E. HoYLE, F.R.S.E., read a paper written in
conjunction with Dr. J. H. Ashworth, and entitled " The
Species of Ctenopteryx, a genus of Dibranchiate
Cephalopoda."

Annual Report of the Council. xxiii

Annual Report of the Council, April, 1906.

The Society began the session with an ordinary membership
of 155. During the present session 14 new members have
joined the Society ; 1 7 resignations have been received, and
there has been one death, viz. : Mr. Charles J. Heywood.
This leaves on the roll 151 ordinary members. The Society
has also lost by death 2 honorary members, viz. : Dr. S. P.
Langley, Secretary of the Smithsonian Institution, Washington,
and Sir J. S. Burdon Sanderson, F.R.S. Memorial notices
of these gentlemen appear at the end of this report.*

The Society commenced the session with a balance in hand
of £zS^' S^' 6d., from all sources, this amount being made up
of the following balances : —

At the credit of General Fund ;^i8 5 4

„ „ Wilde Endowment Fund... 233 16 6

„ ,, Joule Memorial Fund 64 16 6

„ „ Dalton Tomb Fund 34 6 2

;^35i 5 6

The total balance in hand at the close of the session
amounted to ;^426. los. 5d., and the amounts standing at the
credit of the separate accounts, on the 31st March, 1906, are
the following : —

* Memorial notices of Dr. George Wilson, formerly an ordinary member,
Professor Carl Gegenbaur, of Heidelberg, and Mr. Robert Rawson, F.R. A.S.,
honorary members of this society, appear with the others in this volume.
The first two of these were received too late for insertion in the volume of
last year, whilst the death of Mr. Rawson was not known at the time of the
preparation of this report.

xxiv Annual Report of the Con net/.

At the credit of General Fund "^^9 i8 i

,, ,, Wilde Endowment Fund ... 248 16 i

„ ,, Joule Memorial Fund 72 11 4

„ ,, Dalton Tomb Fund 35 4 2

Balance 31st March, 1906 .^^426 to

The Wilde Endowment Fund, which is kept as a separate
banking account shows a balance of ^248 i6s. id. in its
favour, as against ^233. i6s. 6d., at the beginning of the
financial year, the receipts from the invested funds being the
same as last year.

No expenditure has been incurred in respect of the Joule
Memorial Fund and the Dalton Tomb Fund, the balances of
which remain at the amounts stated above. The Dalton Tomb
Fund stands as a separate account at the Manchester and
Salford Savings Bank.

The Librarian reports that during the session 757 volumes
have been stamped, catalogued and pressmarked, 690 of these
being serials, and 67 separate works. There have been written
466 catalogue cards, 365 for serials, and loi for separate works.
The total number of volumes catalogued to date is 29,122 for
which 10,195 cards have been written.

Satisfactory use is made of the library for reference purposes,
but the number of volumes consulted is not recorded. During
the session, 160 volumes have been borrowed from the library,
as compared with 134 in the previous session.

Some attention has continued to be paid to the completion
of sets, 7 volumes or parts having been obtained, which partly
complete two sets. These were presented by the societies
publishing them.

A larger amount of binding has been done this session, 498
volumes having been bound in 418.

Annual Report of the Council. xxv

A record of the accessions to the library shows that, from
April, 1905, to March, 1906, 737 serials and 58 separate works
were received, a total of 795 volumes. The donations during
the session (exclusive of the usual exchanges) amount to 53
volumes and 299 dissertations ; 5 volumes have been purchased
(in addition to the periodicals on the regular subscription list).

During the past session the Society has arranged to exchange
publications with the following : — The Bureau of Standards,
Washington ; The Field Columbian Museum, Chicago ; The
Direccion General de Estadistica de la Provincia de Buenos
Aires, La Plata \ The Washington Academy of Science ; The
Ethnological Survey for the Phillipine Islands, Manila ; The
University of Washington, Seattle ; The Laboratoire Russe de
Zoologie, Villefranche-sur-Mer, and the Academica Polytechnica,
Porto.

The publication of the Memoirs and Proceedings has been
continued under the supervision of the Editorial Committee.

The Society is indebted to the following gentlemen for the
undermentioned gifts : —

Mr. Alfred Brothers, F.R.A.S., for a portrait of the late
Rev. William Gaskell, and for a lantern slide photo-
graphic portrait of Dr, Joule.

Mr. Charles Bailey, M.Sc, F.L.S., for a quarto MS. volume
of the minutes of the meetings of the Manchester
Botanists' Association.

Mr. Francis Nicholson, F.L.S., for "Annals of Electricity,
Magnetism, and Chemistry," conducted by W. Sturgeon.
Vols. I — 9, 1837-42, and "The Annals of Philosophical
Discovery," conducted by W. Sturgeon, i vol., 1843.

The Council arranged for the Wilde Lecture to be delivered
on Tuesday, March 20th, 1906, by Professor H. H. Turner,
D.Sc, F.R.S., of Oxford University.

The Council resolved that an address be presented to the

xxvi Annual Report of the CoiuiciL

American Philosophical Society on the occasion of the celebra-
tion of the 2ooth anniversary (April 17-20, 1906) of the birth of
Benjamin Franklin, and that Dr. F. W. Clarke, Honorary
Member of this Society, be requested to act as its representa-
tive. The following is a copy of the address : —

To The American Philosophical Society.
The Manchester Literary and Philosophical Society
sends greetings to its sister, The American Philosophical
Society for Promoting Useful Knowledge, on the occasion
of the Two Hundredth Anniversary of the birth of its
founder, Benjamin Franklin.

As philosopher, statesman and diplomatist, and as a
pioneer in the scientific fields of capillarity, acoustics, elec-
tricity and meteorology. Dr. Franklin will long be remembered,
and his intimate association with your Society is a circum-
stance of which you may be justly pioud.

(Signed) W. H. Bailey, K.B., President.

Francis Jones, 1 Hon.
Charles H. 'Le^s,) Secretaries.
April 6th, igo6.

Sir J. S. Burdon-Sanderson, Bart., F.R.S. was born at
Jesmond near Newcastle-on-Tyne in December, 1828, and was
educated at home. His family on both sides bore names of
distinction which have formed a study in eugenics by Francis
Gallon. The border country made a strong appeal to Sanderson's
nature, and throughout life he revelled in moorland scenery
and wild life.

Sanderson entered the University of Edinburgh as a medical
student, and soon displayed an inborn faculty for physiological
research. Whilst still an undergraduate he finished two papers
on vegetable irritability and on the metamorphosis of coloured

I

Annual Report of the Council. xxvii

blood corpuscles, and after graduating as M.D. in 185 1 he weret to
Paris to study chemistry under Wurtz, and physiology under the
stimulating guidance of Claude Bernard and Magendie. In the
following year Sanderson began practising as a physician, and
became attached to St. Mary's Hospital, London, as registrar,
lecturer in botany and in medical jurisprudence.

As a pathologist Burdon-Sanderson will be chiefly remembered
for his admirable services to the Paddington district, which he
served as medical officer of health for eleven years, and for his
report as Inspector to the medical department of the Privy
Council (i 856-1 863). From 1870 Burdon-Sanderson's work
took a more definitely physiological turn, and his application of
physiological methods to pathology marked him out as an
investigator of the first rank. Official recognition of his scientific
ability was shewn in appointments and honours. From 1874 to
1882 Burdon-Sanderson held the Jodrell Professorship of
Physiology at University College, London ; the Waynflete Chair
of Physiology at Oxford from 1882 to 1894; and the Regius
Professorship of Medicine at the same University for the last
nine years of his life.

Burdon-Sanderson's name is permanently associated with the
advance of the study of physiology and pathology in this country
during the last thirty years, from a retrograde position to one in
the van of scientific progress. His skill in experimental method,
his striving to render biological experiment exact with the
exactitude of chemistry and physics, and his insight at once
broad and deep into the physiology of organs and organisms,
contributed to this result. Before Pasteur, Burdon-Sanderson
discovered the mode of attenuating the virus of anthrax, and
suggested the use of the attenuated virus as a means of protection
against the disease. His Privy Council reports on Tuberculosis,
Pyaemia and Septicaemia were in advance of general medical
knowledge. His novel experimental methods are now the common-
place of every good physiological department: and the discoveries
he made on the electromotive phenomena of the beating heart,
of the Dionaea plant, of voluntary muscle, and of the electric organ

xxviii A fimtal Report of the Council.

in fish are classic. The Oxford Medical School is his enduring
monument, and in the minds of all who knew him Burdon-
Sanderson's personality, striking, attractive, inspiring, will colour
the memory of his achievements.

