ie
Se
rh
i
ga
|
}

ae

Cited ech dean ae

ee

ae
Saye oe

qe

:

ee

PNET P ent circa earnereiomaitatae
os :

as

a ee

2S

SI

CPN Pte

as
“-
ACTOS
Seen
Se

re

Sree
Ss

SES
eres

Rus.

One:

LG

eee

See

Oe nS

Beane

Sr

ED

aes

Se

Sane
M3

=
~~

i)

i

Ay

yi

x

"ih
aun

A
tris) My

Cy ey
PESTS TS

ee

MURS
PANE

een

MEI)

i bt
yi

R

irk heas
its a $
te Hon
Pay ve
SOOM

RR
a a

om

Nhe
ee

> at AY

sat
Oe
m

uss

Bel aOR RIES
SAE BETES
7 Sar : D

Pot

SMITHSONIAN
MISCELLANEOUS COLLECTIONS

VOL. 85

Ooo CCCong,

“EVERY MAN IS A VALUABLE MEMBER OF SOCIETY WHO, BY HIS OBSERVATIONS, RESEARCHES, AND
EXPERIMENTS, PROCURES KNOWLEDGE FOR MEN’’—SMITHSON

(PUBLICATION 3175)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
1933

THE SCIENCE PRESS PRINTING COMPANY
LANCASTER, PENNSYLVANIA

ADVERTISEMENT

The present series, entitled ‘‘ Smithsonian Miscellaneous Collec-
tions ’’, is intended to embrace all the octavo publications of the In-
stitution, except the Annual Report. Its scope is not limited, and
the volumes thus far issued relate to nearly every branch of science.
Among these various subjects zoology, bibliography, geology, min-
eralogy, anthropology, and astrophysics have predominated.

The Institution also publishes a quarto series entitled “Smith-
sonian Contributions to Knowledge”. It consists of memoirs based
on extended original investigations, which have resulted in impor-
tant additions to knowledge.

C. G. ABzot,
Secretary of the Smithsonian Institution.

(iii )

LO;

LE

CONTENTS

Aspot, C. G. Weather dominated by solar changes. 18 pp., 4
text figs. Feb. 5, 1931. (Publ. 3114.)

WETMoRE, ALEXANDER. The avifauna of the Pleistocene in
Florida. 41 pp., 6 pls., 16 text figs. Apr. 13, 1931. (Publ.
3115.)

Watcott, CuarLes D. (With explanatory notes by Charles E.
Resser.) Addenda to descriptions of Burgess shale fossils.
40 pp., 23 pls., 11 text figs. June 29, 1931. (Publ. 3117.)

Tuériot, I. Mexican mosses collected by Brother Arsene
Brouard—III. 44 pp., 22 text figs. Aug. 25, 1931. (Publ.
2122.)

Brackett, F. S., and LippeLt, Urner. Infra-red absorption
bands of hydrogen cyanide in gas and liquid. 8 pp., 5 text
fies. Angry 5, tosis. ( Publ. 3123. )

Snopcrass, R. E. Morphology of the insect abdomen. Part I.
General structure of the abdomen and its appendages. 128
pp., 46 text figs. Nov. 6, 1931. (Publ. 3124.)

McAtTer, W. L. Effectiveness in nature of the so-called pro-
tective adaptations in the animal kingdom, chiefly as illus-
trated by the food habits of Nearctic birds. 201 pp. Mar.
15,1932. (Publ..3125.)

SwanTon, JoHN R. Modern square grounds of the Creek In-
dians. 46 pp., 5 pls., 15 text figs. Nov. 11, 1931. (Publ.
3126.)

WuLF, OLiver R. The determination of ozone by spectrobolo-
metric measurements. 12 pp., 3 pls., 5 text figs. Nov. 30,
1O¢1. (Publ 2127,)

Micrer, Gerrit S., Jk. Human hair and primate patterning.
12 pp, 5 pls..Wec. 19,1031. 0( Publ. 3130.)

AvpricH, L. B. Supplementary notes on body radiation. 12
pp., 5 text figs. Feb. 2, 1932. (Publ. 3131.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VOLUME 85, NUMBER 1

HHodgkins Fund

and

Roebling Fund

WEATHER DOMINATED BY SOLAR
CHANGES

BY
C. G. ABBOT

(PUBLICATION 3114)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
FEBRUARY 5, 1931

The Lord Baltimore Preas

BALTIMORE, MD., U. 8. A.

Hovgkins Fund and iocbling Fund

WEATHER DOMINATED BY SOLAR CHANGES
By C..G. ABBOT

My title suggests a radical change of view regarding weather and
weather forecasting. Let us contrast, for a moment, weather and
climate. All men realize that it 1s the sun which furnishes the heat
which warms the earth, and that the regular motions of rotation of
the earth upon its axis, and of its revolution in its orbit around the
sun produce those periodic variations of the solar heating which
govern climates. Differences in latitude and of proximity to oceans
and to other great terrestrial features introduce alterations from place
to place in these periodic changes of solar heating; thereby are pro-
duced climatic differences. As regards weather, which consists in
departures from regularity in climate, I suppose that practically all
meteorologists have been holding hitherto that it depends principally
on the complexities of the earth. According to that view, weather
represents, as it were, the changing eddies and whirlpools in the
Niagara of climate, due to the jutting rocks of local circumstances,
and, owing to enormous complexities, 1s essentially unpredictable for
any considerable time in advance.

I shall present evidence to show that weather, on the contrary, is
caused chiefly by the frequent interventions of actual changes of the
emission of radiation within the sun itself. Local conditions, to be
sure, alter the magnitudes and times of the effects of these interven-
tions into terrestrial affairs by the variable sun, but in ways determin-
able by statistical studies. Hopeful indications will be given that
changes of the solar radiation and their weather-consequences may be
predictable long in advance.

Figure 1 shows the daily observations of the solar constant of radia-
tion made at Montezuma, Chile, by the Astrophysical Observatory of
the Smithsonian Institution since 1924. The values give the intensity
of the sun’s radiation as it would be found by an observer in free
space situated at the earth’s mean distance from the sun. As far as
possible, they are independent of any effects of the varying trans-

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No, 1

1924

ia. LAAT
| cet

Eee
eae

Fic. 1.—Daily observations at Montezuma, Chile, of the “ Solar Constant of
changing about the mean value, 1.94 calories. Circles, crosses, dots repr¢

SHOWS ASCENDING SEQUENCES
SHOWS DESCENDING SEQUENCES

* since 1924. Shows that the sun’s gift of rays to warm the earth is frequently
ectively satisfactory, nearly satisfactory, and unsatisfactory observations.

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

» DAYS ZERO 5

| | EL eee eel
Eine aa [eee
AEE RC NS-ANEE Se
RT aS Sa eee
per SZ a eae ae

-1,00 -4 V y

coe eset PST CRS VES ee

Palas ACS arse ae
ea eee DALYAN a Ue adie i
i a

PEA Ne NIE ila
E i CRS
HA Na ees
PAYER

aa aaa

BS ales

AZ yee
I SS , ry [7 |

A aL 14

l\

Fre. 2.
sponding fo average foie eine of 0. 8 per cent. "The solar eee culminate
on the day “zero,” but weather effects sometimes occur much later. Weather
ne corresponding to rising solar radiation, full lines: to falling radiation,
dotted.

Ir-o— t1'o— 6€'£ go'S f0'g zz't te'z 60°! 11‘°O— 69'°0 gz'1 oS'o— 16'1 FL-0o— fg'0— og'o LE°r gg‘ F€-0 «gg'0 «6gt'S =ob bh § 60°z zz°Oo— or'z *** Sue

pe}oa110+5
o£°S ofS oo'S of: Ofb coh oO£€ Off oof o£ Of'z oo'z ofr Of1 CO'r O£'0 Of'0 or*O— OFO— Og‘o— oOII— ob 1— Of"1— 00°Z— OF Z— * "YIU
A[1®I Run

1g'S+ L£1°S+ 6€°g+ 94°64 E€:o1+ze2°L4 b6'S+ 6E°b+ 6g°z+ 6E°F+4 OS'F+4 OS1+4 I9'F4HS°0— Li°0+ OS'1+ Lg*I1+ QS'1+ go°0— 90°0-+ go°H+ OO'E+ O£'0+ ez'z~— 0°O **** SUBETN
O'Iz+ o'S1+ o'ez+ S*Sz+ o'g+ o'f+ S14 oror+ S*Z4+ orf+ S*go+ S'E— S'h+ oror1+ S'o— O'S+ o'h+ S*0+ S*1— o'h— oO'f1+4+ 0'9+ O'S— O'ZI— 90'0
Sto S'§S $:Sg 0-69 SIS S'oh o'Sh S'S orS S:0S o'0S o'oh o'gh SES of S:gh S*Lb ohh o'zh SOE SoS S'Oh S'ge S11 Sep

FI f1 ZI II ol 6 8 £ 9 S r £ z I 1 of 6z gz Lz gz Sz be fz zz Ie? 2 OL OI
Oo'z— S-z— o'f— o'f1— o'S— o'S— S'0+ o'or— $*S— o'6— or1i— S°f1— S°f4 Se S-‘bi— S*1z— $°6 S°9 cae $-6+ o'9+ o'S— o'bI— o'g— o0'0
01S S'0S o'0S ooh o'gh o'gh SES of S:lp ohh o'zb SOE SoS S-6h Sgf SIF S*fh Soh S'0S S:zqg o°6S o'gh O'6€ oO'Sb o'k$

9 S v £ z I 1 of 6z gz Zz ge sz bz fe zz Iz oz 61 gl ZI 91 St FI Ci) oe OLOT
S-o+ S:e— o'zi— S'S— o'zi4+ oroz+ S*Iz+ o°g1+ o°f1+ o'r14+ S'e+ orf— o°f4+ o-f— S11— S*f/— o'z— S'z— 0°64 o%4+ S°zi+ O'ZI+ $-€1+ oo o'o
S-9S SES ohh S:0S ogg oo S'4L ob 069 O'fg S*gS O'6h O'fg O'6h Sth Sgh o'bS SES o'Sg o'gS S99 0°99 $°69 079 o'9S

e1 ZI Il ol 6 8 Z 9 S v £ z I 1f of 6z gz le gz Sz bz he cs Iz OZ, Vase OZOr
o‘o1r— S*6— o°g+ S*14+ S*S— o'gi— S*bi—o-Z— S‘o— oror+ S*hI+ o°h1+ S°S+ o'h— S°6— S-€1— S:gi— S*6— S:Z— S:o1— S'b— 0°64 S1I+ S-€+ o'0
o'ov =S:9h o'bg S'ZS SoS o'gf Sih o'6h SSS o'99 Sof o'0L S'19 O'zS Soh S'zh S'*6E Soh S:gh SSh S18 0°59 S*Z9 S65 0705

gl SI br £1 ZI II ol 6 8 £ 9 S ¥. £ z I 1f of 6z ge le gz Sz bz fz stress 9zO1
S-g+ o'gt+ S*gz+ o'1€+4 S*0f+4 o'zz+ Sz1+ O°L+ o'f+ o0'o o'Z+ 0°6-++ 0°9+ o'zI+ S*Sz+ oge+ oroz S'gr+ S'oI1+ o'1+ O'I— O'b— o'f— S'z— o°o
o'6h S°SS o'99 S:0L o'0Z S*19 o'cS S*oh Seb S*6& Soh S:gh S'Sh GS-16 o'S9 S:4g $°6S 0°95 o'08 S-oF G*gE SSE Sg o'4€ S°6E

6 8 Zz 9 s + £ z I 1 of 6z gz lz gz Sz bz fz zz 1z oz 61 gl £1 Ola sees Ocor
O'SI+ O'zI+ o'gi+ S*1€+ o'b€+ o'gz+ S'zz+ S*g1+ O'£Z4 O'S+ O'Z+ O'f+ S'E+ O'94+ S'zI1+ oO°Z1+ O'g14 S*zI+ S*h+ o'S+ oS+ o'or1+ o'or+ S'I o'o
S-gh S$'Sp S1S oSg S:Lg S*6S o'9S o'0S S'ob S*gE SSE Sof OLE G-6E ooh SoS S'6h ooh og SBE Sgt Seb S°Eh o'z€ S-ee

6z ge lz ge Sz tz ez zz 1z oz 61 gi 41 gt or br €1 ZI II ol 6 8 £ 9 So ostreess*QzOr
o'hi+ S‘or+ S*b+ o's o'Z o°o1— 0°6 S-g. o'9 Sot o'S+ o-h+ S'0+ S*-Z— o-L— o-L— orz— orz— S°€I— o-zI— o*g— S'6— S'g— S'P— o'0

$-6S o'9S o'0S Sob S*gE G:SE Sof o-Z€ $6 o'gh S'0S S:6b o-gh o-gE S'gE S:gf Sb Sb oe Ske SLE og o'Z€ ob S°SP

bz tz zz Iz oz 61 gi ZI gl SI at £1 zi II ol 6 8 c 9 S P 2 z I Gz ss She's oeor
oe o'b— S'h+ orzz+ orfz+ o-git+ S°S4+ S'Z4+ o'94+ S*g+ S*44+ O'f+ O'h— O'I— O'S4 0°64 O'z+ S'z— S*9Q— 0°9— S-o+ o'z+ o'h— S'z— o'0
S‘SE oF€ of S09 S19 S'9S ohh o'gh S'hh “orb o'gh Sib SPE S°LZE Sh S:*Lb Sob oof o'z& Sze o'6€ S‘oh S'rE o'g& Sgt

z I 1£ of 6z gz lz gz Sz bz tz zz Iz oz 61 gI a 91 ar Fr £1 ZI Ir o1 Oy serosa Zor
Seer “Srir o%Z oo ore 0°6 ofr o'9 Sr S:z— o'z— S'bh 09 = 8600'0 Sr o'r ore G'2 o£ S‘Sr ofr $9 One Cit oo
o-lp ooh Strb Spf SLE Seb S:Lb Sob o'g€ o'z€ S'z€& O'6€ Sob SPE og SgE SLE och Sib O'S Goze vottr js*Z0 Ora 1S" re

be tz zz 1z oz 61 gl ZI gl SI FI 3 zI II Ol 6 8 Zz 9 S b £ z I Gel ete POT
Aeq Aeqd Aeqd Aeqd Arq yoie yl
yqoz yysr yqor yas 0197, jo yuoyy

Aep 049z UO UOTJBUIWTND IY} YOVaI saduanbas sejOs Surpusosy

SaanjAvgag adnjosogua [ puo saanjosaquma [ uvapy uojiburyso 4{A—1 Alav |,

og'0o— 0'0 Sh:o— So'z— So'€— oL:0— ofr Shz 06-0 — 0gG— 01 °£— Sz*1— of 0 00°I1— $6°z— 09°7— $S*1— So'0+ oz'1— SL‘0— Sg*f— og*z— Sg'1— of*I— ov'z *** sueaU
peyeri0g
o£:S ofS ooS ol'h ofb ooh o£ off oof OL Of'e oo'c O41 OF'r OO'L 04°0+ Of°0+ O1'0O— OF'O— 0g‘o— OI'I— ob'I— Of*I— 00'z— OF'z— **YoIeW
Ajiea
or'S+ O£'S+ SS:b+ Sg-z+ Sz-1+ o€°€+4 o1°S+ $4°S+ o1'z+ o1*b— og'b— $L°0+4 00°'z+0F'0-+4+ S6°I— 06°1— Sz*I— S0°0— 09'1— cS-1— So'S— oo-b— SE°€— of'€— o'0 '*°"* ean
0'o o:zI— S*or— S*o— o'g+ orz+ 0°64 Szr4- o°S— o-or— S*11—o°f— S°S— o*g— S*9— S-91— S'g— o-f— SG°E1~ S°L— o°6— S:z1— S‘bi— o*ZI— 0'o
S95 S‘th o-gh 9S Sg S:gS $:Sg 0°69 S*1S Soh o'Sh S'S o'1S SOS o'0S ooh o'gh SES of Seb S:Lb oh o'zb SOE S98
gi “1 gI i FI eI zI II OL 6 8 Z 9 ic) v £ z I If of 6z ge lz gz CZ iii LOT
S'1— of€— GS'S+ o'f€+ Sz+ o'z+ o'g— o'0 $-S+ oS— S:o+ S-o— o'h— o'g— S*g— S'gt+ S*14+ S°6— S-g1— S*b— S‘1— S:z+ S*bi-+ o'11+ O'0
Soh o'Sh S'ES o18 S:0S o0S ooh o'gh SES of S:gh S:Lp. ohh o'zh SOF SoS S*Oh Sse G1€ Sb Soh S05 S:zqg 0°65 o'gh
6 8 £ 9 Ss + £ z I 1 of 6c gz lz gz Sz bz &z zz Iz oz 61 gi ZI OL ser OF Or
Spt otbi+ oL— S:Z— G$:11—0°S— S-z1+ 094+ o'1—, $*€1— OoI— S'z— o'h+ S*b1+ O°614 S°gI+ o°-oI1+ S:0+ o'S— o'6— orvzi— o'S— o'f— o'9— 0'0
o79S S:Sg Shh ohh ooh Soh og S°4S S:0oS og S*1b o'6b 5°55 O99 csoZ -O102 “Seto; “oszS) “Seo “Seer “Set. “Scoh Sieh) UScSh Sirs
oz 61 gi “1 gl SI FI e1 zi Il OI 6 8 = Z 9 S v £ z I If of 6z gz Lem cei Oc OT
o:St-+ S'z+ o'g+ S*f1-+ ooz+ S*of+ oSE+ S*hE+4+ o'gz+ S*gI1+ O'1I+ O'LZ+4+ O'F+ O'II+ O*EI+ O'OI+ O'OI+ S-6z+ o'z€+ o'bz+ S'oz+ S'bi+ 0°S+ o'f+ o'0
Goo ores Sc1h o°6h S-SS o799 S*0Z 0-04 S19 0:25 S:oh S:zb S-Of ‘S-oh S:gh S:Sp 5:15 0°89, S29 S:65 o°95 o:0S S-or S-g& &:Se
zI II OI 6 8 iL 9 S P £ z I If of 6z gz lz gz Sz bz tz zz 1z oz Gre sesiee nel OZOr
o'1z4+ Stoz+ orzi+ G:z+ o'f— o'L— o-0I— o'f— o'1— O'h— O'Z+ G-S1+ o-gi+o'or+ S*9+ S*0+ o°6— o'1I— O'hI— o'€1— S*zi— o-or— S*f€— o'1+ o'o
GeoZesos0L) S210). ores) Sob) “Scep = St6e— Sob “S:gh (SeGh° ‘SerS- <Osto) (S229) 5705 o'9S o70S S:oh G*gE S:SE Sof o-46 SOE o'oh S05 - S-6P
9 S P £ z I 1e of 6c gz Le gz Sz bz fz ze Iz oz 61 gl “1 gI SI FI Er sieceiee es Oc6r
or— of+ o'S+ orz+ o'g+ S'1zZ+ o'hz+ Ogi S°zI+ $°9+4 S‘€e— o'S— o'g—o'L— S'g— O'h— S'z4+ O'4+ 0°9+ S:z+ S:S— o'S— o'S— oo 0'o
S-6€ Soh S:gh S'Sh S*1S o:S9 S:49 $*6S og o'0$ Soh S-gE GSE Sof of S°6E ooh S*oS S:‘6h o'gh o'gf S-g& Sg& S:fh Sep
1 of 6z gz lz ge Sz te tz zz Iz oz 61 gi LE gl SI FI oe zI 11 o1 6 8 Lie ois Oc OY
o'6r+ o'zz+ or1z+ O11 O'4Z+4+ $°6+4 Gey eiGsGs Cage Gf ost} G-o4- S=r14-Sen— (Sef— Sto — aS: — Sor Sgt Sse S‘o+ S'i+ o'1+ o'$+ 0°70
S19 S'hg S*fg S°eS S:'6h oS o'bS S‘gh o'Sh o'68 S'Eh o6h ofS orb O'6€ Ozh O0'6€ o'bP o71S Ocoh (02th okey ‘Srth, S:Zb Seep
gt 41 9 SI v1 £1 zi 11 or 6 8 Z 9 S ¥ € z I 1£ of 6z gz z gz Scaecieisisicrieis/ZOL
o'b-+ 0°70 Sot S*L— S-b— o'o1— S*h1— S:-g— S:g— o'g— S‘g1— o'f€— S'z— o'gi— Stiz— S'S1— o'S— or1— S'I— o-zi— o-gi— S'g— o'£— O'1I— 0°0°
ofg 065 $:6S S18 SS o-6b Shh SOS S:zS O18 S:zp ooh S‘oh orb S*LZE Sh o-bS o'gS S45 ofp ofh SoS ozs o-gh 0765
zi 11 ol 6 8 4 9 $ ¥ £ z I It of oz gz lz ge Sz be tz zz 1z oz Or) ssceee i SZOr
Ong ez s.9— <S-V—. O91 SPI — 0211 — 6 O1— 0791 —5°61— fh -E1— 028 — O71) S0- Or o1— o'v1— S*go— o'S— o°6— orz+ S°L— S'11— S°L1— S-gi— 0°0
oor S‘tbh SoS S:zS o1S S*zb ooh Soh orb S°ZE S-Ey o'vS o'gS S48 orZb ofr S:0S o'zS o'gh o'6S S'6h S*Sh S$:6F sgt o:ZS
Z 9 Ss v £ z I If of 6z gz lz gz Sz be 4 zz Iz oz 61 gl £1 gI SI Pies ces ocOL
O°Iet Sgt Sror4- Shr or24- o-P4+ Ste+ o'S-+ of So —- S:4— S'p— Sr++ S54 S°1— o-9— 0-01— S:6— of— S1— S*4— o'9— S'E— S:b— o'0—
ofp S09 S19 SoS ohh ooh S'th o-fh o'gb Sib SPE SLE Seb Selb Soh O79€ orz€ Sze O68 Soh SHE OOF S ge S48 oz
1 of 6z gz lz ge Sz be fz ze Iz oz OI gi 41 QI SI FI f1 ZI II ol 6 8 Lae nese OL
Aeqg seq Aeq seq seq yore
yyoz qyS1 q01 qs 0197 jO yO!

Aep OJ9z UO VOTJEUTWUTND I9Y4} YOeII saduaNnbas Jejos Surpusdsaq

saanjangag, aanqosaduay pun saanyosaquia . uvapy wopburysp A{—ze Adv LL

NO. I WEATHER DOMINATED BY SOLAR CHANGES—ABBOT 7.

parency of our earth’s atmosphere. No appreciable 12-month peri-
odicity appears in the results. This is a good sign of their indepen-
dence of atmospheric influences. Full and dotted curves in Figure 1
mark all the well-supported sequences of rising and of falling solar
radiation. They occur in short intervals, averaging 5 days. All of
those selected exceed 0.4 per cent in range, averaging 0.8 per cent.
These rising and falling sequences are 111 and 106 in number, respec-
tively. Many are lost because of unfavorable observing conditions.

Figure 2 shows average changes in the mean temperature and the
barometric pressure at Washington, D. C., associated with these rising
and falling sequences of solar radiation, during the months of March,
April, September, and October. These meteorological exhibits are
average values representing the work of 7 years, and of about Io cases
of each kind in each month.

The method of computing the curves shown in figure 2 is illustrated
in tables 1 and 2 as regards temperatures of March. The temperatures
(which are the mean of maximum and minimum at Washington as
published by the U. S. Weather Bureau) are arranged in consecutive
series of 25 days each. In each series, the fifth day is that on which
the solar change examined reached its culmination. Departures of
temperatures are always computed from the first day of the series
as the base. The mean values of all the departures occurring in
March in the years 1924 to 1930 are given at the foot of the table.
They are corrected to eliminate the secular rise of temperature which,
of course, occurs during any 25-day interval at that season of the year.
The final result is plotted in figure 2. The reader will see that in all
cases there is a marked opposition between curves corresponding to
rising and falling solar radiation, respectively.

Eleven physicists to whom I have shown these results unanimously
concur in advising me that the constant opposition of the weather
effects following opposite solar causes demonstrates a physical con-
nection between the weather of Washington and the changes in the
solar constant of radiation as observed in Chile. Average changes
of mean temperature of 5° Fahrenheit are found corresponding to
solar changes averaging only 0.8 per cent. Hence we may suppose
that on many occasions temperature effects caused by solar changes
may reach 10°, and sometimes 15° or 20°. That is to say, major
changes in weather are due to short period changes in the sun. So
revolutionary is this conclusion for meteorology, that I hesitated to

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS vot. 85

publish it until the unanimous approval of many competent critics
encouraged me. I am further supported in this view by having found
a similar opposition of relations prevailing not only at Washington
but at Williston, North Dakota, and Yuma, Arizona, in all months
of the year.

By what physical connection are these surprising meteorological
results produced by such small solar changes? We must discard at
once, I think, the idea that changes of ground temperature, directly
produced, communicate the effects to the surface air. For firstly, by
Stefan’s law, in equilibrium conditions radiation varies as the fourth
power of the absolute temperature. Hence a change of 1 per cent in
radiation, if acting directly and in equilibrium conditions, should re-
quire but + per cent change in the earth’s temperature. Actually the
change of temperature observed exceeds 1 per cent, reckoned from
the absolute zero. Secondly, in March and some other months, a tem-
perature effect at Washington is found to be nearly simultaneous with
the solar change. The solid earth has too large a capacity for heat to
follow in temperature thus quickly. Thirdly, large effects occur at
Washington 10 or 12 days, and sometimes 16 or 17 days, after the
solar cause ceases. Not all of these effects can be direct. Fourthly, in
September a reversal of sign is observed.

Admitting that the meteorological effects are produced indirectly,
let us recall: Firstly, that from 10 to 25 per cent of the solar radia-
tion is primarily absorbed in the atmosphere itself, which has a very
small capacity for heat. Secondly, that the atmosphere circulates in
great cyclonic whirls. Thirdly, that the temperature of a station
depends greatly on the prevailing wind direction. May it not be that
the instantaneous changes of heat absorption in the atmosphere tend
to displace centers of cyclones, and thereby to alter the wind direction
at stations, thus altering their temperatures ?

How shall we explain deferred effects occurring 10 or even 17 days
after the culmination of solar sequences? May they not result from
atmospheric waves drifting in a southeasterly direction from distant
centers of action where primary effects are produced? If so, we
must perceive that the average effects shown in tables I and 2 can
form no trustworthy basis for forecasting individual cases. For
primary and secondary effects, treading on each other’s heels, as it
were, must often interfere, and either augment or reduce expected
weather changes.

NO. I WEATHER DOMINATED BY SOLAR CHANGES

ABBOT 9

SOLAR PERIODICITIES

It would be encouraging from a forecaster’s standpoint if definite
periodicities’ should be found in solar variations. In table 3 are given
10-day mean values of solar radiation from 1918 to 1930." A ten-
dency towards the recurrence of a certain form of 8 months’ period
was discovered in the 10-day means. To evaluate this periodicity,
the 10-day mean values were arranged in a table of 9 lines of 24 con-
secutive values each, beginning with May, 1924. Mean values of the
24 columns being computed, they resulted thus:

8-month period

Direct, (Mieanis@” wclsc eanelesie.es 40 41 42 41 44 41 41 #42 «41 43 «+42 «+40 41 42
Smoothed Means ............ 40 41 42 42 43 #42 #42 «42 «42 «42 «#41 «4t)«64t) «(41
Smoothed Departures ........ o +r +2 +2 +3 +2 +2 +2 +2 42 +1 +1 +1 GIL
Direct. Means i. 2..0<50000<. 39 41 4o 38 41 4t 39 38 «37° «37
Smoothed Means ............. 41 40 40 40 39 39 39 «638 © «603706037
Smoocthed Departures ........ +1 0 oO o —I —I —I —2 —3 -—3

« First two figures omitted. Thus for 1.940 calories, I substitute go. Departures are given
from 1.940, omitting three figures.

From these numbers a smoothed curve was drawn which gave the
departures from 1.940 calories. Subtracting these departures, the
original data were cleared of the 8-month periodicity from January,
1924, to December, 1930. It was then perceived that another peri-
odicity of 11 months seemed present. By a similar arrangement
in lines of 33 consecutive revised 10-day means of solar constant
numbers, the following values were computed, representing the
11-month. periodicity :

I1-month period

Direct eiMeans! <24 sh eteasavie ce 4o 41 39 38 «638 «6936 «638 «30-335 37) 37s 34s 38s 400
Smoothed Means ............ 41 40 39 38 38 37 37 37 36 360 «636637 «638 = 630
Smoothed Departures ........ I 0 —I —2 —2 —3 —3 —3 —4 —4 —4 —3 —2 —I
Direct Means ..........00000 40 41 43 44 41 40 38 42 42 40 42 #45 43 «46
Smoothed Means ............ 40 41 42 42 41 40 40 41 4 42 42 «43 «45 ~«46
Smoothed Departures ........ o I 2 2 I oO oO I I 2 2 RB 5 6
DinectoiMieans cance secession 44 45 43 43° «41
Smoothed Means ............ 45 44 43 #42 «4!
Smoothed Departures ........ 5 4 2 2 I

As these two periodicities had been evaluated solely from results
of 1924 to 1930, I desired to see whether they were also in evidence
from 1918 to 1923. For this purpose, I made templates fitting the
smoothed-curve departures for both periodicities. These templates
I traced again and again in their proper phases to fill the entire period

* The best values are those obtained since January, 1924. Prior to August, 1920,
all observations were made in the outskirts of the city of Calama, amid dust and
smoke, and with less perfect equipment than subsequently. Prior to January, 1919,
there was only one observation per day and by the “long” method.

IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

from August, 1918, to December, 1930. I then added their amplitudes
algebraically. This produced a curve which obviously bore a con-
siderable resemblance to the curve A of figure 3 throughout its whole
extent. This indicated that both 8- and 11-month periodicities have
prevailed in solar radiation since 1918.

I now desired to search for longer periodicities. It seemed better
to use monthly mean values for this, as given in table 4 and figure
3, A. Having read from the curve of combined departures of 8-month
and r1-month periodicities the departures for the second decade of
each month from 1918 to 1930, I subtracted these from curve A,
figure 3, and replotted the again-revised data. This curve seemed to
indicate the existence of a periodicity of 45 months. Arranging the
corrected solar values in lines of consecutive 45’s, and proceeding as
previously, the following result appeared:

45-month period

Direct) AMieans serxcis 2 <sieecrec ee 20) 32) “4135 37) 28) (4h) 633) at bag 47 37 edge AS
Smoothed Means ............ 3333) 934) 1835) 590, 37 38.030) dOnt gen ageeasieraa
Smoothed Departures ........ 7 —7 —-6 —-5 —4 —3 —2 —I o +1 +2 +3 +3 +4
DinectedMieans; esjeissnes violetecerate 44-45 AY 38) 46-46: 44" 46" 4t 50)) 743) 437 az vad
Smoothed! WMeans (.iiciices cece 44 45 45 45 46 46 46 46 46 46 45 45 45 45
Smoothed Departures ........ +4 +5 +5 +5 +6 +6 +6 +6 +6 +6 +5 +5 +5 +5
Direct” Means, aie .ciee sic.s.c.cteve oie 44 30 40 40 38 48 40 47 47 4I 4r 39 40 39
Smoothed Means ............ 45 45 45 44 44 43 43 42 4 41 .40 39 37 36
Smoothed Departures ........ See ci Saeed eet Ae eee ee eo o —I —3 —4
Dinect je Mieans) 201 1e1a)s/</a1e1ers eres 36) 35, 83t

Smoothed Means ............ 350 340633

Smoothed Departures ........ —5 —6 —7

After removing the 45-month periodicity as in former cases, there
seemed to exist a periodicity of 25 months, which by similar treatment
resulted as follows:

25-month period

Directs Means: i icewccs ceccecte 30!) © 33) 9530537 8 34 Bese AT aS sooo eso ues 740 Soma
smoothed: Means ...10010...10cis- 32, 32 33, 34 34 35) 350.36. 37) 938) g05) 0 on 40
Smoothed Departures ........ —8 -—8 —7 -6 -6 —5 —5 —4 —3 —2 —I —I o o
Direct, IWeansi sacs cslesie esses 4%. 44 43 “40! 43 420 42 43) 42) 40n 33
Smoothed Means ............ 40, 41 | 4t 42 42°) 42) a2 942° \4t 40. 735
Smoothed Departures ........ o +1 +1 +2 +2 +2 +2 +2 +1 0 —5

Removing the 25-month periodicity, as before, a nearly smooth curve
resulted in which the 68-month period corresponding to a half sun-spot
period was clearly seen. The coordinates of the five periods discovered
are as follows:

Coordinates of Periods

Length Amplitude Date of Zero

in Months in Calories Departure
68 -O14 Dec. 15, 1929
45 013 Sept. 15, 1930
25 .O10 Nov. 15, 1929
II .009 Dec. I, 1929

8 .005 May I, 1930

NO. I WEATHER DOMINATED BY SOLAR CHANGES——ABBOT ligt

005
010
00
O10
.000
O10
1.94:
1.942
19.
1.938)
1.936

me ot We een ey. ee
mosis) mie
mmc) Meier | fea
a ih N a
fees) evi icp- a
hyo) Leys) RT
Bixee, ligne Marie) Foo
ete) ovis) Qo
te es Sere ee ieee
Tore |) Oe oe
mMEeeony (sai hers AC
iaalipalha 3 Toe) A
eee ree. eames as
ssc Hole) el
COS GOAN Pee
EEPE ae as a
PCS ene) es
ie aa Sige sia me Senet
Gas, Wella de) aes
ori | Melee ier AC eals
aa Pete) ge
LES TORS
7 err i a a i Cocco
Paice eo) ab esl ea ee
F ea eet si os
eS Se aa rar
mise laa) A
Sigs Beni Vae al el le
EERE ESERIES (Oe
N \) a
POSS) OPAC CAT Ss ees
ee TEL ar NY
Ss ule cele <p ete
ima MIMS” ees
(oe area q herein
eit Oe ig a ae Hd
a I Be =
Fees Ho ee
iSeeeaeil| Pag a i
ata HP) ee
KN
eae Co ams -
BESne invade) tise
eco ECRCREERE TA sae
q 3S aT\
1 a a
. S707 mice ra eee
Beet Sela CACC) ee
hse ae pee ea
mee Pil £1 ne |

1.94 1.93
1.93
1.92

Fic. 3.—Periodicities in solar radiation.

TZ SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

I next made templates and traced the five periods hitherto described in
a way to cover the entire interval 1918 to 1930. The total effect of the

TABLE 3.—TZen-Day Solar Constant Values, 1918-1930

Decade 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930

Jan. I 1.943 1.968 1.956 1.924 1.946 1.037 1.945 1-944 1.9390 1.041 1.925 1.938
2 1.048 1.967 1.053 1.946 ..... 1.943 1.939 1-943 1.938 1.931 1.941 1.937
3 ieGkis Jeolte) Goode Te Q5 2s sterile 1.944 1.947 1.933 1-93I 1-942 1.939 1.929
Feb. 1 T.OO2! LeO58. aicicie TeOll e934. W935) eciaccr 1.938 1.936 1.947 1.937 1.933
Z 1.951 1.954 1-952 1.947 1-951 1-943 1-951 1.939 1.946 1.941 1.925 1.939
3 1.930 1.956 1.958 1.948 1.923 1.938 1.938 1.929 ...-- 1.934 1.925 1.942
Mar. 1 1.950 1.959 1-954 1-949 1.929 1.947 1.941 1.941 ..... 1.950 1.932 1.940
2 1.942 1.948 1.940 1.939 1.036 1.944 1.036 1.048 1.044 1.945 1.931 1.937
3 T2931 D2032" eee 1.932 1.931 1.942 1.941 1.932 1-941 1.945 1.932 1.941
Apr 1 1.943 1-948 1.951 1.930 1-934 1-942 1.945 1.927 1.941 1.946 1.932 1-941
2 1.957 1.956 1.941 1.937 1-928 1.948 1.950 1.937 1.945 1-940,1.942 1.938
3 1.961 1.952 1.934 1-025 1-934 1-947 1.946 1.939 1.945 1.940 1.938 1.941
May 1 1.953 1.950 1.946 1.924 1.934 1.944 1.946 1.937 1-947 1-943 1.936 1.045
2 1.921 1.961 1.939 1.925 1.935 1.948 1.950 1.938 1.944 1.951 1.941 1.048
3 TeQ45. TisQ50) VLQ4T) frais ae 1.937 1.950 1.954 1.942 1.944 1.949 1.937 1.942
June 1 1.957 1.943 1.933 1.910 1.918 1.957 1.943 1.939 1.950 1.947 1.938 1.949
2 1.938 1.034 1.936 1.913 1.934 1.956 1.943 1.946 1.943 1.948 1.932 1.944
3 1.962 1.938 1.945 1.920 1.933 1-953 1.948 1.945 1.945 1.951 1.932 1.941
July 1 1.951 1.945 1.960 1.904 1.034 1.946 1.052 1.942 1.9049 1.943 1.935 1-945
1.961 1.940 1.957 1.913 1.928 1.951 1.954 1.949 1.942 1.942 1.931 1.949
3. 1.921 1.950 1.951 1.953 1.918 1.944 1.942 1.047 1.944 1.946 1.940 1.935 1-947
Aug. I 1.955 1.961 1.930 1.944 1.919 1.942 1.950 1.949 1.045 1.942 1.943 1.931 1.946
2 L045, 13042 T2927 9s 5.0. 1.916 1.940 1.940 1.941 1.942 1.941 1.937 1.932 1.947
3 1-950) 1-955) 1-932) asec I.92I 1.941 1.933 1-942 1.942 1.941 1.932 1.930 1.943
Sept 042) 1.038) 1c0SI> ean uae 1.945 1.941 1.956 1.942 1.940 ..... 1.926 1.942
2) 1.946 1.942" 1.044) Secu. 1.932 1.9044 1.950 1.946 1.940 1.942 1.938 1.928 1.929
3 1.044 1.937 1.944 1.969 1.916 1.942 1.946 1.950 1.943 1.950 1.921 1.930 1.939
Oct. IT 1.951 1-947 1.942 1.959 1.926 1.940 1.953 1.942 1.938 1.945 1.930 1.928 1.939
2 1.930 1.949 1.950 1.969 1.929 1.942 1.949 1.949 1.937 1.044 1.935 1.933 1.041
3. E933) 1.9601 “150431 1/0660... 1.938 1.048 1.046 1.929 1.943 1.927 1.926 1.939
Nov. 1 1.928 1.958 1.951 1.953 1.929 1.934 1.948 1.944 1.931 1.945 1.924 1.932 1.942
2 1.945 1-951 1.946 1.949 1.035 1.944 1.951 1.948 1.926 1.943 1.932 1.936 1.0943
3 1.047 1.948 1.045 1.952 1.920 1.044 1.045 1.044 1.930 1.944 1.930 1.939 1.049

Dec. 1 1.062 1.944 1.957 1.956 1.912 I 042 1.942 1.944 I
3 : p 5 -935 1.949 1-930 1.941 1.94
2 1.969 1.949 1.957 1.938 1.916 1.942 1.947 1.045 1.931 1.935 1.924 1.939 ean
Zi I960" 2-058) I.056. a. 0 ee 1.912 1.921 1.939 1.046 £.935 1.939 1.027 1.940 1.951

Taste 4.—Monthly Mean Solar Constant Values, 1918-1930

Month 1018 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930

Jan. 1.043 1.064 1.955 1.048 1.946 1.942 1.943 1.941 1.938 1.940 1.938 1.9346
Feb. 1.949 1.956 1.956 1.943 1.930 1.939 1-943 1.938 1.943 1.943 1.929 1.938
Mar. 1.941 1.945 1-949 1.938 1.932 1.945 1.939 1.939 1-942 1.946 1.931 1.939
Apr. 1.953 1-952 1.944 I.93I 1.932 1.946 1.947 1.934 1.944 1.942 1.937 1.940
May 1.940 1.953 1.943 1.925 1.936 1.948 1.950 1.939 1.945 1-947 1.938 1.944
June 1.955 1.939 1-939 1.914 1.928 1.955 1.945 1.944 1.946 1.948 1.934 1.943
July I.92I 1.954 1.945 1.956 1.912 1.936 1.946 1.951 1.944 1.945 1.942 1.933 1.947
Aug. 1.954 1.953 1.930 1.944 1.918 1.941 1.940 1.045 1.944 1.941 1.937 1.931 1.945
Sept. 1.044 1.939 1.947 1.969 1.924 1.944 1.946 1.950 1.942 1.944 1.927 1.928 1.937
Oct. 1.939 1.953 1.944 1.962 1.927 1.940 1.949 1.946 1.034 1.944 1.930 1.929 1.940
Nov. 1.941 1.953 1.948 1.951 1.929 1.941 1.948 1.946 1.929 1.944 1.929 1.936 1.944
ee 1.962 1.950 1.957 1.953 1.915 1.933 1-942 1.945 1.932 1.942 1.926 1.940 1.947

early

Mean .. 1.949 1.948 1.952 1.927 1.937 1.946 1.946 1.938 1.943 1.938 1.034 1.942

five periodicities is summed up algebraically in curve B, which will
be seen to represent the main features and even most details of

Noe I WEATHER DOMINATED BY SOLAR CHANGES—ABBOT 13

curve A of figure 3. Inasmuch as three of the five periodicities which,
combined, yield curve B are determined entirely from the work of
1924 to 1930, and the other two are to a large extent thus deter-
mined, the part of curve B from 1918 to 1923 may be regarded as
if it were a forecast. Its good fit* encourages us to expect to see
these five periodicities continue to hold until 1933, producing the
general march of solar variation forecasted in curve I of figure 3.