Sir John Burdon-Sanderson received many honours from
learned Societies during his lifetime. He was thrice Croonian
Lecturer, twice for the Royal Society, in 1867 and 1877, and
once for the Royal College of Physicians, in 1891. In 1878 he
was Harveian Orator for the College of Physicians, and in 1880
received the Baly Medal. For special research work and for
his general services to physiology and pathology he was awarded
in 1883 one of the Royal Medals of the Royal Society, of which
institution he was a Fellow.

He served on three Royal Commissions — that on Hospitals in
1883, that on the Consumption of Tuberculous Meat and Milk
in i8go, and that on the University of London, 1892-94. In
1893, Sir J. Burdon-Sanderson was President of the Nottingham
meeting of the British Association. He was elected an Honorary
Member of the Manchester Literary and Philosophical Society
in 1894. F. W. G.

Professor Gegenbaur was born in Wiirzburg on August 21st,
1826. He attended the local "Gymnasium," and in 1845
matriculated in the University of his native town as a student of
medicine and natural science. Under the guidance of such
distinguished leaders as Kolliker, Virchow, Leydig, and Heinrich
Miiller, Gegenbaur acquired a thorough grasp of the comparative
method, and his first post-graduate work was an investigation
into the life-histories of Medusae and other marine organisms,
which he carried out partly in the North Sea partly in the
Mediterranean in the stimulating company of Johannes Miiller
and Kolliker. For two years Gegenbaur continued to work out
the marine zoology of Italy, and in this period much of his grasp
of invertebrate anatomy was acquired.

In 1855 Gegenbaur was made Professor of Zoology and allied
subjects in the University of Jena, but in 1862, upon the re-

I

Atinual Report of the Council. xxix

constitution of the chair, Haeckel was appointed to the zoological
professorship whilst Gegenbaur retained the teaching of anatomy,
which, however, in his hands, was always of a morphological and
never of a merely descriptive character. At Jena Gegenbaur
both as teacher and investigator shewed himself to be one of the
foremost German scientists, and to this period most of his
classical discoveries belong.

IMany attempts were made on the part of other Universities
to attract Gegenbaur from his Thiiringian home, but not until he
had spent fourteen years at Jena did he elect to exchange his
position for one at Heidelberg, where he continued to the end
to be the foremost anatomist of his country, and one of the
sanest and shrewdest of its scientists.

The Royal Society elected Gegenbaur a foreign member in
1884, and awarded him the Copley medal in 1896. In similar
ways every country that encourages scientific work has recognised
the influence and importance of Gegenbaur, and he reaped to
an unusual extent the honours that fall to men of the first rank.

He was elected an Honorary Member of this Society on
April 26th, 1892.

In person Gegenbaur was tall and robust, and of the dark
Bavarian type. His strongly developed personality stamped
itself on all he did or said. In affairs Gegenbaur was absolutely
straightforward, stern, terse, and on occasion choleric. From his
pupils he demanded a distressing exactitude, and his intuitive
power of reading character made the existence of many of his
research students a burden.

Gegenbaur applied the comparative method to anatomy with
the keenest criticism of the weak points in phylogenetic specula-
tion. His leading principles were, that function makes and
modifies organisation, that such acquisitions are inherited, and
that no deduction or reconstruction of an ancestral stage can be
recognised unless each stands the test of physiological fitness.

His writings are among the classics of anatomical literature.
His text-books on the anatomy of Vertebrates and of Man are
standard works. His memoirs and papers form the foundation

XXX Annual Report of the Council.

for much of the present knowledge of vertebrate and invertebrate
zoology. But greater than his writings or his influence was the
dominant character of the man himself. A short autobiography
appeared in 1901. Y. W. G,

Samuel Pierpont Langley, who has been one of our
honorary members since 1887, died on February 27th, 1906, at
the age of 72, while still in possession of all his faculties. A brief
mention of the more important of his contributions to scientific
literature is all that can be attempted within the limits at disposal.
By the invention of the bolometer he was enabled to study the
heat radiation from point to point of the sun's disc, to estimate
anew the constant of solar radiation, and to map out the solar
spectrum on the infra-red side, as far as "76/^. More recently the
investigation has been continued at the Allegheny Observatory,
under his direction, and with an improved form of the instru-
ment, to a still lower limit, ■37/x. The discovery that there are
considerable gaps in this portion of the spectrum, due to atmos-
pheric absorption, is likely to prove of service to meteorologists.
As an observer and delineator of solar phenomena, Langley has
probably never been surpassed, while his determination of the
moon's temperature is a model of patient work carried out under
most baffling conditions. To the problem of aerial flight he
also gave much attention, and his aerodrome doubtless embodies
the principles upon which, if ever, the true conquest of the air
will be achieved. In addition to the highest experimental skill
and ingenuity, Langley possessed in an unusual degree the
qualities of an organiser and administrator, the foundation of
the Smithsonian Aslrophysical Observatory, and of the National
Zoological Park being entirely due to his efforts, and his eighteen
years secretaryship of the Smithsonian Institution testifies still
further to his many-sided vigour. C. L. B.

Robert Rawson was born at Brinsley, Nottinghamshire,
July 22nd, 1814. Up to the age of 23 he worked as a miner,
but having acquired a good knowledge of mathematics he then

Annual Report of the Cotincil. xxxi

undertook the duties of clerk and draughtsman in a railway
constructor's office, where he was engaged from 1837 to 1842.
Then for some years he was resident in Manchester as a private
teacher of mathematics. He was elected a member of the
Society in 1845, and was on the Council in 1847, i" which year
he left Manchester, having accepted the offer of the Admiralty
of the position of Head Master of the Dockyard Schools,
Portsmouth. He was elected an Honorary Member of the
Society in 1849.

He retired from the Dockyard Schools in 1875, when he was
the recipient of a Testimonial which was presented to him by
Admiral Maclintock, K.C.B , F.R.S., LL.D. He was a Fellow
of the Royal Astronomical Society, and an Associate Member of
the Society of Naval Architects from its commencement.

His communications to this Society, chiefly on mathematical
subjects, numbered fifteen, and extended over the long period
of nearly forty years. The first appeared in 1844, and the last
in 1883.

He died at his residence, Warblington Villa, Havant, March
I ith, 1906. T. T.

Charles James Heyvvood, who was elected an ordinary
member on January 8th, 1889, was one of a family connected
with this Society for more than a hundred years. A remote
ancestor of our late member was the Rev. Nathaniel Heywood,
M.A., ejected in 1662, for nonconformity, from the Vicarage of
Ormskirk. From the son of the ejected Vicar, also the
Rev. Nathaniel Heywood, M.A., Nonconformist Minister at
Ormskirk, descended two or three generations of Irish and
Liverpool merchants. The family established a bank in Liverpool,
and later on in 1788 two members of the family started the
banking business in Manchester.

In the following year began the long connection of the family
with the Literary and Philosophical Society by the election to
membership on February 6th, 1789, of Mr. Benjamin Heywood,
as he is called in the list of member?, though it should doubtless

xxxii Annual Report of the Council.

have been Benjamin Arthur Heywood. B. A. Heywood was
elected Treasurer in 1791, and was succeeded in that office by
his brother Nathaniel. Though he was Treasurer from 1796
to 1 8 14, Nathaniel Heywood's name does not occur in the list
of members. He married Ann, daughter of Thomas Percival,
M.D., F.R.S., the virtual founder of the Society, its first Vice-
President, and for many years its President. Three of Nathaniel's
sons were members of the Society, Benjamin (elected 18 15),
Richard (elected 1822), and James (elected 1833). The last
was a F.R.S., and M.P. for North Lancashire, and is well-known
for his successful efforts to open the older universities to non-
conformists. The eldest son Benjamin, like his brother James,
was a F.R.S. and a Member of Parliament, and in 1838 was
created a baronet. He served the Society as Treasurer from
1 81 5 to 1850. He continued the traditions of his family in his
acceptance of the responsibilities of wealth, and in the first half
of the last century took part in every movement for the
amelioration of the condition of his fellow citizens. Sir Benjamin's
second son, Oliver Heywood, joined the Society in 1864. By
his public spirit and charity he well earned the honorary freedom
of the city, which was conferred on him in 1888, he being the
first person so honoured, and he is one of the very few Manchester
men commemorated by a statue in his native city. Charles
James Heywood, Oliver Heywood's younger brother, and sixth
son of Sir Benjamin, was born at Acresfield, Pendleton, March
25th, 1835, and died at Chasely, Pendleton, December ist, 1905.
He was educated at Harrow, and at Trinity College, Cambridge,
and graduated B.A. in 1856, and M.A. in 1859. Soon after
leaving the University he became a partner in the family banking
business and was associated with it until Heywood's Bank was
absorbed by the Manchester and Salford Bank. In 1858
Mr. Heywood married Anna Margaret, daughter of William
Langton, F.S.A., a well-known antiquary and banker, and a
member of the Society. Mrs. Heywood survives her husband.