In former publications dealing with possible solar periodicities, [
was indebted to Dr. D. C. Miller for the use of his harmonic analyzing
machine. Two of the periods which I then thought real, namely of
about 25 months and 11 months, are re-discovered by my present
method. I feel better satisfied, however, this time, because there is
nothing arbitrary about my present analysis. It does not assume
periods not indicated by the observations as does the ordinary method
of harmonic analysis, which deals with submultiples of some arbi-
trarily assumed period.

I propose soon to apply a similar method to the individual daily
observations, in the hope of discovering shorter periodicities. Thus
far I have not gone very far in this line, and will reserve it for a later
paper. At present, I will only mention that in the year 1924 there
appeared to be continuing periodicities of 45 days and of the eighth
part thereof, 5.6 days. These are illustrated in curve H of figure 3.
Other periodicities seemed to hold from 2 to 4 months and then
disappear.

So far, I have disclosed in solar radiation continuing periods of
approximately $ and 4 of the 114-year sun-spot cycle, and of 1/16,
1/36, and 1/50 of the Bruckner cycle of 33 years. Besides these there
were periodicities approximating 45 and 5.6 days in the year 1924, of
which it is uncertain whether they belong to these families, though
they approximate to 1/90 and 1/720 of the 114-year cycle.

WEATHER PERIODICITIES

If, as suggested by the title, weather is governed by solar varia-
tion, and if, as has just been shown, the solar variation from 1918
to 1930 comprises five definite continuing periodicities, we should
expect to find these same periodicities in the weather.

For data to investigate this point, I took from ‘ World Weather
Records ”’* the Washington monthly mean temperatures from 1918

* Regarding discrepancies of 1918 to 1920, see footnote on page 9.
* Smithsonian Misc. Coll., Vol. 79, 1927.

T4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to 1923. I supplemented them to 1930 by taking monthly mean values
of “ Max.” plus “ Min.,” as given in the “ Climatological Data.’”’* In
some previous work I had prepared a plot of the average yearly march
of Washington mean temperatures. From this smoothed curve I took
values corresponding to the 15th day of each month, and subtracted
from my monthly mean data. Thus I obtained the temperature-depar-
tures which constitute weather, as freed from the average march of
events which constitutes climate. These results are plotted in curve A
of figure 4 and given in column 9g of table 5.

I then analyzed these temperature-departure data in the manner
already explained regarding the solar data. I employed in my analysis
the same periods of 68, 45, 25, 11, and 8 months used in the solar
work. These were found to represent to a surprisingly close approxi-
mation the variation of Washington temperature-departures since
1918. The agreement with observed data was somewhat improved by
adding a sixth period of 18 months. These six periodicities are shown
graphically in curves C, D, E, F, G, H of figure 4, and their summa-
tion in curve B. The actual data from which these curves are plotted
are given in columns 1 to 8 of table 5.

The reader, I think, will agree with me that the similarity between
curves A and B of figure 4 is both close and significant. Not only are
the main trends of the original observations fairly well reproduced in
the periodic summation, but many of the details also. Discrepancies,
indeed, occur at several times, and unfortunately a principal one is
found in 1930. One, therefore, hesitates to predict that the tempera-
ture departures of 1931 and following years will be defined by the
same six periodicities without modifications of amplitudes or phases.
Nevertheless the discrepancy of 1930 is not much more pronounced
than several preceding ones, after which fair agreements returned.

It may be objected that the five solar periodicities alone were in-
sufficient to give the best representation, without adding a sixth of
18 months not found conspicuously in solar variation. Is not this last
periodicity possibly of terrestrial origin? May it not be due to some
peculiarity of Washington surroundings which lends a predisposition
to a periodicity of 18 months? For analogy, consider an automobile
on a dirt road. It vibrates as the wheels strike the irregularities of
the road, in a manner depending on these outside interferences. But

“Issued monthly by the United States Weather Bureau, Washington, D. C.

WEATHER DOMINATED BY SOLAR CHANGES—ABBOT

I

NO.

‘sainjeroduia} A[YUOU UII UOLSUTYSe AA UT SaI}IOIpoLIog—P “oy

ee || aes Ie eee

tone Bia bce ctallalo eRe)

EN a he Re ae
vt

PCCEEEEEEEEEEE HEREC

Roe ath

CEE Ee et

PECEEPEEE PREECE

ia aaee eee

| |

eee

Lewes sect
HL LN BT AA ta
TATSUYA SAR IMS oe

[| | HH
Ere Re eee

vebl 16
Ree ait “Ne “NYC Nye ‘Neve Nive fier NYC NYP

16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

TaBLe 5.—Periodic Analysis of Washington Temperature Departures

Periodicities
Degrees x 10 ahr.
Original
68 m 45m 25m 18 m 11m 8m Sum data
TOTS VAN: ccisreremecis cre ceesocioss I 5 —7 14 —4 —16 —0.7 —9.7
IED: warstersio siare crstateveiersie o 4 —5 19 —2 —3 1-3 0.8
Mia ri reteicsis clcleists:-feleiciei 1 3 =—3 21 0 —I 1.9 5.1
April, (seceerroe caistetsiors —2 3 —I 16 5 —7 1A) 0.2
eliyammerctoforietereesctelslsrerereiers —3 3 3 a 8 —8 1.0 5-5
AUN eee tere erorndeeciostelet —4 2 8 —4 10 —3 0.9 —2.7
Ata liyjenetererereteveieceroteilereieiereie —s5 2 13 —I2 o 10 0.8 —2.9
AIS ace ete —6 2 17 —I4 —2 6 0.3 2.5
Sians “ddodeoouscadccdas —7 2 21 —14 5 —16 —0.9 —4.0
(OYSEL. USaaoddecusocgncuuc —8 2 23 —9 12 —3 1.7 4.1
ING Wapericrcticiemiaciiscerer —9 I 23 —4 10 —I 2.0 2.0
WGC! (Sores tensa ccrecuereterete —10 I 23 oO —4 —7 0.3 5.2
TOTO) JAN ~ siocetsisisissicistevclereiwtere @ —II oO 21 2 —2 —8s 0.2 4.5
Ive Dae rarstetarsiels\eierarcrstonevercicte —12 —I 17 3 0 —3 0.4 1.2
Mare 2 sisrcrnistelsl syeeyavsteisivele —13 —3 14 4 5 10 3.0 Bur
April Rereewisissinclosisenee —14 —6 II 5 8 6 2.4 0.8
DY same stavatels ctetevore tolerate everss —I4 —9 7. 7 10 —16 0.2 0.5
AMUUTLS area rsioseul stele Sielarsteere.s —I5 —I0 5 10 oO —3 0.2 0.4
Wivtlliy ‘etersieie oete'e «wis wioies assis —I15 —I0 2 14 —2 —I 0.3 —o0.I
WAIST onccoorsisle slvewie lease —16 —Il oO 19 5 —7 0.6 —1.5
Sept eaesciinccctlacser —I15 —Il —3 21 12 —8 Ter 1.2
Oct ua ase eeteneeetece —15 —10 —s5 16 10 —3 0.8 6.8
ING Wan icicteisiateveiniersiercioicteis —I5 —9 —7 7 —4 10 —0.3 2.5
MD CCH erence sie ccleiete mle \eerere 14 7 8 4 2 6 —1.5 —4.2
TOZOMPAI.. aiesicis crs sisiece crests (sieve —I4 —3 —9 —12 0 —16 —5.4 —4.7
eb, atacistee 000 sine veces —13 8 —7 —14 5 —3 —2. —3.3
Maar. oosies sis sienresisie nines —I3 29 —5 —I4 8 —I 0.4 2.2
VA DTI Merete oens.s teers —I3 25 —3 —9 10 —7 0.3 —0.4
IVa tavercicsareeceeisiase steve —I2 7 —I —4 oO —8 —1.8 —3.9
MVIUATLE TS ease levefoxstarsfeyevorestorsrare —12 I 3} oO —2 —3 —1.3 —1.9
Bia yr’ ers tstesale srororo tees ata; zisvere —ilI oO 8 2 5 10 1.4 —2.3
PASM acisicisiversisleierscletele —I0 oO 13 3 12 6 2.4 —0.3
Septat) sacle s:swie cavers ewleete —o oO 17 4 10 —16 0.6 0.6
OGte antisense 5 ceeeitiecicle —8 I 21 5 —4 —3 12 4.8
INGOV es, bsiaisiss2/oisisfevsverersieler svete —7 2 23 7 —2 —I zee 1.6
ID) Oy epareressssis ote locetoisicin eters —6 3 23 10 ° —7 253 2.9
TO2L Mat epercts: sive speresyavejayscctcine —4 5 23 14 5 —8 305 3-2
EE Dee cretecsteisiaitts sisi ctcTovaisints —3 8 21 19 8 —3 5.0 3.0
Mare suis wciciosceee oninele —I 10 17 21 10 ~ 10 6.7 12.2
April, (adissisissilessiaiess oO 12 14 16 oO 6 4.8 5.8
IMiaiy:, Sorretiscpiscieclouneete a 13 II 7 —2 —16 1.5 —1.8
Viti Ctireteicreeiicrtelclesicters siete 4 13 7 —4 5 —3 2.2 0.7
ALY ip, csserctore orvavetore era) erareiers 6 II 5 —12 12 —I 2.1 2.0
AUD Sie cicccis sie saistecse 7 9 2 —I4 10 —7 0.7 —2.5
SEpts sarapsisasine see@ecs 8 a o —I4 —4 —8 —I.1 —3.8
OCEM Saiviea-sctsisistnners sas 9 5 —3 —9 —2 —3 —0.3 0.6
INO Vatoe Section te cists 10 4 —5 —4 0 10 TS 3-2
DD Cciataretstereietatteerisoore ter Il 3 —7 oO 5 6 1.8 1.5
TO22EV AN ee Ncreentemics aes 12 3 —8s 2 8 —16 0.1 —1.4
ebty were ciel setactctere cist atere 12 3 —9 3 10 —3 1.6 2.6
Mar. Ce rverccccccccece I2 2 7 4 Oo —I TO iT
SALDEAL™ (rete steno shies 13 2 —s5 Fi —2 —77 0.6 2.6
AVM oeleleieyroboisteletctorsteistave 13 2 —3 a 5 —8 1.6 2.7
JUNE > Ask misctwmerie tite 13 2 —I 10 12 —3 3:3 1.0
Uitte avcrteerattcre tate cco or 12 2 3 14 10 10 5.1 —0.7
ALi actararate stataverevelchavelane 12 I 8 19 —4 6 4.2 —2.0
Sept cctsiec eee soe 12 I 13 21 —2 —16 2.9 1:7,
OCt) soaienecasinesiioes II o 17 16 oO —3 4.1 3.0
NOVA vostieiiteisetetlonnses II —I 21 7 5 —I 4.2 3.6
DOCS sas wine rnfu te isyevereraynrs ore 10 —3 23 —4 8 —7 27, 1.2
1923 Jan. siatololeterecerelerete Mijetrels 9 —6 23 —I2 10 —8 1.6 3-4
ED: 2 ise tee ost tiins 8 —9 23 —14 oO —3 0.5 —3.4
IM AtY Semitsreterteeintosite 7 —I10 21 —I4 —2 10 IZ 2.1
PADIS sSevootante deminer 6 —10 17 —9 ig 6 1.5 0.6
MIA hii sie crerocrtnseereies 5 —1I 14 —4 12 —16 0.0 —0.7
JUMP o2jn(ato.ciessiveiareieielelie, ete 4 —i1 II ° 10 —3 Tn 21
ANCA Ys sessreie: visite wivueie oveieierete 3 —10 7 2 —4 —f —0.3 —n6
II yc teenie sete sin 2 —9 5 3 —2 —7 —o.8 —0.7
ere Pincitidst ese I —7 2 4 0 —s —o.8 1.4
nee Yeperera te rererorcierste eter ; 0 —3 oO 5 5 —3 0.4 —03
OW scale cvstereainleheiertarsisiaye =I 8 —3 7 8 10 2.9 0.8
DOCH tanioe ios ceieiteciee —2 29 —5 10 10 6 4.8 8.6

NO. I WEATHER DOMINATED BY SOLAR CHANGES—ABBOT 17

TABLE 5.—Periodic Analysis of Washington Temperature Departures—(cont’d)

Periodicities
Degrees x 10 Fahr.
— = — Original

68 m 45m 25m 18m I1m 8m Sum data
EOPAMIAM SY yan alosarararcisrearors naerere —3 25 7 14 0 —16 Ta 1.8
He botgencaceincceee etek —4 7 —8§8 19 —2 —3 0.9 — 7
Mian, seajesciae slew nates —5 I —9o 21 FS —I 1.2 —o.6
ADT lexckers sisiestaisresisistetere —6 0 —7 16 12 —7 0.8 —1.2
May fies caiccscc aa neces —7 oO —5 7 10 —8 —0.3 —3.9
WAIN Ce oercce vee. Sneek —8s oO —3 —-4 —4 —3 —2.2 —1.9
uJintlivaiee steetnencca ceevstecte —9 I —I —12 —2 10 —1.4 —3.0
PAC oO Saerneee ict eelomine vse —10 2 3 —I4 oO 6 —1.3 —2.0
Septimecccic meres —II 3 8 —I4 cS —16 —2.5 —4.4
GE serenrsieaeis area ances —I2 Ss 13 —9 8 —3 0.2 3-3
NOV: odes ates cemcisiases —I13 8 17 —4 10 —I 17, Ley,
IDECH pete ctcals eaterecoes —I4 10 21 oO oO —7 1.0 0.6
TOZG lanl omer alee sees —I4 12 2 2 —2 —8 3 —0.4
Feb. nevis sewelemactiocn —I5 13 23 3 5 3 2.6 7.3
Miatig aescccarscdamatec —I5 II 23 4 12 10 4.5 4.9
PADTIL, els iss wig soa va.ctore —16 9 21 5 10 6 3:5 3.8
May? iets ioc aitsuisiiteles —I5 7 Te 7 —4 —16 —0.4 —3.4
UNE cc ieivistc Seta iealeics se —15 5 14 10 —2 —3 0.9 4.2
uty. Geyeteces care trernara, sarees —I5 4 II 14 oO —I se) O.1
ATI oiioiee dgea aise wrelod —I4 a 7 19 5 —7 ta3 —1.6
Septih casi aus aiyoertdeatas —I14 3 5 21 8 —8s 1.5 5.0
OCEL eae ee ee mnateeses —13 3 2 16 10 —3 1.5 —3.9
INO asthe aca ateecont acs —I3 2 0 7 o 10 0.6 —0.I
IDECEM ais aontnco es ticles —13 2 —3 —4 —2 6 —I.4 1.0
19026: Jans \eicdcsvesiewineiensiaser —I2 2 —5 —I2 5 —16 —I1.4 O.1
Heb perish waicet weasels —I2 2 —7 —I4 12 3 —2.2 1.0
Mars. Wevsiermayetastaaareraisielets —II 2 —8 —I14 10 —I —2.2 —3.2
April eitceewicnatenaases —I10 I —o —9 —4 —7 —3.8 —I.2
AYa Uasree sae tiereeiasee —9 I —7 —4 —2 —8 —2.9 2:
Att Chyna cemetrcrecia st —8 0 —§5 oO oO —3 Ta —4.1
Atl yx" Ta tesc Sviaranstnecs adversities —7 —I —3 ES 5 10 0.6 —0.2
USS weiner saverstveraea ne —6 —3 —I 3 8 6 0.7 1.8
Sept, 4tceaeehtemssan: —4 —6 3 4 10 —16 —0.9 TS
Oct Sisasecs secoociows —3 —9 8 5 oO —3 —0.2 ie
NOV: Giasiacere aie iesareediorarais —I —10 13 ii —2 —1 0.6 0.7
DOCK sakace testes oes oO —10 r7 10 5 —7 r.5 0.4
TOS TMUAT Veet ciee eis ences 2 —II 21 14 TZ —8 3-0 Lad
IMGD:, ciewecieste veeereeaeais 4 —II 23 19 10 —3 2:2 Be,
Mar, aeitadeiscuiseaeeids 6 —10 2 21 —4 10 4.6 4-7
BATT sears orev state overatere<rarere 7 —9o 2 16 — 6 4.1 —1.2
May. siwlasaeeéaiemanisions 8 —7 21 a 0 —16 1.3 TS
MIN? Bs jac ccces ce eeccas 9 —3 17 —4 5 —3 2.1 —3.5
Wutliye: vesssSec are setts ccetdicee 10 8 14 —12 8 —I a7 —0.4
AUG, anjwaciecs creaccacae 11 20 II —I4 10 —7 4.0 —4.7
Sept, seaiuscasneteacas 12 2 a —I4 o —k 2.2 —o.8
OCtR canencce ceateenace 12 7 5 —9 —2 ag 1.0 4.4
NOVA recess sciatic wey i 12 I 2 —4 5 10 2.6 6.6

DEC. © ceed sie tects s anno 13 0 o o 12 6 aur 2)
1928 Jan. Baverailefeexetere eee 13 oO —3 2 10 —i6 0.6 2.8
BieD sc x stiercauicectrecesreacs 13 0 —5 3 —4 —3 0.4 1.9
IMMrev ipa eceresste/e\sceresslaie)s fries 12 I —7 4 —2 —I 0.7 0.9
NTA ee ass ocean at eeae ig 2 —8 5 oO —7 0.4 —I1.4
ME AV "> Fetereiesotiarene/aceiststaverstare 12 3 —9 a 5 —8 1.0 —o.8
WPM, ceyteasniieceen cee. 11 F —7 10 8 —3 2. —3.3
AAU VP) Varescvetcvess cis cis icis ators II 8 —5 14 10 10 4.8 1.4
PAN Getus gis heveawleieistoee 10 10 —3 19 oO 6 4.2 5.7
DEP. adie cse ceewreaenc 9 12 —I 21 —2 —16 2.3 —2.9
Oct? | Biisserodewe cecees 8 13 <4 16 5 —3 ae 4.6
INOW severe aicrstevaxetenarsrsravere oe 7 II 8 Fi 12 —I 4.4 ie
DG Ci: iperresinevas ars apererocnde 6 9 13 —4 10 —7 2.7 3.1
1929 Jan. BiNes ovo) hayes a fore afatonets 5 7 17 —12 —4 — 0.5 1.2
EDs | aacais nae aieelstvsitiasioc 4 5 21 —I14 —2 -3 1.1 —0.5
SVT estes ciete seat eee B 4 ee —I4 oO 10 2.6 3-7
ATL “GacvsAievatenies ac sear 2 3 23 —9 F 6 3-0 Ais
LY cmerletee cio nieicisierersira I 3 23 —4 8 -16 1.5 0.4
UMN, cave acmesneescinecce oO 3 21 oO 10 —3 3.1 —I1.0
VE shen teers odeats ees —I 2 17 2 oO —I 1.9 0.0
PANT G Ss sticraree ale loreranePer Nears —2 2 14 3 —2 —7 0.8 —0.2
Epts a caseeeneeerete. —3 2 II 4 5 —8 Lo. —2.0
OC. caeionenmenees ath —4 a F 5 12 —3 1.9 —0.4
IN GV _ chor mianeectecrnsts —5 2 5 7 10 10 2.9 3.3
DECs Sateen etae eae —6 I 2 10 —4 6 0.0 2.

18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Tas_e 5.—Periodic Analysis of Washington Temperature Departures—(cont’d)

Periodicities
Degrees x 10 Fahr.

Original
68 m 45m 25m 18 m Ir1m 8m Sum data
TOZOWMAMG sete cidec ceva tcc —7 I ° 14 —2 —16 —1.0 3-5
FEDS ateseciewisarisinaciers —8 oO —3 19 oO —3 0.5 6.9
Mise tater ttises sosinewiew sie —9 —I —5 21 5 —I 1.0 2.0
ADT! Feet cce.ciesscreraeinels —I10 —3 —7 16 8 —7 —0.3 0.0
Miaiy? ~ cionicrosiacsiors eisvareloiers —II —6 —s 7 10 —8s —1.6 5.8
ifitirl eM eecrerrstetecietelnee —I2 —9 —9 —4 oO —3 —3.7 1.9
Italy? <cteletavscratevessvsterara sere ers —I3 —10 —7 —I12 —2 10 —3.4 1.9
PAUSE) i srarasicreiocaeve ciate ohn elece —14 —10 —5 —I4 5 6 —3.2 2.1
Septi! Si casiscmeerenrcsr —14 —II —3 —14 12 —16 —4.8 8.3
OCH Meetnoteleryecteicieeiacit= —I5 == ed —I —9 10 —3 —2.9
INO Vomanen seein meie —I15 —I0 3 —4 —4 —I —3.1
ID GCN y cereinerereiatete sYerereincotet= 1s —9 8 oO —2 —7 —2.5

at some special speeds, there are sometimes encountered “ sym-
pathetic ” vibrations due to the make-up of the car itself.

After all, the contribution of the 18-month periodicity to the fit
between curves A and B is a minor feature. Is not their surprising
agreement, which would still be striking if the 18-month curve F were
omitted, significant because related to solar phenomena? Is it not
indeed of promising import from the standpoint of long-range weather
forecasting ?

SUMMARY

1. Contrary to the prevailing view, the weather appears to be gov-
erned by variations in solar radiation.

2. Long-continuing periodicities in solar variation are found which
give promise of value for purposes of long-range weather forecasting.
They appear to be submultiples of 114 and 33 years.

3. All of these periodicities are found in Washington temperature-
departures, and, combined, suffice to represent its main features.

SMITHSONIAN MISCELLANEOUS COLLECTIONS
VOLUME 85, NUMBER 2

THE AVIFAUNA OF THE PLEISTOCENE
IN FLORIDA

(WiTH S1x PLATEs)

BY
ALEXANDER WETMORE

Assistant Secretary, Smithsonian Institution

(PUBLICATION 3115)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
APR 135 1934

The Lord Baltimore Press

BALTIMORE, MD., U. S. A.

Tae AVIFAUNA.OF THE PLEISTOCENE IN FLORIDA

By ALEXANDER WETMORE
ASSISTANT SECRETARY, SMITHSONIAN INSTITUTION

(WitH 6 PLATES)

Pleistocene deposits of fossils containing numerous bones of birds
have been known for years in the western part of the United States
in the Fossil Lake area in Oregon, and in the asphalt beds and caverns
of California, but such material in other sections of our country to
date has been decidedly rare and of limited amount. It is of interest,
therefore, to discuss recent discoveries of abundant avian remains in
Pleistocene beds in several localities in Florida, with representation of
a far larger number of species than has been found at any previous
time in the East.

Early report of birds in the Pleistocene in Florida came from the
excavations at Vero on the east coast which initiated the argument re-
garding the antiquity of man in that area (see fig. 1). There were
found here remains of a jabiru described by E. H. Sellards, and later
there came another collection from which Shufeldt named as new a
gull, a teal, and a heron. More recent excavations by J. W. Gidley and
by F. B. Loomis, and subsequent work by Doctor Gidley and C. P.
Singleton near Melbourne, not far from Vero, have brought to light
many bird bones, while investigations initiated by Walter Wetmore
Holmes near St. Petersburg on the west coast, in what is known as
the Seminole Field, have uncovered the most extensive series of fossil
bird bones that have as yet been found in the eastern part of our
country. This series is supplemented by bones collected in several
localities in Manatee County by J. E. Moore, by a few bird bones
secured by Mr. Holmes from a Pleistocene cave deposit near Lecanto
in central Florida, and by specimens from several localities in the
collections of the Florida State Geological Survey.

The geologic conditions under which these fossils, other than those
from the cave, are found are briefly as follows: At or below sea level
on the east coast of Florida is a bed of cemented sand and broken
marine shells that has been called the Anastasia formation, the Num-
ber One stratum, or the Coquina layer. At the Seminole Field near
St. Petersburg, the corresponding layer is of fine white sand con-
taining many mollusks, less compact than the beds at Melbourne and

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 2

2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Vero. This lower bed is overlaid by a stratum of fine white to light
brown sand, from a few inches to several feet thick, containing occa-
sional lenses or groups of marine shells and, locally, accumulations of
fossilized bones. This is the Number Two layer, usually referred to
as the bone bed, a deposit that is generally thicker on the east coast
than on the west. From this bone bed have come fossil vertebrate

Fic. 1.—Localities where collections of fossil birds have been
made in Florida.

remains. Above this bone bed appear Recent deposits of sand or
humus that form the present surface, though in places the bone bed
is exposed.

The actual age of the specimens from the beds in question has been
subject to some discussion. Dr. O. P. Hay” holds that the Number

* Journ. Washington Acad. Sci., vol. 20, August 10, 1930, p. 335; and in earlier
papers.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 3

Two bed or stratum dates back to the early part of the Pleistocene,
probably to the Aftonian period. In this he seems to be upheld by
Dr. Wythe Cooke.’ Dr. G. G. Simpson’ considers it more probable
that these deposits are of late Pleistocene age than that they date from
the earlier part of that period. Doctor Simpson further concludes that
the faunas from the Number Two bed of the east coast, from Saber-
tooth Cave, and from the Seminole area “ represent a single phase of
geologic time.”

The writer cites these diverse opinions here without attempt to offer
evidence from the bird material in favor of either one.

DISCUSSION OF THE AVIFAUNA

The five principal localities here considered with their fossils may
be now treated briefly, but before taking these up in detail it is of
interest to note that though grebes, cormorants, herons, ducks and
geese, jabirus and other water loving birds are represented among the
birds of these deposits, there have been found as yet no sandpipers,
plovers, or other shorebirds, nor any terns or gulls (Larus vero of
Shufeldt being the yellow-crowned night heron). The lack of gulls is
of interest particularly since gulls are absent also from the Pleistocene
of California, where only one bone of a gull has been identified in
several hundred thousand specimens examined.’

In the present studies there have been identified 65 forms of birds
from the Pleistocene of Florida. Of these three are fossil species of
the Pleistocene, two of them, a teal, Querquedula foridana, and a
turkey, Meleagris tridens, being known only from Florida, while the
third, Teratornis merriami, was described originally from the deposits
of Rancho La Brea in California.

There are nine forms that have not been reported from modern
Florida. Among these is a shearwater, Puffinus puffinus, a pelagic
species of wide range in the Atlantic Ocean and the Mediterranean Sea
that comes here in all probability merely as a casual straggler. The
trumpeter swan, Cygnus buccinator, now nearly extinct, bred formerly
in the interior of the continent, ranging south in migration to Texas.
The whooping crane, Grus americana, a breeding form of the interior
of North America, now nearly extinct, has been reported uncertainly
from Florida. A small gray crane may be the Cuban bird or the little
brown crane of western North America. The California vulture,

* Amer. Journ. Sci., vol. 12, 1926, pp. 449-452.
* Bull. Amer. Mus. Nat. Hist., vol. 54, February 19, 1920, p. 572.
*See Miller, Loye, Condor, 1924, pp. 173-174, and 1930, p. I17.

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Gymnogyps californianus, comes only from the West Coast in modern
times, while an eagle, Geranoaétus sp., has a modern form in South
America and fossils in the Pleistocene of California. The jabiru,
Jabiru mycteria, Mexican turkey vulture, Cathartes aura aura, and
wood-rail, Aramides cajanea are species known today from tropical
America.

The 53 forms remaining that are found:in modern Florida are listed
below. Most of them are common today in the areas under study
either as permanent residents or as migrants from the north during
winter.

Colymbus auritus
Podilymbus podiceps
Phalacrocorax auritus
Anhinga anhinga
Ardea herodias
Casmerodius albus
Egretta thula
Hydranassa tricolor
Florida caerulea
Butorides virescens
Nycticorax nycticorax
Nyctanassa violacea
Botaurus lentiginosus
Plegadis sp.

Guara alba

Cygnus columbianus
Branta canadensis
Branta canadensis hutchinsi
Anas platyrhynchos
Anas rubripes

Anas fulvigula

Nettion carolinense
Nyroca valisineria
Nyroca affinis
Erismatura jamaicensis
Lophodytes cucullatus

Cathartes aura septentrionalis
Coragyps urubu

Buteo jamaicensis

Buteo lineatus

Buteo platy pterus
Haliaeetus leucocephalus
Pandion haliaétus
Polyborus cheriway
Falco sparverius
Colinus virginianus
Meleagris gallopavo
Grus canadensis (large form)
Aramus pictus

Rallus elegans

Rallus longirostris
Gallinula chloropus
Fulica americana
Zenaidura macroura
Tyto alba

Otus asio

Strix varia

Corvus brachyrhynchos
Corvus ossifragus
Agelaius phoeniceus
Megaquiscalus major
Ouiscalus quiscula

Among these species there are 26 that have not before been recorded
in the Pleistocene age, a considerable addition to the 114 modern
species known previously from deposits of that period.

The fact that at this writing 140 species of the birds found living
today in that area of North America included in the limits of the
official Check-list of the American Ornithologists’ Union are known
as fossils in the Pleistocene illustrates clearly the stability in form of
our existing species of birds, since this number is more than 15 per
cent of the total living list (not counting subspecies) for the region

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 5

in question. Progress in our knowledge of these matters has been so
rapid that it may be confidently predicted that eventually all of our
existing species, except those of small size, will be identified in Pleis-
tocene deposits. As conditions are seldom favorable for preservation
of small bones of fragile texture, not much can be known of the
smaller birds as fossils, for their preservation in that state is highly
fortuitous. We may dream, however, of the discovery of ancient
caves, inhabited long ago by Pleistocene owls, with great accumula-
tions of bones of small birds from the pellets of these nocturnal
predators—caves that have been hermetically sealed for tens of thou-
sands of years that chance may bring to attention and so give us
unexpected information on a fascinating subject.

As his studies in avian paleontology have progressed the writer has
become convinced that evolution of our existing birds so far as
differentiation of species is concerned has taken place principally in
the late Tertiary, and that variation since that time has been of slight
degree, confined apparently to minor differences (in color and dimen-
sion) such as are used in our modern studies to distinguish the less
definitely marked of geographic races or subspecies. As our informa-
tion increases it appears that some of the differences that we consider
today as of subspecific value were in existence in birds of the Pleisto-
cene, for example in the gray cranes and in the turkey vultures, and
have persisted to the present without apparent change, a striking ex-
ample of stability in these groups.

The diversity in the bird life of North America at the time of the
coming of the rigors of the Ice Age must have been truly remarkable
since it would seem to have included most of our modern forms to-
gether with a host of others now extinct that are slowly becoming
known from the fossil record. The entire period since the opening
of the Pleistocene has been one of extermination rather than of evolu-
tion, a process that continued steadily until men appeared as the most
active factor contributing to its progress.

THE SEMINOLE AREA

The region surrounding the small settlement of Seminole, not far
from St. Petersburg, Pinellas County, Florida, has been designated as
the Seminole area (see pl. 1). In 1924 Mr. Walter Wetmore Holmes
discovered here a scute from the glyptodon Chlamytherium septen-
trionale, and through continued search during the succeeding 5 years
unearthed numerous other fossil bones including among them many
remains of birds. It is the Holmes collection of fossil birds that

initiated the writer’s present studies on the Pleistocene avifauna of
Florida.

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

According to data secured from Mr. Holmes and from a paper on
the mammals of these beds by Doctor Simpson* of the American
Museum of Natural History, the fossil bones come originally from
one deposit in the area in question, many of them being obtained by
excavating in the original deposit, some coming from erosional wash
and redeposit by Joe’s Creek which runs through this area, and some
from the dump along a drainage canal cut through this region. The
bone-bearing layer is from 1 to 2 feet thick. According to Simpson
“the lowest bed exposed is of white sand, with numerous marine
shells, correlated by Cooke with the Anastasia formation of the east
coast, and hence the equivalent of stratum No. 1 at Vero and Mel-
bourne. Above this, sometimes with a barren sand layer intervening,
is the bone bed, equivalent in age and character with stratum No. 2.
This is generally overlain by a sandy soil, derived from it by weather-
ing.” Gidley has distinguished above this in places deposits of sand
and muck that he considers equivalent to stratum Number Three of
the east coast.

The list of mammals from this area as determined by Doctor Simp-
son is extensive and includes among its 49 species a capybara, a bear,
Arctodus floridanus, a saber-tooth tiger, two ground-sloths, two glypto-
dons, tapirs, peculiar pigs, camelids, mastodon, and elephant that are
considered typically Pleistocene species, in addition to opossums,
moles, rabbits, rodents, skunks, minks, and deer of the same form
as those occurring in the modern fauna.

The bird remains collected by Mr. Holmes include 52 forms, this
being the largest assemblage of fossil species secured to date at any
point in the eastern portion of North America. The importance of the
collection is very evident from examination of the list that follows.
A teal, Querquedula floridana, a huge condor, Teratornis merriami,
and a turkey, Meleagris tridens, are extinct species of the Pleistocene,
the first and last being known only from Florida. The jabiru, Jabiru
mycteria, the Mexican turkey vulture, Cathartes aura aura, and the
wood-rail, Aramides cajanea, are forms that at the present time range
in tropical America and are not now known in the present limits of
the United States. An eagle, Geranoaétus sp., has its only living repre-
sentative in South America though Pleistocene forms have been found
in, California. Most remarkable are remains of the California condor,
Gymnogyps californianus, and a larger condorlike vulture, Teratornis

*Simpson, George Gaylord, Pleistocene mammalian fauna of the Seminole
Field, Pinellas County, Florida, Bull. Amer. Mus. Nat. Hist., vol. 54, Febru-
ary 19, 1929, pp. 501-500, 22 figs.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—-WETMORE 7

merriamt. Bones of the former are abundant in the Pleistocene asphalt
beds of California and the species is known today from northern
Lower California north into California. It has never been recorded
before in the eastern part of the country. Teratornis has been known
previously only from the Pleistocene of California, its occurrence in
beds of similar age in Florida indicating a former broad range that
carried it clear across the continent. The occurrence of the whooping
crane, Grus americana, and of two forms of the brown crane, one
large and one small, is also of interest. There are 20 modern forms
in the collection that have not previously been recorded as fossils.
Following is the list of species from this area:

Colymbus auritus
Podilymbus podiceps
Phalacrocorax auritus
Anhinga anhinga
Ardea herodias
Casmerodius albus
Hydranassa_ tricolor
Florida caerulea
Butorides virescens
Nyctanassa violacea
Botaurus lentiginosus
Jabiru mycteria
Plegadis sp.

Guara alba

Cygnus columbianus
Branta canadensis
Branta c. hutchinsi
Anas fulvigula

Anas sp.

Nettion carolinense
Ouerquedula floridana
Nyroca affinis
Nyroca sp.

Cathartes aura aura
Coragyps urubu

Teratornis merriamt
Buteo jamaicensis

Buteo lineatus

Buteo platypterus
Geranoaétus sp.
Haliaeetus leucocephalus
Pandion haliaétus
Polyborus cheriway
Meleagris gallopavo
Meleagris tridens

Grus americanus

Grus canadensis (large form)
Grus canadensis (small form)
Aramus pictus

Rallus elegans

Rallus longirostris
Aramides cajanea
Gallinula chloropus
Fulica americana
Zenaidura macroura
Strix varia

Corvus brachyrhynchos
Corvus ossifragus
Agelaius phoeniceus
Megaqutscalus major

Gymnogvyps californianus Ouiscalus quiscula

MANATEE COUNTY

From Mr. J. E. Moore of Sarasota, Florida, there have come three
small collections of bones made at as many points in Manatee County.
The first of these was forwarded to me through Dr. George Gaylord
Simpson, and comes from deposits near the mouth of Hog Creek near
Sarasota, Florida. These are said’ to have been found in a stratum

* Simpson, G. G., Florida State Geol. Surv., 20th Ann. Rep., 1920, p. 274.

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

of blue clay 8 to 14 inches thick overlying limestone from 11 to 12
feet below the surface. The birds are accompanied by remains of
Smilodon floridanus, Megalonyx, Mylodon, Chlamytherium, Bore-
ostracon, Tapirus, elephant, mastodon, and other Pleistocene mammals.
The bird bones are dull black in color and are heavily fossilized. The
presence of the California condor is notable. The following species
are represented :

Phalacrocorax auritus Gymnogyps californianus

Botaurus lentiginosus Meleagris gallopavo

A second collection forwarded by Mr. Moore was obtained from a

canal within the city limits of Bradenton, the locality being known
as the Florida Avenue pit. The type material of Parelephas floridanus
Osborn* came from this point. Mr. Moore (in a letter) states that
remains of Chlamytherium, Glyptodon and Bison latifrons were ob-
tained here also. The bird bones examined vary from dull brown to
dull black in color and are well fossilized. The night heron and ruddy
duck are here first recorded from the Pleistocene of Florida and from
these excavations come the best remains of Teratornis. The following
species are represented :

Ardea herodias Teratornis merriam
Egretta thula Meleagris gallopavo
Nycticorax nycticorax Grus canadensis
Anas fulvigula Fulica americana

The third collection was obtained by Mr. Moore at Venice Rocks,
two miles south of Venice, Florida. The material is fragmentary and
varies from light brown to black in color, some bones being more
heavily mineralized than others. Following is the complete list of
species :

Buteo jamaicensis Ouerquedula floridana
Buteo lineatus Nyroca affinis
Haliaeetus leucocephalus Erismatura jamaicensis
Casmerodius albus Jabiru mycteria

Anas sp.

SABER-TOOTH CAVE

According to information supplied by Mr. W. W. Holmes, and a
published account by Dr. George Gaylord Simpson * the sink known
as Saber-tooth Cave (see pl. 2) is located in a bed of Ocala limestone

* Amer. Mus. Nov., No. 393, December 24, 1920, p. 20.
* Pleistocene mammals from a cave in Citrus County, Florida, Amer. Mus.
Nov., No. 328, October 26, 1928, pp. 1-16, 11 figs.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—-WETMORE 9

(Eocene) 1 mile northwest of Lecanto, Citrus County, Florida, on
property belonging to Mr. D. J. Allen. Several years ago Murray
Davis with some companions obtained there the canine tooth of a
saber-tooth tiger (Similodon sp.) which was sent to the Florida State
Geological Survey. Subsequently Mr. Holmes made arrangements
for a thorough examination through the cooperation of Mr. Herman
Gunter, Mr. J. E. King, and Mr. Allen. The work was done under
Mr. Holmes’ direction in February and March, 1928.

The entrance to this cave is described * as being “ through a broad
sink terminating in two vertical shafts. Immediately under these the
floor of the cave was from 25 to 40 feet below the shaft mouths and
there apparently has never been an entrance practicable for large
living mammals. On the floor below the sink and in pockets elsewhere
was a deposit of red earth or clay in which were found numerous re-
mains of Pleistocene animals, apparently representing a distinct unit
fauna. There also occurred in the cave a younger bed of sand and
humus containing no extinct mammals but with numerous remains
of the recent white-tailed deer of the region.”

The bird bones from these deposits are relatively few in number
and are all fragmentary. They are light in color, somewhat stained
by the reddish earth in which they were found, and present the chalky
appearance usual in bones from limestone caves.

The 10 species of birds identified offer little worthy of remark as
all are found in the same area at the present time. The barn owl,
Tyto alba, regularly inhabits caves. The two vultures, Cathartes and
Coragyps, sometimes nest in or about caverns. Occurrence of the
other species must be considered as due to chance except that possibly
the screech owl and barred owl may have sought shelter in the cave.

The bird bones are associated with remains of the saber-tooth tiger,
a capybara, a fossil dog, ground sloth, horse, tapir, a camelid, and
mastodon among Pleistocene species, together with a number of mam-
mals that occur at the present time in this area.

The list of species follows:

Nyroca affinis Colinus virginianus
Cathartes aura septentrionalis Meleagris gallopavo
Coragyps urubu Tyto alba
Haliaeetus leucocephalus Otus asio

Falco sparverius Strix varia

COLUMBIA COUNTY DEPOSITS

From collections in the Florida State Geological Survey obtained in
Columbia County about 3 miles northwest of Fort White, Mr. Her-

Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

man Gunter has forwarded a number of bird bones for examination.
Dr. G. G. Simpson, who has reported on the mammals from this
deposit, states’ that part of this material was collected by J. Clarence
Simpson of High Springs, with additional specimens obtained by L. G:
Getzen and J. D. Lowe. The site is about a mile below the springs
at the head of the Itchtucknee River. Following these earlier collec-
tions a considerable number of bird bones were obtained by Mr. and
Mrs. H. H. Simpson of High Springs during the latter months of
1930. These latter specimens came to hand as the present report was
being completed and have added decidedly to information concerning
birds from this area. The bird bones are brown in color and are
heavily fossilized. Associated mammalian remains include Hydro-
choerus, Mylodon, Equus, Tapirus, Mylohyus, Mastodon, and Archi-
diskidon columbi.

Notable among the birds are the jabiru, the extinct teal, Querque-
dula floridana, and the trumpeter swan. Curiously enough the collec-
tion contains no remains of hawks or vultures.

Following is the list of identified species :

Colymbus auritus Querquedula floridana
Podilymbus podiceps Nyroca valisineria
Phalacrocorax auritus Nyroca affinis

Ardea herodias Lophodytes cucullatus
Nycticorax naevius Pandion haliaétus
Jabiru mycteria Meleagris gallopavo
Cygnus buccinator Grus americanus
Branta canadensis Aramus pictus
Branta canadensis hutchinsi Rallus elegans

Anas platyrhynchos Gallinula chloropus
Anas rubripes Fulica americana

Anas fulvigula

VERO AND MELBOURNE DEPOSITS

The deposits at Vero have attracted the greatest public attention
because remains of man were found there associated with bones of
mammals currently considered of Pleistocene age. Among other speci-
mens secured at this locality were a few bones of birds that were
described by Shufeldt in 1916." In this account there are listed the
turkey vulture, Cathartes aura, barn owl, Tyto alba, great blue heron,
Ardea herodias, and several other species not certainly identified.
Three forms were described as new, a teal, Querquedula floridana, a

*Florida State Geol. Surv., 20th Ann. Rep., 1929, p. 270.
* Florida State Geol. Surv., oth Ann. Rep., 1917, pp. 35-41.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE Tet

supposed heron, Ardea sellardsi, and a supposed gull, Larus vero, the
two latter proving invalid (see beyond under Meleagris gallopavo
and Nyctanassa violacea).

Two bones forwarded by Mr. Gunter to the present writer include
the cormorant, Phalacrocorax auritus, and turkey, Meleagris gallo-
pavo. Mammal remains from this area include Hydrochoerus, Canis
ayerst, Smilodon floridanus, Felis veronis, Megalonyx jeffersonii,
Mylodon harlani, Chlamytherium septentrionale, Tatu bellus, three
species of Equus, Tapirus veroensis, Mylohyus, a camelid, Archi-
diskidon columbi, and Mastodon americanus.

Near Melbourne (see pl. 3) in Brevard County, Dr. F. B. Loomis
of Amherst College located a further deposit which was worked partly
in cooperation with Dr. J. W. Gidley of the United States National
Museum. Subsequently Doctor Gidley carried on extensive work in
this general region during three winters, being assisted by Mr. C. P.
Singleton of Melbourne. Later Mr. Singleton worked in these exca-
vations for the Museum of Comparative Zoology. Bird material ob-
tained has been scattering but has included some important finds. All
of the specimens secured have been available for the present study.
The specimens vary in state of preservation, some being heavily
fossilized and others having a chalky texture. The latter are quite
fragile. The bones vary from light to dark brown in color.

The shearwater that comes first on the list may be a species of acci-
dental occurrence as it ranges regularly at sea or about islands. The
jabiru and the extinct teal, Querquedula floridana, are represented,
as are a large and a small form of the gray crane, Grus canadensis,
These birds accompanied species of mammals of supposed Pleistocene
age.

Following is a list of the birds that have been identified :

Puffinus puffinus Buteo lineatus
Phalacrocorax auritus Haliaeetus leucocephalus
Ardea herodias Polyborus cheriway
Casmerodius albus Colinus virginianus
Nyctanassa violacea Meleagris gallopavo

Jabiru mycteria Grus americanus

Branta canadensis hutchinsi Grus canadensis (large and
Ouerquedula floridana small forms )

Cathartes aura septentrionalis Strix varia

Buteo jamaicensis

ACKNOWLEDGMENTS

In the identification of these specimens the writer is indebted to the
American Museum of Natural History for the loan of a skeleton of

I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the jabiru, and to Dr. Hildegarde Howard, of the Los Angeles Mu-
seum of History, Science and Art, for bones of Teratornis and
certain other important material for comparison.

Mr. W. W. Holmes, in addition to his specimens which he has
most generously placed in the United States National Museum, has
supplied much important information both in letters and in personal
conversations. Ornithologists stand greatly indebted to him for the
care and effort that he has given to the collection of his material which
has caused such an increase in our knowledge of the fossil birds of
this area. Photographs illustrating the Seminole area and Saber-
tooth Cave were obtained from him. Thanks are due to Mr. J. E.
Moore for important specimens that he has forwarded for the national
collections. Dr. Thomas Barbour has kindly forwarded for study
material collected by C. P. Singleton. Dr. J. W. Gidley has furnished
data with regard to the deposits on the east coast and has supplied
certain photographs. Drawings illustrating this report have been
made by Mr. Sidney Prentice.

ANNOTATED LIST

Order COLYMBIFORMES
Family COLYMBIDAE
COLYMBUS AURITUS Linnaeus
Horned grebe
Colymbus auritus Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 135.

In the Holmes collection from near St. Petersburg there are one
complete and several broken humeri. Part of another humerus is
contained in collections from the Itchtucknee River, Columbia County,
in the Florida State Geological Survey. All are similar to those of
the modern birds. The species is known previously from the Pleisto-
cene of Fossil Lake, Oregon, and from cavern deposits in Tennessee.

PODILYMBUS PODICEPS (Linnaeus)
Pied-billed grebe
Colymbus podiceps Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 136.
In the Holmes collection from the Seminole Field there are limb
bones of several individuals of this species which are similar to those
of modern individuals. Other bones are found in the series from the

Itchtucknee River, Columbia County, in the Florida State Geological
Survey. This grebe today ranges throughout North and South

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—-WETMORE 13

America. As it has been recorded in Pleistocene beds in Oregon and
California (McKittrick) its presence in deposits of that age in Florida
indicates a similar wide distribution for North America during the
ice Ace:

Order PROCELLARITFORMES
Family PROCELLARIIDAE
PUFFINUS PUFFINUS (Briinnich)
Manx shearwater
Procellaria puffinus Brtinnich, Orn. Bor., 1764, p. 29.

A left metacarpal with the fourth metacarpus missing was secured
by J. W. Gidley near Melbourne, March 18, 1929. This shearwater
as a species now ranges from Norway south into the Mediterranean,
breeding in Iceland, the Azores and other islands, and at least casually
on Bermuda. At the present time it occurs rarely along the coasts of
North America. It has not been recorded previously as fossil nor has
it been known before from Florida.

Order PELECANIFORMES
Family PHALACROCORACIDAE
PHALACROCORAX AURITUS (Lesson)
Double-crested cormorant
Carbo auritus Lesson, Traité Orn., 1831, p. 605.

Cormorants of this type apparently were as widely distributed in
Florida during the Pleistocene as they are today, for in the collections
here under review there are found the lower end of a tibio-tarsus and
part of an ulna from stratum Number Two at Melbourne secured by
J. W. Gidley, part of an ulna from Hog Creek, near Sarasota, obtained
by J. E. Moore in 1928, and a sacrum and three fragments of humeri
from the Holmes collections in the Seminole Field. In the collections
of the Florida State Geological Survey there are a complete humerus,
part of an ulna and other bones from the Itchtucknee River, Columbia
County, another humerus, white in color, from Rock Springs in
Orange County that is very doubtfully Pleistocene in age, and still
another humerus from the north bank of the canal west of the rail-
road bridge at Vero.

The resident cormorant of this group found now in Florida, Phala-
crocorax auritus foridanus, is smaller than the bird from farther north
and west, Phalacrocorax auritus auritus, which comes to Florida as

14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

a migrant during the winter season. It is interesting to observe that
the distal ends of two humeri in the Holmes collection from the west
coast differ decidedly in size, the transverse breadth across the troch-
leae in one being 15.8 mm. and in the other 18.2 mm., thus exhibiting
differences similar to those that mark the larger and smaller modern
races. It appears possible that differentiation between these two f orms
may have occurred in the Pleistocene, though on the other hand these
two specimens may be merely extremes of individual variation existing
at that time. On this scanty material the writer does not venture to
identify the two as belonging certainly to distinct subspecies.

Family ANHINGIDAE
ANHINGA ANHINGA (Linnaeus)
Snake-bird, water-turkey
Plotus anhinga Linnaeus, Syst. Nat., ed. 12, vol. 1, 1766, p. 218.

The snake-bird is represented by the distal end of a left humerus
collected in the Number Two bed near Melbourne by Doctor Gidley
on May 3, 1929.

This species has not been recorded previously as a fossil.

Family ARDEIDAE

ARDEA HERODIAS Linnaeus
Great blue heron
Ardea herodias Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 143.

The genus Ardea is represented by four cervical vertebrae and
two fragmentary metatarsi from Melbourne, collected by Gidley; by
the lower end of a metatarsus and the upper part of a coracoid from
the Seminole Field collected by Holmes; and by the lower end of a
tibio-tarsus from Bradenton collected by J. E. Moore. The upper
and lower ends of metatarsi and a broken tibio-tarsus are included in
collections in the Florida State Geological Survey from the Itchtucknee
River, Columbia County. All are referred here to the species herodias
without consideration of the possible occurrence of the great white
heron, Ardea occidentalis, confined today to southern Florida, since
so far as present information goes these two supposed species are
indistinguishable in their skeletons. The two specimens from the
Seminole Field are larger than any modern bird seen, suggesting that
possibly there was a larger heron of this type in existence in the
Pleistocene. The differences are shown in the following measure-

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 15

ments: Modern Ardea herodias (seven specimens, including two from
Florida) ; metatarsus, transverse breadth of trochleae 16.2-17.3 mm.,
smallest transverse breadth of shaft 5.9-7.0 mm.; coracoid, trans-
verse breadth of head 13.8-15.7 mm. Fossils from Seminole Field:
metatarsus, transverse breadth of trochleae 18.4 mm., smallest trans-
verse breadth of shaft 8.2 mm.; coracoid, transverse breadth of head
17.3 mm. The material is considered too fragmentary for further
consideration at this time.

CASMERODIUS ALBUS (Linnaeus)
Egret
Ardea alba Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 144.

The lower end of the left tibio-tarsus of a young individual is found
in the collections made by W. W. Holmes in the Seminole Field.
Another was identified in material collected near Venice by J. E.
Moore. An ulna of an individual of large size was included in col-
lections made near Melbourne by C. P. Singleton, in 1928, for the
Museum of Comparative Zoology.

This species has not been recorded previously as a fossil, the only
other reference to its possible occurrence in the Pleistocene being that
of Shufeldt,’ where a fragmentary metatarsus is listed as possibly
from this species.

EGRETTA THULA (Molina)
Snowy heron
Ardea thula Molina, Sagg. Stor. Nat. Chili, 1782, p. 235.

A partly complete metatarsus collected at Bradenton by J. Ek. Moore
is the first record of this species as a fossil. The specimen comes from
a small individual. The snowy heron is fairly common in Florida at
the present. time, and formerly existed there in large numbers. It
has not been recorded previously as a fossil.

HYDRANASSA TRICOLOR (Miiller)
Louisiana heron
Ardea tricolor Muller, Vollst. Naturs. Suppl., 1776, p. 111.

The lower end of a right metatarsus comes from the Seminole Field
near St. Petersburg.
The present species is here first recorded as a fossil.

* Florida Geol. Surv., Ninth Ann. Rep., 1917, pp. 40-41.
2

16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

FLORIDA CAERULEA (Linnaeus)
Little blue heron
Ardea caerulea Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 143.

This species, common in modern Florida, is represented by the
lower end of a right tibio-tarsus and the distal portion of a right femur
secured by W. W. Holmes in the Seminole Field.

This heron is here first reported certainly in fossil deposits.

BUTORIDES VIRESCENS (Linnaeus)
Little green heron
Ardea virescens Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 144.

A right metatarsus and the lower end of a left tibio-tarsus were
collected in the Seminole Field deposits by W. W. Holmes. The
metatarsus measures 51.7 mm. in length.

The green heron is here found fossil for the first time.

NYCTICORAX NYCTICORAX (Linnaeus)
Black-crowned night heron
Ardea nycticorax Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 142.

The lower end of a tibio-tarsus was obtained by J. E. Moore,
at Bradenton, and a similar specimen was collected at the Itchtucknee
deposit by Mr. and Mrs. H. H. Simpson. This heron, which is com-
mon in Florida now, has been found previously in the Pleistocene
of Rancho La Brea in California.

NYCTANASSA VIOLACEA (Linnaeus)
Yellow-crowned night heron

Ardea violacea Linnaeus, Sys. Nat., ed. 10, vol. 1, 1758, p. 143.
Larus vero Shufeldt, Journ. Geol., vol. 25, Jan.-Feb. (Jan.), 1917, p. 18;
Florida State Geol. Surv., Ninth Ann. Rep., 1917, p. 40, pl. 2, fig. 21.

The proximal ends of two left coracoids represent this heron in
the material obtained by W. W. Holmes in the Seminole Field.

The type of Larus vero Shufeldt, a left metacarpal secured at Vero,
Florida (U.S. Nat. Mus. Div. Vert. Pal. No. 8832), on examination
proves to be the yellow-crowned night heron. This species has not
* been recorded before in the Pleistocene.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—-WETMORE 17

BOTAURUS LENTIGINOSUS (Montagu)
American bittern
Ardea lentiginosa Montagu, Suppl. Orn. Dict., 1813, text and plate.

arts of two humeri were obtained by W. W. Holmes in the
Seminole area, and of another by J. E. Moore on Hog Creek near
Sarasota, Florida. This inhabitant of marshes is widely distributed
in Florida at the present time.

Family CICONIIDAE
JABIRU MYCTERIA (Lichtenstein)
Jabiru
Ciconta mycteria Lichtenstein, Abhandl. Kon. Akad. Wiss. Berlin (Phys.
Klass.), for 1816-1817, 1819, p. 163.
Jabiru? weillsi Sellards, Florida State Geol. Sury., 8th Ann. Rep., 1916, p. 146;
pl. 26, figs. I-4, text-fig. 15.

Apparently the great jabiru stork was common in Florida during
the Pleistocene as it is represented in the present collections by many
fragments of bones from a number of localities. In the Seminole
Field near St. Petersburg W. W. Holmes obtained a number of frag-
mentary specimens, including parts of the tibio-tarsus, coracoid,
scapula, ulna, and metacarpus. Most of these are well fossilized
though one fragment appears quite modern. A fragment from the
head of a tibio-tarsus was obtained by J. E. Moore near Venice. A
perfect metacarpal, a coracoid, and part of a metatarsus are found
in collections from the Itchtucknee River, Columbia County, in the
Florida State Geological Survey. At Melbourne in the excavations
on the golf links J. W. Gidley secured the lower end of a right
metatarsus, and parts of an ulna and a metacarpus from the Number
Two stratum. In the collection made at Melbourne by C. P. Singleton
for the Museum of Comparative Zoology there are parts of three
right and one left tibio-tarsi and both extremities of a right meta-
tarsus with the central part of the shaft gone.

After careful comparison of the type»specimen of. Jabiru weillsi,
a right humerus obtained at Vero, Florida, there is nothing evident to
separate it from the modern Jabiru mycteria. It was differentiated
in the original description principally on larger size, but, though large,
it 1s equalled by modern birds in dimension, and is similar to them
in its conformation. The original description gives the total length
of the type humerus as 280 mm. Since then the bone has been broken
and restored, in this process being lengthened until now it is 293 mm.

18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

long, the extra length being obviously due to crushing of the shaft
and the separation of these parts. In a previous publication* the
present writer indicated that sellardsi was probably equivalent to
modern mycteria, a belief that is now substantiated.
The fossil material indicated above shows the same range in in-
dividual variation in size as is found in the modern material examined.
The jabiru is also known from the Pleistocene of Cuba.’

Family THRESKIORNITHIDAE
PLEGADIS sp.
Glossy ibis
The lower end of a left tibio-tarsus from the Seminole Field col-
lected by W. W. Holmes is from a juvenile bird and shows little
evidence of fossilization. It is not practicable to determine whether
it represents Plegadis falcinellus or P. guarauna, both of which occur
in Florida.
GUARA ALBA (Linnaeus)
White ibis
Scolopax alba Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 145.
The distal ends of right and left humeri, of two right ulnae, and
the lower end of a left tibio-tarsus were collected by W. W. Holmes
in the Seminole Field. The white ibis is locally common in Florida

at the present time.
This ibis is here first reported as a fossil.

Order ANSERIFORMES
Family ANATIDAE
CYGNUS COLUMBIANUS (Ord)
Whistling swan

Anas columbianus Ord, in Guthrie’s Geogr., 2d Amer. ed., 1815, p. 319.

The proximal ends of right and left coracoids were obtained by
W. W. Holmes in collecting in the Seminole Field. In modern times
this swan is found in winter occasionally in Florida, mainly along the
Gulf Coast, its principal winter range in eastern North America being
farther north.

The whistling swan has been known previously as a fossil only from
the Pleistocene beds at Fossil Lake, Oregon.

1 Amer. Mus. Nov., No. 301, Feb. 29, 1928, pp. 2-3.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE IQ

CYGNUS BUCCINATOR Richardson

Trumpeter swan
Cygnus buccinator Richardson, Faun. Bor.-Amer., vol. 2, 1831, (1832), p. 464.

One entire and three fragmentary humeri, a metacarpal, a coracoid,
and a tibio-tarsus are included in the Florida State Geological Survey
collections from near the head of the Itchtucknee River in Columbia
County (Catalog nos. V-4576; V-4589; V-4599; V-4598, 2 specimens ;
V-4599 and V-4826). These bones are in an excellent state of preser-
vation, part being dark and part light in color. They agree perfectly
with the modern bird, and are larger and stronger than the correspond-
ing bones in the whistling swan. This fine bird, known previously from
the Pleistocene of Fossil Lake, Oregon, formerly ranged widely
through interior and western North America but at the present time
is at so low an ebb of abundance as to be nearly extinct. It wintered
formerly from southern Indiana and southern Illinois to Texas but
has not been reported previously from Florida.

BRANTA CANADENSIS (Linnaeus)

Canada goose
Anas canadensis Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 123.

From the Seminole Field this goose, a common species on the
northern part of the Gulf Coast of modern Florida, is represented by
the lower end of a left tibio-tarsus and the distal ends of right and
left ulnae. Two ulnae and a broken metacarpal are found in collec-
tions in the Florida State Geological Survey from near the head of
the Itchtucknee River in Columbia County.

BRANTA CANADENSIS HUTCHINSI (Richardson)

Hutchins’ goose
Anser hutchinsti Richardson, Faun. Bor.-Amer., vol. 2, 1831 (1832), p. 470.

In the Holmes collection from the Seminole Field near St. Peters-
burg there is a distal end of a left ulna of the Branta type that is a
counterpart of B. canadensis except for its smaller size. Part of an
ulna comes from the Itchtucknee River, and material secured near
Melbourne for the Museum of Comparative Zoology by C. P. Single-
ton includes a right humerus that also has the same characters. These
are identified as from the Hutchins’ goose, a species rarely recorded
from modern Florida. Current custom in recognizing this bird as a
subspecies of the Canada goose is here followed though some doubt
may be expressed as to whether the two are not specifically distinct.

20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

This form is here identified certainly for the first time in a fossil state,
the only previous records being open to question. Shufeldt* has re-
corded it uncertainly from the Pleistocene of Fossil Lake, Oregon,
listing it in his final table with a query.”

ANAS PLATYRHYNCHOS Linnaeus
Mallard

Anas platyrhynchos Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 125.

Two humeri of the mallard, strong and robust bones, were collected
in the Itchtucknee River deposits in Columbia County by Mr. and
Mrs. H. H. Simpson.

Remains of other ducks of the mallard type from the Seminole
Field, collected by W. W. Holmes, include parts of humeri, an ulna,
a metacarpal, and parts of several coracoids, all in such fragmentary
form that it is not expedient to attempt to identify them specifically.
Parts of two humeri of similar status were obtained by J. E. Moore
near Venice. Probably the black duck and mallard are both repre-
sented.

The mallard is a regular migrant to Florida.

ANAS RUBRIPES Brewster
Black duck

Anas obscura rubripes Brewster, Auk, 1902, p. 184.

Collections from the Itchtucknee River, Columbia County, made by
Mr. and Mrs. H. H. Simpson, include a humerus and a metatarsus of
this species, which has not been recorded previously as a fossil.

Like the mallard the black duck comes regularly to winter in Florida.

ANAS FULVIGULA Ridgway
Florida duck

Anas obscura var. fulvigula Ridgway, Amer. Nat., vol. 8, February, 1874,
Dako
A metacarpal and the proximal and distal ends of two humeri are
equal in size to a female of this species and are identified as this bird.
These specimens come from the Holmes collection from the Seminole
Iield. In material obtained by J. E. Moore at Bradenton there is part
of another humerus.

*Bull. Amer. Mus. Nat. Hist., vol. 32, July 9, 1913, pp. 147, 156, pl. 33, fig.
414.

* See also Shufeldt, Auk, 1913, p. 30, and Science, vol. 37, February 21, 1013,
p. 307, where this same record is given as Branta canadensis hutchinsi (?).

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 21

Collections in the Florida State Geological Survey from near the
head of the Itchtucknee River, Columbia County, contain two entire
and four broken humeri, a coracoid, and two radii (the last being
identified tentatively on basis of agreement in size and contour).
Apparently the species was common in that area. These represent the
first records for the Florida duck in a fossil state. This species breeds
abundantly in Florida at the present time.

NETTION CAROLINENSE (Gmelin)
Green-winged teal

Anas carolinensis Gmelin, Syst. Nat., vol. 1, pt. 2, 1780, p. 533.

In material collected by W. W. Holmes in the Seminole Field there
are parts of two right and two left humeri, of right and left ulnae, and
a left coracoid. The humerus in this species is distinctly shorter than
in the blue-winged teal.

This duck has been reported previously from the Pleistocene of
Oregon, and from several localities in California.

QUERQUEDULA FLORIDANA Shufeldt

Querquedula floridana Shufeldt, Florida State Geol. Surv., Ninth Ann. Rep.,
1917, p. 36, pl. 1, fig. 4, pl. II, fig. 25.

The type specimen of this teal (figs. 2-3) was collected in stratum

Number Two at Vero, Florida, and is now in the collections of the

Tics. 2-3.—Type of Querquedula floridana
Shufeldt (natural size).

22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

United States National Museum. While closely similar to the humerus
of the living blue-winged teal, Querquedula discors, the fossil is
heavier and stronger throughout both in the shaft and in the proximal
and distal ends. It thus bears out the characters assigned to it in the
original description.

In the Holmes collection from the Seminole Field there 1s one right
and one left humerus nearly complete, and the fragments of three or
more others that correspond very closely to the type specimen. With
them are three broken metacarpals. A portion of a humerus was
secured by J. E. Moore near Venice. The proximal half of a right
humerus was collected by C. P. Singleton at Melbourne for the Mu-
seum of Comparative Zoology. There are also a number of bones
from the Itchtucknee River deposits, Columbia County, in the collec-
tions of the Florida State Geological Survey.

Following are measurements of the four complete humeri at hand.

Type, Vero, Florida, total length 65.8, transverse diameter through
trochleae 9.9, transverse diameter through head 16.1, transverse di-
ameter of shaft at center 5.4 nim.

Two specimens, Seminole Field, total length 65.7-70.2, transverse
diameter through trochleae 10.3-10.5, transverse diameter through
head 15.3-15.6, transverse diameter of shaft at center 5.I-5.3 mm.

Three modern Querquedula discors, two males and one female,
measure as follows: total length 65.4, 65.9, 61.8, transverse breadth
through trochleae 9.7, 10.0, 9.2, transverse breadth through head 13.8,
14.0, 13.1, transverse diameter of shaft at center 4.7, 4.8, 4.6 mm.

NYROCA VALISINERIA (Wilson)
Canvasback
Anas valisineria Wilson, Amer. Orn., vol. 8, 1814, p. 103, pl. 70, fig. 5.

A complete ulna is found in collections made by Mr. and Mrs. H. H.
Simpson on the Itchtucknee River in Columbia County.
This species in Florida is a winter migrant from the north.

NYROCA AFFINIS (Eyton)
Lesser scaup duck
Fuligula affinis Eyton, Monogr. Anatidae, 1838, p. 157.

A left humerus collected in the Number Two bed at Melbourne by
J. W. Gidley in 1926, with a right metatarsus in the Holmes collection
from the Seminole Field, part of an ulna obtained near Venice by
J. E. Moore, and four complete and one fragmentary humeri, two

NO; 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—-WETMORE 23

ulnae, a metacarpal, and a tibio-tarsus from the head of the Itchtucknee
River, Columbia County, constitute definite record of this duck. Some
miscellaneous fragments from near St. Petersburg belong probably
to this species but cannot be certainly identified. There is also a
complete ulna from Saber-tooth Cave near Lecanto, collected in 1928
by W. W. Holmes. These form the only certain Pleistocene records
for this duck which is found in abundance in Florida during winter
at the present time.

ERISMATURA JAMAICENSIS (Gmelin)
Ruddy duck
Anas jamaicensis Gmelin, Syst. Nat., vol. 1, pt. 2, 1780, p. 510.

The ruddy duck is represented by a metacarpal collected near
Venice by J. E. Moore. This species is a common visitor to Florida
during the winter.

LOPHODYTES CUCULLATUS (Linnaeus)
Hooded merganser
Mergus cucullatus Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 120.

A humerus was obtained on the Itchtucknee River, Columbia
County, by Mr. and Mrs. H. H. Simpson.

Order FALCONIFORMES
Family CATHARTIDAE
CATHARTES AURA AURA (Linnaeus)
Mexican turkey vulture
Vultur aura Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 86.

In the Holmes collection from the Seminole Field there is the distal
end of a right tibio-tarsus and the shaft of a left coracoid of a turkey
vulture that are decidedly smaller than these bones in the modern
bird of the United States, but that agree exactly in dimension with a
specimen of the Mexican form from Matamoras. They are identified
as typical aura which is here first recorded from within the limits of
the United States, its modern range being from Mexico south to
Panama and Columbia, including Cuba and Jamaica. The transverse
breadth of the distal end of the tibio-tarsus in the Pleistocene speci-
men is 11.0 mm. The modern specimen of aura from Matamoras
(U. S. N. M. 1442) is exactly similar. In a series of eight modern
birds of septentrionalis from Florida, Virginia, Maryland, and Penn-

24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

sylvania this measurement ranges from 12.2 to 13.1 mm. with an
average of 12.6 mm. The fossil bone in form is exactly like the
modern aura.

The existence of remains of two evidently distinct races of the
turkey vulture in the Pleistocene deposits of Florida, of which one is
now northern and the other southern in range is puzzling. Since the
two forms are found in different deposits, septentrionalis coming
from Vero and Melbourne on the east coast and Lecanto in the central
part of the state, and aura from near St. Petersburg, if it is assumed
that the record of aura is not due to a stray individual out of its
normal range, it seems probable that the bone deposits in question
were formed at different periods during the Ice Age.

CATHARTES AURA SEPTENTRIONALIS Wied
Turkey vulture
Cathartes septentrionalis Wied, Reis. Nord-Amer., vol. 1, 1830, p. 162.

The distal part of a left ulna obtained near Melbourne by C. P.
Singleton in 1928 for the Museum of Comparative Zoology, and a
fragment of a left metacarpal collected by W. W. Holmes in Saber-
tooth Cave at Lecanto, Florida, are similar in size to the turkey
vulture found today in Florida. Shufeldt* has reported this bird from
Vero, Florida, and from his figured specimen it is evident that the
large northern bird was, the one represented.

This form ranges today throughout the greater part of the United
States, being absent only in the north, and has had extended range for
a long period of time, since it is known from the Pleistocene deposits
of California.

CORAGYPS URUBU (Vieillot)
Black vulture
Vultur urubu Vieillot, Ois. Amer. Sept., vol. 1, 1807, p. 23, pl. 2.

Numerous fragments of bone from the Seminole Field include
parts of metatarsus, tibio-tarsus, coracoid, humerus, and metacarpal,
while from Saber-tooth Cave at Lecanto there are two bones, the
distal part of a tibio-tarsus and the upper portion of a metatarsus, the
latter from a juvenile individual. All this material was collected by
W. W. Holmes. These remains are similar in size and form to those
of modern individuals.

‘Journ. Geol., 1917, p. 18; Florida State Geol. Surv., Ninth Ann. Rep..
TOL, p. 36, pl. 1, fie. 2:

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 25

The greater abundance of remains of the black vulture over those
of the turkey vulture in these deposits is worthy of comment as indi-
cating possibly the relative abundance of these two birds during the
Pleistocene. The black vulture is first known as a fossil from this
area.

GYMNOGYPS CALIFORNIANUS (Shaw)
California condor
Vultur californianus Shaw, Nat. Misc., vol. 9, 1797, pl. 301.

In the material secured on Hog Creek, near Sarasota, Florida, by
J. E. Moore in 1928, there is part of the distal end of a left meta-
tarsus (see pl. 4, and figs. 4-5) that agrees so exactly in form and

Fics. 4-5.—Fragmentary metatarsus of Cali-
fornia Condor (Gymnogyps californianus) from
near Sarasota (natural size).

dimension with two modern specimens of the California condor that
there is no hesitancy in identifying it as that species. It may be
remarked that Gymnogyps has the middle trochlea of the metatarsus
decidedly smaller than the South American condor ultur, this serving
to distinguish the metatarsus in these two genera without difficulty.

In collecting in the Seminole area W. W. Holmes obtained a bit
of a right humerus comprising the ulnar trochlea with the adjacent
external parts, and the distal end of a right radius that are identified
as remains of this species.

The previously known range of the California condor has been
entirely western as it has been found living in the coast ranges of
California from Santa Clara County south into northern Lower Cali-
fornia, ranging in earlier days north to the Columbia River. Though

26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

it has been reported casually east to Arizona, according to Swarth*
these records lack confirmation. Miller * has identified an ulna from
cave deposits of uncertain age near Las Vegas, Nevada. The same
author * in writing of the occurrence of this species in the Pleistocene
asphalt beds of Rancho La Brea at Los Angeles, California, says that
it is the most common of the American vultures in these deposits, its
remains occurring in almost incredible abundance. With large series
of Pleistocene material for examination he found remarkable uni-
formity when compared with bones from modern specimens.

The report of this species from the Pleistocene of Florida is the
first note of the occurrence of a condor-like bird in eastern North
America and gives an entirely unexpected extension of range for
this species during the Ice Age. Discovery of further remains will be
awaited with interest.

Family TERATORNITHIDAE
TERATORNIS MERRIAMI L. H. Miller
Teratornis

Teratornis merriamt L. H. Miller, Univ. California Publ. Geol., vol. 5, Sep-
tember 10, 1900, p. 307, figs. I-09.

Among fossils collected by W. W. Holmes in the Seminole area
there were found a number of small fragments of bones from what
was evidently a very large form of bird. After some study it was
clear that these were from some large vulture so that on prelimi-
nary examination they were placed among remains assigned to the
condors. Critical study indicated certain puzzling differences from
both the California and the South American condors and it was with
much surprise that they were found to come from the great Tera-
tornis known previously only from California where its remains have
been found in the asphalt deposits at Rancho La Brea, McKittrick
and Carpinteria.

As stated above the remains from the Holmes collection are all
highly fragmentary. The distal end of a left ulna (figs. 9-11), one
of the most characteristic bits, agrees minutely in its somewhat intri-
cate details with two specimens of Teratornis from California. Two
bits of humeri include the articular surface of the head and the
radial trochlea of a left humerus. There are further the distal ends

* Pac. Coast Avif., No. 10, May 25, 1914, p. 83.

* Condor, 1931, p. 32.

“Carnegie Inst., Washington, Publ. No. 349, August, 1925, p. 81.

NO. 2 AVIFAUNA OF PLEISTOCENE IN: FLORIDA—WETMORE

bo
Sy

of left and right radii and the lateral facets from the head of a left
coracoid. These likewise in size and detail are like the corresponding
parts in Teratornis. In fact the agreement is so close that there is
no basis for differentiating the Florida bird from that of California.

Fics. 6-8.—Metatarsus of Teratornis merriami from Bradenton
(natural size).