In strictly municipal affairs Mr. Heywood took no part, but
he was a generous supporter of local charities, and had an active

Annual Report of the Council. xxxiii

share in their management. He was a Churchman of the
hberal Anghcan type and for very many years was a Sunday
Schoolteacher. Of the Gentlemen's Concerts he was Treasurer
and a Director, and at one crisis in the history of the Concerts
came to their rescue by buying the Gentlemen's Concert Hall and
letting it to the Directors at a nominal rent. When the Midland
Railway Company bought the site of the Hall for their Midland
Hotel, instead of making for himself a handsome profit on the
transaction, he was content to stipulate that there should be
accommodation for the Gentlemen's Concerts provided in the
Hotel.

Mr. Heywood was a man of charming personality, and was
a worthy member of a family remarkable during many genera-
ations for its high character, public spirit, and great beneficence.
Though he never took any personal part in the work of the
Society, still we must all feel regret that by Mr. Heywood's death
a family name that has ever been held in high esteem now
disappears from our list of members.

F. N.

By the death of Dr. George Wilson, which took place on
February i6th, 1905, engineering science in this country has
lost one of its most promising workers, and the Society one of
its most active contributors. Dr. Wilson was born on February
17th, 1871, being the only son of Mr. George Wilson, of Man-
chester, and grandson of Mr. George Wilson who was chairman
of the Anti Corn Law League, and chairman of directors of the
Lancashire and Yorkshire Railway Company. He was educated
privately, and at The Owens College, which he entered in 1887,
passing through the engineering course, and graduating with
First Class Honours in Engineering in 1891. After spending
two years in the offices of the Manchester Ship Canal, he was
appointed Junior Demonstrator in Engineering in The Owens
College, relinquishing this appointment to become Lecturer in
University College, Cardiff, in 1895. In 1896 he was offered
the post of Senior Demonstrator at Manchester, together with a

xxxiv Annual Report of the Cotmcil.

Resident Tutorship at Hulme Hall, which posts he held until
1904, when he was appointed Lecturer on Civil Engineering in
the University of Leeds. He obtained the D.Sc. degree in the
Victoria University in 1900.

Dr. Wilson was a most effective and popular teacher, and
his powers of organisation were displayed to a marked degree in
the successful conduct of the large classes in the Whitworth
Laboratory. As an investigator he achieved great success in
that field of engineering research which requires the application
of m.athematical analysis of an advanced character, and it was
this application which was undoubtedly the strongest feature of
his work. Although he had barely attained the age of 34, he
had published the results of a considerable number of investiga-
tions which he had successfully carried out, and there can be
no doubt that his early death cut short a career of the highest
promise. His loss will be widely regretted by the large number
of engineers who have had the good fortune to attend his classes
in the Whitworth Laboratory, and still more by those of his
contemporaries who have been indebted to him for his
invaluable advice and help in professional matters, aid which
was always readily and ungrudgingly bestowed.

Of the considerable number of papers on engineering sub-
jects which he has published, the following may be mentioned :
" On the determination of the reaction at the points of support

of Continuous Beams." Froc. Koyal Society, vol. 62.
" On the relation between Uniform Stress and Permanent
Strain in Annealed Copper Bars and Wires." Manchester
Memoirs, vol. 43, part 4.
" Experiments on Conjugate Pressures in Fine Sand." Froc.

Inst. Civil Engifieers, vol. 149, part 3.
'• On the Failure of certain Cast Steel Dies used in the
manufacture of Drawn Tubes." Manchester Memoirs,
vol. 46, part 2.
"A Factor in the safety of High Speed Torpedo Boat
Destroyers." Manchester Memoirs, vol. 47, part 5.

T. E. S.

Treasurer's Accounts. xxxv

Note. — The Treasurer's Accounts of the Session 1905-
1906, of which the following pages are summaries,
have been endorsed as follows :

April 1 2th, 1906. Audited and found correct.

We have also seen, at this date, the certificates of the following Stocks
held in the name of the Society: — ;,i 1,225 Great Western Railway Company
5% Consolidated Preference Stock, Nos. 12,293, 12,294, and 12,323 ; ;i^258
Twenty years' loan to the Manchester Corporation, redeemable 25th March,
1914 (No. 1564) ; ;^7,500 Gas Light and Coke Company Ordinary Stock
(No. 6,389) ; and the deeds of the Natural History Fund, of the Wilde
Endowment Fund, those conveying the land on which the Society's premises
stand, and the Declaration of Trust.

Leases and Conveyance dated as follow : —
22nd Sept., 1797.
23rd Sept., 1797.
25th Dec, 1799.

)> )» >»
22nd Dec, 1820.
23rd Dec, 1820.

Declarations of Trust : —

8th Jan., 1878.
24th June, 1801.
23rd Dec, 1820.
30th April, 185 1.

We have also verified the balances cf the various accounts with the
bankers' pass books.

.R. S. HUTTON.
(Signed)

C. S. ALLOTT.

XX XVI

Dr.

Treasurer's Accounts.

MANCHESTER

I

LITERARY A1

Arthur McDougall, Treasurer, in Account rvith

I at fi\ IS. od.

Subscriptions :-

J[^i. as. od.

To Cash in hand, ist April, 1905

To Members' Subscriptions : —

Half Subscriptions, 1901-02,

II i> 1903-04, I

II >> i904-°S, 3

1905-06, 13

1900-01, I

1902-03, I

1903-04, 2

1904-05, 15

1905-06, 118

1906-07, I

To Transfers from the Wilde Endowment Fund

To Sale of Publications

To H. Sidebottom for Plates. .

To Dividends :—

Natural History Fund

Joule Memorial Fund

To Income Tax Refunded : —
Natural Historj- Fund
Joule Memorial Fund

To Donation from Mr. H. Bolton
To Discount on Knowles' bill

3 3
13 13

4 4

31 10

247 16

£. a

83

308
79
13

65 10

;{^557 16

To Balance ist April, 1905 ..

To Dividends on ;£i,225 Great Western Railway Company's Stock

To Remission of Income Tax, 1905 . .

NATURAL HISTOl"

£. s.

6 i6
58 3
3 1

£68

To Balance, ist April, 1905 .. .. .. _ ..

To Dividends on ^^258 Loan to Manchester Corporation
To Remission of Income Tax, 1905..

JOULE MEMORl.

L s.

64 16

7 7

07

Ct^ II

WILDE ENDOWMEP

£ s.

To Balance ist April, 1905 .. .. .. .. .. .. .. .. .. .. .. 233 i6

To Dividends on ;C7, 500 Gas Light and Coke Company's Ordinary Stock 31310

To Remission of Income Tax, 1905 .. .. .. .. .. .. .. .. .. 16 lo

To Bank Interest .. .. .. .. .. .. .. .. .. .. 2 14

To Legal Expenses refunded 11

D ALTON

To Balance, ist April, 1905
To Bank Interest

TOBI

I s.
34 7
o 17

^^11

Treasurers Accounts. xxxvii

.PHILOSOPHICAL SOCIETY.

-ociely,froin ist April, igoj, to jist March, igo6. Q,x,

3y Charges on Property : —

Chief Rent (Income Tax deducted)

Income Tax on Chief Rent

Insurance against Fire
By House Expenditure : —

Coals, Gas, Electric Light, Water, Wood, &c.

Tea, Coffee, &c., at Meetings

Cleaning, Sweeping Chimneys, &c.

Crockery
['.V Administrative Charges : —

Housekeeper

Postages, and Carriage of Parcels and of " Memoirs "

Stationery, Cheques, Receipts, and Engrossing

Printing Circulars, Reports, &c.

Extra attendance at Meetings, and during housekeeper's holidays

Miscellaneous Expenses ..

By

|By

JBy
|By
SBy

o

12

II

'3

17

6

32

14

I

15

13

oi

5

13

II

I

35

3

62

8

0

3^

4

ici

9

0

7

I J

b

0

3

6

6

£. s. d.

26 15 9

Publishing : —

Printing " Memoirs and Proceedings"

Illustrations for "Memoirs" (except Nat. Hist, papers) ..

Binding "Memoirs''
Library : —

Books and Periodicals (except those charged to Natural History Fund)

Periodicals formerly subscribed for by the Microscopic il and Natural History

Section . .
Natural History Fund : —

(Items shown in the Balance Sheet of this Fund below) ..
joule Memorial Fund :—
(No Expenditure this Session)

Balance at Williams Deacon's Bank, ist April, 1906
in Treasurer's hands

OS O 13

060
360

48 15 9

4 14 o

53 16 3^

53 9 9
66 4 9

142 10 2

.£557 '6 3

FUND, 1905 — 1906. (Included in the General Account, above.)

By Natural History Books and Periodicals
By illustrations for papers on Nat. Hist, in " Memoirs "
,, Balance, 1st April, 1906

(No expenditure this Session).
By Balance, ist April, 1906 .

FUND, 1905— 1906. (Included in the General Account, above.)

L s. d.

72 n 4

£>!- I' 4

FUND, 1905— 1906.

By Assistant Secretary's Salary, April, 1905, to .March
By Maintenance of Society's Library: —

Binding and Repairing Books
By Repairs to Society's Premises
By Cleaning Carpets..
By Legal Expenses
By Honorarium to Lecturer. igo5
By Transfers to Society's Funds
By Cheque Book
By Balance at District Bank, ist .April, 1906

£ s. d.
135 o o

61 II 10

S 4 9
324
15 1 1 6
15 15 o
79 8 o
026
248' 16 I

Ls^^ 12 o

FUND, 1905-1906.

(No Expenditure this Session).

By Balance at Manchester and Salford^Savings Bank, ist April, 1906

j: s. d.

35 4 2
i-i5 4 2

xxxviii TJic Council .

THE COUNCIL
AND MEMBERS

CF THE

MANCHESTER
LITERARY AND PHILOSOPHICAL SOCIETY.

(Corrected to August 14TH, 1906.)

Sir WILLIAM H. BAILEY, M.I.Mech.E.

l9icc-|3rc0ilicuts.

H. B. DIXON, M.A., F.R.S., F.C.S.

HORACE LAMB, M.A., LL.D., D.Sc, F.R.S.

W. BOYD DAWKINS, M.A., D.Sc, F.R S.

FRANCIS NICHOLSON, F.Z.S.

FRANCIS JONES, M.Sc, F.R.S. E., F.C.S.
F. W. GAMBLE. D.Sc. i

■AlEvcasurcr.
ARTHUR McDOUGALL, B.Sc.

^ibvaviau.

W. E. HOYLE, M.A., D.Sc, F.R.S.K.

(Dthev iVUmbcrs of the QTouucil.

WILLIAM THOMSON, F.R.S.E., F.I.C.

THOMAS THORP, F.R.A.S.

CHARLES BAILEY, M.Sc, F.L.S.

R. L. TAYLOR, F.C.S., F.I.C.

CHARLES OLDHAM.

ERNEST F. LANGE, F.C.S.

Assistant (Secretary aub |E;ibiaviau.

A. p. HUNT, B.A.

Ordinary Members. xxxix

ORDINARY MEMBERS.

Date of Election.

1901, Dec. 10. Adamson, Harold. Oaklands Cottage, Godley, near Man-

chester.

1902, Mar. 18. Allen, J. Fenwick. 147, Withiugton Road, Whalley

Range, Manchester.
IQ02, Jan. 21. Allott, Charles S., M.Inst.C.E. 519, Stretfotd Road, Old

Traffbrd, Manchester.
1S70, Dec. 13. Angell, John, F.C.S., F.I.C. 6, Beaconsfteld, Derby Road,

Withiugton, Manchester.
1S96, Jan. 31. Armstrong, Frank. 88, Deansgate, Manchester.

1865, Nov. 14. Bailey, Charles, M.Sc, F.L.S. Atherstone House, North
Drive, St. Annes-on-the-Sea, Lanes.

1888, Feb. 7. Bailey, Alderman Sir William H., M.I.Mech.E. Sale

Hall, Sale, Cheshire.
1S95, Jan. ^' Barnes, Charles L., M. A. ^, Swinton Avenue, Chorlton-on-
Medlock, Manchester.

1903, Oct. 20. Barnes, Jonathan, F.G.S. South Cliff House, 301, Great

Clowes Street, Higher Broughton, Manchester.

1896, April 14. Behrens, George B. Caerfedwcn, Trefnant, North IFales.

1895, Mar. 5. Behrens, Gustav. Holly Royde, IVithington, Manchester.
1898, Nov, 29. Behrens, Walter L. 22, Oxford Street, Manchester.
1868, Dec. 15. Bickham, Spencer H., F.L.S. Underdown^ Ledbury.

1 901, Nov. 12. Bles, Abraham J. S. Palm House, Higher Broughton,

Manchester.

1896, Oct. 6. Bowman, F.H., D.Sc, F.R.S.E. 4, Albert Square,

Manchester.
1896, Feb. 18. Bowman, George, M.D. ^()\, Stretford Road, Old Ira ford,

Manchester.
1S75, Nov. 16. Boyd, John. Barton House, 11, Duisbury Park, Didsbury,

Manchester.

1902, Oct. 21. Bradley, Henry Went worth. IVoodside. Wilmstow,

Cheshire.

1889, Oct. 15. Bradley, Nathaniel, F.C.S. Suiinyside, Whalley Range,

Manchester.

xl Ordinary Members.

Date of Election.

1861, April 2. Brogden, Henry, F.G.S., M.I.Mech.E. Hale Lodge,

Altriiichatn, Cheshire.
1889, April 16. Brooks, Samuel Herbert. Slade House, Leveushtilme,

Manchester .
i860, Jan. 24. Brothers, Alfred. Handforth, near Majiches/er.
1886, April 6, Brown, Alfred, M. A., M.D. Sandycroft, Higher Brotigh-

ton, Manchester.
1889, Jan. 8. Brownell, Thomas William, F.R.A.S. 64, Upper Brook

Street, Manchester.
1889, Oct. 15. Budenberg, C. F., M.Sc, M.I.Mech.E. Bozvdon Lane,

Marple, Cheshire.

1905, Nov. 14. Burgess, Charles H., M.Sc, Demonstrator of Chemistry

in the University of Manchester. Shaftesbury
House, Cheadle Hulme, Majtchester.

1906, Feb. 27. Burton, Joseph, A. R.C.S., Dublin. Tile Works, Clifton

Junctio7i, near Manchester.
1894, Nov. 13. Burton, William, F.C S. 77^1? Hollies, Clifton function,
near Manchester.

1904, Oct. 18. Campion, George Goring, L.D.S. 264, Oxford Street,
Manchester.

1903, Nov. 3. Capper, Stewart Henbest, M.A., Professor of Architecture
in the Victoria University of Manchester. 337, Moss
Lane East, Manchester.

1899, Feb. 7. Chapman, D. L., M.A., Assistant Lecturer and Demon-
strator of Chemistry in the Victoria University of
Manchester. The University, Manchester.

1901, Nov. 26. Chevalier, Reginald C, M.A. , Mathematical Master at the

Manchester Grammar School. 43, Lansdoivne Road,
West Didslmry, Manchester.

1902, Nov. 4. Clerk, Dugald, M.Inst.C.E., F.C.S. 18, Sonihampton

Bnildings, Chancery Lane, London, W. C.
1901, Nov. 12. Coignou, Caroline, M. A., Science Mistress at the Manchester

High School for Girls. 60, Cecil Street, Greenheys,

Manchester.
1895, April 30. Collett, Edward Pyemont. 8, St. John Street, Manchester.
1884, Nov. 4. Corbett, Joseph. Town Hall, Salford.

1903, Oct. 20. Core, William Hamilton, M.Sc. Grooinbridge House,

Withington, Manchester.
1895, Nov. 12. Crossley, W. J., M.I.Mech.E. Openshaw, Mamhesier.

Ordinary Members. xli

Datt ef Election.

1904, Oct. 18. Crosthwaite, Robert, M.A. Camb., B.Sc. Lond., Head
Master of the Municipal Secondary School, Whitworth
Street. 25, Rathen Road, Withingtou, Manchester.

1904, Jan. 5. Darbishire, Arthur D., M.A., B.Sc. ^i, Eardley Crescent,

London, S. IF.

1901, Nov. 26. Darbishire, Francis V., B.A., Ph.D., Demonstrator and

Analyst at the South Eastern Agricultural College. TAe
College, Wye, Kent.