With the material described above at hand it has been highly grati-
fying to find in specimens collected by J. 2. Moore at Bradenton a
nearly complete metatarsus (pl. 5, and figs. 6-8) and a broken femur
(fig. 12) that likewise agree in close detail with the bird of California.

an ANI SS

9 10 11

Fics. 9-11.—Distal end of ulna of Teratornis merriami from the Seminole
area (natural size).

Fic. 12——Femur of Teratornis merriami from
Bradenton (natural size).

28

NO. 2. AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 29
The femur was in a fragile condition and was received in several
pieces. These have been so skillfully joined by N. H. Boss that they
illustrate well the form of the bone.

The identification of this form from Florida is one of the most
unexpected results of this study of the Pleistocene from Florida and
indicates a wide range in Pleistocene times for this peculiar bird.

Bamily, ACGUIPT TRIDAI
BUTEO JAMAICENSIS (Gmelin)
Red-tailed hawk
Falco jamaicensis Gmelin, Syst. Nat., vol. 1, pt. 1, 1788, p. 266.

The red-tailed hawk is represented in the W. W. Holmes collection
from the Seminole Field by the distal end of a left metatarsus, the
lower ends of two left tibio-tarsi, and a left coracoid. J. IX. Moore
secured part of an ulna near Venice. Gidley collected the lower por-
tion of a left humerus in the lower part of the Number Two bed on
the golf links at Melbourne. The species occurs today in Florida, and
has been recorded previously as fossil in the Pleistocene of California.

The red-tailed hawk has been known for many years as Buteo
borealis, the original reference being Falco borealis Gmelin, Syst.
Nat., vol. 1, pt. 1, 1788, p. 266, where it is species No. 75. The pre-
ceding species, No. 74, Falco jamaicensis on the same page is based
on the cream-colored buzzard of Latham’ described from a specimen
from Jamaica, evidently an immature of the red-tailed hawk. As the
name jamaicensis comes first on the page in question in Gmelin’s
work it will replace the familiar borcalis as the specific name for
this hawk.

BUTEO LINEATUS (Gmelin)
Red-shouldered hawk

Falco lineatus Gmelin, Syst. Nat., vol. 1, pt. 1, 1788, p. 268.

The red-shouldered hawk was apparently as common in Pleistocene
times as today, for it is represented by a number of fragmentary
bones. Holmes obtained a left humerus lacking the head and the
distal ends of two tibio-tarsi from the Seminole area. J. E. Moore
secured part of a metatarsus near Venice. In excavations on the golf
links at Melbourne Gidley secured a nearly complete left metatarsus
in 1926, a fragment of another in 1928, and a broken left femur
in 1930. The red-shouldered hawk is represented in modern Florida
by a resident form Buteo lineatus alleni that besides differing in color

* Gen: Syn. Birds, vol. 1, pt. 1, 1781, p. 40:

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

from the northern race is slightly smaller, and by a migrant form
Buteo lineatus lineatus that is present in winter and is slightly larger.
The specimens from the west coast are from slightly smaller birds
than those from Melbourne, suggesting that possibly two forms are
presented in the fossil material. This cannot be definitely decided
from the bones now at hand.

The red-shouldered hawk is here first recorded in fossil form.

BUTEO PLATYPTERUS (Vieillot)
Broad-winged hawk
Sparvius platypterus Vieillot, Tabl. Encycl. Meéth., vol. 3, 1823, p. 1273.

In the Seminole Field in Pinellas County, W. W. Holmes obtained
the distal end of a left humerus, and the proximal part of a left meta-
carpus. This species has not been recorded previously as a fossil.

The broad-winged hawk today is a winter visitor to Florida ar-
riving from the north in October and departing in March.

GERANOAETUS sp.
Eagle

The distal ends of three ulnae obtained by W. W. Holmes in the
Seminole Field come from an eagle of this genus, which was repre-
sented in the Pleistocene of California by two species G. fragilis and
G. grinnelli and of which there is one living species G. melanoleucus
in South America. The bones from Florida agree in size with the
latter. The material is considered too fragmentary to warrant specific
determination at the present time. The genus is here first recorded
from eastern North America.

HALIAEETUS LEUCOCEPHALUS (Linnaeus)
Bald eagle
Falco leucocephalus Linnaeus, Syst. Nat., ed. 12, vol. 1, 1766, p. 124.

The bald eagle is represented by fragments from the Seminole
Field, and by two broken radii from Saber-tooth Cave, near Lecanto,
collected by W. W. Holmes, as well as by part of an ulna collected
near Venice by J. E. Moore and a number of bones from near Mel-
bourne, obtained by Gidley and Singleton. The collection made by
Singleton for the Museum of Comparative Zodlogy contains part of
a metacarpal. Several of the Melbourne specimens are practically
complete, and show no differences from the modern bird which is
common at present in Florida.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 31

PANDION HALIAETUS (Linnaeus)
Osprey
Falco haliaétus Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. Ol.

A nearly complete left metatarsus was obtained by J. W. Gidley on
April 5, 1929, in the Number Two bed at Melbourne where it was
associated with remains of a peculiar extinct bear, Arctodus flori-
danus. In 1930 a femur was secured in the Itchtucknee deposits by
Mr. and Mrs. H. H. Simpson. These are the first reports of the
osprey in the Pleistocene of our continent.

Family FALCONIDAE
POLYBORUS CHERIWAY (Jacquin)
Audubon’s caracara
Falco (cheriway) Jacquin, Beytr. Gesch. Vogel, 1784, p. 17, pl. 4.

On the golf course near Melbourne, Gidley collected an ulna that is
identical with that of modern birds. On the west coast W. W.
Holmes obtained numerous remains in the Seminole Field, these in-
cluding parts of humeri, ulnae, a femur, a tibio-tarsus, and several
metatarsi. Two fragmentary humeri and the lower end of a tibio-
tarsus are similar in size to small modern specimens of the caracara
from Florida. Parts of four metatarsi agree in having the distal troch-
leae distinctly smaller than in any of the three modern birds seen.
Ulnae and part of a femur also seem smaller than usual. It will be
recalled that two subspecies of this caracara are now recognized,
Polyborus cheriway cheriway of northern South America, and Poly-
borus cheriway auduboni of Florida and the southwestern part of the
United States south into Mexico, the former being smaller in size.
The smaller fossil bones here under discussion seem to show approach
to the modern race of South America.

Remains of the caracara are common in the Pleistocene deposits
of California but are here reported for the first time outside that State.
In Florida the species at the present time is peculiar to the prairies of
the Okeechobee and Kissimmee regions, where it is locally common.

FALCO SPARVERIUS Linnaeus
Sparrow hawk
Falco sparverius Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 90.
Material collected in Saber-tooth Cave at Lecanto, Florida, in 1928
by W. W. Holmes includes parts of right and left tibio-tarsi of this

species, a common bird in this area at the present time.
3

32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Order GALLIFORMES
Family PERDICIDAE
COLINUS VIRGINIANUS (Linnaeus)
Bob-white
Tetrao virginianus Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 161.

In material collected by W. W. Holmes in the Seminole area re-
mains of the quail may be said to be common since this species is
represented by six humeri more or less complete and part of an ulna.
It is likewise common in the collection from Saber-tooth Cave near
Lecanto where two humeri, a metatarsus and two femora were ob-
tained. In excavations on the golf links at Melbourne in February,
1928, Gidley secured two humeri at the line of contact between _
stratum Number One and stratum Number Two.

These bones all appear similar to those of modern quail. The
species is abundant in Florida, and has been previously reported as a
fossil from Pleistocene cavern deposits in Tennessee.

Family MELEAGRIDIDAE
MELEAGRIS GALLOPAVO Linnaeus
Turkey

Meleagris gallopavo Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 156.
Ardea sellardsi Shufeldt, Journ. Geol., Jan—Feb. (publ. Jan.), 1917, p. 109.

In the series of bird bones obtained in the Seminole area, Pinellas
County, by W. W. Holmes, remains of the wild turkey far out-
number those of any other species represented ; they include 98 frag-
ments of metatarsi, tibio-tarsi, femora, coracoids, humeri, ulnae, and
metacarpi. Most of these have been broken into small bits. The speci-
mens exhibit the usual variation in size found in series of wild turkey
bones, and do not differ from modern specimens. In the material
collected on Hog Creek, near Sarasota, by J. E. Moore in 1928 there
is included parts of a femur and a tibio-tarsus, the latter considerably
worn. The proximal end of a metatarsus has been forwarded by
Mr. Moore as taken at Bradenton. In Saber-tooth Cave near Lecanto
Holmes obtained a single spur core from the metatarsus of a male
individual. Collections in the Florida State Geological Survey from
near the head of the Itchtucknee River, Columbia County, include
metatarsi, femora, humerus, ulnae, and other bones, all more or less

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—WETMORE 33

fragmentary, and part of a sternum. There is part of a metatarsus
in the same collections from the north bank of the canal between the
Florida East Coast Railroad and the highway at Vero. Gidley col-
lected a broken metatarsus near Melbourne March 1, 1928, and
Singleton in the same year working in this same deposit for the
Museum of Comparative Zoology secured parts of humerus, meta-
tarsus, and tibio-tarsus. The wild turkey must have been abundant in
Florida during the Ice Age.

The type specimen of Ardea sellardsi Shufeldt,’ the distal end of
a right tibio-tarsus, proves on examination to be from a wild turkey.
The bone is from an individual apparently barely adult and of small
size, possibly from a young female. The condyles are worn and
abraded in such a manner as to mask their true form, leading to error
in the earlier identification. The type in question is equalled in size
by the smallest in a considerable series of modern wild turkey bones
examined. Ardea sellardsi thus becomes a synonym of Meleagris
gallopavo. The specimen was taken in Pleistocene deposits in stratum
Number Three, near Vero, Florida.

MELEAGRIS TRIDENS sp. nov.

Characters—Metatarsus (pl. 6, and fig. 13) similar to that of
Meleagris gallopavo Linnaeus * but male with three-pointed spur core.

Description—Type, U.S. Nat. Mus. No. 12052. Central portion of
shaft of right metatarsus, collected by W. W. Holmes, in the Semi-
nole area, Pinellas County, Florida. Shaft strong, flattened antero-
posteriorly below, and more rounded above ; anterior face with a wide,
shallow groove that becomes obsolete at level of central spur; below
this the anterior face is ridged and shallowly grooved by tendons
leading to the toes; external side of shaft rounded; internal side more
flattened, spurs rising from a common base in a broad buttress of
bone projecting obliquely inward from the inner side of the posterior
surface; central spur strong and heavy (tip partly broken away) ;
with an accessory spur above and below of smaller size, the upper one
slightly more acute than a right triangle in outline, relatively broad
transversely, with the distal extremity widened laterally so that in
form it is like a cog in a cogwheel; distal accessory spur longer, more
slender, with a conical, rather sharp point; outer surface of buttress
supporting spurs broadly grooved for the passage of tendons that in
life passed down the back of the metatarsus ; a distinct, rather narrow,

‘Journ. Geol., Jan-Feb. (publ. Jan.), 1917, p. 19. See also Florida State
Geol. Surv., 1917, Ninth Ann. Rep., pp. 38-39, pl. 2, fig. 15.
* Meleagris gallopavo Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 156.

34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

shallow groove across the base of the spur core buttress on the an-
terior side, to allow passage of another tendon. Bone brown in color,
well fossilized.

Remarks.—In size and form the specimen here described is similar
to the ordinary turkey, Meleagris gallopavo, except for the three
points of the spur core. Description of it as a new species has been
withheld for 2 years to allow careful consideration of its peculiarities.
These are susceptible of three interpretations: (1) that the bone is
pathological and therefore aberrant, (2) that it is simply an individual
variant, and (3) that it represents a distinct species.

Fic. 13.—Fragmentary metatarsus (type) of Meleagris
tridens (natural size).

With regard to the first it needs only casual inspection to determine
that the entire bone, including the spur cores is entirely normal and
without any indication of disease. The very regularity and symmetry
of its development indicate that the increased number of spurs is not
due to any injury. As for the second supposition, in the past two years
the writer has examined critically all of the specimens of wild turkeys
that have been available to him, has seen the tarsal bones of a con-
siderable number, one hundred or more, that have come from Indian
pueblos in the Southwest and elsewhere, has seen several hundreds of
domestic turkeys, and has talked with persons who have reared do-
mestic turkeys for years without learning of any instance where a
male turkey had more than a single spur. Under these circumstances

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—-WETMORE 35

it has seemed entirely logical to consider this specimen as representa-
tive of a peculiar species. Among numerous other tarsal bones in the
material from the Florida Pleistocene it stands unique, indicating
probable rarity. It is possible that some of the bones from other parts
of the skeleton that have been identified as Meleagris gallopavo belong
to M. tridens but this cannot be determined.

It may be remarked that multiple spurs are not unusual among
gallinaceous birds of the Old World, though hitherto unknown in any
American species. The pheasants of the genus /thaginis regularly
have two or more pairs of spurs in the male. The same is true of
Polyplectron, while according to Ogilvie-Grant * duplicate spurs occur
in Haematortyx, Caloperdix, and Galloperdix. The vulturine guinea
fowl, Acryllium vulturinum, frequently has two to four lumpy spur-
like processes on the tarsus.

The type of Meleagris tridens is so fragmentary that it affords few
measurements. The transverse diameter of the shaft just below the
spurs is 9.0 mm. The buttress supporting the spurs is 30.6 mm. long.
The form may be ascertained from the accompanying figure.

Order GRUIFORMES
Family GRUIDAE
GRUS AMERICANA (Linnaeus)
Whooping crane
Ardea americana Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 142.

Remains of cranes obtained by W. W. Holmes in the Seminole
Field, Pinellas County, include fragments of metacarpus, tibio-tarsus
and femur, and of three ulnae. Two fragmentary ulnae and one
radius are in the collections of the Florida State Geological Survey
from the Itchtucknee River area in Columbia County. Gidley collected
part of an ulna from stratum Number Two at Melbourne in 1930, and
Singleton secured part of another ulna in the same beds in June, 1929,
when collecting for the Museum of Comparative Zoology. All are
easily distinguished from the bones of other cranes found with them
by their much greater size.

Though the whooping crane was recorded from Florida by early
ornithologists, in recent years doubt has been cast upon these reports
and the species seems not to have been certainly found in modern

*Cat. Birds Brit. Mus., vol. 22, 1893, pp. 221, 222, 260.

36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

times south of Georgia. These records from the Pleistocene are
therefore of exceptional interest. This species is now nearly extinct,
only a few individuals being known to exist in the interior of our
country. It has not been recorded previously as a fossil.

GRUS CANADENSIS (Linnaeus)

Gray crane
Ardea canadensis Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 141.

Remains of gray cranes are common in the deposits at Melbourne
and in the Seminole area, and contain representatives of two forms,
one large in size and the other small. The large form has the dimen-
sions of Grus canadensis pratensis Meyer, the Florida crane, which is
resident in Florida today, and is supposed to be that race. There is
now recognized however another race, Grus canadensis tabida Peters,
of more northern and western range that resembles pratensis in size,
but differs in coloration, so that there is no certainty as to the form
that ranged in Florida during the Pleistocene. This larger bird is
represented in the Pleistocene collections by a coracoid, a femur, and
part of a metacarpal, all secured by Gidley near Melbourne, and the
head of a metatarsus and the symphysis of a lower mandible collected
by Holmes in the Seminole area, as well as by part of a tibio-tarsus
secured by Moore at Bradenton.

The smaller race from the Florida Pleistocene has the dimensions
of the little brown crane, Grus canadensis canadensis, that now ranges
in the western half of the United States, and might be supposed to
be that form were it not that the Cuban crane, Grus canadensis
nesiotes, is a bird of equally small dimension. In fact the differences
between G. c. canadensis and G. c. nesiotes seem to rest on color
characters that appear not to have been definitely worked out. The
small form is represented in the Pleistocene collections at hand by the
distal end of a humerus, parts of two radii, and two coracoids from
Melbourne, obtained by Gidley, and the distal end of a humerus
secured by Holmes in the Seminole area.

The occurrence of these two races in the Pleistocene of Florida is.
suggestive of the modern condition in the western part of the United
States, where a large gray crane and a small one occur together during
migration over a considerable area.

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA

WETMORE 37

Family ARAMIDAE

ARAMUS PICTUS (Meyer)
Limpkin
Tantalus pictus Meyer, Zool. Ann., vol. 1, 1794, p. 287.

The distal end of a left humerus was collected by W. W. Holmes
in the Seminole area. Parts of five metatarsi and a broken tibio-
tarsus are found in collections from the Itchtucknee River deposits
in Columbia County, the specimens being in the Florida State Geo-
logical Survey. All these are similar to the corresponding bones in
modern birds. The species is of regular occurrence in Florida at the
present time. It has not been recorded previously as a fossil.

Family RALLIDAE
RALLUS ELEGANS Audubon

King rail
Rallus elegans Audubon, Birds Amer. (folio), vol. 3, 1834, pl. 203.

In the Seminole area W. W. Holmes secured a complete right
femur, and Mr. and Mrs. H. H. Simpson obtained a humerus in the
Itchtucknee beds in Columbia County. These bones of this species
are distinguished from the clapper rail by larger size.

This rail, common in Florida now, inhabits mainly fresh-water
marshes. It has not been recorded previously as a fossil.

RALLUS LONGIROSTRIS Boddaert

Clapper rail
Rallus longirostris Boddaert, Tabl. Planch. Enl., 1783, p. 52.

The distal end of a humerus comes from the Seminole area, col-
lected by W. W. Holmes. The clapper rail is a sedentary species
inhabiting salt-water marshes that is common at the present time along
the coast of Florida, where several subspecies, slightly differentiated
from one another, occur in different geographic areas.

It has not been reported previously as a fossil.

ARAMIDES CAJANEA (Miiller)

Wood rail
Fulica cajanea Miller, Vollst. Nat. Suppl., 1776, p. 1109.
The determination of two fragmentary metatarsi (see figs. 14-16)
and a nearly complete femur collected by W. W. Holmes in the Semi-

nole area as belonging to a form of wood rail, a group of birds com-
prising several forms that range now from southeastern Mexico south

38 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

to Argentina, has been another of the unexpected finds in the present
collection. The two metatarsal bones are well fossilized, one being
black in color and the other brown. The femur contains somewhat
less mineral.

Identification to species of these bones has been difficult owing to
lack of material for comparison. That they are not related to the
large A. ypecaha and A. saracura of the area from southern Brazil
southward is obvious. Likewise it is evident on close study that they

Fics. 14-16.—Metatarsi1 of the wood rail Aramides cajanea from the Seminole
area (natural size).

are from a bird larger than A. axillaris and its allies, which are among
the smallest forms of the group. They are smaller than albiventris,
but agree with A. cajanea, which now ranges in two or more sub-
species from Panama southward into Brazil, and are identified as of
that group on this basis. The genus has not been previously recorded
north of southeastern Mexico nor has it been previously encountered
as a fossil. Its occurrence in the Pleistocene of Florida is quite in
keeping with the various types of mammals of South American
affinity that come from these same beds.

NO. 2 AVIFAUNA OF PLEISTOCENE IN. FLORIDA—WETMORE 39

GALLINULA CHLOROPUS (Linnaeus)
Gallinule

Fulica chloropus Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 152.

Parts of four humeri were collected by W. W. Holmes in the
Seminole field on the west coast, and another humerus was secured
on the Itchtucknee River, Columbia County, by Mr. and Mrs. H. H.
Simpson. The species, abundant in present day Florida, has not been
recorded previously as a fossil.

FULICA AMERICANA Gmelin
Coot

Fulica americana Gmelin, Syst. Nat., vol. 1, pt. 2, 1780, p. 704.

Bones of this species collected in the Seminole area by W. W.
Holmes include one entire and two fragmentary humeri and the distal
ends of two tibio-tarsi. Parts of a humerus and a coracoid were
obtained by J. E. Moore at Bradenton. A number of other limb bones
are found in the collections of the Florida State Geological Survey
from the Itchtucknee River area in Columbia County.

The coot is found now in abundance in Florida in winter and a few
remain to nest during summer. The species has been reported pre-
viously from the Pleistocene of Oregon.

Order COLUMBIFORMES
Family COLUMBIDAE
ZENAIDURA MACROURA (Linnaeus)
Mourning dove
Columba macroura Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 164.

Two metacarpals and the proximal end of an ulna from the Semi-
nole Field, obtained by W. W. Holmes, are in size similar to the
modern forms of the mourning dove of North America, being larger
than the bird of the West Indies.

Order STRIGIFORMES
Family TYTONIDAE
TYTO ALBA (Scopoli)
Barn owl
Strix alba Scopoli, Annus I. Hist.-Nat., 1769, p. 21.

In the collection obtained by W. W. Holmes in Saber-tooth Cave
at Lecanto in 1928 there are a number of fragmentary bones of the

40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

barn owl, including parts of the femur, tibio-tarsus, metatarsus, and
ulna. The species is quite common in modern Florida.

Family STRIGIDAE
OTUS ASIO (Linnaeus)
Screech owl
Strix asio Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 92.
Parts of two humeri of the screech owl were collected by W. W.

Holmes in Saber-tooth Cave near Lecanto. This bird is common and
widely distributed in Florida.

STRIX VARIA Barton
Barred owl

Strix varius Barton, Fragm. Nat. Hist. Penn., 17909, p. II.

The barred owl, a common species in Florida at the present time,
seems to have had equally wide distribution during the Pleistocene.
In the Seminole area W. W. Holmes obtained a number of fragments
including parts of the metatarsus, humerus, ulna, and metacarpus.
In the excavation of Saber-tooth Cave at Lecanto Mr. Holmes further
obtained a nearly complete femur. J. W. Gidley secured part of a
metatarsus in the golf links area at Melbourne. The species has not
been previously recorded as a fossil.

Order PASSERIFORMES
Family CORVIDAE
CORVUS BRACHYRHYNCHOS Brehm
Crow
Corvus brachyrhynchos Brehm, Beitr. Vogelk., vol. 2, 1822, p. 56.

Numerous remains of the common crow were secured by W. W.
Holmes in the Seminole area, indicating that this species was as
common during the Pleistocene as it is in Florida at the present time.
Crows have been recorded previously from Ice Age deposits in Cali-
fornia but not before from the Pleistocene of eastern North America.

CORVUS OSSIFRAGUS Wilson
Fish crow
Corvus ossifragus Wilson, Amer. Orn., vol. 5, 1812, p. 27, pl. 37, fig. 2.

A humerus, lacking the head, was obtained by W. W. Holmes in
the Seminole area, this being the first report of this species for the

NO. 2 AVIFAUNA OF PLEISTOCENE IN FLORIDA—-WETMORE 41

Pleistocene. The bone is similar to that of the common crow but is
decidedly smaller. The fish crow is widely distributed through the
Florida Peninsula today.

Family ICTERIDAE
AGELAIUS PHOENICEUS (Linnaeus)
Red-winged blackbird
Oriolus phoeniceus Linnaeus, Syst. Nat., ed. 12, vol. I, 1766, p. 161.

A right humerus lacking the distal end, and a left one with part of
the head missing, were secured in the Seminole area by W. W.
Holmes. This is a common resident of marshes throughout much of
North America and abounds today in Florida. It has not been identi-
fied certainly before from the Pleistocene.

’

MEGAQUISCALUS MAJOR (Vieillot)
Boat-tailed grackle
Ouiscalus major Vieillot, Nouv. Dict. Hist. Nat., vol. 28, 1819, p. 487.

The proximal portion of a right humerus found by W. W. Holmes
in the Seminole area comes from an individual of small size. These
grackles are common in Florida, ranging mainly about water. The
species has not been recorded before from the Pleistocene.

QUISCALUS QUISCULA (Linnaeus)
Crow blackbird
Gracula quiscula Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 109.

A nearly complete left humerus secured by W. W. Holmes in the
Seminole Field comes from an individual of small size. This grackle,
common in modern Florida, has not been recorded previously from
the Pleistocene.

nm ue Wen

af SG FF ae aie ia ae Hil Sain ae
— — Wy BAG, 2 ae apes ae ee BV Th vi fee mea: ae
Pe a at ce Sb inieattl iain a
— * | Aan ts at uh: “Se LE ig
, ; i . Q a, Ais i t Eas a a i ie
oy i ae AE, a rsa oe vai

E - y ef ey By
BP eet a i ey in id gr vas oe: ee
f erage 78 ena sf i ‘i j i [ i
Pass in es ' 2B ub Die Na ay i} ua a a Na ms had ran

rot! Tee i N nC: nah iH My (a; OMG eae Signs |
hd . Suge: aii! sa UA PEMD Me ina .
i an 1 Seen da eg Ee ai.

et ; oN shah aot onl din
tr 3 ne ih

i-2 a $ > ‘ ! : ithe a ihe aN ie ee
) 4 vi Sai ne .
15, 1 -
’ A a aan \ig os ar ane
= ee 5 ty as ri ae
¥ ! '
7 iy i , A
ae
- 7 t a Atte }
F bea ne ‘I ; mn
"1 iy \
i } en Bt ‘Ar ; At cs
a ' f ~ ac
i - ‘ 5 ite ea)
RY oi
— 1 a i“
- 7 i ” ‘
7 % ene i a
wd pica
j ‘pe a outa a ae :

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL.“85)\NO. 2,. PL. 7

1. General view of the Seminole area with excavation in foreground.

Photograph by W. W. Holmes.

~
i i

25

2. Stratification in excavation at Seminole with the bone bearing layer indi-
cated by two trowels at center. Photograph by W. W. Holmes.

4

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. (85, NO= 2), ‘RE. 2

1. Entrance to Saber-tooth Cave near Lecanto, Fla.

2. Excavations in Saber-tooth Cave.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 2, PL. 3

<y

1. Stratification at Melbourne, the bone bearing layer
being the dark band through the center. Photograph by

J. W. Gidley.

2. Excavations on the golf links at Melbourne, Fla. Photograph by

J. W. Gidley.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. "85, NO. 2); IPL. 4:

At left fragmentary metatarsus of California condor, collected by J. E.
Moore at Venice, Fla., compared with modern specimen at right. (Natural size.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLS 855 NO. 2, PEL 5

i.

iM

‘
Ry

2

J. E. Moore at Bradenton, Fla.

1
Metatarsus of Teratornis merriami obtained by |
( Natural size. )

2. Femur of Teratornis merriami obtained by J.
(Natural size. )

ie
EE. Moore at Bradenton, Fla.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLE. 85; NO; 2) (PE. 6

At left type of Meleagris tridens, compared with metatarsus of modern male
Meleagris gallopavo merriami. (Natural size.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VOLUME 85, NUMBER 3

ADDENDA TO DESCRIPTIONS OF
SUK Ghoo SotAle Osos

(WiTH 23 PLATEs)

BY
CHARLES D. WALCOTT

(With Explanatory Notes by Charles E. Resser)

(PUBLICATION 3117)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
JUNIE 295 1931

Tbe Lord Waitimore Press

BALTIMORE, MD,, U. 8 A.

ADDENDA TO DESCRIPTIONS OF BURGESS
SHALE. FOSSILS

By CHARLES D. WALCOTT
(Witn ExpLanatory Notes By CHARLES E. RESSER)
(WitTH 23 PLATES)
PREFATORY STATEMENT

Shortly after his discovery of the remarkable Burgess shale fossils
in British Columbia in 1910, Dr. Charles D. Walcott described the
more striking species of the major classes of animals and plants
represented, to bring to the attention of the scientific world their
exceptionally well preserved anatomic details. Subsequent quarrying
at the locality yielded many additional specimens of the described
forms as well as examples of rarer species not secured in the first
season’s work.

During my 13 years’ association with Doctor Walcott he frequently
dwelt upon the fact that he considered his papers on the Burgess
shale forms rather in the nature of announcements than as completed
studies of these wonderfully preserved fossils. He always intended
to return to the study of the described species and to publish more
detailed descriptions and interpretations of their form and structure.
However, the stress of war times and advancing years prevented a
realization of this hope. Nevertheless, from time to time, he had
photographs prepared or made notes of his observations regarding
structure, all of which were preserved with the collections.

At the request of the National Museum authorities I have assembled
these notes and illustrations for publication so that they may not be
lost to science. It must be remembered that none of the statements,
and particularly none of the interpretations, in the following pages
should be regarded as Doctor Walcott’s final opinion, since he recog-
nized many of them as tentative. He more than once stated that
fully 15 years’ work remained to be done on the 35,000 Burgess shale
specimens in the National Museum’s collections.

All generic and specific names, having been created by Doctor
Walcott, are, of course, to be credited to him, and not to us jointly.

In order to show clearly exactly what Doctor Walcott wrote and,
on the other hand, what I have added—chiefly by way of explana-
tion—two type faces are used. Doctor Walcott’s manuscript is printed
in 10-point type, while the explanations added by me appear in the
aUatlSe DOME type: Cuartes E. Resser.

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 3

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Nv

INTRODUCTION

It is obvious that one person cannot cover, even in a very cursory manner,
the entire field of Cambrian stratigraphy and paleontology, especially with such
rich faunas as the Burgess shale extant. Again, in the case of the Burgess
shale faunas, none but a trained student in biology can do more than merely
assign specific and generic names to the gross forms. Further, it is doubtful if
any biologist, however versatile, could by long years of study perceive every-
thing to be learned from these wonderful fossils. In view of these facts it has
seemed advisable to encourage specialists to study the various classes repre-
sented rather than attempt to monograph the subject as a whole. Accordingly,
such a study by Dr. G. E. Hutchinson has recently been printed* and one by
Dr. Rudolf Ruedemann has been submitted for publication.*? In this way we
may hope to carry on the work suspended by the passing of the discoverer of
these unique forms.

The following descriptions either were prepared entirely by Doctor Walcott
or are based on notes and illustrations left by him.

Formation and locality—For every species described in this paper, the for-
mation and locality are as follows: Middle Cambrian, Burgess shale: (Loc.
35k) on the west slope of the ridge between Mount Field and Wapta Peak,
1 mile (1.6 km.) northeast of Burgess Pass, near Field, British Columbia.

DESCRIPTION OF GENERA AND SPECIES
MARGARETIA, new genus

The characters of this strange organism are presented in the specific descrip-
tion or shown in the illustrations.
Genotype.—M. dorus, new species.

MARGARETIA DORUS, new species
Plate 1, figs. 1-6

More than 70 specimens of this peculiar organism have been assembled from
the Burgess shale collections. In the following description comparisons are
made with algae. Other notes by Doctor Walcott, apparently his latest, together
with suggestions by Mr. A. H. Clark, and particularly the presence in the same
drawer of specimens of Titanideum suberosum, indicate that Doctor Walcott’s
latest opinion was that M. dorus might really be an Alcyonarian.

Description——Mass forming a thin membranous perforated sheet,
narrow at the base and expanding to a width of 1.5 cm. in 2 cm.
distance ; length of narrow base about 1.5 cm, and of wider portion
4 cm.; the perforations are elongate oval and apparently arranged on
longitudinal and obliquely transverse lines; tegument presumably

*Restudy of Some Burgess Shale Fossils. Proc. U. S. Nat. Mus., vol. 78,

art. II, pp. I-24, pl. 1, year ?
* Some New Middle Cambtian Fossils from British Columbia. To be printed

in the Proc. U: S. Nat. Mus.

NO. 3 BURGESS SHALE FOSSILS—-WALCOTT 3

leathery as it is a black film with irregular, broken, longitudinal lines
with more or less scaly edges.

Microscopic structure undetermined.

Observations.—The perforations in the tegument are not unlike
those of the living alga Agarum turneri Post and Ruprecht, and one
might imagine that a fragment of the strong frond of this species
is similar to M. dorus, but the resemblance is only general ; the per-
forations of M. dorus are more uniform than those of the beautifully
perforate living alga Kallymenia perforata Agardh, which also has a
far more delicate tegument.

Holotype and paratypes —uvU. S. N. M., Nos. 83922, 83923a-e.

REDOUBTIA Walcott 1918
REDOUBTIA POLYPODIA Walcott
Plate 2, figs. 2-3

Redoubtia polypodia Walcott, 1918, Smithsonian Mise. Coll., vol. 68, no. 12,
p. 5, fig. 5.

The holotype is refigured since it was first published in the popular account
of field explorations issued annually by the Smithsonian Institution, which does
not reach paleontologists generally.

Accompanying the original figure is the following statement, “An elongate
creeping holothurian with numerous tube feet and tentacles.”

Whether the second specimen really represents the same species appears some-
what doubtful inasmuch as the tube feet are smaller and more numerous. The
larger appendages above the specimen, as posed on the plate, are parts of
another animal.

Holotype and paratype.

U.S. N. M., Nos. 83924 and 83925.

PORTALIA Walcott 1918
PORTALIA MIRA Walcott
Plate 3, figs. 2-3

Portalia mira Walcott, 1918, Smithsonian Misc. Coll., vol. 68, no. 12, p. 6,
figs. 6, 7.

Another holothurian first figured in the explorations account for 1918 is also
refigured, to give it wider availability. This form differs from the preceding
Redoubtia polypodia in having fewer and longer tube feet and in their apparently
different grouping.

Holotype-—U. S. N. M., No. 83027.

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

MISKOIA Walcott 1911’
MISKOIA PLACIDA, new species
Plate 2, fig. I

Comparing this form with M. preciosa Walcott, the genotype, the new species
is stouter and shorter. As none of the specimens referred to the type species
preserves the rear end, it is not possible to determine whether the lobate termi-
nation of M. placida is also characteristic of the first species. The annulations
of the body are clearly marked in the specimen illustrated, particularly on the
counterpart, which is a mold of the exterior. Teeth are shown around the mouth
as bright, shiny, curved, chitinous hooks.

Holotype—U. S. N. M., No. 83028.

CANADIA Walcott 1911
CANADIA SPINOSA Walcott

Plate 4; plate 5; plate 6, figs. 1-2

Canadia spinosa Walcott, 1911, Smithsonian Misc. Coll., vol. 57, no. 5, p. 118,
pl. 23, figs. 4-7.

Original description — Body slender, formed of 20 to 21 segments
that, when flattened on the shale, are a little longer than wide; each
segment has a pair of parapodia with a dorsal and ventral bundle of
strong non-jointed setae. The setae are finely illustrated by figures 4,
6, and 7. Head minute, with a pair of large tentacles curving outward
from the front anterior margins; a bundle of fine setae occurs on
each side of the head back of the base of the large tentacles. A
straight slender enteric canal is indicated on several specimens. Mouth
and anus not seen, but probably at or near the end of the annelid.

“ Dimensions.—The largest adult specimen has a length of 34 mm.,
with a width of the body at the seventh segment from the head
of. 1.5 mm.”

Fortunately many additional specimens of this interesting spiny worm were
found after 1911. Those first illustrated give a fairly good conception of the
general features; however, illustrations of additional specimens may show
features particularly desired by the biological student at places where the other
specimens are faulty.

Plesiotypes—U. S. N. M., Nos. 83929a-e.
Walcott, C. D., Smithsonian Misc. Coll., vol. 57, no. 5, p. 114, pl. 18, figs.
1-5, IOI.

NO. 3 BURGESS SHALE FOSSILS—WALCOTT 5

CANADIA SETIGERA Walcott
Plate 7, figs. 1, 4; plate 8, fig. 3

Canadia setigera Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 110, pl. 23,
figs. 1-3.

Canadia setigera Walcott, 1916, Ann. Rep. Smithsonian Inst., 1915, pl. 12,
figs. I-3.

The original description states that “this species differs from C. spinosa in
being more elongate, slender, and with much smaller bundles of finer setae.”

It is further stated that a series of 36 specimens shows gradation between the
two types originally illustrated. It seems, however, that in reality several dis-
tinct forms are included in the species as now constituted.

Plesiotypes.——U. S. N. M., Nos. 83930a-c.

CANADIA GRANDIS, new species
Plate 0, fig. 10

A single wide Canadia that shows the body annulations very well and that
has numerous setae seems to differ from C. spinosa mainly in the larger bundles
of setae.

Holotype-——vU. S. N. M., No. 83932.

CANADIA IRREGULARIS Walcott
Plate 6, figs. 4-6; plate 7, fig. 3
Canadia irregularis Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 120.

Original description.—“A slender species not over 20 mm. in length.
The setae are irregular in size and appearance and suggest partially
worn macerated specimens of the slender forms of C. setigera.”

The specimens on which this description was based are now illustrated for
the first time. A study of the figures, however, causes some doubt to arise
regarding specific differentiation from C. grandis.

Lectotype and paratypes—uU. S. N. M., Nos. 83933 and 83934a
and b.

CANADIA SPARSA Walcott
Plate 6, fig. 3
Canadia sparsa Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 119.

Original description.— A slender form with only two strong setae
on each very short parapodia. Finer setae may occur but they are
not shown in the one specimen.”

This form is another that was not illustrated in 1911. In this case two ques-

tions may be raised: First, the specific identity of all the specimens seems
doubtful, and second, the generic reference to Canadia is also uncertain.

Holotype —U. S. N. M., No. 83935.

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

CANADIA DUBIA Walcott
Plate 7, fig. 2; plate 8, figs. 1-2; plate 0, fig. 8
Canadia dubia Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 110.

Original description‘ This species is proposed to include a small
chaetiferous annelid not over 10 mm. in length. One specimen shows
a bundle of very fine setae on each side near the head.”

Four of the original specimens are illustrated.

Cotypes—U. S. N. M., Nos. 83936a-d.

CANADIA SIMPLEX, new species
Plate 0, fig. 9
A tiny organism that appears as a slender tube with a termination surrounded

by a ring of setae constitutes the material to which Doctor Walcott attached

this name.
Holotype —U. S. N. M., No. 83937.

WIWAXIA Walcott 1911
WIWAXIA CORRUGATA (Matthew)
Plate 3, fig. 1

Orthotheca corrugata Matthew, 1800, Trans. Roy. Soc. Canada, 2d ser., vol. 5,
sec, 4; -p. 42) ple 1, lige 3:

Orthotheca corrugata Walcott, 1908, Canadian Alpine Journ., vol. I, no. 2,
p. 246; plat figs ti:

Wiwaxia corrugata Walcott, 1911, Smithsonian Misc. Coll., vol. 57, no. 5,

p. 123, pl. 21, figs. 1-4.
A particularly fine example of this remarkable spined worm turned up in
some of the more recent collections. Its picture is included since it may repre-
sent a relatively undistorted specimen with most of the plates retained.

Plesiotype.—U. S. N. M., No. 83038.

OTTOIA Walcott 1911
OTTOIA MINOR Walcott
Plate 9, figs. 1-7

Ottoia minor Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 120, pl. 22,
figs. 5-6.
Original description—* This species differs from O. prolifica in
its proportionally more slender form when elongated and straighter
outline both when elongated and contracted. The hooks are also much

NO 3 BURGESS SHALE FOSSILS WALCOTT 7
finer and extend farther back on the anterior end. The annular lines
and interspaces are also finer and more irregular.”

Several specimens, among the many found subsequent to I91I, preserve some
of the structure features very well, and illustrations were prepared by Doctor

Walcott to show them. However, it is very doubtful whether the forms shown
on plate 9, figures 2 and 4, belong to this species.

Plesiotypes—U. S. N. M., Nos 83939a-g.

PIKAIA Walcott 1911
PIKAIA GRACILENS Walcott
Plate 8, figs. 4-5
Pikaia gracilens Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 132, pl. 20,
figs. 1-2.

Original description— Body elongate, slender, and tapering at
each end. It is formed of many segments that are defined by strong
annular shiny lines. Head small with two large eyes and two tentacles
as shown by figure 1. Back of the head the first five segments carry
short parapodia that appear to be divided into two parts.

“The enteric canal extends from end to end without change in
character. It is relatively large along the central portions and tapering
toward the ends. Judging from such specimens as the one illustrated
by figure 2, its annulations correspond in size with those of the body.

“Surface apparently smooth. Two entire adult specimens and
several fragments of others indicate a length of about 5 cm.”

Two additional figures are presented at this time.

Plesiotypes-—U. S. N. M., Nos. 83940a-b.

SELKIRKIA Walcott 1911
SELKIRKIA MAJOR (Walcott)
Plate 10

Orthotheca major Walcott, 1908, Canadian Alpine Journ., vol. 1, p. 246, pl. 1,
fig. II.

Selkirkia major Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 120, pl. 10,
fig. 6.

This species was first described from the Stephen formation on Mount Stephen.
Later Doctor Walcott found apparently the same shell at the Burgess Pass
quarry, but in this instance the soft body of the animal was preserved and there-
fore, in the 1911 discussion, he removed it from the Hyolithidae to the poly-
chaetous annelids.

Photographs of two exceedingly well preserved individuals with the body
extending beyond the shell are printed here for the first time.

Plesiotypes.—U. S. N. M., No. 83041a-b.

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

AYSHEAIA Walcott 1911
AYSHEAIA PEDUNCULATA Walcott
Plate 11

Aysheaia pedunculata Walcott, 1911, Smithsonian Misc. Coll., vol. 57, p. 117,
pl. 23, figs. 8-0.

Aysheaia pedunculata Hutchinson, 1930, Proc. U. S. Nat. Mus., vol. 78, art. 11,
p. 14.

This fossil has perhaps attracted wider attention than any other. Recently
G. E. Hutchinson, of Yale University, studied this peculiar form and concluded
that Aysheaia is an extinct Onychophora. Unfortunately, Mr. Hutchinson did not
see the two individuals here illustrated, as they were buried away among
numerous specimens of Ottoia. As the present assemblage of material progressed
they came to light and are now illustrated, especially as they are perhaps the best
preserved specimens available.

At the time these photographs were found among the notes Doctor Walcott
intended using in further publications relating to this animal, two letters were
discovered, both suggesting that Aysheaia may be an Onychophora or a Perip-
atus. The first letter, dated September 21, 1911, was written by Prof. W. M.
Wheeler of Harvard University, and reads as follows:

“T wish to thank you for your very interesting publications on the Middle
Cambrian Annelids. On plate 23, I noticed two figures of Aysheaia pedunculata.
This creature bears the most extraordinary resemblance to Peripatus, except for
the head, and judging from the figures the ‘head’ may be something which
does not belong to the fossil. I have just shown these figures to Mr. C. T. Brues,
who has been working on Peripatus, and he also was struck with the remarkable
resemblance. Is there any possibility that it might be Peripatus instead of an
Annelid? If this should prove to be the case it would be a matter of the very
greatest interest.”

The second letter dated October 25, 1911, was written by Prof. Charles
Schuchert, of Yale University, and contains the following :

“The other point is one that Lull has called my attention to and refers to
figures 8 and 9 of plate 23 which you call Aysheaia pedunculata. The question
that I want to ask is, have you considered it as a possible Onychophora or
related to Peripatus? Of course if one looks at your illustrations and compares
them with the illustration of Peripatus given by Parker and Haswell in their
Text-book of Zoology, page 607, in the edition of 1910, one can see considerable
differences and yet there are in your figures several points in common to
make one wonder whether you have not a marine ancestor of this land-living
arthropod.”

This rather lengthy historical account is presented to show that at least three
authorities arrived at the same conclusions independently.

Plesiotypes—U. S. N. M., Nos. 83942a-b.

LEANCHOILIA Walcott 1912
LEANCHOILIA SUPERLATA Walcott
Plate 12; plate 13, figs. 1-2; plate 14, figs. 4-5

Leanchoilia superlata Walcott, 1912. Smithsonian Misc. Coll., vol. 57, p. 170,
pl. 31, fig. 6.

Original description.— Body elongate, with clearly defined head
shield and nine strong body segments up to the point where the

NO. 3 BURGESS SHALE FOSSILS—WALCOTT 9

posterior part of the body is broken off. The anterior pointed end of
the head is broken off in such a manner that the presence of a frontal
appendage is suggested. The large opening on the side of the head
indicates a large pedunculated eye comparable with that of Opabinia
regalis (pl. 28, fig. 1).

“ Appendages.—Of the head appendages, the antennae are the best
preserved. These are large and composed of several strong joints,
of which three now show from beneath the carapace; the second of
these bears a long slender branch on its inner margin, and the third
two branches, one of which is similar to that of the second joint.
These two branches appear to be composed of one very long slender
joint followed at the end by several short small joints that curve
upward and presumably gave the branches flexible extremities ; the
third and lower branch has a similar slender proximal joint that at
its outer end has three slender, jointed branches. This structure makes
a very effective clasper of each of the antennae. Back of the right
antennae are two narrow appendages that may be the ends of one of
the third and fourth pairs of head appendages.

“The thoracic legs terminate in flat, elongate, broad, lanceolate
joints. The terminal joint is about three-fifths the entire length of
the leg, and has a fringe of strong setae on its outer and posterior
margin. The condition of preservation is such that the details of struc-
ture of the other portions of the leg cannot clearly be determined.”

The illustrations presented herewith apparently were prepared by Doctor
Walcott to exhibit the detailed structure of this interesting crustacean.

Plesiotypes—U. S. N. M., Nos. 83943a-g.

LEANCHOILIA MAJOR, new species

Plate 13, fig. 3

Several specimens, of which the best is illustrated, were labeled Leanchoilia
major by Walcott. Just why he should have chosen this specific name is not
readily apparent as these individuals are not sufficiently larger than L. superlata
to warrant the designation. In fact there is but little difference between this
form, which happens to be flattened out horizontally, and the specimen shown
in figure 2, plate 13, referred to the genotype.

Holotype-—U. S. N. M., No. 83044.

NARAOIA Walcott 1912
NARAOIA COMPACTA Walcott
Plate 13, fig. 4; plate 14, figs. 1-3; plate 15, figs. 2-3
Naraoia compacta Walcott, 1912, Smithsonian Misc. Coll., vol. 57, no. 6, p. 175,
ple2s8y hesasea.
Many specimens of this interesting form have been found since its
preliminary description in 1912, but none shows the cephalic ap-

IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

pendages in position or an uninjured trunk limb. The carapace is
thicker than that of Burgessia bella and the body is firmly attached to
the fused segments forming the posterior dorsal shield, and there is a
close union between the body of the cephalic region and the carapace

st
\
\

-
-

ewonee

\
H

|
a
\\ \
i
Alp
aay
i
4} be

Fic. 1.—Naraoia compacta Walcott.

a, antennae; as, anal segment; ds, reflex margin of posterior carapace; en,
endopodite; ex, exopodite; 7, intestine; m, mandible; m’, maxilla; mm”, maxillula;
seg, segmented posterior carapace; st, stomach; tf, telson.

(About 5.) Diagrammatic outline of ventral view of appendages, etc.

that extends from the anterior ventral margin of the latter back to |
the line of the anterior margin of the posterior dorsal shield. In the
restoration (text figs. 1, 2) I have made an attempt to incorporate
all information available.

NO. 3 BURGESS SHALE FOSSILS—WALCOTT Tet

Exoskeleton.—The dorsal exoskeleton as seen from its dorsal side
is formed of a carapace and a posterior segmented shield. The true
cephalic carapace or shell fold is attached to the cephalic somites near
its anterior portion, probably as in the recent Apodidae or the asso-
ciated Burgessia and also along the line of the body as far back as
the anterior margin of the posterior dorsal carapace. The carapace 1s
not known to have had a reflected anterior margin with a labrum at-

hor eee |e —T=S-~.__ \\ WYS=-- 4. h
WaT CRS i ! :

tl, pa ae SEL.

Fic. 2.—Naraoia compacta Walcott.

c, Carapace; g, digestive glands; d.g., lateral digestive glands; c, eye; h.c.,
hepatic caeca; st, stomach; ¢./., thoracic limb. ;
(About * 5.) Diagrammatic outline of the digestive organs.

tached as in Burgessia, but it may have had, as none of the specimens
shows the ventral side of this part in an uninjured condition. The
carapace is broader than long, with a somewhat uniformly rounded
outline except posteriorly where it is nearly transverse; it probably
had a ventral as well as dorsal membrane between which the great
hepatic caeca were located very much as are the shell glands in the
Apodidae.”. For some unknown reason the anterior portion of the
carapace is usually distorted by being crowded back so as to wrinkle
and shorten it.

The large shield of the posterior part of the exoskeleton is com-
posed of 14 fused segments with a narrow border. It has the ap-
pearance of the many-segmented pygidium of the trilobite belonging

"In no other manner can I explain the wonderful preservation of the digestive
tubes and caeca.

I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to the genera Ogygopsis or Orria,’ except that the median lobe is not
as strongly marked. These two genera are mentioned as they occur in
correlative Middle Cambrian formations and one of them in the
Stephen formation which is nearly contemporaneous with the Burgess
shale. The dorsal carapace and shield also appear similar in outline
to the dorsal exoskeleton of the freshly hatched young of Limulus
polyphemus as figured by Packard.* The posterior shield has each of
the thoracic body segments, excepting the two posterior which pro-
ject beyond it, attached directly to it, and it forms the dorsal side
of the exoskeleton of the body segments. The exoskeleton of the free
posterior segments and telson, of the cephalic segments, and of the
ventral side of the trunk segments was so exceedingly delicate as to
leave only a slight trace on the shale. Both the anterior carapace and
posterior segmented shield were very thin and readily distorted.

The anterior carapace slightly overlapped the posterior shield and
the two were closely held together by the strong body as evidenced
by their rarely being found separated. The telson is short and marked
by fine short spines.

Dimensions. —The largest specimen had a length of about 40 mm.,
the carapace being shortened by compression. A specimen that has
been slightly narrowed by compression has the following dimensions :

Menoth of carapace ccs. sere a sso cise ears pale ates 16
Wadth oficarapacesiccasccme atene cities nite ae ile iol eters 15
Length of posterior dorsal: shieldai..: ssc. a0 someteeien lens 17
Width-of posterior dofsalishteldis: «sccm... soe ocrdeeie sen 14

Eyes.—The eyes are represented by two small crescent-shaped
bright spots a little in advance of the anterior end of the stomach as
illustrated by the restoration (fig. 2). The position and form corre-
spond quite closely to the paired eyes of the recent Apus lucasanus
Packard.’

Cephalic appendages.—The antennae are uniramous, short jointed,
and slender in their distal portion, and have a large proximal joint ;
the intermediate joints are unknown. Of the cephalic limbs only slight
indications were found of the proximal joints of three pairs, and a
few terminal joints extending from beneath the carapace, nothing
of their original form being preserved. All traces of cephalic ap-

* Smithsonian Misc. Coll., vol. 64, no. 5, figs. 1 and 2, pl. 66, 1916.

* Mem. Boston Soc. Nat. Hist., vol. 2, pl. 5, figs. 25, 25d, 1871.

* Twelfth Ann. Rep., U. S. Geol. and Geog. Surv. Territories, Hayden, Pt. 1,
pl. xvi, fig. 2, 1883.

NO. 3 BURGESS SHALE FOSSILS—WALCOTT 13

pendages are posterior to the hepato-pancreatic tubes passing from
the stomach to the hepatic caeca.

There is no clearly defined line between the cephalic and trunk
limbs, but from the relations of the limbs in Burgessia and Marrella
it is assumed that it is between the third pair of cephalic limbs and
the supposed first pair of trunk limbs. The specimens are too much
obscured by the compression they have undergone to permit of recog-
nition of the detailed structure of the limbs.

Thoracic limbs.—The specimen represented by figure 3, plate 14,
has the distal ends of 17 thoracic or trunk limbs projecting beyond the
left margin of the posterior shield; the shield in this specimen has
not more than 14 fused segments outlined on it, so it is probable that
the three anterior limbs belong with the body segments between the
anterior segment of the posterior shield and the third pair of cephalic
limbs. Another alternative is that the distal portion of the two anterior
limbs extending beyond the margin of the shield belong to the maxilla
and maxillula, which would leave only one pair of limbs from the seg-
ment anterior to the posterior shield and posterior to the cephalic
limbs. The limbs were so subject to displacement, however, that any
deduction is very uncertain. The distal portion of the thoracic or
trunk limbs shows an endopodite with a slightly curved terminal spine
with a slender section back of it corresponding to the slender distal
joint of the endopodite of Marrella and Burgessia; and then the joints
broaden towards the coxopodite with slight indications of five joints
between the distal joint and coxopodite.

The exopodite is represented by many slender filaments that were
attached to a multi-jointed arm or support similar in appearance to
that of the exopodite of Marrella. The filaments are relatively broad,
as they occur flattened on the shale. There are strong indications of
large coxopodites, but none show their original form or the exact
point of attachment of endopodite or exopodite, and the joints of
the endopodites have been so crushed down as to be no longer definitely
recognizable. The exopodites were nearly as long as the endopodites,
and the filaments of the former are usually extended out to the end
of the endopodites or beyond.

Digestive organs—The exact location of the mouth is unknown,
but from the apparent position of the antennae and proximal joints of
the cephalic limbs, it was posterior to the point of entrance of the
hepatic tubes, back of which the intestine was large with minor hepatic
caeca opening into it through four small tubes, all of which are anterior
to the posterior dorsal shield as indicated in the diagrammatic res-
toration (fig. 2) ; beneath the posterior dorsal shield the intestine is

14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

slightly constricted at the union of the trunk segments and extends
back to the anal segment which protrudes posteriorly from beneath
the shield ; the gullet connecting the mouth and the stomach must have
extended forward and upward.

The large hepatic glands and caeca are somewhat similar to those
of Burgessia bella (text fig. 5); a short, strong tube with a well-
marked anterior and posterior tube leads out from the stomach, and
branching from these lateral tubes are a series of hepatic caeca. The
small hepatic caeca are located between the long posterior tubes of the
hepatic glands and the intestine and posterior to the main hepatic tube.

Owing to the excellent state of preservation of many of the speci-
mens showing the hepatic caeca it is probable that they were situated
between the dorsal and ventral membrane of the carapace and thus
held in position and protected from destruction; that they are pre-
served at all is one of the wonders of this remarkable Burgess shale
fauna.

Functions of appendages——These were presumably the same as
for similar organs in Marrella and Burgessia, and the mode of occur-
rence is essentially the same.

Diagrammatic restorations —The diagrammatic restoration (fig. 2)
presents the outline of the carapace and posterior shield with the
stomach, intestine, hepatic tubes and caeca of the digestive system
outlined, also the thoracic trunk, the telson and the thoracic limbs
as far as known and interpreted. The data for this diagram are very
good except the jointing of the endopodites and the exact form of the
coxopodites and proximal joints of the exopodites. In figure 1, the
endopodites on each side have been omitted and the exopodites drawn
in so as to show their structure and position above the endopodites
and below the ventral membrane of the posterior dorsal shield. The
sixth limb has both the endopodite and exopodite attached ; this should
be compared with the thoracic limb of Marrella splendens (text fig. 9).

We know so little of the cephalic limbs of Naraoia compacta that I
hesitate to give a diagrammatic sketch of them, and it would not be of
even tentative value if we did not have the cephalic limbs of Marrella
and Burgessia for suggestion; from the latter and from the evidence
afforded by a few specimens, the outline of figure I is drawn.

Comparison with crustaceans—WNaraoia has many characters in
common with the trilobite and some in common with Marrella, Bur-
gessia, and Waptia, which will be spoken of in the discussion of this
group of genera.

Plesiotypes—U. S. N. M., Nos. 83945a-e.

NOU 3 BURGESS SHALE FOSSILS—-WALCOTT 15

NARAOIA SPINIFER, new species
Plate 15, fig. 1

Three specimens referred to this species are known, of which the
one figured shows best the spines on the margin of the posterior
dorsal shield; another preserves both the carapace and dorsal shield,
the latter having marginal spines while the carapace has a smooth
margin. On the third specimen the test of the dorsal shield is nearly
all exfoliated on the left side so as to expose the body, several of the
fringed exopodites, and coming from beneath them the distal portions
of the endopodites.

This species differs from Naraoia compacta in having eight short,
small spines on the outer margin of the dorsal shield equally spaced
between the anterior margin and a large posterior median spine ; all
three of the known specimens of the posterior dorsal shield are also
larger than those of N. compacta, as they average 25 mm. in length
exclusive of the posterior median spine. The one specimen preserving
the carapace indicates that it was similar to the carapace of N. com-
pacta, the recognized differences between the two species being con-
fined to the posterior dorsal shield.

Holotype —U. S. N. M., No. 83946.,

BURGESSIA Walcott 1912
BURGESSIA BELLA Walcott
Plate 15, figs. 4-7; plate 16; plate 17; plate 18, fig. 1

Burgessia bella Walcott, 1912, Smithsonian Misc. Coll., vol. 57, no. 6, p. 177,
pl. 27, figs. 1-3; pl. 30, figs. 3-4.

Since the publication of the original description of Burgessia bella
Walcott, a large number of more or less well preserved specimens
have been collected from the Burgess shale, a few of which preserve
details of structure that make it possible to draw a diagrammatic resto-
ration indicating the increase in our information of the cephalic and
thoracic appendages (text figs. 3, 4).

Exoskeleton—The exoskeleton is very delicate and the carapace
is so thin as to be almost membranaceous. The segment to which the
telson is attached appears to be partly covered ventrally by a heart-
shaped plate that is attached to the anterior margin of the segment
or to the posterior limb-bearing segment of the thorax; it suggests a
supra-anal plate.

There appear to be five cephalic, eight thoracic, and one abdominal
segment, also a long, slender telson with numerous joints. One ex-
ample 21 mm. in length has 30 joints.

2

16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

o
{\

1

if

|
a
a
a
|
| -}-
|
i
i
i
tl
I
I
i
i
U
I

H

|

i

Fic. 3.—Burgessia bella Walcott.

a, antennae; ap, anal plate; as, anal somite; c, carapace; d, doublure; en,
endopodite; ex, exopodite; /b, labrum; m, mandibles; m’, maxillae; m”, maxil-
lulae; ¢, telson.

(About 7.) Diagrammatic outline of ventral side with appendages. Much
of the data on which this figure is based are shown by the specimens illustrated
on plates 16 and 17. The exact form and position of the proximal joints of the
cephalic limbs is unknown, but their general outline and position are about as
outlined. The thoracic limbs, especially the endopodites, are well preserved in
several specimens, but the exopodites are rarely seen and then only as faint and
delicate impressions on the shale. The protopodites are fairly well defined, also
the trunk segments and telson. The posterior endopodite and exopodite both
differ from those anterior to them.

NOS BURGESS SHALE FOSSILS—WALCOTT 17

Carapace.—Carapace semicircular, with a deep notch on the poste-
rior side. It appears to have had an upper (dorsal) and lower (ven-
tral) membrane between which the irregular ramifications of the
hepatic caeca were located. The shell glands, so conspicuous in the
Apodidae, have not been recognized in Burgessia.

Labrum.—tThe labrum is attached to the reflected anterior rim
(doublure) of the ventral side of the carapace and extends back
nearly one-third its length ; the labrum is rounded posteriorly and has
a shallow obliquely transverse furrow on each side just in advance of
its posterior margin; it was thin, readily distorted by pressure and is
rarely preserved; one of the best exantples is illustrated by figure 3,
plate 17 ; it appears to have covered the anterior portion of the mouth.

Eyes.—The eyes are indicated by a minute round spot on each side
of the dorsal median axis of the carapace and a short distance within
the anterior margin.

Fic. 4.—Diagrammatic outline of a thoracic limb of Burgessia.

pr, protopodite ; en, endopodite ; ex, exopodite. ;
The outline is based on the examination of many specimens aided by the known
form of the endopodite of Marrella which is somewhat similar.

Dimensions.—The average length of the larger specimens is about
1o mm. A few are 12 and many are 6 to 8 mm. in length exclusive
of the long telson. The relative proportions of the carapace, thorax,
and abdomen are indicated by the diagrammatic restoration (text
fig. 3).

Cephalic appendages.—These consist of well-marked antennae (figs.
3, 4) and three pairs of limbs situated between the antennae and the
hepato-pancreas tubes; there is evidence that the basal or proximal
joints of the cephalic limbs are relatively large and the remaining
joints slender, but their exact position in relation to the labrum and
their details of form and structure are not determined. It is quite
probable that they represent the mandibles, maxillulae, and maxillae
very much as in Marrella, and I have so represented them in the
diagrammatic restoration of the ventral view of the species (text
fig. 3).

Thoracic limbs.—The ten pairs of biramous thoracic limbs are uni-
form in character with the exception of the posterior pair, which are

18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

relatively smaller and more slender. Each limb has a strong proximal
joint (coxopodite) to which the endopodite is attached; the latter is
formed of four stout joints and two slender joints with two or three
short spines at the end of the distal joints ; the four joints between the
proximal and slender sixth joint may have a flattened extension on
the ventral side as in the endopodite of Marrella that gives them a
greater transverse diameter, and this may also occur in the sixth joint ;
the distal joint is slender and probably cylindrical; the exopodite has
not been seen attached to the protopodite, but from a number of speci-
mens showing their position there is little doubt of their having been
attached as on the thoracic limb of Marrella. The exopodite is an elon-
gate oval, apparently unjointed lobe as seen in the specimen repre-
sented by figure 4, plate 17; a fringe of fine, short filaments occurs
on the ventral and outer margins; the delicate structure and small
size makes it difficult to determine its exact nature, but as far as known
it recalls the exopodite of Neolenus. One specimen indicates that there
may have been an anterior support for the exopodite that extended
beyond the flat filamentous lobe and terminated in two minute spines ;
the proximal portion of the endopodites has been flaked off in this
specimen so as to expose the exopodites ; the slender distal extensions
may belong to the exopodites or they may be the ends of the endopo-
dites of the opposite side flexed under. I am inclined, however, to
think that they belong with the exopodites. What may be a modified
exopodite has been seen in one specimen; it projects from beside the
posterior thoracic endopodite and consists of a central axis with seven
sharp spines projecting from its posterior side and a terminal spine ;
or it may be an endopodite showing the edges of plate-like joints in the
same manner as those of Marrella splendens (pl. 22, figs. 6, 7).
Digestive organs ——The mouth was situated at the ventral side and
probably bounded in front by the labrum and on the sides by the
mandibles ; the mouth presumably opened into a gullet that passed
into a large stomach apparently divided or forked anteriorly ; from
the rear of the stomach a straight intestine extended back to the anus.
A strong, relatively large tube is given off from each side of the
stomach at about the fifth segment; these have strong branches at
the proximal end, one extending forward and another backward, both
of which have short bifurcating branches on both the outer and inner
sides. In nearly all well-preserved specimens the large tube and often
the large connecting tubes are rounded as though they were distended
when buried in the sediment; this would accord with the view that
these were large digestive glands that contained food in process of

NOT 3 BURGESS SHALE FOSSILS—-WALCOTT 1g

digestion, the ultimate or hepatic caeca secreting a digestive juice as
in Lepidurus and other crustaceans having such glands.’

oat

/\

/

ANN AS

iW
al sf
a7
)
f

Fic. 5.—Burgessia bella Walcott.

ap, anal plate; as, anal somite; cs, central stomach; d.g., digestive glands;
d.g'., lateral digestive glands; e, eye; h.c., hepatic caeca; 1, intestine; s’ and s”,
anterior lobes of stomach; sc, anterior central lobe of stomach; f¢, telson.

(About 7.) Diagrammatic outline of digestive organs. Most of the data
on which this figure is based are shown by the specimens illustrated on plate 16,
figures I, 3 and 4. The exact relations of the anterior central lobe of the stomach
to the central stomach are unknown as there appears to be a line separating them.

The anus is supposed to have been at the last thoracic
beneath the platelike structure shown on figure 3, plate 17.

segment

Parker and Haswell, Text-book of Zool., vol. 1, p. 491, 1897.

20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Functions of appendages.—The functions of the cephalic and
thoracic limbs were probably similar to the functions of those of
Marrella splendens.

Mode of occurrence.—This delicate and beautiful little crustacean
occurs abundantly in association with Marrella splendens and Waptia
fieldensis, and is quickly recognized, even when distorted and crushed,
by its carapace and strongly marked hepatic caeca. The carapace is
almost always outlined on the shale, as are frequently the body and
telson; the large strong endopodites are usually more or less clearly
marked, although their jointed structure has generally been lost in the
flattening in the shale; the exopodites were so delicate that they are
rarely preserved, and the same is true of the labrum and eyes.

Comparison with crustaceans —Burgessia has certain characters in
common with Marrella and Naraoia and belongs in a group with them
which will be discussed later (p. 37).

Diagrammatic restorations of ventral surface ——I have endeavored
to present, in text figures 3, 4, and 5, interpretations of the structure
of my Burgessia bella.

Plesiotypes—U. S. N. M., Nos. 83947a-0.

WAPTIA Walcott 1912
WAPTIA FIELDENSIS Walcott

Plate 18, figs. 2-5; plate 19; plate 20; plate 21, fig. 2
Waptia fieldensis Walcott, 1912, Smithsonian Misc. Coll., vol. 57, no. 6, 1912,
p. 181, pl. 27, figs. 4, 5.

The general characters of this species were described in 1912, since
when a large number of specimens have been collected from the Bur-
gess shale, a few affording data from which a fairly accurate dia-
grammatic restoration of the animal may be drawn (text fig. 6).

Exoskeleton—The exoskeleton of the carapace, trunk, and caudal
furca was very thin and readily distorted. The trunk consists of 5 to 7
short cephalic segments ; 8 narrow thoracic segments, each bearing a
pair of uniramous appendages; 6 long abdominal segments and 2
broad lobelike terminal caudal furca or rami; the latter have three
transverse lines indicating four fused segments. The posterior margin
of the abdominal segments bears four or more strong spines with a
fringe of small, short, sharp spines between them. The last or anal
segment has a minute anal opening on a slightly rounded elevation
near its posterior ventral margin. The abdominal segments have often
been narrowed and lengthened, or broadened and shortened by dis-
tortion in the shale.

NO. 3 BURGESS SHALE FOSSILS—-WALCOTT 21

Carapace.—The carapace when viewed from its dorsal side is elon-
gate, narrowed anteriorly, expanded posteriorly, and has the outline
of two broad lobes by the incurving of the rounded posterior margin

oS
aS
sa

: rE

BL eiae on

Fic. 6.—Waptia fieldensis Walcott.

a, antennae; a’, antennules; a./., abdominal limbs; a.o., anal opening; ¢,
carapace; c.f., caudal furca; c./., cephalic limbs; e, eye; en, endopodite (??) ;
ex, exopodite; h.c., hepatic caeca; i, intestine; 7. p., rostral plate; sh. gl., shell
gland; st, stomach; th./., thoracic leg.

(About X 3.5.) Diagrammatic outline of ventral view of appendages and
digestive organs. Most of the data on which this figure is based are shown by
specimens illustrated on plates 19 and 20.

towards the median line; when folded over on its longitudinal axis
each side is a long semi-oval with the narrow end in front; there is

no evidence of the presence of a longitudinal median line or hinge
in the many specimens collected from the Burgess shale.

22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Rostral plate-——A small triangular rostral plate with narrow, sharp,
longitudinal median ridge has been seen in four specimens (see fig. 3,
pl. 19; fig. 2, pl. 20); it is located in the median line between the
antennae,

Dimensions.—The average length of an entire adult specimen is
from 40 to 50 mmn., the carapace being about 16 mm. in length with
a width when flattened of 15 mm. The general proportions of the
various parts are shown by figure 4, plate 27, of the 1912 paper.

Eyes.—tThe eyes are relatively large and placed at the end of a
stalk or peduncle that projects from beneath and on each side of the
rostral plate as seen from above. The peduncles are slender at their
proximal end and expanded in a broad oval outline on the distal third
of their length, the expanded section carrying the elongate oval visual
surface ; the peduncle appears to have had at least one joint at about
the inner third of its length and to have been attached to a prostomium
at its proximal end.

Cephalic appendages——These consist of a pair of long jointed
antennae that project forward beside and beneath the median rostral
plate, and a pair of short lobelike antennules appear to be represented
close to the eye and above the antennae (see fig. 3, pl. 19; fig. 2, pl. 20)
in several specimens.. Traces of three pairs of cephalic limbs have
been observed but their structure and form are unknown.

Thoracic limbs ——A number of specimens have five strong thoracic
limbs that extend from their union with the body trunk forward and
outward beyond the edge of the carapace (see fig. 2, pl. 18) ; the distal
joint is short and has three strong and two small curved spines project-
ing from its outer end and fine spines along its margin; the three next
joints are rather short and spiniferous, but the detailed character of
the remaining joints is unknown. The limb observed is assumed to be
the endopodite of a biramous limb, but the exopodite was not de-
veloped or it was so small and delicate as not to be preserved in the
fossil state.

Abdominal limbs —Each of the abdominal limbs is represented by
long, multi-jointed exopodites bearing long, slender filaments (see
fig. 3, pl. 20). The proximal joint was probably short and without
fringing filaments, but none of the specimens proves this to have been
the case ; the exopodites are rather large at the proximal end, tapering
gradually to a slender, flexible terminal section ; the filaments of the
terminal section are sometimes gathered in tufts or bundles as shown
by figure 3, plate 20. The filaments are usually flattened and matted
together, but a few specimens show them to have been slender,
cylindrical tubes similar to the filaments on the exopodite of Marrella

=

NO. 3 BURGESS SHALE FOSSILS—WALCOTT 23

Splendens (see restorations). The presence of a rudimentary endop-
odite is suggested on some specimens by an elongate, triangular,
light-colored space on the proximal portion of the exopodites as shown
by figure 3, plate 20; these light areas may be the outline of a space
inside the broad arm of the exopodite, but they usually cross the axis
of the arm diagonally ; if they do represent the endopodite they were
exceedingly delicate and attached by a broad base beside the exopodite
in such a manner as to be held almost rigidly in place, and they are
always in the fossil state pressed against the proximal section of the
exopodite, and they have a silvery sheen so characteristic of the con-
tents of the inside of the limbs of all crustaceans of the Burgess shale
preserving the limbs. I do not think that they represent the endop-
odites, but they are the only suggestion of the latter thus far observed
in connection with the abdominal limbs of Waptia.

rp--

ie}
i
\\)
|

A

\\
i

\
\\\
LL

i

<I +
ee
\ !
[Ue

“<i

So

I
Is

s

Fic. 7.—IWaptia fieldensis Walcott.

a, antennae; a.o., anal opening; c, carapace; c.f., caudal furca; e, eye; ex,
exopodite; /i.c., hepatic caeca; 1, intestine; r./., rostral plate; sf, stomach;
th. 1., thoracic leg.

(X 3.) Diagrammatic side view of a section of the animal, illustrating the
appendages, digestive tract, etc.

Functions of appendages.—The functions of the antennules and
antennae were presumably sensory as in recent Malacostracans, as
they do not appear to have been modified for any other purpose, and
the proximal joint, as far as known, did not function as a manducatory
organ.

The mandibles, maxillulae, and maxillae are unknown; the five
pairs of thoracic limbs may have been used for crawling on the bottom,
but with short joints and spinous distal joint they could not have been
very effective; the exopodites of the eight pairs of abdominal limbs
served as natatory organs and also as branchiae, the long delicate fila-
ments presenting an extended surface area to the water.

Digestive organs—What is known of the digestive system of
Waptia indicates that it was somewhat similar to that of the living

24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Apus. The elongate globular stomach (st), with the small digestive
or hepatic glands (d. gl.), long simple intestine (7) terminating on the
last segment at the anus (am) all suggest corresponding organs in
A pus, and it is highly probable that the mouth was ventral and com-
municated with the stomach by a gullet extending upward and for-
ward. The shell gland (sh. gl.) or renal organ is distinctly marked
in several specimens and, as far as comparison is possible, is not
unlike that of Apus.

Observations —Waptia was a pelagic, free and active swimming
animal using its abdominal limbs and the broad terminal rami for
propulsion. The fact that it is found in association with algae and
sponges is explained by the conclusion that the sessile forms of life
were detached and drifted into the Burgess pool and deposited along
with the pelagic forms that dropped to the bottom of the sea.’

The carapace of Waptia is much like that of Hymenocaris except
that it is not separated into two equal parts by a median longitudinal
hinge line, and there is no evidence of the presence of an adductor
muscle scar on each side as in Hymenocarts.

DIAGRAMMATIC RESTORATION OF VENTRAL VIEW OF THE
BODY AND APPENDAGES, WITH OUTLINE OF
DIGESTIVE ORGANS

Most of the data on which the restoration is based is shown by
the specimens illustrated on plates 18, 19 and 20. The form and
position of the cephalic appendages are unknown with the exception
of the antennae and possibly antennules; the proximal joints of the
thoracic and abdominal limbs are outlined on the specimens though
their form is not preserved, but otherwise the limbs are fairly well
known. The body cavity is outlined by figure 3, plate 18, but it is
not included in this diagrammatic figure. The shell glands (sh. gl.),
stomach, intestine, and hepatic caeca are outlined, as they represent
what is known of the digestive organs.

Plesiotypes—U. S. N. M., Nos. 83948a-e.

WAPTIA CIRCULARIS, new species

Plate 21, hg. 3

A single specimen with a short, rounded carapace was labeled W. circularis by
Doctor Walcott. As far as the rather poor preservation permits a determination
it would seem that otherwise it is similar to WW’. fieldensis.

Holotype —vU. S. N. M., No. 83449.

* Smithsonian Misc. Coll., vol. 67, no. 5, pp. 219, 220, 1919. Idem, no. 6, p. 265,
under Habitat.

NO. 3 BURGESS SHALE FOSSILS—-WALCOTT 2

on

SKANIA, new genus

Description.—Dorsal shield thin, broadly rounded in front and
tapering from the postero-lateral angles of the cephalic carapace to
the posterior end of the shield.

Cephalic carapace transverse with the postero-lateral angles ex-
tended into spines ; posterior margin arched forward; frontal margin
reflected to form a doublure to which a small elongate labrum is
attached. Eyes unknown but indicated by a bright spot on the carapace
a short distance outward from the side of the labrum. No traces of
facial sutures.

Posterior dorsal shield, elongate and formed of 14 or 15 fused
segments with a more or less distinctly marked border. There is a
short transverse segment or telson (pygidium) outlined, but whether
it is free from the next anterior segment is unknown.

Surface of test apparently smooth.

Dimensions. —-This genus is based on a small animal, S. fragilis,
5 to 17 mm. in length.

Appendages.—There are indications of antennae, three pairs of
cephalic limbs, and a pair of limbs for each segment of the posterior
dorsal shield.

Digestive organs.—An intestine extends from the posterior segment
forward to the central part of the cephalic carapace where it widens
out to form an elongate oval stomach. There are traces of hepatic
caeca adjoining the stomach.

Genotype—Skania fragilis Walcott.

Stratigraphic range-—The stratigraphic range is limited to a band
of dark siliceous shale about 4 feet in thickness forming a part of the
Burgess shale member of the Stephen formation.

Observations—The generic name is derived from Skana, the name
of a glacier in the Mount Robson District, Alberta, Canada.

The specimens representing the species of this genus are small and
so thoroughly flattened in the shale that little more than a black film
remains. This makes it very difficult to obtain details and also leaves
some doubt as to whether the posterior dorsal shield is formed of
fused or free segments.’

SKANIA FRAGILIS, new species
Plate 21, fig. 1

Description —General outline irregularly heart-shaped but subject
to wide variation owing to distortion by compression.

*At the head of these notes, Doctor Walcott later wrote, “A_ trilobite,
CDs W.”

26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Dorsal shield Owing to its extreme tenuity there is usually little
more than a dark film on the shale that has definite outlines, and shin-
ing through it are traces of the digestive organs and the ventral limbs.
The transverse cephalic carapace recalls that of Marrella without the
great median spines; it is often incurved at the center of its anterior
margin and laterally projects into long backward-curving, spine-like
extensions that are so tenuous as to suggest that the cephalic carapace
was formed of a delicate membrane.

Fic. 8.— Skania fragilis Walcott.

_ (X7.) Diagrammatic outline of ventral side showing the various parts as
interpreted from several specimens. No details of the segments of the posterior
dorsal shield are preserved, but the segments are clearly outlined. The intestine
is quite definite, also the fact that it contracted at each segment and expanded
into a stomach beneath the cephalic carapace. Only the proximal portion of
the limbs is outlined, although fragments of the distal portion are preserved on
one specimen.

The thoracic portion of the dorsal shield is clearly segmented in two
specimens, and traces of segmentation occur in others, but usually
there is only a black smear on the shale with the outline of the intestine
showing through it. There appear to be 14 or 15 fused segments and
possibly a minute terminal segment or telson. None of the 29 speci-
mens in the collection show the outlines of the median lobe, although
one has a slight elevation along the line of the intestine.