1895, April 9- Dawkins, W. Boyd, M.A., D.Sc, F.R.S., Professor of

Geology in the Victoria University of Manchester.

Falloivjield House, Fallowfield, Manchester.
1894. Mar. 6. Delepine, A. Sheridan, M.B., B.Sc, Professor of

Pathology in the Victoria University of Manchester.

The University, Manchester.
1887, Feb. 8. Dixon, Harold Baily, M.A., M.Sc, F.R.S., F.C.S.,

Professor of Chemistry in the Victoria University of

Manchester. The University , Manchester.

1905, Jan. 10. Duffield, W. Geoffrey, B.A., B.Sc, Research Fellow in

the University of Manchester. The University, Man-
chester.

1906, Jan. 30. Dunkerley, Stanley, D.Sc, Professor of Engineering in the

University of Manchester. The University, Manchester.

1902, May 13. Ellison, Robert William. ^ Brookside,^ Crofts Bank Road,

Urmston, Manchester.

1883, Oct. 2. Faraday, F. J., F.L.S., F.S.S. Carshalton House,
Heaton Road, Withington, Manchester.

1905, May 2. Fearon, Ernest, Chemist to the Salford Corporation Gas
Works. Hundon House, Arnold Road, Alexandra Park,
Manchester.

1903, Dec. 15. Fishenden, Richard B., Lecturer in Photo-Mechanical

Processes at the Municipal School of Technology,
Manchester. 311, Moss Lane East, Manchester.

1898, Nov. 29. Gamble, F. W., D.Sc, Assistant Lecturer and Demonstrator
of Zoology in the Victoria University of Manchester.
The University, Manchester.

1896, Nov. 17. Gordon, Rev. Alexander, M.A. Swnmerville, Victoria

Park, Manchester.
1905, Nov. 20. Grimshaw, William Edwin, B.A., Physics Master, Man-
chester Grammar School. 46, Broadway Street, Oldham.

xlii

Date of Election.

1902, April

29.

1905, Oct.

17-

1902, Jan.

7-

1895, Mar,

5-

1884, Jan.

8.

1905. Nov.

14.

1898, Nov.

29.

1896, Nov.

3-

1889, Oct.

IS.

1900, Oct.

16.

, Ordinary Members.

Herbert, Arthur M., B.A. Frankwyn, Hale, Cheshire.
Hewitt, Charles Gordon, B.Sc, Assistant Lecturer and

Demonstrator in Zoology in the Victoria University of

Manchester. The University , Manchester.
Hewitt, David B., M.D. Oakleigh, Norlhivich, Cheshire.
Hickson, Sydney J., M.A., D.Sc, F.R.S., Professor of

Zoology in the Victoria University of Manchester. The

Un iversity, Manchester '.
Hodgkinson, Alexander, M.B., B.Sc. 18, St. lohn Street,

Alanchester.
Holt, Alfred, M.A., Research Fellow of the University of

Manchester. Croft on, Aigbiirth, Liverpool.
Hopkinson, Alfred, K.C., M.A., LL.D., Vice-Chancellor

of the Victoria University of Manchester. Fairjield,

Victoria Park, Manchester.
Hopkinson, Edward, M.A., D.Sc, M.Inst.C.E. Ferns,

Aider ley Edge, Cheshire.
Hoyle, William Evans, M.A., D.Sc, F.R.S.E., Director

of the Manchester Museum. The University , Manchester .
Hutton, R. S., D.Sc, Lecturer on Electro-Chemistry in the

Victoria University of Manchester. Jhe University,

Manchester.

J, Oct. 17. Ingleby, Joseph, M.LMech.E. Sunimer Hill, Pendleton,

Manchester.

1901, Nov. 26. Jackson, Frederick. 14, Cross Street, Manchester.
1870, Nov. r. Johnson, William H., B.Sc. IVoodleigh, Altrincham.
1878, Nov. 26. Jones, Francis, M.Sc, F.R.S.E., F.C.S. Manchester

Grammar School, and Beaufort House, Alexandra Park,
Manchester.

1886, Jan. 12. Kay, Thomas. Moor field, Stockport, Cheshire.
1895, Nov. 12. Kirkman, William Wright. The Grange, Tiniperley,
Cheshire.

1903, Feb. 3. Knecht, Edmund, Ph.D., Professor of Tinctorial

Chemistry at the Municipal School of Technology,
Manchester. 5, Station Road, Cruvipsall, Manchester.

1904, Oct. 18. Knowles, H. B., M.A., Principal of the Royal Technical

Institute, Salford. ' Norcote,' Swinton Park, Salford.

1902, Feb. 4. Kolp, Noah, Woodthorpe, Victoria Park, Manchester.

Ordinary Members. xliii

Date 0/ Election.

1893, Nov. 14. Lamb, Horace, M.A., LL.D., D.Sc, F.R.S., Professorof

Mathematics in the Victoria University of Manchester.
6, Wilhraham Road, Fallozvfield, Manchester.

1902, Jan. 7. Lange, Ernest F., F.C.S. Fairholine, 3, Willow Bank,

Fallowfie/d, Manchester.

1904, Mar. 15. Lea, Arnold W. VV., M.D. 246, Oxford Road, Manchester.

1903, Nov. 17. Leigh, Charles W. E , Librarian of the University. The

University , Manchester.
1902, Nov. 4. Leigh, Sir Joseph Egerton. The Towers, Didsbury,

JManchesler.
1902, Jan. 7. Longridge, Michael, M.A., M.Inst.C.E. Linkvretten,

Ashley Road, Bozvdon, Cheshire.
1857, Jan. 27. Longridge, Robert Bewick, M.I. Mech.E. Yew 7 ree House,

Tabley, Knutsford, Cheshire.

1S66, Nov. 13. McDougall, Arthur, B.Sc. Lyndhttrst, The Park, Buxton.

1905, Oct. 31. McNicol, Mary, B.Sc, Research Scholar in the Victoria

University of Manchester. 182, Upper Chorlton Road,
Manchester.

1904, Nov. I. Makower, Walter, B.A., B.Sc. 214, Upper Brook Street,

Manchester.

1902, Mar. 4. Mandleberg, Goodman Charles. Carlton House, Broom

Lane, Higher Broughton, Manchester .
1875, Jan. 26. Mann, J. Dixon, M.D., F.R.C.P. (Lond.), Professor of
Medical Jurisprudence in the Victoria University of
Manchester. 16, St. John Street, Manchester.

1901, Dec. 10. Massey, Herbert. Izy Lea, Biirnage, Didsbury,

Manchester.
1864, Nov. I. Mather, Sir William, M.P., M.Inst.C.E., M.LMech.E. Iron

Works, Salford.
1873, Mar. 18. Melvill, James Cosmo, M.A., F.L.S. Meole-Brace Hall,

Shreivshiiry.
1 88 1, Oct. 18. Mond, Ludwig, D.Sc, Ph.D., F.RS., F.C.S. Winnington

Hall, Northwich, Cheshire.

1894, Feb. 6. Mond, Robert Ludwig, M. A., F.R.S.E., F.C.S. Winning-

ton Hall, Nofthivich, Cheshire.

1903, Oct. 20. Moore, Frederick Craven, M.D., M.Sc, 61, Ardwick

Green, Manchester.
1899. Mar. 7. Morris, Edgar F., M.A., F.C.S. Grey House, Barrington
Road, Altrincham, Cheshire.

1902, P'eb. 18. Moss, William E., B.A. C\o Messrs. Davies, Benachi

iSr' Co., 7, Rumford Street, Liverpool.

xliv Ordi?iary Memters.

Date of Election.

1873, Mar. 4. Nicholson, Francis, F.Z.S. The KnoU, Windermere,

Westmoreland.
1900, April 3. Nicolson, John T., D.Sc, Professor of Engineering at the

Municipal School of Technology, Manchester. Nant-y-

Glyn, Marple, Cheshire.
1889 April 16. Norbury, George. Hillside, Preshvick Paik, Prestwich,

Lanes.

1884, April 15. Okell, Samuel, F. R.A.S. Overley, Langham Road,

Boivdon, Cheshire.
1903, Jan. 6. Oldham, Charles. Brook Cottage, Ktmtsford, Cheshii'C.

1 901, Nov. 26. Paine, Standen. Devisdale, Boivdon, Cheshire.

1S92, Nov. 15. Perkin, W. H., jun., Ph.D., M.Sc, F.R.S., Professor of
Organic Chemistry in the Victoria University of Man-
chester. The University, Manchester.