Labrum.—Traces of a narrow doublure and small labrum have been
seen on two specimens; the labrum appears to have been elongate with

an outline similar to that of the labrum of Burgessia.

NO. 3 BURGESS SHALE FOSSILS—WALCOTT 27

Dimensions.—The largest specimen has a length of 17 mm., but
the average length is from 5 to 8 mm.

Appendages—One specimen has several thoracic endopodites out
of place on one side, and other specimens show the proximal joint
obscurely but sufficiently well to recognize them; another specimen
has what may have been slender antennae projecting from beneath
the flattened labrum and posterior to it three pairs of slender ap-
pendages in which all traces of joints have disappeared ; there are also
on this specimen several threadlike, silvery lines extending from the
central axis out to a margin which indicates that the limbs were long
and slender ; none of the specimens clearly show the exopodites or any
details of the limbs. Specimens of Marrella and Burgessia often
have threadlike, silvery lines representing the limbs, these lines being
the pyritized contents of the joints, the test having disappeared in the
process of mineralization of the original specimen.

Digestive organs.—The stomach is represented by an enlargement
of the anterior portion of the intestine within the cephalic carapace,
and the intestine extends back to the last segment; traces of hepatic
caeca also occur beneath the cephalic carapace adjacent to the stomach.

Observations —This very delicate form was placed, when sorting
the collections, among specimens of the young of Marrella splendens,
but it became evident upon close examination that they were quite
distinct. They have a dorsal shield resembling that of Naraoia. I
have examined all the specimens in hopes of finding free segments,
but without results. There is no well-defined border about the pos-
terior dorsal shield as in Naraoia, but there is a definite margin that
is unbroken by the extension of the fused segments beyond it.

The almost complete flattening of all the specimens prevents any
comparison with the median lobe of the trilobites, and there is no
indication of facial sutures although there are slight traces of eyes
on the cephalic shield at about the same place as in Nathorstia.’

Holotype —U. S. N. M., No. 83950.

MOLLISONIA Walcott 1912
MOLLISONIA ? RARA Walcott
Plate 21, fig. 4
Mollisonia ? rara Walcott, 1912, Smithsonian Misc. Coll., vol. 57, no. 6, p. 198,
pl. 24, figs. 6, 7.
Original description.—* Of this species there are several fragmen-
tary specimens. The species differs from M. gracilis, with which it is

* Smithsonian Misc. Coll., vol. 57, no. 6, pl. 28, fig. 2, 1912.

28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

associated, in the character of the thoracic segments and pygidium ;
also, so far as we can determine from this superficial study, there are
seven segments and the pygidium shows distinct segmentation with
a denticulated border.”

A complete individual was found after the two fragments were described in
1912. The angularity of the shield at the bottom of the specimen as mounted on

the plate is characteristic as its essential angles and curves are repeated in all
the specimens referred to the species.

Plesiotype—U. S. N. M., No. 83951.

MARRELLA SPLENDENS Walcott
Plate 22, figs. 1-9

Marrella splendens Walcott, 1912, Smithsonian Misc. Coll., vol. 57, no. 6,
p. 193, pl. 25, figs. 1-6, pl. 26, figs. 1-6.

In the preliminary note of 1912 the general form and appearance
of the carapace and appendages of Marrella splendens were described
and illustrated. Since then a large number of specimens have been
collected, some of which have added to our information both of the
carapace and ventral side.

Exoskeleton—The exoskeleton with the exception of the cara-
pace is very delicate and formed of a series of 31 segments or somites,
to 24 of which a pair of biramous appendages are attached; also a
terminal segment of the body forming a minute plate-like telson and
five segments of the head indicated by the presence of four pairs of
free appendages and one segment incorporated in the body of the
carapace ; this is indicated by the anterior lateral free spines of the
carapace with a pair of sessile eyes. As far as may be determined from
the compressed fossil specimens the section of the body segments was
broadly oval with a dorsal stergite and a ventral sternite section, the
appendages being attached on the lower side on the ventral sternite
below the margins of the dorsal stergite.

This does not mean that the eyes necessarily represent the anterior
segment but that they represent one segment whatever may have been
its original position.

Carapace.—Carapace strong, subquadrangular, and with two large
dorsal postero-lateral, spinelike lobes (fig. 9) comparable with the
postero-lateral lobes of the carapace of the Apodidae. At each antero-
lateral angle a strong, backward-curving spine is attached by a close
suture. These spines complement the great dorsal thoracic spines and
may be compared with the movable or free cheeks of the trilobite. A
narrow median carina or ridge extends the entire length of the lateral
spines and the postero-lateral lobes.

NO 3 BURGESS SHALE FOSSILS—WALCOTT 29

Labrum.—The labrum is attached to the strong frontal rim
(doublure) of the ventral side of the carapace (fig. 9), and extends
back to within a short distance below the posterior median margin of
the carapace ; its posterior lateral angles are extended into short, spine-
like projections and the posterior margin appears to have been pro-
vided with two short points. The labrum appears to have covered the

Fic. 9.—Marrella splendens Walcott.

Restoration as described in text: a, antennae; c.sp., carapace spine; d, doub-
lure or reflex of anterior margin of carapace; e, eye; en, endopodite; er,
exopodite ; m, mandible; mm’, maxillula; m”, maxilla; ¢, telson.

anterior portion of the mouth ; this is indicated by the proximal end of
the protopodite or basal joint of the large mandible and that of the
antennae passing beneath it in many specimens (see pl. 22).
Eyes—The eyes (fig. 9) have not been seen from the dorsal side
of any of the several hundred specimens preserving the carapace, and
rarely from the ventral side; this leads to the conclusions (a) that
they were located on the lower anterior margin in such a position
as to be concealed when the carapace was flattened by compression ;

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

(b) that they were of a very delicate structure and readily destroyed ;
(c) that when preserved they were likely to be distorted and displaced
by compression in the shale and could only be seen from the ventral
side of the flattened carapace when they projected beyond the margin
and were outlined on the shale. At first I thought that the eyes were
situated on the carapace just within the line of its union with the
large antero-lateral spines. Later I re-examined all the specimens
showing the eyes, and found two that indicated that the visual surface
of the eye was on one side of the suture separating the spines from
the carapace, and the cap or palpebral lobe on the other, and one that
quite clearly indicated that it was attached to the proximal end of the
great spines, the latter being equivalent to the free cheeks of the
trilobite. The interpretation of this is that the visual surface of the
eye was attached to the great spine outside of the suture that outlined
the spine from the carapace, and that the cap or palpebral lobe of the
elevated visual surface of the eye was attached to the carapace as
in the trilobites with elevated eyes and free cheeks.

It is difficult to determine the extent of the elevation of the eye
above the carapace, but from its inconspicuous position in the fossil
state I strongly suspect that it was only slightly raised and that its
field of vision was largely forward and downward; this would be in
accord with the needs of a small, active, free-swimming animal that
spent little time on the bottom.

Among the trilobites the eye of Deiphon forbesi Barrande’* is at the
proximal end of a large genal spine forming the free cheek, and the
great eye of Bohemilla stupenda Barrande’* occupies nearly the entire
width of the proximal end of the free cheek which is extended into a
long strong spine. |

Digestive organs.—The intestinal canal extends from the posterior
margin of the labrum back to the small, platelike termination of the
body ; it is contracted a little opposite the line of union of each of the
segments (see figs. 6, 7, and 9, pl. 22) ; anteriorly the intestine widens
out between the labrum and carapace to form what may have been
the stomach; the narrow canals of the dorsal lobes passed into the
space between the carapace and labrum and probably entered the en-
larged intestinal canal as did the canals of the antero-lateral spines
which appear to pass without interruptions through the close sutures
that unite them with the carapace; the canals of the postero-dorsal
lobes may represent the shell-glands or excretory organs of the recent
Apodidae.

* Syst. Silur. de Boheme, vol. 1, pl. 2B, fig. 4, 1852; suppl., pl. 2, fig. 19, 1872.

“Idem, vol. 1, Suppl., pl. 14, fig. 3o.

INKOG, 6) BURGESS SHALE FOSSILS—-WALCOTT al

Cephalic appendages.—These consist of antennae, mandibles, simple
slender maxillulae, and slender maxillae. The proximal joints of the
cephalic appendages are so badly crushed and matted together beneath
the labrum or just back of it that it has been very difficult to determine
exactly their form and relations to each other, but it is highly probable
that they were arranged as in the restoration (text fig. 9).

Antennae.—The antennae are long, slender, and many-jointed, with
fine spines at the distal end of each joint. As far as may be determined,
the proximal joint was attached to the ventral surface beneath the
postero-lateral angle of the labrum just in advance of the mandible.
There is no evidence that it served as a jaw or manducatory organ
except that in specimens preserving them their inner (proximal) end
is in front of and adjoining the large proximal joint of the mandibles
(see figs. 3, 6, and 7, pl. 22).

Mandibles—The mandibles are formed of a strong proximal joint
with four short, strong joints followed by five slender, elongate joints
(see fig. 9), the latter being almost covered with very fine setae that
give a plumose appearance to the appendage as it extends out beyond
the great backward curving spines of the carapace. I examined hun-
dreds of specimens before finding a proximal joint with its inner end
sufficiently well preserved to suggest the character of its masticatory
surface; two specimens indicate that it is as shown in figure 7,
plate 22, and in the restoration. There is no evidence as to whether
the proximal joint is composed of one long joint or two closely united
short joints. The usual location of the mandibles in well preserved
specimens is shown by figures 1 and 2, plate 22.

Maxillulae—These are long, slender, and with about Io slender
joints. They look like thoracic legs (endopodite) but their position
and slender joints serve to distinguish them. Portions of them may be
seen in figures I and 2, plate 22.

Maxillae—As far as known the maxillae are formed of joints a
little longer than those of the maxillula and about the same diameter ;
both appear to have been slender, rather closely jointed, simple ap-
pendages as far as the endopodite was concerned; there is strong
evidence that an exopodite was present, similar to those of the exop-
odites of the trunk appendages, but they have not been seen directly
attached to the protopodite; where the parting of the shale is on the
plane of the exopodites they are usually present next to the mandible
and directly over the position of the maxillulae and maxillae, which
suggests strongly that they were present.

The maxillulae and maxillae were so slender that they are usually
absent as the result of having been torn off or crushed between the

3

32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85
strong mandibles and the thoracic limbs. In figure 1, plate 22, their
exopodites are shown on the left side, and on the right side the endop-
odites of the maxilla with the exopodite of the maxillula, the endop-
odite of the latter having apparently been pushed under and a little
forward of the mandible. Sometimes the endopodite is present but
the joints are indistinguishable or only a few can be seen.

Thoracic limbs.—The biramous thoracic limbs appear to be uniform
in character from the cephalon to the minute plate-like telson at the
posterior end of the body. Each limb is formed of a protopodite, a
jointed endopodite, and a jointed fringed exopodite.

Protopodite.—The large protopodite is attached by its inner end
to the lower side of the body segment about half way between the

Vi © ---\---Sectron of Body

Fic. 10—Diagrammatic outline of the posterior side of one of the anterior
thoracic limbs: pr, protopodite; en, endopodite; e+, exopodite; f, filaments of
exopodite ; 7, intestine.

This figure indicates the point of attachment of the limb to the body, also
approximate position of the intestine.

ventral median line and the rounded outer side of the body apparently
in the same manner as the trunk limbs of Apus, except that in the
latter there is no evidence that the protopodite served as a gnathobase.
The protopodite is elongate, apparently cylindrical at its inner end,
and flattened somewhat at the distal end; it is strong, and supports an
endopodite and an exopodite. It is usually flattened so as to appear
of about the same width throughout its length; a few specimens in-
dicate that it narrowed at its proximal end, essentially as shown in the
restoration.

Endopodites—The endopodite or leg is formed of six joints. The
first five joints of the anterior limbs are rather flat and broad at the

‘T find that at many places Doctor Walcott changed “ protopodite” to
“coxopodite.” Whether this term was supposed to have been changed here also
I was unable to ascertain —C. E. Resser.

NO. 3 BURGESS SHALE FOSSILS—WALCOTT ns

sides; narrow and slender on the dorsal and ventral view; short, very
fine spines occur at their distal end and along the side of the joints.
The slender distal joint is more nearly cylindrical and has a short,
strong, slightly curved spine with one or two fine spines beside it ex-
tending out from the end of the joint. The anterior legs appear to have
been delicate and slender, but usually they have retained their natural
position remarkably well. Usually the first joint of the endopodite
of the fourth pair of limbs is slightly expanded, and the first and sec-
ond joints of the fifth to seventh pairs of limbs, and the first five joints
of the eighth to twentieth pairs of limbs. The expanded joints vary in
degree of expansion from slight enlargement on the fourth limb to
where the transverse diameter is considerably greater than the length
of the joint. The latter recall the transverse flattened joints of the
endopodite of the trilobite Triarthrus becki.’

Fic. 11.—Diagrammatic enlargement of a section of the exopodite showing the
body and the attached cylindrical filaments.

On some specimens showing the expanded joints the extended por-
tion is very narrow from base to point, and gives the effect of a strong
spine projecting from midway of the joint; in other specimens the
base is as long as the joint and the apex is obtuse, which is the pre-
vailing form. When in a natural condition the expansion of the joint
was undoubtedly on the lower or the ventral side, and the fact that
in the fossil state specimens occur with all the expansions pointing
forward means only an accident of preservation; some occur with
scarcely a trace of the enlarged joint, owing to the fact that the ven-
tral side of the endopodite is buried in the shale, leaving the narrow
dorsal side in view; in the restoration (fig. 9) I have outlined the flat,
vertical posterior side of the endopodite.

Exopodite—The exopodite is attached to the protopodite about
midway of the length of the latter. It is formed of a long, strong
proximal joint to which is attached a long, slender, multi-articulate
appendage, each segment of which supports a long, slender, flat

* Smithsonian Misc. Coll., vol. 57, p. 137, pl. 30, fig. 20, 1912.

34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

(formerly round) filament, which is beautifully preserved in some
specimens (fig. 8, pl. 22). The fringe of filaments often overlap
those of the adjoining exopodite so as to form an imbricating series
of fringes the entire length of the body.

An exopodite 3 mm. in length has attached to it 42 filaments that
average about 12 to 14 to the millimeter in the proximal portion,
and 14 to 16 towards the distal end; they increase gradually in length
from the proximal end until on the distal section they may be as long
as the entire exopodite. A rusty specimen laid aside as of little value
proved on cleaning to have the filaments preserved as long, slender
cylinders or tubes (see fig. 8, pl. 22).

Most of the fossil specimens. have the fringes extending forward
and outward, but when the animal was living they undoubtedly ex-
tended outward and backward so as not to impede its forward move-
ment.

As mentioned under Cephalic appendages, there is good reason to
think that the two posterior oral appendages (maxillulae and maxillae)
have in addition to the simple jointed endopodite an exopodite similar
in structure to that of the thoracic limbs.

E pipodite —A single specimen shows what I thought in rg12 to be
a large epipodite or branchiae, but which I have now decided to be
several of the fringed exopodites pressed down together and more or
less macerated in the contents of the body which were squeezed out
on that side. It was only after finding a number of examples showing
the fringed exopodites arranged in this manner but not pressed into
each other that I gave up the view that a large epipodite was present.

Several specimens have been found since 1912 that indicate the
presence of a small oval flattened lobe attached to the dorsal side of
the protopodite, or it may be to the proximal joint of the exopodite,
but it may be that this appearance is caused by the manner in which
the protopodite and the segments of the body are matted down to-
gether ; some of the thin oval bodies, however, are so clearly defined
that they suggest the presence of a small epipodite, but I do not
consider the evidence sufficient to warrant representing them on the
restoration of the thoracic limb.

Functions of appendages.—The proximal joints of the antennae
may have assisted in mastication and may have had a sensory func-
tion. The proximal joints of the mandibles undoubtedly served in
mastication, but whether those of the slender leglike maxillulae and
maxillae aided is undetermined, as nothing has been seen of either

* Smithsonian Misc. Coll., vol. 67, pls. 20, 32, 1918.

NO. 3 BURGESS SHALE FOSSILS—WALCOTT 35

gnathobasic spines or lobes. It is probable, however, from their posi-
tion and form they were of service in directing food to the mouth.

The long, flat outer joints of the mandibles may have been used in
swimming as an aid to the thoracic limbs.

The endopodite of the biramous trunk appendages probably served
both as a natatory and ambulatory leg, although from their delicate dis-
tal joints they evidently were little used in walking. The fringed exop-
odites may have assisted in swimming and they undoubtedly served as
gills. The absence of a channel formed of gnathobases on the protop-
odites of the trunk limbs such as occurs in the Apodidae and Trilobita,
and also of any known organ for seizing food, leads to the conclusion
that the exopodite may have served to direct a current of water bear-
ing food towards the mouth. The long, strong filaments attached to
the segments of the exopodite are comparable with the filaments of
the exopodite of the trilobite Neolenus,’ and the remarks on the latter
should be read in this connection.

Diagrammatic restoration of the ventral appendages.—This is shown
in text figure g and presents my interpretation of the arrangement of
the appendages. The long body, with its numerous segments, is at-
tached to the carapace in its cephalic region only. The antennae (a),
mandibles (m), maxillulae (m’), and maxillae (1m”) are drawn in
from the evidence given by many specimens, no one of which has
all the limbs in place; several of the best specimens are illustrated on
plate 22. The eyes (¢) are somewhat theoretically placed, but their
general position is known despite the displacement most of them have
been subjected to. On the right side the entire series of protopodites
and endopodites forming the thoracic limb are represented ; the form
of the inner end of the protopodite is based on indications afforded by
several specimens, although usually the protopodite is crushed flat and
appears to be of the same size throughout its length. On the left side
the protopodite is cut away so as to show the approximate point of
attachment of the proximal end of the exopodite. The latter are
drawn from such specimens as are represented on figures 1-9, plate 22,
and many others not illustrated. The relative position and form of
the exopodite, endopodite and protopodite is shown by text figure 10.

Mode of occurrence—Marrella splendens occurs abundantly in the
compact, hard shale but there are few really fine specimens. This
free-swimming, delicate little crustacean dropped down on the surface
of the bottom and was speedily buried by fine mud settling over it;
the mass of gradually hardening mud pressed the rounded body into a

* Smithsonian Misc. Coll., vol. 67, no. 7, p. 370, 1920.

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

thin film and matted those parts resting on each other together unless
there was a thin film of mud between them. When there was such a
film of mud, it later hardened into shale and formed a plane of weak-
ness along which the shale parted when, from the action of weathering
or of force applied with hammer or chisel, the shale was split open.
Sometimes the parting is between the matrix and the ventral or dorsal
surface of the specimen, or it may be between the series of fringed
exopodites and the endopodites ; a number of specimens in the col-
lection show a part of the endopodites with the exopodites above or
below them, and again the parting may have been above or below the
exopodites on one side of the body and the reverse on the other side
(see pl. 22). The structure of the body and the thoracic appendages is
very clearly exhibited, but the cephalic appendages, labrum and cara-
pace, are usually so matted together that it is difficult to distinguish
the details of structure.

OBSERVATIONS

Marrella and the trilobite—Marrella has several characters in com-
mon with the trilobite and others that are dissimilar.

SIMILAR CHARACTERS

A cephalic shield supporting a labrum.

Sessile eyes on the proximal end of a great spine equivalent to

the free cheek of the trilobite.

3. A labrum (hypostoma) with the proximal joints of the cephalic
limbs gathered at its posterior end in a manner comparable with
that of the trilobite.

4. A pair of biramous limbs for each trunk segment formed of a
protopodite, jointed endopodite (leg), and a jointed exopodite,
but without any known epipodite.

5. Expansion of the joints of the endopodites on some of the thoracic

limbs.

Ne

DISSIMILAR CHARACTERS

1. Absence of a thoracic dorsal shield.

2. Almost total absence of an abdominal section or pygidium.

3. Position of proximal joint of antennae.

4. A large third cephalic appendage (mandible).

5. The manner of attachment of the coxopodite of each trunk limb
directly by its proximal end to the side of the ventral surface
of the body.

ENO’ 3 BURGESS SHALE FOSSILS—-WALCOTT 3

“I

6. The coxopodite did not serve as a gnathobase.
1 and 2 are considered to be more primitive characters.
3, 4, 5, and 6 less primitive.

My present conclusion is that Marrella is a less primitive form than
the Apodidae, and while a more primitive form than the trilobite it is
nearer the latter than the Apodidae, and should be grouped near it
but not with the Trilobita. At the time of my preliminary examination
of the crustaceans then known to me from the Burgess shale I placed
Marrella and Nathorstia as progenitors of the trilobite,” but with our
present information Marrella will be placed with Burgessia, Nathorstia
being left under Trilobita.

COMPARISON WITH CRUSTACEANS

Marrella and the Branchiopoda.—Marrella, with its sessile eyes,
carapace-like cephalic shield, labrum attached to the doublure, numer-
ous trunk limbs, and the large mandible, suggests the Apodidae, but
when we consider the well-developed antennae, large removable spine
attached to the cephalic shield, biramous trunk limbs on each body
segment consisting of a fully developed endopodite and exopodite,
and the absence of caudal rami, the conclusion is that Marrella repre-
sents a more advanced stage in the evolution of the Crustacea than
Apus and its allies. The biramous limb of Marrella, like that of the
trilobite, undoubtedly passed through the foliaceous or multiramous
limb stage in its evolution, probably in pre-Cambrian time.

Marrella differs from the Branchiopoda in:

8

Absence of lobed multiramous foliaceous trunk limbs with gnatho-
bases and in the presence of biramous trunk limbs with protop-
odite, jointed endopodite (leg), and jointed exopodite.

b. Absence of furcal rami.

c. Presence of a pair of biramous limbs on each trunk segment back

to the telson.

Marrella includes the following characters of the Branchiopoda:

a. A true carapace arising from a fold of the integument.

b. A labrum attached to the reflected margin or “ doublure ” of the
carapace.
c. A large mandible serving as a jaw in the process of mastication.

Plesiotypes—U. S. N. M., Nos. 83486a-1.

* Smithsonian Misc. Coll., vol..57, p. 161, 1912.

38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

HELMETIA Walcott 1918
HELMETIA EXPANSA Walcott
Plate 27

Helmetia expansa Walcott, 1918, Smithsonian Misc. Coll., vol. 68, no. 12, p. 7,
fig. 8.

This is another of the species first published in the Smithsonian explorations
pamphlet for 1917, and for that reason is here reprinted.
Holotype —U. S. N. M., No. 83952.

EXPLANATION OF PLATES

PLATE I

Margaretta dorus new SpecieSeirmaccere ince oie ee eee eee 2
Fic. 1. Impression of outer surface.

2. A wider and less well defined specimen. Possibly a mutilated
fragment.

3. Impression of surface, on which the elevations apparently were
more conical and more numerous, which may indicate another
species.

4. Unretouched photograph by reflected light of the typical form.
This is again the impression of the outer surface. Compare
the elongate elevations with the more conical ones in fig. 3.

5. Unretouched photograph of the holotype.
6. (Xx 2.) Another, less perfect specimen on which Doctor Walcott
apparently laid stress.
PLATE 2
PAGE
Miuskowe placida:, New \SPeCieS) Ss.cnoan cushion a sh ee ee ee 4

Fic. 1. The holotype showing the annulations, the teeth around the
mouth, and the manner in which the digestive tract seems to
run along one side of the body, as it is compressed in the
shale.

Redoubtia poly podia: NV alcott s oco5 585s cias siace aig a 74 oie 02 © sure eleiieys ons reat 3
Fic. 2. A specimen, possibly of a different species, since it has smaller
appendages. Parts of a Hymenocaris? lie above it.
3. (X1.5.) Holotype, as illustrated previously.

Wawania corrugaia NVialCOttes case cal oli mie oreiocre) cist crcnevetors Lovelelstersicielere 6
Fic. 1. Photograph of a very fine individual retouched to bring out the
striae on the scales.

PortalazmiracNValcott csc oon eee eins see cee AC Se SO OPE ie etree 3
Fics. 2, 3. Counterparts of the holotype. A specimen of Miskota lies
nearby.

NO. 3 BURGESS SHALE FOSSILS—-WALCOTT 39

PLATE 4

Ht SPHOSE) NV AICOU soaccaisie grrr cin sails ane eels ala ea E Seaview Wins ss eve 4
Fic. 1. (X 1.5.) Specimen having an unusually straight position, with
all the spines turned backward and thus crowded together.
2,3. Natural size and enlarged (3) views of a fine complete
individual. The enlarged picture has been retouched.

AREOLA ESP INOSO, NVAICOEE 5 2s cc ccrs Ss ecwie Soe a's « fiw sis sie aay oo mais Sie winigiaua ola Scie aie 4
Fics. 1, 2. Natural size and enlarged (X 4) photographs. The larger one
has been retouched.

PLATE 6

AAA SPinOSD NV AICO sacar Or cca e os cess ae tee We sate ss he dleore needs 4
Fic. 1. (XX 2.) Photograph of a well preserved specimen.
2. (X2.) Unretouched illustration of another specimen whose
specific identity Doctor Walcott doubted somewhat.

MPC PAT SAN ALCOULS oo. 313 vite Ad OS Sew Da eae Tailed Maw Raed Saws Qed ee 5
Fic. 3. (X 2.) Unretouched photograph of this peculiar worm.

CP AIUAIET COULATIS WV. ACO natin see utn Samir nd coos acosened beneege wanes 5
Fic. 4. Small, somewhat broken specimen. (See pl. 9, fig. 3, for enlarge-
ment. )

5. (X2.) An unretouched photograph of the lectotype.
6. (2.) Coiled specimen doubtfully referable to the species.

RPA CAS OTIC CT @- VV BICOUR <aue tc cfc oa s'dpars Sarena sea WG igs Se are e's cos ed Hel we ties 5
Fic. 1. (2.) Unretouched photograph of a rather complete curled
specimen.

4. (X 2.) Unretouched photograph showing the attachment of the
setae.

(DOM EATOMO UO 2Oe WVIAICOtE 5 sa. ible ec nectaya ever sin reais ars ciai ers: fede gene S aie ele Sochareiar es 6
Fic. 2. (X 3.) A good illustration of this small form.

pata regulars. WalCOtts: 0... ics cwcs cds cadawnecensn bes sume cael secs 5
Fic. 3. (X3.) Enlargement of specimen illustrated as fig. 4 on preced-
ing plate.

40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

PLaTE 8
PAGE
Canadia dubtas Walcott) ccc.226 6 oe ne eoreke Cee nae | eR eee 6
Fic. 1. A fairly large specimen.
2. (X2.) <A specimen showing the intestinal tract.
Canadiavsetiger am NVialcotta- see tee ee ee ee EEC COE Eee 5

Fic. 3. (X3.) Retouched figure of a specimen showing the bundles of
spines particularly well.

Pikaia.gracilens. Walcottsaacnisnane ee ee eee Cee eee 7
Fic. 4. (X 2.) A wide specimen. The illustration is slightly retouched.
5. (X2.) Another rather well preserved example.
PLATE 9
PAGE
Ottoialmmor Walcottic: cca oacon See ce leeks viele cone Ie Oe EEE 6
Fic. t. (X2.) <A short form that may have its shape because the mouth
parts are retracted.
2. (2.) Small specimen, showing intestinal tract, doubtfully
referred to the species.
3. (X2.) Apparently the anterior end of a poorly preserved indi-
vidual.
4. (x 2.) Another form like fig. 2.
5. (X2.) <A specimen with a peculiar restriction that may be
accidental.
6,7. Retouched photographs of a particularly fine individual. The
whole specimen is enlarged somewhat (1.5) and the an-
terior end considerably (X 4).
GanadiavdubiaaNWVialcottaces eee cee eee eee 6
Fic. 8. (x 2.) Unretouched photograph of a small individual.
Canadia simples, mew SpeCieS.\seieiee dae ellie ete teed eis helt eee eee 6
Fic. 9. ( 3.) Retouched photograph of the holotype.
Ganadial grandis; new species sans: eee ee te ene ae eee 5
Fic.10. (X 2.) Retouched figure of the holotype.
PLATE 10
PAGE
Selkirkia: major \@Wialcott)) 2 sckc se cor cctetie seit etna ee eae ae 7

Fic. 1. (circa X 3.5.) A large individual, possibly preserving the origi-
nal proportions of the shell.

2. (circa X 3.5.) Another specimen whose shell is more or less

crushed, but with even better preservation of the soft parts.

NO. 3 BURGESS SHALE FOSSILS—W ALCOTT 41
PLATE.

PASEO a PEGUNCUIUTO VWalCOtl oss aAnie cack dais senda 4 kaka lndn bos 3480.0 dels 8
Fic. 1. (x 2.) An unusually perfect specimen.
2. Large coiled specimen that Doctor Walcott regarded as possibly
representing a different animal, but which may appear odd
only because of its unusual attitude.

PLATE 12
PAGE
ECONGHOIMNG ASUPCTIALde Wi alCOtt jc cero oie aicie suvlei o-oo ei aicl da, nese Ss aw ade on ss 8
Fic. 1. (X 2.) Retouched figure of a fragmentary specimen.
2. Natural size unretouched photograph of a well preserved indi-
vidual.
3. Enlarged, retouched figure of the counterpart of the preceding
figure.
PLATE 13
PAGE
Leanchoilia superlata Walcott.............0. cee cee cc cecsenvececcsecees 8

Fic. 1. (3°). An appendage apparently belonging to this species.
2. An individual flattened out in the horizontal plane, thus giving it
an unusual aspect.

NECOMGHOUIGE MAT ONss NEW {SPCCIES ayes crest ie Seuss bc Gries ea siseree ieee ices sme es 10
Fic. 3. Illustration of the holotype.

WWaraoia compacta, Walcott. ... 62c<.0c ss ce cscs ec wes reece secs severe ees 10
Fic. 4. (X 3.) An unretouched photograph of a good specimen. The
appendages are only faintly shown extending beyond the
carapace at the rear of the specimen. (See pl. 15, fig. 3.)

PLATE 14
PAGE
ON COMPACTION VV ALCOLGS 5 cena sn bere oo gag adie ood eb ennes pagans dosages 10
Fic. 1. (X 2.) Dorsal view of a compressed and somewhat distorted
specimen, on which the outline of the body trunk is well
shown and, on the left side, the ends of 16 exopodites pre-
serving their fine filaments.
2. (xX 2.) Dorsal view of a specimen with the carapace shortened
nearly one-half by compression. The central portion showing
the hepatic glands or caeca is unusually well preserved. The
posterior dorsal shield has the outlines of the fused segments
and the outer border clearly indicated, and on the left side,
where a piece is broken out, two of the exopodites are seen,
and at the posterior margin, traces of endopodites.
3. (X 3.) Enlarged view of a specimen with peculiar features
along the axis.

Eeanchowha superlata Walcott... 6s <siecs0ccscevecoevacccvsngusboeccecese 8
Fic. 4. ( 2.) Details of a separate appendage.
5. (x 2.) Another appendage of a different type.

42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

PLATE 15

PAGE

Naraota spintioronew iSpeCieS.2 achaiiack sae se See ee ee eee ee
Fic. 1. (X2.) Posterior dorsal shield with traces of segmented body
and thoracic limbs; a dorsal view of another specimen shows
clearly defined exopodites with the slender, distal ends of

endopodites projecting from beneath them.

15

Naraoia compacta. Walcotts.. isos naciciecinee ties Sei ae ree eee 10
Fic. 2. (> 2.) Smaller, retouched picture of the specimen shown in
fig. 4, pl. 13.
3. (X 1.5.) Incomplete individual clearly showing the axis.
Burgéessia bella Walcott): 47s seine & see eee so oes eee 1
Fic. 4. (XX 3.) Specimen showing the outlines of 10 thoracic legs
formed of the six joints of the endopodite and the large
protopodite, also the annulated intestine and fragments of the
crushed carapace.

5. (x 4.) A dorsal view of a specimen in which the stomach and
portion of the intestine as well as the large hepatic tubes are
distended so as to be moderately convex. The specimen also
shows the position of the antennae and portions of the hepatic
caeca.

6. (xX 4.) A macerated specimen indicating the position of the
eyes, the antennae, a portion of the labrum, the mandibles,
the maxillulae, and the maxillae; also the proximal portion
of seven pairs of the thoracic limbs, with a fair indication of
the point of attachment of the limbs to the body.

7. (X3.) Another very fine individual.

PLATE 16
PAGE
Bungessiaivelia, Wialcottaana tes oe oe eee ee eee eer IS

Fic. 1. ( 2.) Fragmentary specimen indicating structure of limbs.
(Possibly oriented incorrectly.)

2. (x4.) A specimen preserving 10 pairs of thoracic limbs, show-
ing their approximate place of attachment to the body, and
also their expanded joints and general form.

The carapace has been crowded back and crushed, but the
antennae project from its anterior side in an almost natural
position.

3. (X2.) <A badly decayed specimen that shows the manner of
attachment of the limbs.

4. (x 4.) Dorsal view of a specimen with the digestive organs
beautifully preserved. These include the intestine posterior to
the large hepatic tubes, the stomach anterior to the latter,
and also the anterior and posterior branches of the main
hepatic tubes and the numerous finer hepatic caeca, which
occur on both the outer and inner sides of the main branches.

This specimen gives most of the data for the restoration
of the digestive organs as shown by text fig. 5.

NO: 3 BURGESS SHALE FOSSILS—WALCOTT 43

5. (XX 4.) Specimen showing the expanded joints of the endopo-
dites of the cephalic limbs.

6. (> 3.) Partly side view of a crushed and distorted specimen
illustrated for the purpose of showing the exopodites that
occur near the body, the proximal part of the endopodites
with the protopodites having been flaked off from above the
exopodites in the specimen; in their natural position the
exopodites were probably above and between the endopodites.
The outer ends of the long, strong endopodites are well
shown in this specimen, although the joints have been ob-
scured. The slender jointed leglike structure associated with
the exopodites may be the distal part of the endopodites of the
right side.

PLATE 17

TaRP a SEL OCIA NN AICOUE erate cise ee viahaye 6.6: o'ale cisiqin Wikre pantie doh Side oyna d arden wren 15
Fic. 1. (X3.) Another specimen preserving mainly the * skeleton.”

2. ( 2.) A small individual crushed obliquely.

3. (X3.) Ventral view of a specimen preserving the reflected
anterior margin and labrum, also the antennae, outlines of
the inner portions of the cephalic limbs, and more or less
distinctly outlined thoracic endopodites with transversely ex-
panded joints. The anal plate is clearly indicated, also the
anterior end of the telson.

4. (4.) Side view of a crushed specimen preserving on the left
lower side the outlines of 10 thoracic legs, on the right side
the outlines of four entire lobelike exopodites and the distal
end of four posterior to them. The exopodites still show
slight traces of fine filaments (?) along the posterior and
outer margins.

PLATE 18
PAGE
PSSA Dela WV BICOtts ma. v a othe soe ete SAG A PUES POSNMA <ribak dias wa 15
Fic. 1. (X2.) An incomplete specimen showing the manner in which
the limbs project beyond the carapace.

Vie iICMn GLA GTISUS NV a lCOLtc. qatsicncse aie vinic etkde-oies citi. sii oh sate asi eieie easels 6 20
Fic. 2. (X 4.) Side view of a flattened specimen preserving six or more
cephalic limbs, the limbs from both sides being more or less
crushed down together.
3. (X2.) Specimen showing on the right side a flattened uninjured
antennae in its natural form, and on the left side only the out-
line of the interior of the joints. The latter mode of occur-
rence is quite common for the antennae and other appendages.
This specimen also has the stalked eye preserved on the
right side.

44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

4. (x 2.) A few thoracic exopodites in which the flattened fila-
ments are unusually well preserved.

5. (x 4.) Side view of a flattened specimen showing the outlines
of body through the carapace, the stalked eye, antennae, four
cephalic limbs, and several thoracic limbs.

This specimen is particularly instructive, as it shows the
outline of the thoracic limbs from the body to their distal end.

PLATE 19
PAGE
Waepna fieldensis Walcotts. 2.2 seoe onto oe oe Co ero ene Oe 3 20
Fic. 1. (X 2.) Anterior portion of an individual showing the limbs.

2. (Xx 2.) Unretouched photograph of a distorted specimen show-
ing the eye particularly well.

3. (X 2.) Dorsal view of the carapace of a specimen showing the
outline of the body, also the small rostrum between the
the antennae and the eyes.

4. (X 3.) Specimen with details of eyes and antennae.

PLATE 20

Wapiti feldensis: Walcott s,. css sacar cee oe iets ee ao eee eee 20
Fic. 1. (xX 2.) Posterior portion of specimen preserving the three seg-
ments, the anal opening, and the lobed, segmented caudal
furca.

2. (3 3.) Anterior portion of a carapace, median rostral plate,
antennae, stalked eyes, and a palp on the left side.

3. (X< 2.) Side view of a flattened specimen in which a number of
the exopodites retain the fringing filaments, some of the latter
being gathered in a cluster at the distal ends. This specimen
also shows the outlines of elongate triangular, light-colored,
shiny places that may indicate the interior of the exopodite
or possibly a rudimentary endopodite.

PLATE 21
PAGE
SRaniaafragtliss Walcott: couccsyoccleoc tine oe ees oo ae oe eos 2
Fic. 1. (X 4.) Specimen showing the flattened intestine and outline
of thoracic segments.

Weaptia: feldensts. NValcotty ..s.a5 ws iesinage © ehcdetavers arp arncio esis te ene eis telerayaieret 20
Fic. 2. (> 2.) Dorsal view of a flattened specimen in which the cara-
pace has been removed from over the cephalic area and por-
tion of the thoracic region. The proximal portion of eight
thoracic limbs is clearly shown, also of three pairs of cephalic
limbs.

NOD 3 BURGESS SHALE FOSSILS—WALCOTT 45

Te MICN GIT EUIGLIS TOW ‘SPECIES aise sch aie mans «cis ould a diaibe's sie ee secre ue ss 4 24
Fic. 3. ( 2.) Side view of a flattened specimen illustrating the form
of the carapace, one side of which has been crushed upward
and backward. The exopodites of the thoracic limbs appear
to have the same structure as those of Waptia fieldensis.