1901, Oct. 29. Petavel, J. E., B.A. The University, Manchester.

1885, Nov. 17. Phillips, Henry Harcourt, F.C. S. Lynwood, Turton,

nr. Bolton, Lanes.

1902, Oct. 21. Pope, W. J., F.R.S., F.C. S., Professor of Chemistry at the

Municipal School of Technology, Manchester. Corchester,
Bramhall, Cheshire.
1901, Nov. 12. Pratt, Edith M., D.Sc. Peak House, Dukinjieid, Cheshire.

1903, Dec. 15. Prentice, Bertram, Ph.D., D.Sc, Lecturer in Chemistry,

Royal Technical Institute, Salford. Primrose Villa,
Snowdon Road, Eccles.

1903, Feb. 3. Radcliffe, L. G., F.C.S., Lecturer in Chemistry at the

Municipal School of Technology, Manchester. 6, Alim
Terrace, Old Traffbrd, Manchester.

1904, Feb. 2. Radford, Catherine, B.Sc. 31, Caivdor Road, Fallowjield,

Matichester.

1900, P'eb. 20. Ragdale, John R. The Beeches, Whitejield, near Man-

chester.

1901, Dec. 10. Ramsden, Herbert, M.D. (Lond.), M.B., Ch.B. (Vict.).

Sunnyside, Dobcross, near Oldham, Lanes.
1888, Feb. 21. Ree, Alfred, Ph.D., P'.C.S. 15, Mauldeth Road, With-

ington, Manchester.
1869, Nov. 16. Reynolds, Osborne, M.A., LL.D., F.R.S., M.Inst.C.E.

19, Ladybarn Road, Fallowfield, Manchester.

Ordinary Members. xlv

Date of Election,

1880, Mar. 23. Roberts, D. Lloyd, M.D., F.R.S.E., F.R.C.P. (Lend.).
Ravenswood, Broughton Park, Manchester.

1897, Oct. 19. Rolhwell, William Thomas. Heath Brewery, Newton

Heath , near Manchester.

1905, Oct. 31. Saxelby, Edith Mary, B.Sc, Research Scholar in the

Victoria University of Manchester. 3, Alexandra Road

South, Alexandra Park, Manchester.
1873, Nov. 18. Schuster, Arthur, Sc.D., Ph.D., F.R.S., F.R. A. S., Professor

of Physics in the Victoria University of Manchester. Kent

House, Victoria Park, Manchester.

1898, Jan. 25. Schwabe, Louis. Hart Hill, Eccles Old Road, Pendleton,

Manchester.

1902, Jan. 21. Shann, Sir Thomas Tnornhill. Meadoiv Bank, Heaton

Not r is, Stockport.
1890, Nov. 4. Sidebotham, Edward John, M.A., M.B., M.R.C.S.
Erlesdene, Boivdoi, Cheshire.

1903, April 28. Sidebottom, Henry. The Hall Cottage, Cheadle Huline,

near Stockport.
1901, Oct. 29. Sinclair, Sir W. J., M.D., Professor of Obstetrics and
Gynaecology in the Victoria University of Manchester.
Garvock House, Dudley Road, Whalley Range, Man-
chester.

1895, Nov. 12. Southern, Frank, B.Sc. 6, Pa)k Avenue, Timperley,

Cheshire.

1896, Feb. 18. Spence, David. Honeyhanger, Hasleinere, Surrey.
1901, Dec. 10. Spence, Howard. Audley, Broad Road, Sale, Cheshire.

1904, Nov. I. Stansfield, Herbert, B.Sc, A.LE.E. 20, Every Street,

Ancoats, Manchester.

1905, May 2. Stopes, Marie C, D.Sc, Ph.D., Demonstrator of Botany

in the University of Manchester. 11, Kensington
Avenue, Victoria Park, Manchester.

1897, Nov. 30. Stromeyer, C. E., M.Inst.C.E. Steam Users' Association,

9, Mount Street, Albert Square, Manchester.

1905, Nov. I. Sutcliffe, William Henry, F.G.S. Shore, Littleborough,
Lanes.

1895, April 9. Tatton, Reginald A., M.Inst.C.E. Engineer to the
Mersey and Irwell Joint Committee. 44, Mosley Street,
Manchester.

1893, Nov. 14. Taylor, R. L., F.C.S., F.LC. Central School, Whitivorth
Street, and 37, Mayfield Road, Whalley Range, Man-
chester.

xlvi Ordinary Members.

Date of Election.

1873, April 15. Thomson. William, F.R.S.E., F.C.S., F.I.C. Royal
Institniion, Manchester.

1896, Jan. 21. Thorburn, William, M.D., B.Sc. 2, St. Peters Square,
RIanchester.

1896, Jan. 21. Thorp, Thomas, F. R.A. S. Moss Bank, IVhitefield, near
Manchester.

1899, Oct. 31. Thorpe, Jocelyn F., Ph.D., Demonstrator in Organic
Chemistry in the Victoria University of Manchester.
The University , Afanchester.

1899, Oct. 17. Todd, William Henry. Greenfield, Flixton, near Man-
chester.

1873, Nov. 18. Waters, Arthur William, F.L.S., F.G.S. '' Alder ley,"

Mc Kinky Road, Bournemouth.
1892, Nov. 15. Weiss, F. Ernest, D.Sc, F.L.S., Professoi of Botany

in the Victoria University of Manchester. 20, Brunswick

Road, Withington, Manchester.
1895, April 9. Whitehead, James. Lindfield, Fulshaiv Park, Wilms low,

Cheshire.
1901, Oct. I. Wild, Robert B., M.D., M.Sc, M.R.C. P., Professor of

Materia Medica and Therapeutics in the Victoria

University of Manchester. Broome House, Fallowfield,

3Ianchester.
1859, Jan. 25. Wilde, Henry, D.Sc, D.C.L., F.R.S. The Hurst, Alderley

Edge, Cheshire.
1905, Oct. 31. Willis, Ethel G., M.A., B.Sc, Science Mistress, Man-
chester High School for Girls. 22, Cawdor Road,

Fallowfield, Manchester.

1901, Nov. 26. Wilson, William, M.A. Carron Vale, 80, Fitzivarren

Street, Pendleton, Manchester.
1903, Oct. 20. Wood, Harry EdwiO; B.Sc. The Physical Laboratory,

The University, Manchester.
1905, Oct. 31. Woodall, Herbert J., A.R.C.S. Z'2, Market Place, Stockport.

1902, Oct. 21. Woollcott, Walter. IVestinghouse Works, Trafford Park,

A/anchester.
i860, April 17. Woolley, George Stephen. Victoria Bridge, Manchestei:

1903, Nov. 17. Worthington, John Henry William, B.A., Assistant Master

at the Manchester Grammar School. 60, Filey Road,
Fallowfield, Manchester.
1863, Nov. 17. Worthington, Samuel Barton, M.Inst.C.E., M.I.Mech.E.
Mill Hank, Bowdon, and 37, Princess Street, Manchester.

Ordinary Members. xlvii

Date of Election.

1865, Feb. 21. WorLhington, Thomas, F.R.I. B.A. 46, Brown Street,

Manchester.
1895, Jan. 8. Worthington, Will. Barton, B.Sc, M.Inst. C.E. Kirkslyks,

Diiffiehi, near Derby.
1897, Oct. 19. Wyatt, Charles H., M.A., Ckel/ord, Cheshire.

N.B. — Of the above hst the following have compounded fur their
subscriptions, and are therefore life members : —

Bailey, Charles, M.Sc, F.L.S.
Bradley, Nathaniel, F.C.S.
Brogden, Henry, F.G.S.
Ingleby, Joseph, M.I.Mech.E.
Johnson, William II., B.Sc.
Worthington, Wm. Barton, B.Sc.

xlviii Honorary Members.

HONORARY MEMBERS.

Date of Election.

i8y2, April 26. Abney, Sir W. de W., K.C.B., D.Sc, F.R.S. Rathtnore

Lodge, Bolton Gardens South, South Kensington, London,

S.IV.
1892, April 26. Amagat, E. H., For. Mem. R.S., Memb. Inst. Fr.

(Acad. Sci.), Examinateur a I'Ecole Poly technique.

Avenue d'' OrUans, 19, Paris.

1894, April 17. Appell, Paul, Membre de I'lnstitut, Professor of Theoretical

Mechanics. Factilti des Sciences, Paris.
1892, April 26. Ascherson, Paul F. Aug., Professor of Botany. Universitdt,

Berlin.
1889, April 30. Avebury, John Lubbock, Lord, D.C.L., LL.D., F.R.S.

High Elms, Down, Kent.

1892, April 26. Baeyer, Adolf von, For. Mem. R. S., Professor of Chemistry.

I, A}-cisstrasse, Munich.
18S6, Feb. 9. Baker, Sir Benjamin, K.C.M.G., LL.D., F.R.S. 2, Queen

Square Place, Westminster, London, S. fV.
1886, Feb. 9. Baker, John Gilbert, F.R.S., F.L.S. 3, Cumberland

Road, Kezv.

1895, April 30. Beilstein, F., Ph.D., Professor of Chemistry. 8th Line,

N. 17, St. Petersburg, IV. 0.
1886, Feb. 9. Berthelot, Marcelin P. E., For. Mem. R.S., Membre de

I'lnstitut, Professor of Chemistry, Secretaire perpetual

de I'Academie des Sciences. Paris.
1892, April 26. Boltzmann, Ludwig, For. Mem. R S., Professor of Physics.

Liirkenstrasse J, Vienna, IX. i.
1886, Feb. 9. Buchan, Alexander, M.A., LL.D., F.R.S., F.R.S.E.

42, Heriot Row, Edinburgh,

1888, April 17. Cannizzaro, Stanislao, For. Mem. R.S., Corr. Memb.

Inst. Fr. (Acad. Sci.), Professor of Chemistry. Reale
Universith, Rome.

1889, April 30. Carruthers, William, F.R.S., F.L.S. 14, Vermont Road,

Norjvood, London, S.E.

1903, April 28. Clarke, Frank Wigglesworth, D.Sc. United States
Geological Survey, IVashington, D.C., U.S.A.

1866, Oct. 30. CHfton, Robert Bellamy, M.A., F.R.S., F.R.A.S., Pro-
fessor of Natural Philosophy. 3, Bardzvell Road,
Banbury Road, Oxford.

Honorary Members. xlix

Date ef Election.

1892, April 26. Curtius, Theodor, Professor of Chemistry. Universitdt,
Kiel.

1892, April 26. Darboux, Gaston, Membra de I'lnstitut, Professor of
Geometry, Faculte des Sciences, Secretaire perpetuel de
I'Academie des Sciences. 36, Rue Gay Lussac, Paris.

1894, April 17. Debus, H., Ph.D., F.R.S. 4. Scklangenweg, Cassel,

He s sen, Germany.

1888, April 17. Dewalque, Gustave, Professor of Geology. University,

Liige.

1900, April 24. Dewar, Sir James, M.A., LL.D., D.Sc, F.R.S., V.P.C.S.,

Fullerian Professor of Chemistry, Royal Institution,

Albemarle Street, London, IV.
1892, April 26. Dohrn, Dr. Anton, For. Mem. R.S. Zooloqische Station,

Naples.
1892, April 26. Dyer, Sir VV. T. Thiselton, K.C.M.G., C.I.E., M.A.,

F.R.S., Director of the Royal Botanic Gardens. Kew.

1892, April 26. Edison, Thomas Alva. Orange, N.[., U.S.A.

1895, April 30. Elster, Julius, Ph. D. 6, Lessingstrasse, Wolfenbiittel.
1900, April 24. Ewing, James Alfred, M.A., F.R.S.,Professor of Mechanism

and Applied Mechanics. Royal Naval College, Greenwich.

1889, April 30. Farlow, W. G., Professor of Botany. Harvard College,

Cambridge, Mass., U.S.A.

1900, April 24. Forsyth, Andrew Russell, M. A., Sc.D., F.R.S., Sadlerian
Professor of Pure Mathematics. Trinity College, Cam-
bridge.

1889, April 30. Foster, Sir Michael, K.C.B., M.P., M.A., M.D., LL.D.,
Sec. R.S., Professor of Physiology. Trinity College,
Cambridge.

1892, April 26. Fiirbringer, Max, Professor of Anatomy. Grossherz.
Universitdt, Jena.

1900, April 24. Geikie, James, D.C.L., LL.D., F.R.S. , Murchison Pro-
fessor of Geology and Mineralogy. Kilmorle, Coiinton
Road, Edinburgh.

1895, April 30. Geitel, Hans. 6, Lessingstrasse, IVolfenbUttel.

1894, April 17. Glaisher, J. W. L., Sc.D., F.R.S., Lecturer in Mathematics.
Trinity College, Cambridge.

1894, April 17. Gouy, A., Corr. Memb. Inst. Fr. (Acad. Sci.), Professor
of Physics. Faculty des Sciences, Lyons.

1 Honorary Members.

Date ef Election.

1900, April 24. Haeckel, Ernst, Ph. D. , Professor of Zoology. Zoologisches

Institut, Jena.
1894, April 17. Harcourt, A. G. Vernon, M.A., D.C.L., F.R.S., V.P.C.S.

St. Clare, Ryde, Isle of Wight.
1894, April 17. Heaviside, Oliver, F. R.S. Bradley Vieiv, Newton Abbot,

Devon.
1892, April 26. Hill, G. W. West Nyack, N. Y., U.S.A.
1888, April 17. Hittorf, J ohann Wilhelm, Professor of Physics. Polytech-

nicum, AlUnUer.
1892, April 26. Hoff, J. van't, Ph.D., For. Mem. R.S., Professor of

Chemistry. 2, Uhlandstrasse, Charlotfetiburg, Berlin.
1892, April 26. Hooker, Sir Joseph Dalton, G.C.S.I., C.B., D.C.L.,

F.R.S., Corr. Memb. Inst. Fr. (Acad. Sci.). The Camp,

Sunningdate, Berks.
1869, Jan. 12. Huggins, Sir William, O.M., K.C.B., LL.D., D.C.L.,

P.R.S., F.R.A.S., Corr. Memb. Inst. Fr. (Acad. Sci.).

90, Upper Tulse Hill, Brixton, London, S. IF.

1S51, April 29. Kelvin, William Thomson, Lord, O.M., G.C.V.O., M.A.,

D.C.L., LL.D., F.R.S., F.R.S.E., For. Assoc. Inst.

Fr. (Acad. Sci.). Netherhall, Laigs, Ayrshire.
1892, April 26. Klein, Felix, Ph.D., For. Mem. R.S. , Corr. Memb. Inst.

Fr. (Acad. Sci.), Professor of Mathematics. 3, Wilhelm

Weber Strasse, Goltingen.
1894, April 17. Konigsberger, Leo, Professor of Mathematics. Universitd',

Heidelberg,

1892, April 26. Ladenburg, A., Ph.D., Professor of Chemistry. 3, Kaiser

Wilhelm Strasse, Breslati.
1902, May 13. Larmor, Joseph, M.A., D.Sc, LL.D., Sec. R.S., F.R.A.."^.

St. John'' s College, Cambridge.
1892, April 26. Liebermann, C, Professor of Chemistry. 29, Alatthii/-

Kirch Strasse, Berlin.
5^1887, April 19. Lockyer, Sir J. Norman, K.C.B., F.R.S,, Corr. Mem^».

Inst. Fr. (Acad. Sci. ). Science School, South Kensington,

London, S. W.
1902, May 13. Lodge, Sir Oliver Joseph, D.Sc, LL.D., F.R.S., Principal

of the University of Birmingham. The University,

Birmingham.
1900, April 24. Lorentz, Henrik Anton, Corr. Memb. Inst. Fr. (Acad. Sci.),

Professor of Physics. Hooigracht, 48, Leyden.

Honorary Members. li

Date of Election.

1892, April 26. Marshall, Alfred, IVl.A., Professor of Political Economy.

Balliol Croft, Madingley Road, Caml'ridge.
1S92, April 26. Mascart, E. E. N., For. Mem. R.S., Membre de I'Institut,

Professor at the College de France. 176, Rue de

r University, Paris.
1889, April 30. Mendeleefif, D., Ph.D., For. Mem. R.S. Univeisiti, St.

Petersburg.

1901, April 23. Metchnikoff, Elie, D.Sc, For.Meni.R.S. Institut Pasteur,

Paris.
1895, April 30. Mittag-I.effler, Gosta, D.C.L. (Oxon.), For. Mem. R.S.,

Professor of Mathematics. Djtirsholm, Stockholm.
1892, April 26. Moissan, H., Membre de I'Institut, Professor of the Faculte

des Sciences a la Sorbonne. 7, Rue Vauquelin, Paris.
1894, April 17. Murray, Sir John, K.C.B., LL.D., D.Sc, F.R.S

Challetiger Lodge, Wardie, Edinburgh.