MUNG MUSONTOL SONG. NVAlCOUE sera.) aelees.ere cyst cis 0 2 4S sid orsiaiwg so G's aiesdieibicieie’ere auc 27

Fic. 4. (X 2.) <A retouched figure of a most excellent specimen.

PLATE, 22

ARRCG SPIEndenS WalCOtl a. isc ccscuwicrs sos cdsisies de gees ee oe esa cee tes 28
Fic. 1. (X 4.) Dorsal view of a specimen showing the mandibles, the
antenna on the left side extending down beside the mandible,
and the exopodites referred to the maxillae and maxillulae ;
what may be the endopodite of the maxillulae is shown on the
right side below the fringing filaments of the exopodite of the
maxilla.

(x 4.) In this ventral view the mandibles have been pushed
forward so that the mandible and antenna on the right side
are in a vertical position and side by side, and the endopodites
of the maxilla and maxillula are in advance of their natural
position; on the left side two endopodites of thoracic limbs
are faintly outlined.

3. (xX 4.) Ventral view of a specimen in which the joints of the
mandibles have been crowded together and somewhat short-
ened; the proximal joint on the right side is well exposed by
the labrum having been pushed forward; the inner margin
of the joint is serrated but not quite as clearly shown as on
the specimen represented by fig. 0.

4. (XX 4.) The dorsal view of this specimen is illustrated to show
the position of the antenna on the right side which is the
same as the antenna on the left side in fig. I, and on both
sides in fig. 5; the position of the mandibles is also well
shown.

5. (X4.) Ventral view showing the position of the mandibles
beside the labrum, also the exopodites of the maxillae and
maxillulae posterior to them; the antennae appear to have
been torn away in the crowding forward of the mandibles.

6. (x 4.) Ventral view of a specimen showing the mandibles, the
one on the right side preserving the proximal joint with a
serrated inner margin. This mandible has been colored white
in order to bring it out more clearly in the reproduction; the
thoracic exopodites are very clearly shown on both sides, the
endopodites having been largely exfoliated.

(X< 4.) Ventral view of a specimen preserving a complete man-
dible on the left side, also several endopodites of the thoracic
limbs, and on the right side six joints of the mandible and a
few imperfect thoracic endopodites, which are exposed on both
sides by the exfoliation of the exopodites.

ty

NI
.

46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

8 (x 6.) Enlargement of the filaments of one of the thoracic
exopodites in which the filaments retain their original round,
slender tubular form. This is the only specimen among
several hundred that I have examined which has escaped
flattening by compression. This is due to the pyritization of
the filaments of this particular specimen.

9. (X4.) Ventral view of a specimen illustrated to show what
appears to be a small oval flattened lobe attached to the
dorsal side of a protopodite or it may be to the proximal
joint of the exopodite; a number of the thoracic exopodites
with the filaments projecting forward occur on both sides,
also fragments of the large postero-lateral spines of the
carapace which lie above the exopodite of the thoracic limbs.

PLATE 23

Helmetia expansa Walcott. cnass 2258) so -0s Joe oe ene ee eee 38
Photograph of the holotype.

1

VOL. 85, NO. 3, PL.

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 38.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOEI"85),, INO: 3; (PES

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 38.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3, PL. 3

5 ae Pact i
—
Se on a

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 38.

VOL. 85, NO. 3, PL. 4

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 39

VOL. 85, NO. 3, PL. 5

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 39,

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3, PL. 6

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 39.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3. PL.7

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 39.

VOL.85, (NO-°3), PES 8

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 40.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3, PL. 9

cy * S73] ores
SRS 4 is ;

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 40.

on

VOL. 85, NO. 3, PL. 70

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 40.

SMITHSONIAN MISCELLANEOUS COLLECTIONS

—

a

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 41.

VOE. 85, 7NO. 3, PES 14

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOES85, NO 3) PEaat2

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 41.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL 85, NO. .3;, PE. 13

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 41.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3, PL. 14

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 41.

NO*. 3; PEE 15

VOL. 85;

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 42.

VOES 85) \NO- 3; (REZaG

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 42.

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 43.

VOL. 85,

NO. 3, PL.

17

VOL. 85, NO. 3, PL. 18

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 43.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3, PL. 19

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 44.

VOL. 85, NO. 3, PL. 2

SMITHSONIAN MISCELLANEOUS COLLECTIONS

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 44.

| SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, (NO. 3); PL. 21

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 44.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3), PE, 22

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 45.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 3,\RL. 23

BURGESS SHALE FOSSILS

FOR EXPLANATION SEE PAGE 46.

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VOLUME 8&5, NUMBER 4

MEXICAN MOSSES COLLECTED BY
BROTHER ARSENE BROUARD — II]

BY
I. THERIOT
Fontaine la Mallet, France

S
' % he
es)

(PUBLICATION 3122)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
AUGUST 25, 1931

The Lord Baltimore Presa

BALTIMORE, MD., U. S. A.

oe

MEXICAN MOSSES COLLECTED "BY BROTHER
ARSENE BROUARD—III

By I. THERIOT
FONTAINE LA MALLET, FRANCE

With the present paper the study of Brother Arscne’s mosses comes
to an end. In addition to this collection a rather large number of
species collected by Brother Amable in 1926-27 are included. These
last are from the Valley of Mexico and adjacent regions. Together
the two collections comprise about 300 species, among which 18 had
not been found previously in Mexico, and 44 are entirely new. Brother
Arséne’s and Brother Amable’s explorations have thus enriched the
bryological flora of Mexico by more than 60 species, These valiant
botanists are deserving of great praise and also of thanks from all
those who are interested in bryology. Referring to the figures given
by Cardot* the Mexican flora now includes, with the Arséne and
Amable contributions, about 700 species of mosses. We are therefore
well above the 400 species enumerated in the Prodrome of Bescherelle.

It remains now for me to fulfill the very agreeable duty of ex-
pressing my gratitude to Brother Arsene and Edwin b. Bartram,
who have assumed with such kindness the task of translating my
French text into English, and to the Smithsonian Institution for
publishing my studies.

The two previous parts were published in the Smithsonian Miscel-
laneous Collections as follows: Vol. 78, no. 2, pp. 1-29, June 15, 1926;
vol. 81, no. I, pp. 1-26, August 15, 1928. In the introduction to the
first of these there will be found a list of localities in the states of
Michoacan and Puebla from which specimens are cited; this will
prove useful also for the present and final contribution.

DITRICHACEAE (continued)
DISTICHIUM CAPILLACEUM (Sw.) Bry. Eur.

Valle de México: Desierto (Bro. Amable 1500).
Very probably new to Mexico.

eikey: Bryol. 38:97. 1011; ‘40: 33. 1013.

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 4

2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

BRYOXIPHIACEAE
BRYOXIPHIUM MEXICANUM Besch. Journ. de Bot. 6: 180. 1892
Eustichium norvegicum Schimp. in Besch. Prodr. Bryol. Mex. 29. 1871.

Valle de México: Desierto (Bro. Amable 1629).

DICRANACEAE (continued)
CAMPYLOPUS (Eucampylopus) ANGUSTI-ALATUS Thér., sp. nov.

Distr. Federal: Rio Frio, alt. 3,000 meters (Bro. Amable 1724).

Habitu, colore, rete foliorum C. Pittiert R. S. W. sat similis, sed
differt : caulibus elongatis usque 6 cm., foliis magis concavis, angustis
(basi 0.5 mm.), acumine angustiore, cellulis laminae minus incrassatis,
costa angustiore (0.36 mm.).

| 2 3
\
ae
KS
| C
x 12

Fic. 1—Campylopus angusti-alatus Thér. 1, leaf; 2, acumen; 3, lamina ceils
in the upper third; 4, alar and suprabasal cells; 5, cross-section of a leaf near
the base; 6, cross-section of costa.

In size this species suggests C. A pollinairei Ther.,’ from Colombia,
but may be distinguished from it by the leaves, which are erect-
appressed when dry, little or not flexuous, shorter and only half
as wide, very narrow toward the base (8 to to rows of cells), and
by the non-incrassate cells of the lamina.

CAMPYLOPUS DESTRUCTILIS (C. M.) Jaeg.

Dicranum destructile C. M. Bot. Zeit. 17: 220. 1850.
Campylopus Arsenei Thér. Smithsonian Misc. Coll. 787: 5. 1926.

Having seen from the Berlin Museum the types of C. Chrismari
(C. M.) and C. destructilis (C. M.), both Mexican species, I have

* Archiv. Bot. 2: 187. 1928.

NO. 4 MEXICAN MOSSES—THERIOT 3

come to the conclusion that they are distinct, the first belonging to
the subgenus Eucampylopus, the second to the subgenus Pseudocampy-
lopus; furthermore, that C. Arsene: Thér, does not differ from the
latter.

CAMPYLOPUS (Eucampylopus) SAINT-PIERREI Theér., sp. nov.

Hidalgo: El Chico, on rocks, alt. 2,600 meters, leg. Marius Saint-
Pierre (Bro. Amable 1589).

Sterilis. Caespites densi. Caulis 2-3 cm. altus, simplex vel parce
ramosus, inferne radiculosus, basi terra obrutus, dense foliosus. Folia
sicca erecta, parum flexuosa, humida erecto-patula, lanceolata, longis-
sime subulata, canaliculata, marginibus integris, apice denticulatis,

Fic, 2.—Campylopus Saint-Pierrei Thér. 1, leaf; 2, acumen; 3, upper cells
at a; 4, suprabasal cells toward b; 5, alar cells; 6, cross-section of leaf near
the base; 7, section from the acumen; 8, cross-section of costa near the base.

5-6 mm, longa, 0.7 mm. lata; costa basi 0.3 mm., breviter excurrente,
dorso superne sulcata, haud lamellosa ; auriculis distinctis, valde exca-
vatis, cellulis alaribus numerosis, minutis, vesiculosis, rete supra-
basilari chlorophylloso, cellulis marginalibus (3-4) linearibus, hyalinis,
sequentibus quadratis, internis rectangularibus, parietibus incrassatis,
sinuatis; cellulis laminae rhomboidalibus valde incrassatis, juxta-
marginalibus minutis, juxtacostalibus sensim majoribus.

At first sight a cross-section of the costa seems to indicate the
subgenus Palinocraspis. Under a moderate magnification this section

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

appears as a thin opaque slice, in which is seen only the median arc
of eurocysts covered on both sides by cells with a small lumen.

But, using a higher magnification for the same section, one notices
that the ventral surface is formed of a unique layer of cells with
very thick walls showing a lumen like a “ cat-eye”’ (not punctiform).
These cells, which are not stereids, remind one exactly of those found
in C. matarensis Besch., from La Réunion Island. Like the last named
species, C. Saint-Pierrei belongs to the group “ C.b.”*

Another peculiarity is that the comal leaves often end in a rather
long hair point, which is discolored or subhyaline, recalling to mind
that of the group Trichophylli.

POTTIACEAE
MOLENDOA OBTUSIFOLIA Broth. & Par. in Card. Rev. Bryol. 40: 36. 1913

Puebla: Hacienda. Alamos (4631, 4697); Rio Alseseca (574).
Vera Cruz; Cordoba (10978). Distr. Federal: Mixcoac (9432, 9434,
9462, 9465, 9466, 9468). Valle de México (Bro. Amable): Desierto
(1244) ; San Juanico (1261).

The leaves of this species show a peculiarity which has not yet
been pointed out: the margins of the acumen are a little thickened
by the bistratose marginal cells, such as may be seen in some other
species of the family Pottiaceae.

MOLENDOA OBTUSIFOLIA Broth. & Par. var. INCRASSATA Thér., var. nov.

Puebla: Hacienda Batan (5007 p. p.), associated with Trichosto-
mopsis crisptfolia Card.

Caulis 1 cm. altus. Folia breviora latioraque, I mm. X 0.25 mm.;
rete opaco, cellulis incrassatis, costa 50 p.

Here the thickening of the cell walls is more accentuated than
-in the type form: toward the apex the lamina is composed of 2 layers
of cells throughout, while in the middle the thickening is confined
to 2 or 3 rows of marginal cells.

ANOECTANGIUM LIEBMANNI Schimp. in Besch. Prodr. Bryol. Mex. 15. 1871

Morelia: Jesus del Monte (7965).

ANOECTANGIUM COMPACTUM Schwaegr. Suppl. 1: 36, pl. rr. 1811

Puebla: (4800).

*E. & P. Nat. Pflanzenfam. ed. 2, 10: 187. 1924.

NO. 4 MEXICAN MOSSES—THERIOT 5

ANOECTANGIUM APICULATUM Schimp. in Besch. Prodr. Bryol. Mex. 16.
1871

Puebla: Hacienda Alamos (4770). Morelia: Cerro San Miguel

(5042). Valle de México: Puente de la Venta (Bro. Amable 1388).

ANOECTANGIUM EUCHLORON (Schwaegr.) Mitt. Musc. Austr. Amer. 176.
1869

Morelia: Loma del Zapote (7508).

ANOECTANGIUM CONDENSATUM Schimp. in Besch. Prodr. Bryol. Mex. 16.
1871
Valle de México (Bro. Amable): Contreras (1483); Desierto
(11627 )'.

HYMENOSTOMUM (Kleioweisia) SEMIDIAPHANUM Thér., sp. nov.

Morelia: Cerro San Miguel, on earth (5040).

Dioicum? fl. masc. haud vidi. Tenellum ; caulis 1 mm. altus, pauci-
foliatus. Folia sicca crispula, madida patula, inferiora minuta, caetera
sensim majora, oblonga-lanceolata, breviter acuminata, acuta, concava,

Fic. 3.—Hymenostomum semidiaphanum Thér. 1, entire plant, dry; 2, moist
plant; 3, 4, 5, leaves; 6, acumen; 7, basal cells; 8, cross-section of leaf toward
the base; 9, cross-section of acumen; 10, fragment of cross-section from acumen.

superne canaliculata, marginibus integerrimis, valde involutis, folia
media 1 mm. longa, 0.3 mm. lata; costa basi 40 », percurrente vel
breviter excurrente ; rete basilari in dimidio inferiore hyalino, cellulis
rectangularibus, parietibus tenuibus, cellulis laminae opacis, densissime

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

papillosis, papillis elatis, diam. 6 ». Pedicellus 3 mm. longus, pallidus,
siccus valde arcuatus, humidus erectus; calyptra cucullata, dimidium
partem capsulae obtegens; capsula oblonga, clausa; operculum haud
secedens, e basi conicum oblique rostratum ; sporae papillosae, 18-20 wm.

Easily distinguished from the other species of the subgenus Kleio-
weisia by the length of the pedicels.

HYMENOSTOMUM MEXICANUM Card. Rev. Bryol. 36: 70. 1909

Puebla: Road to Cholula. (4849). Valle de México: Tenayuca
(Bro. Amable 1380).

These plants conform closely to Cardot’s type except in the in-
florescence. The type specimen has a paroicous inflorescence (teste
Cardot) ; here it is autoicous (male flower on a short pedicel under
the female flower). However, I do not think it is wise to base a
new species on this difference alone; it seems more logical, in my
opinion, to conclude that in H. mexicanum the inflorescence is variable,
occasionally paroicous but more often autoicous, as is normal among
the species of the genus Hymenostomum.

GYMNOSTOMUM CALCAREUM Bryol. Germ. 1: 153, pl. 10, f. 15. 1823

Valle de México: Morales (Bro. Amable 1595), c. fr.
This species is new to Mexico.

GYROWEISIA OBTUSIFOLIA (Hampe) Broth. in E. & P., Nat. Pflanzenfam.
E3389. 1902

Puebla: Cerro Guadalupe (666); Hacienda Alamos (4634).
Morelia: Loma del Zapote (7507).

The leaves of this species are identical as to form and areolation
with those of Husnoticlla revoluta. But this is the only likeness, the
leaves of G. obtusifolia being exactly plane on the margin and having
the costa half as narrow, not dilated at the apex.

GYROWEISIA PAPILLOSA Ther., sp. nov.

Morelia: (4927) ; Loma Santa Maria (7887).

Pusilla, caulis vix 1 mm. Folia erecto-appressa, parum crispula,
difficile emollita, oblonga vel oblongo-lanceolata, summo late obtusa,
valde concava, integerrima, marginibus planis, costa basi 30 », superne
dilatata, sub apicem evanescente, cellulis basilaribus hyalinis, rectangu-
laribus, cellulis mediis et superioribus quadrato-hexagonis, chloro-
phyllosis, papillosis, 8-10 », parietibus tenuibus. Pedicellus pallidus,

NO. 4 MEXICAN MOSSES—THERIOT

ae

erectus, 8-10 mm. longus; capsula oblonga, erecta, basi attenuata ;
operculum conico-rostratum ; annulus latus, peristomium sub orificio
insertum, dentes papillosi.

Py
3
EL)

Eby

at
ed

CET)
EL)
Ey

PE
)=]
Cr

amas
[ze
yy,

(a)
ss

Fic. 4.—Gyroweisia papillosa Thér. 1, 2, leaves; 3, cells in upper part at a;
4, middle cells near ); 5, basal cells; 6, cross-section of a leaf; 7, 8, perichaetial
leaves; 9, capsule; 10, fragment of peristome; 11, annulus.

Differs from G. obtusifolia by the very unequal, narrower leaves,
whose cells are more chlorophyllose and papillose, also by the capsule
with a broader annulus.

HUSNOTIELLA REVOLUTA Card. Rev. Bryol. 36: 71. 1909

Puebla: Guadalupe (662, 671, 800); Hacienda Santa Barbara
(4519, 4520, 4595 p. p.). Morelia: (4922, 7653) ; Andameo (4829,
4841, 4842, 4845). Distr. Federal: Mixcoac (9441, 9447, 9475 P. P.,
9483). Valle de México (Bro. Amable): Desierto (1249); San
Juanico (1227) ; Penon (1255) ; San Borja (1259).

Forma ELATA Ther., form. nov.—Stems longer, up to 10 to I5
mm., green above, discolored below. Leaves dimorphous, the lower
short, oval as in the type, those of the young shoots elongated, almost
twice as long as the lower stem leaves. Valle de México: Contreras
(Bro. Amable 1333).

HUSNOTIELLA REVOLUTA Card. var. PALMERI (Card.) Thér., comb. nov.
Husnotieclla Palntert Card. Rev. Bryol. 37: 121. 1910.

Puebla: (698) ; Rancho Santa Barbara (4594b). Morelia: Loma
Santa Maria (7880, 4902). Distr. Federal: Mixcoac (9471, 9475
Pp. p., 9490).

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

In reducing H. Palmeri Card. to a mere variety, following my own
observations, I feel in perfect agreement with Mr. R. S. Williams
and also with Cardot himself.’

HUSNOTIELLA TORQUESCENS (Card.) Bartr. Bryologist 29: 45. 1926

Didymodon torquescens Card. Rey. Bryol. 36: 83. 1900.

Morelia: (7652); Bosque San Pedro (4577, 4583); La Huerta
(7967) ; Loma Santa Maria (4913, 7882, 7886).

These specimens differ slightly from the type in being green, not
glaucous, above ; but the agreeinent is complete in so far as the fruit
and the form and structure of the leaves are concerned.

HYMENOSTYLIUM CURVIROSTRE (Ehrh.) Lindb.

Puebla: Boca del Monte (4737) ; Hacienda Batan, with var. sca-
brum Lindb. (4960) ; Esperanza (7976).

HYMENOSTYLIUM INCURVANS (Schimp.) Broth. in E. & P. Nat.
Pflanzenfam. 1°: 389. 1902

Gymnostomum incurvans Schimp. in Besch. Prodr. Bryol. Mex. 15. 1871.

Puebla: Esperanza (4740). Valle de México: Desierto (Bro.
Amable 1500 p. p.).

For want of good capsules the determination remains uncertain.
The male and female flowers are lateral, an abnormal character for the
genus Hymenostylium. This fact is, however, not unique; I have
previously mentioned a similar case in connection with H/. curvirostre.’

TRICHOSTOMUM CLINTONI C. M. Linnaea 38: 636. 1874

Morelia: Jesus del Monte (7613 p. p.) ; Loma Santa Maria (7860,
7874) ; Campanario (7643).

TRICHOSTOMUM INVOLVENS Card. Rev. Bryol. 40: 34. 1913

Puebla: Cerro Guadalupe (674, 802) ; Rancho Guadalupe (4612).

E. G. Paris distributed as T. lamprothecium C. M. a moss collected
by Bro. Arséne at Puebla which is very different from this species
and should be referred to T. involvens.

TRICHOSTOMUM CHLOROPHYLLUM C. M. var. BREVIFOLIUM Thér.,
var. nov.

Puebla: (4994, 4995).

Folia breviora latioraque.

* Rev. Bryol. 40: 34. I913.

* Bull. Acad. Internat. Geogr. Bot. 11: 319. 1902.

NO. 4 MEXICAN MOSSES—THERIOT 9

TIMMIELLA SUBANOMALA (Besch.) Broth. in E. & P. Nat. Pflanzenfam.

I’: 392. 1902

Trichostomum subanomalunm Besch. Prod. Bryol. Mex. 33. 1871.

Puebla: Hacienda Alamos (582, 4513, 4635, 4769) ; Finca Guada-
lupe (730). Morelia: Campanario (4773, 7523) ; Loma Santa Maria
(4907). Distr. Federal: Mixcoac (9448). Valle de México (Bro.
Amable): Desierto (1215, 1251, 1271) ; Rio Frio (1711). Hidalgo:
EL Chico (1579).

TIMMIELLA ANOMALA (Bry. Eur.) Limpr. Laubm. Deutschl. £: 590. 1888

Valle de México: Tenango (Bro. Amable 1686).

This species, which grows also in California, Arizona, Louisiana,
and Florida, has not previously been recorded from Mexico. The
specimens are identical with the European moss,

T. subanomala is very close to T. anomala, and to my mind seems
to be only a subspecies or a species of secondary rank. It differs
from JT. anomala by the position of the male flowers, the form of the
acumen (which is more gradually narrowed and more acute), and by
the much broader costa, up to two and a half times as wide.

TRICHOSTOMOPSIS CRISPIFOLIA Card. Rev. Bryol. 36: 74. 1900

Puebla: Rancho Santa Barbara (4593, 4594a, 4595, 4599) ; Rancho
Posadas (4807). Morelia: Cerro Azul (4934, 4936). Valle de
Mexico (Bro. Amable): San Jeronimo (1593); Tenango (1688).

Forma CRASSIRETIS Thér., form. nov.—Cellulae laminae valde
incrassatae——Puebla: Hacienda Batan (5007 p. p.).

TRIQUETRELLA FERRUGINEA (Schimp.) Thér., comb. nov.
Barbula ferruginea Schimp. in Besch. Prodr. Bryol. Mex. 37. 1871.

San Cristobal (F. Muiiller), original specimen, Puebla: Boca del
Monte (Purpus, in hb. Cardot), (Arséne 4726) ; Esperanza (4742).
Valle de México: Desierto (Bro. Amable 1419).

The leaves are triquetrous, squarrose-spreading when moist; the
areolation and papillae suggest the genus Leptodontium. I found in
no. 4726 a single capsule. The fruit of this species not being known,
I describe it:

Folia perichaetialia erecta, vaginantia; pedicellus flavidus, 1 cm.
altus ; capsula parum inclinata, oblongo-cylindrica ; operculum oblique
rostratum.

Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL: 85

LEPTODONTIUM ANGUSTINERVE Thér., sp. nov.

Puebla: Esperanza (4741 ; 4743).

Sterile. Caespites sat densi, intense virides. Caulis 4-5 cm. longus,
simplex vel parce ramosus, laxe foliosus. Folia sicca erecta, parum
patula, flexuosa, madida patula, oblongo-lanceolata, sensim acuminata,
acuta, carinata, decurrentia, marginibus integerrimis, in medio reflexis,
1.2 mm. longa, 0.3 mm. lata; costa basi 45 », percurrente, dorso papil-

Fic. 5.—Leptodontium angustinerve Thér. 1, 2, cauline leaves; 3, acumen;
4, middle cells; 5, basal cells; 6, papillae from the dorsal side; 7, cross-section
near the lower third of leaf; 8, 9, fragments of 7.

losa; cellulis laminae quadrato-hexagonis, incrassatis, 12X10 p, pa-
pillosis, papillis 1-2, elatis, obtusis, cellulis apicalibus laevibus, cellulis
basilaribus rectangularibus, incrassatis, laevibus.

This species is rather close to L. filescens Hampe, but is distin-
guished by the narrow acute leaves, the narrow costa, the incrassate
cells with only 1 or 2 large papillae, the smooth apical cells, and the
more compact basal areolation.

LEPTODONTIUM ARSENEI Ther., sp. nov.

Morelia: Cerro San Miguel (5073).

Sterile. Caespites sat densi, compacti, fusco-virides. Caulis 4-5 cm.
altus, dense radiculosus, regulariter foliosus. Folia sicca crispula,
humida patulo-squarrosa, vaginantia, e basi erecta, breviter oblonga,
superne dilatata, sat abrupte in acumen angustum, canaliculatum, sub-
obtusum contracta, marginibus integris, in medio foli revolutis,
1.3 mm. longa, 0.5 mm. lata; costa basi 70 m, percurrente, dorso pa-

INOS MEXICAN MOSSES—THERIOT Il

pillosa, cellulis laminae opacis, vix distinctis, quadratis, 7-8 », dense
et tenuiter papillosis, cellulis basilaribus juxtamarginalibus linearibus,
laevibus, hyalinis, internis rectangularibus, papillosis, paulum chloro-

phyllosis.

Fic. 6.—Leptodontium Arsenci Thér. 1, 2, 3, cauline leaves; 4, cross-section
in the acumen; 5, fragment of 4; 6, upper cells; 7, cells in the upper part of
leaf base toward a; 8, marginal cells of leaf base at b.

This species also belongs to the group of L. filescens. It may be
separated from it by the more elongated stems, by the leaves with a
short sheathing base, by the narrower acumen (subobtuse and not
apiculate), and the smaller median cells.

LEPTODONTIUM FILESCENS (Hampe) Mitt. Musc. Austr. Amer. 50. 1869
Valle de México ; Desierto, upon bark associated with Rozea stricta
Besch. (Bro. Amable 1444).
A Colombian species, new to Mexico.

LEPTODONTIUM SQUARROSUM (Hook.) Par. Ind. Bryol. 732. 1896
Holomitrium serratum (Schimp.) C. M. Syn. 2: 587. 1851.
Valle de México (Bro. Amable): Desierto (1431); Rio Frio

(1396).
The last plant is a form with strongly undulate leaves.

LEPTODONTIUM EXASPERATUM Card. Rev. Bryol. 36: 74. 1909
Valle de México: Rio Frio (Bro. Amable 1690).

LEPTODONTIUM sp.

Valle de México (Bro. Amable): Puenta de la Venta (1391):
Desierto (1439 p. p.).

The present species, which unquestionably is new, was recognized
as such independently by my friend Edwin B. Bartram, who will
shortly publish a description.

I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

TORTULA SUBNIGRA Mitt. Musc. Austr. Amer. 164. 1869

The position of this moss has remained uncertain until now. Jaeger
classified it as a Barbula. Brotherus* thought that perhaps it was a
Didymodon. Lastly, J. Cardot, who saw its capsules in good condition,
verified the absence of a peristome and asked~ if it might not be
convenient to separate Mitten’s species from true Tortula to constitute
the type of a new subgenus.

Mr. Bartram and myself think that we must go further. Trans-
verse sections of the leaves show that the costa possesses the structure
of the Trichostomoideae and not of the Pottioideae. Another peculiar
fact is that the lamina is composed of two layers of cells, except at
the margin. These noteworthy characters, in addition to that presented
by the sporophyte, justify the creation of a new genus. We are
happy to dedicate this Mexican genus to our eminent friend J. Cardot,
whose last and precious works contributed so much to a better under-
standing of the bryological flora of Mexico, and who has enriched
it with such a large number of new species.

NEOCARDOTIA Thér. & Bartr., gen. nov.

Caulis erectus. Folia sicca crispata, humore patulo-squarrosa, cari-
nata, acuta, marginibus inferne revolutis, superne serratis ; rete opaco,
bistratoso ; cellulis basilaribus linearibus, laevibus, hyalinis, caeteris
minutis, dense papillosis ; costa breviter excurrente, in sectione trans-
versali e strato medio cellulis eurycystis et stereidis ventralibus et
dorsalibus composita. Folia perichaetialia perlonga, erecta, vaginantia ;
capsula erecta, cylindrica, symmetrica, gymnostoma, annulata ; opercu-
lum breviter conicum, cellulis recte seriatis ; sporae laeves.

The affinities of the genus Neocardotia are with the genus Lepto-
dontium, as much through the habit of the plant as by its serrated
cauline leaves and strongly sheathing perichaetial leaves, 2 to 3 times
larger than the stem leaves.

A single species :

NEOCARDOTIA SUBNIGRA (Mitt.) Thér. & Bartr., comb. nov.
Tortula subnigra Mitt. Muse. Austr. Amer. 104. 1860.

Caespites nigrescenti-virides ; caulis 2-4 cm. altus. Folia 3.5-4 mm.
longa, 1 mm. lata; costa basi 60-80 », dorso rugulosa ; cellulis laminae
6-7 », paulum incrassatis, papillis minutis, obtusis; pedicellus 1 cm.

a ap & Pe Nat. Pflanzentam. Tee OF; L902'5 ed. 2 10. 2/735 1024.
S /
* Rev. 3ryol. 38: IOI. TOLL.

NO. 4 MEXICAN MOSSES—THERIOT L3

longus, sporae diam. 12-18 ».—Often mixed with other mosses, and
almost always with Braunia secunda.

Mexico: Loc. class. (Humboldt) ; La Cima (Barnes & Land 373
p. p.). Valle de México (Bro. Amable): Salazar (1294, 1716) ;
Contreras (1470). Hidalgo: Mineral del Chico (Orcutt 6841).
Arizona: Santa Rita Mts. (Bartram 813).

The plant from Arizona shows some differences as compared with
the Mexican plant, slight differences it is true, but worthy of mention;
it is a little more siender, the leaves are more shortly acuminate, and
their cells less regularly bistratose.

HYOPHILA DENTATA Card. Rev. Bryol. 40: 36. 1913
Morelia: (7896).

HYOPHILA MEXICANA Thér., sp. nov.

Valle de México: Tizapan, on earth (Bro. Amable 1613 p. p.).
Sterilis. Caulis perbrevis ; folia 2 mm. longa, 0.6 mm, lata, margini-
bus integerrimis, planis, superne paulum involutis; cellulis laminae
> 7, ’
hexagonis, papillosis, parietibus tenuibus, diam. 6-7 ,, costa latissima,
S Tat
go-100 pw, in mucronem brevem acutum excurrente.

Fic. 7.—Hyophila mexicana Thér. 1, leaf; 2, acumen; 3, cross-section toward
the middle; 4, median cells; 5, basal cells. Hyophila subangustifolia Thér.
6, 7, cauline leaves: 8, median cells; 9, marginal cells toward a; 10, basal cells;
II, perichaetial leaf; 12, 13, capsules.

In form, size, and areolation of the leaves this species may be com-
pared with H. Bescherellei C. M. It differs in the very short stems
and perfectly entire leaves, and in the hyaline basal cells, with thin,
soft walls, These last characters distinguish our species from H. frag-
ilis Card.

14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

HYOPHILA SUBANGUSTIFOLIA Thér., sp. nov.

Valle de México: Tizapan, on earth, associated with the preceding
species and other mosses (Bro. Amable 1613 p. p.).

Dioica. Caespites incohaerentes laxiusculi, virides. Caulis perbrevis,
2-3 mm. altus. Folia sicca crispula, humida erecto-patentia, 1.3-1.6
mm. longa, 0.40—0.45 mm. lata, oblongo-lanceolata, obtusa, breviter
mucronata, marginibus planis, integris, superne parum involutis ; costa
60 p» lata, breviter excedente, dorso laevi; cellulis basilaribus hyalinis,
ad costam elongate rectangulis, ad marginem brevioribus, superiori-
bus minutis, quadratis vel hexagonis, papillosis, obscuris, diam. 7-8 p.
Pedicellus erectus, 4 mm, longus ; capsula oblonga, gymnostoma, annu-
lata; operculum oblique et longe rostratum, capsulam subaequans ;
sporae papillosae, diam. 18 ». Flos masculus ignotus.

I can find no better comparison for this than H. angustifolia Par. &
Ren., from Madagascar. It differs from the latter in its shorter stems,
oblong-lanceolate leaves (wider, scarcely involute above, and not
cucullate at the apex), more compact areolation, and longer opercu-
lum.

WEISIOPSIS STENOCARPA Thér., sp. nov.

Valle de México: Desierto (Bro. Amable 1205 p. p.). Growing
as isolated stems among other mosses, such as Didymodon oeneus and
Campylium hispidulum var. Sommerfeltii.

Jl

Fic. 8.—Weisiopsis stenocarpa Thér. 1, 2, cauline leaves; 3, acumen; 4, me-
dian cells; 5, basal cells; 6, antheridial bud; 7, perichaetial leaf; 8, capsule;
g, annulus; 10, fragment of peristome.

Autoica, pusilla. Caulis ascendens, 2-5 mm. altus. Folia sicca cri-
spula, humore patula, lanceolato-acuminata, acuta, basi plicatula,
marginibus planis, integris, 2-2.2 mm. longa, 0.3-0.4 mm. lata, costa

NO. 4 MEXICAN MOSSES—THERIOT I

an

basi 60 », percurrente vel breviter excedente; rete opaco, papilloso,
papillis densis, minutis, cellulis quadrato-hexagonis, 8-9 p, cellulis
basilaribus laxioribus, hyalinis, rectangularibus. Folia perichaetialia
longiora (3 mm.) ; pedicellus erectus, pallidus, 5-6 mm. longus; cap-
sula angustissime cylindrica 1.5 mm. longa, 0.26 mm. crassa, annulata,
peristomium sub ore insertum, dentes lineares, integri, tenuiter pa-
pillosi; sporae laeves, 12 » crassae. Calyptra? Operculum?

Comparable to WW. stomatodonta (Card.) Broth. in form and size
of capsule, but very different in its autoicous inflorescence, longer,
more narrowly acuminate leaves (with margins not involute), looser
basal areolation, and entire peristome teeth.

Furthermore, does Cardot’s species, which has the peristome teeth
divided into two branches, really belong to the genus Weisiopsis?

DIDYMODON (Erythrophyllum) PATENTIFOLIUS Thér., sp. nov.

Valle de México: Xoquiapan (Bro. Amable 1676); Mixcoac
(Arséne 9442).

Dioicus, tenellus, obscure viridis. Caulis erectus, simplex, vix
2 mm. altus. Folia sicca crispula-patula, humida patentia, carinato-

Fic. 9.—Didymodon patentifolius Thér. 1, 2, cauline leaves; 3, median cells:

x

4, basal cells; 5, 6, cross-sections in acumen; 7, cross-section of costa near
base; 8, perichaetial leaf; 9, dry capsule; 10, moist capsule; 11, peristome; 12,
fragment of peristome.

concava, lanceolato-ligulata, acuta, marginibus anguste revolutis, in-

tegris, 1.5-1.6 mm. longa, 0.3 mm. lata; costa papillosa, basi 60

crassa, percurrente ; cellulis laminae hexagonis, chlorophyllosis, obscu-

ris, papillosis, 8-9 », basilaribus laxis, hyalinis, teneris, oblongo-hexa-

gonis, vel rectangularibus. Folia perichaetialia similia sed majora,
2

16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

haud vaginata ; pedicellus purpureus, erectus, 10-12 mm, longus ; cap-
sula oblonga (2 mm. longa c. operculo), annulata, peristomii dentes
e€ membrana basilari humillima erecti (0.6 mm. alti), fere usque ad
basin in 2 crura filiformia papillosa divisi; sporae laeves, 12-15 p;
operculum conico-rostratum.

Distinguished at a glance from D. oeneus by its small size. The leaf
margin is so closely revolute that at first sight the border seems to be
thickened and formed of two layers of cells.

DIDYMODON OENEUS (C. M.) Schimp. in Besch. Prodr. Bryol. Mex. 28. 1871
Trichostomum Oeneum C. M. Syn. 2: 628. 1851.

Puebla: Esperanza (4802). Valle de México: Desierto (Bro.
Amable 1205 p. p., 1417).

DIDYMODON CAMPYLOCARPUS (C. M.) Broth. in E. & P. Nat. Pflanzenfam.
I°: 405. 1902

Trichostomum campylocarpum C. M. Syn. 2: 628. 1851.
Valle de México: Desierto (Bro. Amable 1248).

DIDYMODON INCRASSATO-LIMBATUS Card. Rev. Bryol. 36: 81. 1909

Morelia: (7914); Rincon (4567); Loma Santa Maria (4886) ;
Cerro San Miguel (5085). Valle de México: Tlalpan (Bro. Amable
1348, 1349).

According to these specimens the peristome teeth are a little twisted,
and not exactly straight. Furthermore, the cells of the operculum are
arranged in oblique rows, a character which is also found in Pringle’s
no. 10588. No. 4567, from Rincon, is a form with acute leaves whose
margins are less strongly thickened.

DIDYMODON FUSCO-VIRIDIS Card. Rev. Bryol. 36: 83. 1909
Valle de México: San Juanico (Bro. Amable 1264).

DIDYMODON PUSILLUS Card. Rev. Bryol. 36: 82. 1909
Valle de México (Bro. Amable): Desierto (1207 p. p., 13am
1615); Rio Frio (1746) ; Xoquiapan (1752).
A robust form, the stems tall, up to 2 cm., leaves wider at the base,
inargins plane or slightly reflexed. These plants seem to mark a
transition toward the last preceding species.

DIDYMODON DIAPHANOBASIS Card. Rev. Bryol. 37: 125. 1910
Valle de México (Bro. Amable): Contadero (1363); Rio Frio
(1704).

NO. 4 MEXICAN MOSSES—THERIOT Dy,

DIDYMODON MEXICANUS Besch. Prodr. Bryol. Mex. 28. 1871

Puebla: Hacienda Alamos (4764). Valle de México: San Juanico
(Bro. Amable 1322, 1332).