1894, April 17. Neumayer, Professor G., For. Mem. R.S., Director of the

Seewarte. HoheiizoUo-n Strasse, 9, Neustadt an dtr

Haardt, Germany.
1887, April 19. Newcomb, Simon, For. Mem. R.S., For. Assoc. Inst. Fr.

(Acad. Sci.), Professor of Mathematics and Astronomy.

1620, P Street, Washington, D.C., U.S.A.

1902, May 13. Osborn, Henry Fairfield, Professor of Vertebrate Paleon-

tology. Cc'Iujnhia College, New York, U.S.A.
1894, April 17. Ostwald, W., Professor of Chemistry. Groszbothen, Kgr.
Sachscn.

1899, April 25. Palgrave, R. H. Inglis, F.R.S. , F.S.S. Belton, Great

Yarmouth.
1S92, April 26. Perkin, Sir W. H., LL.D., Ph.D., F.R.S., V.P.C.S. 7he

Chestnuts, Sudbwy, Harrow.
1S94, April 17. Pfeffer, Wilhelm, For. Mem. R.S., Professor of Botany.

Botanisches Institut, Leipsic.
1S92, April 26. Poincare, H., For. Mem. R.S., Membre de I'Institut,

Professor of Astronomy. 6^, Rue Clatide Bernard, Paris.

1S92, April 26. Quincke, G. H., For. Mem. R.S., Professor of Physic^.

Un ivers it at. He iu el berg.

1899, April25. Ramsay, Sir William, K.C.B., Ph.D., F.R.S., Professor of
Chemistry. ' 12, Arundel Gardens, Notting Hill,
London, VV.

Hi Honorary Members.

Date of Election.

1886, Feb. 9. Rayleigh, John William Strutt, Lord, O.M., M.A.., D.C.L.
(Oxon.), LL.D. (Univ. McGill), F.R.S., F.R.A.S.,
Corr. Memb. Inst. Fr. (Acad. Sci.). Terling Place,
Wit ham, Essex.

1900, April 24. Ridgway, Robert, Curator of the Department of Birds, U.S.
National Museum. Brookland. District of Columbia,
U.S.A.

1897, April 27. Roscoe, Sir Henry Enfield, B.A., LL.D., D.C.L., F.R.S.,
V.P.C.S., Corr. Memb. Inst. Fr. (Acad. Sci.). 10,
Bramhain Gardens, EarPs Court, London, S. IV.

1889, April 30. Routh, Edward John, D.Sc, F. R.S. Neivnhain Cottage,
Qzieen's Road, Cambridge,

1902, May 13. Scott, Dukinfield Henry, M.A., Ph.D., F.R.S., F.L.S.,

Honorary Keeper of the Jodrell Laboratory, Royal

Botanic Gardens, Kevv. Old Palace, Richmond, Surrey.
1892, April 26. Sharpe, R. Bowdler, LL D., F.L.S., F.Z.S. British

Uluseum (Natural History ), Cromiuell Road, London,

S. IV.
1892, A[)ril 26. Solms, H., Grafzu, Professor of Botany. Univirsitiil,

Strassbitrg.
1869, Dec. 14. Sorby, Henry Clifton, LL.D., F.R.S., F.L.S., F.G.S.

Broomfield, Sheffield.
1886, Feb. 9. Strasburger, Eduard, D.C.L., For. Mem. R.S., Professor

of Botany. Universitdt, Bonn.
1895, April 3°- Suess, Eduard, Ph.D., For. Mem. R.S., For. Assoc. Inst.

Fr. (Acad. Sci.), Professor of Geology. 9, Africaner-

gasse, Vienna.

1895, April 30. Thomson, Joseph John, M.A., Sc.D., F.R.S., Professor of

Experimental Physics. 6, Scrope Terrace, Cambridge.
1894, April 17. Thorpe, T. E., C.B., Ph.D., D.Sc, LL.D., F.R.S.,

V. P.C. S. Governmmt Laboratory, Clement's Inn

Passage, Strand, London, W. C.
1894, April 17. Turner, Sir William, K.C.B., M.B., D.C.L., F.R.S.,

F. R.S. E., Professor of .■\natomy. 6, Eton Terrace,

Edinburgh.
1886, Feb. 9. Tylor, Edward Burnett, D.C.L. (Oxon), LL.D. (St. And.

and McGill Colls.), F. R.S., Professor of Anthropology.

Museum House, Oxford.

Corresponding Member. Hu

Daie of Election.

1894, April 17. Vines, Sidney Howard, M.A„ D.Sc, F.R.S., Sherardian
Professor of Botany. Heading/on Hill, Oxford.

1894, April 17. Warburg, Emil, Professor of Physics. Phynkalisches
Insfitut, Nene IVilhel/nslrasse, Berlin.

1894, April 17. Ward, II. Marshall, D.Sc, F.R.S., Professor of Botany.
Botanical Laboratory, New Museums, Cambridge.

1894, April 17. Weismann, August, Professor of Zoology. Universitdt,
Freiburg i. Br.

1886, Feb. 9. Young, Charles Augustus. Hanover, New Hampshire,
U.S.A.

1888, April 17. Zirkel, Ferdinand, For. Mem. R.S., Professor of Mineralogy.
Thralstrasse, 33, Leipsic.

CORRESPONDING MEMBER.

1850, April 30. Harley, Rev. Robert, Hon. M.A.(Oxon),F. R.S.,F.R. A.S.,
Hon. Memb. R.S. Queensland. Rosslyn, Westbourne
Road, Forest Hill, London, S.E., and 7 he Athenaum
Club, London, S. W.

liv Aivards of Medals ana Preiniunis.

Awards of the Wilde Medal under the conditions of the
Wilde Endowment Fund.

1896. Sir George G. Stokes, Bart, F.R.S.

1897. Sir William Huggins, K.C.B., RR.S.

1898. Sir Joseph Dalton Hooker, G.C.S.I., C.B.,

F.R.S.

1899. Sir Edward Frankland, K.C.B., F.R.S.

1900. Rt. Hon. Lord Rayleigh, F.R.S.

1901. Dr. Elie Metchnikoff, For.Mem.R.S.
T903. Prof. Frank W. Clarke, D.Sc.

1905. Prof. Charles Lapworth, LL.D., F.R.S.

Awards of the Dalton Medal.
1898. Edward Schunck, Ph.D., F.R.S.
1900. Sir Henry E. Roscoe, F.R.S.
1903. Prof Osborne Reynolds, LL.D., F.R.S.

Aivards of the Premium under the conditions of the
Wilde Endoivment Fimd.

1897
1898
1899
1900,
1901

Peter Cameron.
John Butterworth, F.R.M.S.
Charles H. Lees, D.Sc
Prof A. W. Flux, M.A.
Thomas Thorp.

The Wilde Lectures. Iv

THE WILDE LECTURES.

1897. (July 2.) " On the Nature of the Rontgen Rays."

By Sir G. G. Stokes, Bart, F.R.S. {28pp.)

1898. (Mar. 29.) "On the Physical Basis of Psychical

Events." By Sir Michael Foster, K.C.B.,
F.R.S. {46 pp.)

1899. (Mar. 28.) "The newly discovered Elements;

and their relation to the Kinetic Theory of
Gases." By Prof. William Ramsay, F.R.S.
{19 pp.)

1900. (Feb. 13.) " The Mechanical Principles of Flight."

By the Rt. Hon. LORD Rayleigh, F.R S.
{26 pp.)

1901. (April 22.) " Sur la Flore du Corps Humain."

By Dr. Elie Metchnikoff, For.Mem.R.S.

isspp.)

1902. (Feb. 25.) " On the Evolution of the Mental

Faculties in relation to some Fundamental
Principles of Motion." By Dr. Henry Wilde,
F.R.S. {34PP-,3P^-)

1903. (May 19.) *' The Atomic Theory." By Professor

F. W. Clarke. D.Sc. {32 pp.)

1904. (Feb. 23.) " The Evolution of Matter as revealed

by the Radio-active Elements." By Frederick
SODDY, M. A. {4.2 pp.)

1905. (Feb. 28.) " The Early History of Seed-bearing

Plants, as recorded in the Carboniferous Flora."
By Dr. D. H. Scott, F.R.S. {32 pp.,3pi-)

1906. (March 20.) "Total Solar Eclipses." By Pro-

fessor H. H. Turner, D.Sc, F.R.S. {32pp.)

J