DACTYLHYMENIUM PRINGLEI (E. G. Britt.) Card. Rev. Bryol. 36: 72. 1909

Puebla: (9493); Rancho Santa Barbara (4598, 4811).
These specimens represent a form with less papillose leaves and
a nearly smooth costa.

BARBULA BESCHERELLEI Sauerb.

Puebla: Cerro Guadalupe (686a, 799). Morelia: Punguato
(5048) ; Campanario (7564) ; Cerro San Miguel (5087). Tlaxcala:
(613). Valle de México: Desierto, Contadero (Bro, Amable).

BARBULA BESCHERELLEI Sauerb. var. CRASSINERVIA Theér., var. nov.

Distr. Federal: Mixcoac (9470, 9473).

Folia basi cordata, sat abrupte contracta, costa lata, 90 »; folia
perichaetialia late ovata vel oblonga, breviter acuminata; capsula
anguste cylindrica.

BARBULA BESCHERELLEI Sauerb. var. STENOCARPA Card.

Valle de México: Xoquiapan (Bro. Amable 1749).

BARBULA ALTISETA Card. Rev. Bryol. 36: 83. 1909

Tlaxcala: (621, 720).
A robust form, the stems longer and the leaves less strongly revo-
lute than usual.

BARBULA GRACILIFORMIS Schimp. in Besch. Prodr. Bryol. Mex. 35. 1871
Puebla: Cerro Guadalupe (668, 669, 680, 801). Distr. Federal:

Mixcoac (9460, 9461). Tlaxcala: Acuitlapilco (725).
Nos. 680 and 8o1 are more robust forms.

BARBULA GRACILESCENS Schimp. in Besch. Prodr. Bryol. Mex. 34. 1871

Puebla: (600). Morelia: (7946, 7948) ; Bosque San Pedro (4577
P- p., 4578, 4581, 4925, 4926) ; Loma Santa Maria (4888, 4904) ;
Jests del Monte (7622). Distr. Federal: Mixcoac (9431, 9435, 9454,

9456, 9458, 9469, 9480).

18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Barbula altiseta Card., B. graciliformus Schimp., and B. gracilescens
Schimp. are very closely allied species, especially the last two.

According to authentic specimens 8. gracilescens is distinguished
from B. graciliformis by its flexuose leaves when dry (not stiff and
subimbricated), often narrower, with the acumen a little longer and
thinner. I have not detected any differences in areolation, costa, or
recurvature of the leaf margins. And inasmuch as I have found forms
that could not be definitely connected with either species, | am much
inclined to believe that these so-called species are in reality but forms
of a single one.

BARBULA TERETIUSCULA Schimp. in C. M. Syn. 1: 614. 1849

Puebla: (g07); Fort de Lorette (4622). Morelia: Loma Santa
Maria (7864 p. p.).

BARBULA SUBTERETIUSCULA Card. Rev. Bryol. 36: 85. 1909

Puebla: Rancho Posadas (4808).

BARBULA BOURGEANA Besch. Rev. Bryol. 36: 35. 1909

Puebla: (4996); Rio San Francisco (4999); Hacienda Alamos
(4637).

BARBULA ORIZABENSIS C. M. Linnaea 40: 638. 1876

Puebla: Hacienda Alamos (578); Cerro Guadalupe (4616) ;
Rancho Guadalupe (4590 p. p.). Distr. Federal: Muixcoac (9464,
9467 ).

All of these plants are sterile but are identical with the type, which
I have been able to examine.

The author compares this species with B. spiralis Schimp. It is
distinguished, he says, by its less twisted leaves and its cylindrical
capsule. These characters are rather intangible and valueless to one
who has seen a series of specimens of B. spiralis, Happily other
important and obvious characters are available: In B. orizabenstis the
margins of the leaves are merely reflexed and not revolute; they are
plane at the apex, the costa is thinner and not widened in the acumen,
and finally the areolation is chlorophyllose and papillose almost to
the base. These characters establish the true position of B. orizabensis
by the side of B. unguiculata and not of B. spiralis.

| have noticed in all my specimens, as well as in Pringle’s no. 10574,
the presence of abundant moniliform propagula in the leaf axils.

NO. 4 MEXICAN MOSSES—THERIOT 19

BARBULA SPIRALIS Schimp. in C. M. Syn. 1: 622. 1849
Puebla: (892) ; Mayorazgo (4673, 5975) ; Cerro Guadalupe (655,
663, 675, 681, 690, 691, 692, 693, 4619, 4620) ; Rancho Guadalupe
(728, 4591, 4602 p. p., 4604 p. p., 4607, 4609 p. p.); Rio Alseseca
(7o1); Malinche (6003). Distr. Federal: Tlaquecomeca (9478) ;
Mixcoac (9446, 9447, 9449, 9482). Tlaxcala: Acuitlapilco (741,
742). Morelia: Campanario (7922) ; Andameo (4830, 4843, 4844) ;
Cuincho (5082) ; Cerro Azul (5053) ; Loma del Zapote (7509). Valle
de México (Bro. Amable): San Juanico (1231, 1262); Texcoco
(1287) ; Guadalupe (1228) ; Pefion de los Bafios (1256).
Of all the mosses known to Mexico this species seems to be the
most common.

BARBULA EHRENBERGII (Lor.) Fleisch. var. MEXICANA Thér., var. nov.

Nuevo Leon: Monterrey (Bro. Abbon 10969).
A forma typica differt foliis valde revolutis,

BARBULA (Hydrogonium) RUBRICAULIS Thér., sp. nov.
Nuevo Leon: Monterrey (Bro. Abbon 10968).
Sterilis. Caespites densi, glauco-virides. Caulis erectus, simplex,
ruber, 1.5-2 cm. altus, basi terra obrutus, laxe foliosus. I[*olia sicca

Fic. 10.—Barbula rubricaulis Thér. 1, 2, leaves; 3, acumen; 4, median and
marginal cells; 5, basal cells. Barbula Abbonti Thér. 6, 7, leaves; 8, cross-
section of leaf; 9, acumen; 10, median cells; 11, basal cells.

erecto-flexuosa, madida erecta, paulum patula, oblongo-ligulata, sub-
obtusa, concava, decurrentia, marginibus integris, planis vel uno
latere parce reflexis ; costa basi 60-70 p, apicem attingente, rete chlo

20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

rophylloso, cellulis subquadratis, 10 x 8 p, laevibus, parietibus incrassa-
tis, rete basilari laxiore, cellulis juxtacostalibus rectangularibus, hya-
linis, margines versus quadratis vel breviter rectangularibus. Caetera
desunt.

In color of the stems and in form, size, and areolation of the
leaves B. rubricaulis is very close to B, dialytrichoides Thér., from
China, differing only in its nearly smooth areolation. I have not seen
the fruit.

BARBULA (Hydrogonium) ABBONII Thér., sp. nov.

Nuevo Leon: Monterrey (Bro. Abbon 10970).

A B. rubricauli proximo differt: caule viride, foliis siccis valde
patulis, humidis subsquarrosis, brevioribus, 1.4 mm. longis, 0.60-0.65
mm. latis, costa latiora, go-100 p crassa, cellulis mediis majoribus,
15-20 wX 10-15 p.

The leaves show the median areolation of B. Ehrenbergiana var.
mexicana and the basal areolation of B. rubricaulis; the leaf margins
are narrowly revolute three-fourths of the way up from the base.

BARBULA (Streblotrichum) CALCAREA Thér., sp. nov.

Morelia: Loma Santa Maria, on calcareous rocks (4891). Valle
de México: Desierto (Bro. Amable 1620).

Fic. 11.—Barbula calcarea Thér. 1, 2, 3, cauline leaves; 4, median cells;
5, basal cells; 6, 7, perichaetial leaves; 8, deoperculate capsule.

Pusilla. Caulis gracilis, simplex, 3-5 cm. altus, laxe foliosus. Folia
mollia, sicca appressa, humida patula, elliptica vel oblonga, late ro-
tundata, decurrentia, marginibus integerrimis, inferne planis, superne

NO. 4 MEXICAN MOSSES—THERIOT 21

valde revolutis, 0.9 mm. longa, 0.4 mm. lata; costa basi 60 », continua,
dorso papillosa; cellulis mediis opacis, indistinctis, dense papillosis,
diam. 10-12 p, superioribus minoribus, rete basilari laxiore, pellucido,
cellulis rectangularibus, chlorophyllosis, plus minus papillosis, infimis
laevibus. Folia perichaetialia pauca, intima 2 duplo longiora, convo-
iuta, longe vaginantia, apice lingulata, obtusa; pedicellus tenuis, palli-
do-luteus, 7-8 mm, longus ; capsula (immatura) minuta, anguste-cylin-
drica ; operculum rostratum, Caetera ignota.

By its slender habit, its loosely foliate stems, and leaves revolute
in the upper two-thirds, the present species is immediately distin-
guished from B. hypselostegia Card. and B. Muenchii Card., both of
which also have obtuse leaves.

BARBULA (Streblotrichum) STENOTHECA Thér., sp. nov.

Valle de México; Rio Frio, on earth (Bro. Amable 1726).
Dioica. Caespites sat densi, obscuro-virides. Caulis erectus, flexu-
osus, gracilis, remote foliosus, 10-15 mm. altus. Folia sicca incurvato-

, (i

i.

Hy,
A

Fic. 12—Barbula stenotheca Thér. 1, cauline leaf; 2, acumen; 3, median
leaf cells; 4, basal cells; 5, 6, cross-sections of leaf; 7, cross-section of costa
near base; 8, 9, perichaetial leaves; 10, dry capsule; 11, calyptra; 12, peristome.

crispata, madida patula, oblongo-lanceolata, subobtusa, breviter apicu-
lata, marginibus integris, usque ad apicem revolutis, 2 mm. longa,
0.6 mm. lata, costa valida, basi 60 » crassa, dorso laevi, breviter ex-

22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

currente; rete opaco, cellulis hexagonis, 10 pm, parietibus tenuibus,
dense et minute papillosis, cellulis inferioribus rectangularibus, pellu-
cidis, laevibus, parietibus firmis. Folia perichaetialia numerosa, ex-
terna patulo-subsquarrosa, intima convoluta, longe vaginantia, decolo-
1ata, longissima, 4 mm. longa, obtusiuscula, apiculata ; pedicellus ruber,
15 mm, longus; capsula erecta, angustissime cylindrica vel arcuato-
cylindrica, 4 mm. longa, 0.4 mm. crassa; calyptra } partem capsulae
obtegens ; operculum longe conicum, 1.3 mm. longum; annulus sim-
plex; peristomium elatum, 1 mm. altum, dentibus valde contortis,
membrana basilari brevi; sporae laeves, 8-9 vp.

Differs widely from the other Mexican species of the section Stre-
blotrichum in size and habit, and especially in the dimensions of the
capsule.

MORINIA EHRENBERGIANA (C. M.) Thér., comb. nov.

Barbula Ehrenbergiana C. M. Syn. 1: 636. 1849.
Barbula trichostomoides Besch. Prodr. Bryol. Mex. 38. 1871.
Morinia trichostomoides Card. Rev. Bryol. 37: 124. I910.

Valle de México: Desierto, on earth (Bro. Amable 1240).

While studying this specimen I recognized, by a happy chance, its
identity with Barbula Ehrenbergiana C. M. and with Morinia tri-
chostomoides (Besch.) Card. The name established by Muller has
priority, hence the above new combination.

ALOINA CALCEOLIFOLIA (Spruce) Broth. in E. & P. Nat. Pflanzenfam.
I°: 428. 1902

Puebla: (704) ; Mayorazgo (4672).

ALOINELLA CATENULA Card. Rev. Bryol. 36: 76. 1909

Valle de México (Bro. Amable) : Desierto (1207 p. p., 1217 p. p.) ;
Salazar (1296 p. p.).
Terrestrial, in isolated bits, always associated with other mosses.

TORTULA PARVA Card. var. LATIFOLIA Thér., var. nov.

Puebla: (4509); Rancho Santa Barbara (4593, 4600) ; Hacienda
Alamos (4720). Morelia: Cerro Azul (4933). Valle de México:
Cartridge Factory (Bro. Amable 1459 p. p.).

A forma typica differt statura robustiore, foliis longioribus et duplo
latioribus, 1.2-1.8 mm. X 0.6-0.8 mm.

NO. 4 MEXICAN MOSSES—THERIOT 23

‘“TORTULA AMPHIDIACEA (C. M.) Broth. in E. & P. Nat. Pflanzenfam.
I°: 424. 1902

Barbula? amphidiacea C. M. Linnaea 38: 639. 1874.

Puebla: Rancho Santa Barbara (4810 p. p.). Morelia: Pare San
Pedro, c. fr. (4587) ; Cerro Azul (4932). Valle de México: Conta-
dero (Bro. Amable 1300, 1308 p. p., 1360).

The plant from San Pedro bears capsules. The fruit being un-
known, I describe it: Pedicel short, 6-7 mm., almost hidden by the

6
iS 'O BD ww og OOSQYT
CX OEY FS ESSeE

2 ie cy Y (s @ \ )
<90 an , OOOGS&

: 0
ib OO S 9090
', . A ay ee Oo
2OQ*gOO0G

x 200 a

&

Fic. 13.—Tortula amphidiacea. 1, 2, stem leaves; 3, cross-section of a leaf;
4, the same, costa; 5, upper marginal cells toward a; 6, median cells; 7, basal
cells; 8, propagulae.

long and numerous innovations ; capsule oblong-cylindrical. All the
capsules are old and have neither operculum nor peristome.

The species is well characterized in other particulars: The leaf
is acute and entire, more or less marginate at the base with several
rows of narrow cells, the lamina with differentiated, incrassate and
smooth cells ; the costa is percurrent or short-excurrent ; the stems bear
oblong propagula.

TORTULA RIPICOLA Ther., sp. nov.

Valle de México: Morales, on earth, bank of a small stream, asso-
ciated with Fissidens tortilis Hampe & C. M. (Bro. Amable 1596
Das)

Sterilis, pusilla. Caulis inferne denudatus, superne rosulato-folio-
sus, 0.5-I cm. altus, in axillis foliorum propagula numerosa, fusca,
sphaerica gerens. Folia sicca paulum crispula, humore patula, ovato-
lanceolata, obtusa vel raro subacuta, breviter mucronata, marginibus

24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

integerrimis, toto ambitu revolutis, 1.5 mm. longa, 0.6-0.7 mm. lata;
rete opaco, dense et minute papilloso; cellulis quadrato-hexagonis,
chlorophyllosis, haud incrassatis, diam. 6 p, basilaribus paucis, hyalinis,
laevibus, breviter rectangularibus, costa basi 60 », dorso minute pa-

Fic. 14.—Tortula ripicola Thér. 1, 2, 3, leaves; 4, 5, acumens; 6, median leaf
cells; 7, basal cells; 8, 9, 10, 11, cross-sections of a leaf; 12, propagulae.

pillosa, in mucronem brevem excedente, structura in sectione trans-
versali ut in genere.

Group of T. papillosa Wils. Well characterized by its small size, its
leaves broadened at base, revolute all around and short-mucronate,
and its very compact areolation.

TORTULA FRAGILIS (Tayl.) Mitt. Musc. Austr. Amer. 172. 1869
Tortula confusa Card. Rey. Bryol. 36: 87. 1900.
Tortula Pringlet Card. Rev. Bryol. 36: 87. 1909.
Puebla: (4510); Esperanza (4940). Distr. Federal: Tlalpam
(9494).

For the synonymy of this species the reader is referred to a recent
note by Mr. E. B. Bartram.’ While studying Bro. Arséne’s collection
I also formed a clear idea that Cardot’s two species could not be main-
tained, as the author himself apparently suspected.

TORTULA OBTUSISSIMA (C. M.) Mitt. Musc. Austr. Amer. 174. 1869

Puebla: Cerro Guadalupe (667, 673). Tlaxcala: (606). Valle de
México (Bro. Amable): San Juanico (1232) ; Tenayuca (1376) ;
Xoquiapan (1675).

‘ Bryologist 29: 53. 1926,

NO. 4 MEXICAN MOSSES—THERIOT

to

cn

TORTULA OBTUSISSIMA (C. M.) Mitt. var. CONNECTENS (Card.) Thér.,
comb. nov.

Tortula connectens Card. Rey. Bryol. 36: 87. 1900.

Puebla: Fort Guadalupe (4621). Morelia: Cerro Azul (4531).
Distr. Federal: Mixcoac (9485).

The characters indicated to separate T. connectens from T. obtu-
sissima do not appear constant, and I combine the two species.

GRIMMIACEAE (continued)
GRIMMIA INVOLUCRATA Card. Rev. Bryol. 36: 105. 1909

Valle de México (Bro. Amable): Tlalpam, c. fr. (1448) ; Zaca-
tenco, ster. (1352).

GRIMMIA PRAETERMISSA Card. Rev. Bryol. 36: 105. 1909

Valle de México (Bro. Amable): Rio Frio, on rocks (1401 p. p.,
1681).

The capsule is sometimes pale and scarcely exserted, sometimes
brown, longer-pedicellate, and well exserted.

GRIMMIA CALIFORNICA Sull. in U. S. Rep. Expl. Miss. Pacif. 4: 187, pl. 4.
1856

Valle de México (Bro. Amable): Rio Frio, intimately mixed with

the preceding species (1401 p. p.) ; Llano Grande, alt. 3,700 meters

(1724, 1734).

GRIMMIA PULLA Card. Rev. Bryol. 36: 106. 1909

Valle de México: Contreras, on rocks (Bro. Amable).

The plants are fruited, but the over-ripe capsules have lost their
peristomes. Sporophyte pseudo-lateral, because of the 1-3 rather
elongated innovations borne under the male flower ; pedicel 2-2.5 mm.
long, suberect when dry, arcuate when moist ; capsule oblong, strongly
furrowed.

SPLACHNACEAE
TAYLORIA (Eutayloria) TORTELLOIDES Thér., sp. nov.

Hidalgo: El Chico, 2,600 meters (Bro. Amable 1587 p. p.). Grow-
ing as isolated stems among other mosses, especially with Bryuwi
Ehrenbergianum.,

26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Dioica? Flos masculus ignotus. Caulis brevis, vix 1 cm. altus,
simplex vel parce ramosus, inferne denudatus, radiculosus, pauci-
foliatus, apice rosulato-foliosus. Folia sicca valde crispata, nitida,
humore patentia, oblongo-spathulata, e basi contracta, decurrentia,
apice rotundata, apiculata, apiculo brevi, obliquo, marginibus planis,
inferne paulum reflexis, integris vel remote et obtuse denticulatis, 3
mm. longa, 2 mm. lata; costa basi 120 p, raptim attenuata, sub apicem
evanescente, in sectione transversali ut in genere; cellulis mediis
hexagonis, chlorophyllosis, parietibus tenuibus, 60 x 30 », marginalibus

Fic. 15.—Tayloria tortelloides Thér. 1, lower leaf; 2, comal leaf; 3, upper
cells at a; 4, median cells at Db; 5, marginal cells toward c; 6, basal cells; 7,
young dry capsule; 8, moist capsule; 9, wall of capsule orifice; 10, fragment ot
peristome.

(2-3 ser.) elongatis, inanis, cellulis basilaribus rectangularibus, parce
chlorophyllosis. Folia perichaetialia similia, intima minora; pedicellus
erectus, perbrevis, 1.5 mm. longus, laevis, pallidus; capsula sub-
cylindrica, brevicollis, 2 mm. longa; operculum obtuse conicum, colu-
mella inclusa, peristomii 16 dentes liberi, opaci, dense papillosi, 0.4
mm, longi; sporae laeves, 15-18 » crassae. Calyptra?

The extremely short pedicel and the entire leaves, rounded apicu-
late, broadly spatulate, and shrivelled when dry (like some Tortula),
readily distinguish this plant from the other species of the subgenus
Eutayloria.

NO. 4 MEXICAN MOSSES—THERIOT

No
N

BRYACEAE (continued )
MIELICHHOFERIA SAINT-PIERREI Thér., sp. nov.

Valle de México: Lerma; leg. Marius Saint Pierre (Bro. Amable
1685).

Paroica, laxiuscula caespitosa, tenella, viridis. Caulis julaceus,
2-3 mm. altus, ramis erectis, vix 5 mm. longis. Folia caulina conferta,
imbricata, ovato vel oblongo-lanceolata, acuminata, acuta, 0.8-1.2 mm.
longa, 0.4-0.5 mm. lata, marginibus parce et anguste reflexis, superne
remote denticulatis vel sinuolatis, costa basi 30-36 » crassa, subpercur-

Fic. 16.—Mielichhoferia Saint-Pierrei Thér. 1, 2, 3, stem leaves; 4, leaf from
an innovation; 5, median cells; 6, apical cells; 7, basal cells; 8, moist capsule;
9, fragment of peristome and annulus.

rente, rete membranaceo, cellulis elongate rhomboideis, 60-70 px 12 p,
marginibus angustioribus, basilaribus rectangularibus; folia ramea
angustiora, marginibus erectis. Folia perichaetialia caulinis similia ;
pedicellus erectus, 8-12 mm. longus; capsula inclinata vel subhorizon-
talis, symmetrica, oblongo-cylindrica, collo attenuato instructa ; annu-
lus latus ; peristomium simplex, membrana basilari subnulla, processus
angusti, 10 p lati, granulosi haud appendiculati; sporae sublaeves,
12-15 »; operculum convexum, mamillatum.

LEPTOBRYUM PYRIFORME (L.) Wils.

Valle de México: Tlalpam (Bro. Amable 1246 p. p.).

EPIPTERYGIUM MEXICANUM (Besch.) Broth.

Valle de México (Bro. Amable): Puente de la Venta (1400) :
Santa Rosa (1513); Desierto (1642).

28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

EPIPTERYGIUM MEXICANUM (Besch.) Broth. var. ANGUSTIRETE Thér.,
var. nov.
Valle de México: Contreras (Bro. Amable 1659).
Folia angustiora, cellulis chlorophyllosis, angustioribus,

MNIOBRYUM INTEGRUM (Card.) Broth. in E. & P. Nat. Pflanzenfam.
ed. 2, 10: 363. 1924
Webera integra Card. Rey. Bryol. 40: II. 1913.
Valle de México: Contreras (Bro. Amable 1478).

WEBERA SPECTABILIS (C. M.) Jaeg.

Webera cylindrica (Mont.) Schimp. in Besch. Prodr. Bryol. Mex. 52. 1871.

I have received from Bro. Amable rather numerous collections
of Webera of the present relationship, but frankly, I have not suc-
ceeded in distinguishing W. cylindrica from W. spectabilis. The
characters I had considered distinctive are rarely combined on the same
plant and all of them show a rather wide variability, as, for instance,
in the width of the leaf, the recurvature of the margin, the width
of the costa and of the cells, and the length of the capsule. My con-
‘clusion is that we must unite the two species. Webera spectabilis has
priority.

WEBERA PSEUDO-BARBULA Thér., sp. nov.

Valle de Mexico (Bro. Amable): Desierto (1630, 1643) ; Con-
treras (1658) ; Lerma (1684).—In all these localities the plants grow
in company with Anomobryum filiforme var. mexicanum, a remarkable
fact of association.

Fic. 17.—Webera pseudo-Barbula Thér. 1, stem leaf; 2, marginal and median
cells; 3, apical cells; 4, leaves of innovations; 5, apical cells of 4; 6, marginal
and median cells of 4; 7, propagula; 8, perichaetial leaf; 9, dry capsule; 10,
moist capsule; 11, peristome (fragment) ; 12, fragment of annulus.

NO. 4 MEXICAN MOSSES—-THERIOT 29

Dioica. Caespites laxi, virides. Caulis brevis, 5 mm. longus, superne
innovationibus elongatis, 10-15 mm. longis, in axillis foliorum superi-
orum propagula fusca, numerosa, subglobosa gerens. Folia sicca
erecta, parum flexuosa, humore erecto-patula, ovato-acuminata, 1.2-1.5
mm. longa, 0.5 mm. lata, marginibus planis, interdum anguste revo-
lutis, elimbatis, integerrimis, apice denticulatis ; costa basi 60 p, sen-
sim attenuata, ante apicem evanescente ; cellulis anguste rhomboideis,
chlorophyllosis, 70-90 » x 8-9 pw, ad marginem angustioribus ; folia in-
novationis similia sed minora. Folia perichaetialia longiora, ovato-
lanceolata, acuminata, intima anguste lanceolata, tenui-acuminata,
marginibus revolutis, costa percurrente; pedicellus flexuosus, 20-25
“mm. altus; capsula suberecta vel inclinata, oblonga collo breviore
attenuata; operculum convexum, mamillatum; annulus latus; exo-
stomii dentes pallidi, haud marginati, dorso inferne laeves, supernc
papillosi, 0.27 mm. alti, membrana ad 4 dentium producta, processus
lineares, fugaces, cilia rudimentaria; sporae diam. 12-15 p.

Very close to W. didymodontia (Mitt.) Broth., which is distin-
guished at a glance by its globular capsule.

BRACHYMENIUM (Dicranobryum) SAINT-PIERREI Thér., sp. nov.

Valle de México: Contreras, on earth; leg. Marius Saint-Pierre
(Bro. Amable 1338 p. p.).

Dioicum. Caulis brevis, 2-3 mm. altus, inferne denudatus, inno-
vationibus numerosis, clavatis. Folia sicca appressa, oblonga, breviter

x 120

Fic. 18.—Brachymenium Saint-Pierret Thér. 1, plant at natural size; 2, stem
leaf; 3, median cells; 4, lower leaf of innovation; 5, upper leaf of innovation;
6, median cells; 7, marginal cells; 8, basal cells; 9, moist capsule; 10, oper-
culum; 11, fragment of annulus,

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

acuminata, mucronata, elimbata, marginibus integris, planis; costa
flexuosa, basi 40 p, breviter excurrente, cellulis longe hexagonis,
36-60 »X10p, basilaribus quadratis; folia innovationis inferiora
minuta, caetera sensim majora, valde concava, apice congesta, Folia
perichaetialia majora, deltoidea ; pedicellus pertenuis, flexuosus, 12-15
mm. longus ; capsula suberecta vel horizontalis, microstoma, oblongo-
cylindrica, collo longo attenuata; annulus latus; operculum conico-
convexum, mamillatum; peristomium externum normale, 0.32 mm.
altum, internum? (capsulae immaturae) ; sporae laeves, 18 crassae.

In the size and form of its capsule this species recalls B. rubricarpum
(Besch.). It differs in its leaves, which are of another form, short-
mucronate, with the hair-point not spreading when dry, in its areola-
tion, the median cells of which are a little shorter and the marginal
cells not differentiated, and in its paler capsule.

BRYUM BOTTERII C. M. Linnaea 38: 622. 1874

Valle de México (Bro. Amable): Desierto (1619, 1633, 1637) ;
Rio Frio (1709); Salazar (1714); Llano Grande, 3,700 meters

(1736).
BARTRAMIACEAE (continued)
BARTRAMIA ITHYPHYLLA (Hall.) Brid.

Valle de México: Rio Frio (Bro. Amable 1405, 1407 p. p.).
This species is new to Mexico.

BARTRAMIA THRAUSTA Schimp.; C. M., Nuov. Giorn. Bot. Ital. 4: 41. 1897

Valle de México: On rocks (Bro. Amable 1502 p. p., 1503 Pp. p.)-

An interesting discovery. This species belongs to the South Amer-
ican flora and was known previously only from Bolivia and Argen-
tina. Unfortunately I found only two specimens, these among tufts
of Anacolia intertexta.

BARTRAMIDULA MEXICANA Schimp. in Besch. Prodr. Bryol. Mex. 58. 1871
Valle de México: Desierto (Bro. Amable 1624).

PTY CHOMITRIACEAE

PTYCHOMITRIUM LEPIDOMITRIUM Schimp. in Besch. Prodr. Bryol.
Mex. 41. 1871

Valle de México (Bro. Amable) : Contreras (1443, 1469) ; Salazar
(1719).

NO. 4 MEXICAN MOSSES—THERIOT 31

HEDWIGIACEAE (continued)
HEDWIGIDIUM IMBERBE (Sm.) Bry. Eur.

Valle de México: Salazar, on trunks of trees (Bro. Amable 1294
p.p:).

A new genus for Mexico, also, I believe, for all of North America.
This number (1294) was made up of a close intermixture of four
species, two of them predominant: Hedwigidium imberbe and Nco-
cardotia subnigra; the two others, Hedwigia albicans var. viridis and
Braunia secunda, were represented by a few plants only.

It was an easy matter to separate Neocardotia subnigra and Hed-
wigia albicans, but quite another thing with regard to Braunia secunda,
whose presence I did not even suspect. If some fruiting plants had
not been present this species would have been overlooked, its size and
appearance being so similar to those of Hedwigidium imberbe.

It is rather unusual to find, associated in such a manner, two species
that are indistinguishable either to the naked eye or by the use of a
hand lens except by their fruit; and it is even more unusual to be
unable to find morphological and anatomical characters by which to
separate them. The form and size of the leaves, recurvature of the
borders, plication of the lamina, and areolation, all are identical. I do
not know of another example of such an association and such a
similarity.

AMBLYSTEGIACEAE
CAMPYLIUM CHRYSOPHYLLUM (Brid.) Bryhn, Explor. 61. 1893

Hypnum chrysophyllum Brid. Musc. Rec. 2°: 84. pl. 2. 1801.

Morelia: Loma Santa Maria (7870).

CAMPYLIUM HISPIDULUM (Brid.) Mitt. var. SOMMERFELTII (Myrin)
Lindb. Musc. Scand. 38. 1879
Puebla: Rancho Guadalupe (4602). Morelia: Cerro Azul (4561) ;
Loma Santa Maria (5103, 5105, 7859 p. p.); Campanario (7462
p.p.). Valle de México (Bro. Amable): Desierto (1222 p. p., 1238,
1432) ; Contreras (1462, 1490).

AMBLYSTEGIUM SERPENS (L.) Bry. Eur.

Distr. Federal: Tlalpam (9493).
This species seems to be new to Mexico.
3

32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

AMBLYSTEGIUM VARIUM (Hedw.) Lindb. var. ARSENEI (Par. & Broth.)
Thér., var. nov.

Amblystegium Arsenci Par. & Broth., Ms.

Puebla: Rio San Francisco (5004). Valle de México: Tlalpam
(Bro. Amable 1453).

I had previously received this plant from E. G. Paris under the
name A. Arsenei Par. & Broth., sp. nov., likewise from Rio San
Francisco. Indeed, at first sight it would appear different from
A. varium in several respects: (1) Its narrower leaves; (2) its
greatly developed perichaetium, the perichaetial leaves being almost
4 times longer than the cauline leaves; (3) the capsule not arcuate
when dry.

These characters, however, fade out to some extent upon close
examination: (1) In A. varium the form of the leaves is very vari-
able ; (2) if specimens of A. varium are found with an inconspicuous
perichaetium and short perichaetial leaves, there are others whose
perichaetium is as well developed as in A. Arsenei; (3) there remains
only the character afforded by the form of the capsule. This last is
not sufficient to justify the recognition of a species.

AMBLYSTEGIUM ORTHOCLADUM (Beauv.) Jaeg.

Puebla: Finca Guadalupe (737) ; Hacienda Alamos (4723, 4725).
Morelia: Bosque San Pedro (4569).

The last plant, probably half submerged, is a form with greatly
elongated stems and branches and a thicker nerve (60 4).

AMBLYSTEGIUM RADICALE (Beauv.) Mitt. Musc. Austr. Amer. 569. 1869

Puebla: ‘Hacienda Batan (934); Rio San Francisco (5000).

Plants sterile, the determination only probable. The plant from
Rio San Francisco has the stems and branches laxly foliate and
the larger leaves widely spreading, either dry or moist.

AMBLYSTEGIUM JURATZKANUM Schimp.
Valle de México: Tlalpam (Bro. Amable 1346 p. p.).

The nerve extends well into the apex of the leaf. This is almost
the only difference I could find, as compared with the preceding
plants identified as A. radicale.

AMBLYSTEGIUM HYGROPHILUM (Jur.) Schimp.

Puebla: Hacienda Batan (5008).
A new species for Mexico.

NO. 4 MEXICAN MOSSES—THERIOT 33

HYGROHYPNUM PALUSTRE (Huds.) Loesk.

Puebla: San Felipe (4504, 4505).
This species was not known previously from Mexico.

DREPANOCLADUS EXANNULATUS (Giimb.) Warnst. var. MEXICANUS
(Mitt.) Card. Rev. Bryol. 37: 54. 1910

Puebla: Hacienda Batan (4961). Querétaro: Cienaga de la
Canada (11002).

The var. mexicanus seems close to var. pinnatus (Boul.), from
which it may be distinguished by its almost entire leaves, with thinner
costae. The most conspicuous character of this variety consists in the
very marked apical prominence of the cells. It is, perhaps, the first
time this peculiarity has been noted in connection with D. exannulatus.

PLATYHYPNIDIUM SUBRUSCIFORME (C. M.) Fleisch. Laubm. Fl. Jav. 4:
1537- 1922

Hypnum subrusciforme C. M. Linnaea 38: 658. 1874.

Rhynchostegium malacocladum Card. Rev. Bryol. 37: 71. 1910.

Puebla: (699); banks of Alseseca (700); Cerro Guadalupe
(676); Hacienda Alamos (4626, 4629, 4761). Valle de México
(Bro. Amable): Morales (1597) ; Tenango (1687).

I have noticed the variability of this species with regard to the
form of the acumen and of the capsule.

PLATYHYPNIDIUM PRINGLEI (Card.) Broth. in E. & P. Nat. Pflanzenfam.
ed. 2, II: 347. 1925

Rhynchostegium Pringlei Card. Rev. Bryol. 37: 70. 1910.

Puebla: Hacienda Alamos (4628, 4769 p. p.); Hacienda Batan
(5006). Morelia: Pare San Pedro (4589) ; Andameo (4822) ; Cam-
panario (7534) ; Loma Santa Maria (4908, 4910).

PLATYHYPNIDIUM OBTUSIFOLIUM (Besch.) Broth. in E. & P. Nat.
Pflanzenfam. ed. 2, II: 347. 1925.

Rhynchostegium obtusifolium Besch. in Card. Rev. Bryol. 37: 71. 1910.

Morelia: Cerro San Miguel (4870, 5041, 5071); Campanario
(7631); Loma Santa Maria (4890, 4917). Distr. Federal: Tlal-
pam (9492).
PLATYHYPNIDIUM OBTUSIFOLIUM (Besch.) Broth. var. SUBACUTUM

Thér., var. nov.

Leaves subacute and contracted at the apex.
Valle de México: Tlalpam, in water (Bro. Amable 1450, 1452).

34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

BRACHY THECIACEAE

PLEUROPUS BONPLANDII (Hook.) Broth. in E. & P. Nat. Planzenfam. 1’:
1136. 1908

Leskea Bonplandtu Hook. in Kunth Syn. Pl. Aequin. 1: 61. 1822-28.

Puebla: Esperanza (4745, 4754). Valle de México (Bro. Ama-
ble): Desierto (1438); Santa Rosa (1504). Hidalgo: El Chico
(1580).

BRACHYTHECIUM

I must confess that the study of the Mexican specimens belonging
to the genus Brachythecium has been an extremely laborious task :
The sterility of many of them on the one hand and, on the other,
the difficulty if not the impossibility of obtaining, for the sake of com-
parison, good and complete specimens of the types, are among the
more important contributory causes.

I studied nearly 60 numbered specimens and drew almost all of
them. They belong, excepting five or six, to the sections Acuminata
and Salebrosa. Now except for B. salebrosum (Hoffm.) and B. lai-
reticulatum Card., the Mexican species of this group are very diffi-
cult to identify. For one specimen that agrees with the type there are
many others which combine characters common to several species and
which one hesitates to attribute to one rather than the other, Hence
I gave three different names in succession to the same specimen with-
out being entirely satisfied with any of them. My conclusions are as
follows :

(1) Several of my determinations remain uncertain; they are
merely probable.

(2) Some of the Mexican species are very polymorphous, like
our B. rutabulum, and their forms have been taken for new species.
Therefore one must not be surprised to find indeterminable specimens
which in turn seem like new species.

It will be a task for future bryologists, those who will have the
privilege of studying the flora im situ, to weigh these variable species
and to make, with a thorough knowledge of the facts, whatever re-
ductions are necessary.

BRACHYTHECIUM TENUINERVE Card. Rev. Bryol. 37: 65. 1g10

Puebla: Xuehitl, near Esperanza (7988). A form which by its
laxer areolation marks a tendency toward B. lanceolifolium Card.

Valle de México (Bro. Amable): Contreras (1221, 1483); De-
sierto (1618) ; Salazar (1717); Llano Grande (1738). This is a
form with very elongate branches.

NO. 4 MEXICAN MOSSES—THERIOT 35

BRACHYTHECIUM ALBULUM Besch. in Card. Rev. Bryol. 37: 66. 1910

Morelia: Bosque San Pedro (4582).

I have seen only a very incomplete specimen of the type. The
present specimen seems to differ from it by the longer and more
slender acumen of the leaves.

BRACHYTHECIUM LANCEOLIFOLIUM Card. Rev. Bryol. 37: 66. 1910

Puebla: Cerro Guadalupe (796) ; Hacienda Alamos (4760) ; Rio
San Francisco (5003). Morelia: Loma Santa Maria (5089). Distr.
Federal: Mixcoac (9453). Valle de México (Bro. Amable): Santa
Teresa (1339) ; Contadero (1364) ; Tizapan (1612).

BRACHYTHECIUM LANCEOLIFOLIUM Card. var. GRACILE Card. Rev.
Bryol. 37: 66. 1910
Puebla: Hacienda Batan (935) ; Hacienda Alamos (4799). More-
lia: Campanario (7452).

BRACHYTHECIUM CLADONEURON (C. M.) Par. Ind. Bryol. 132. 1894
Hypnum cladoneuron C. M. Linnaea 38: 652. 1874.
Puebla: Hacienda Alamos (4696).

BRACHYTHECIUM COMTIFOLIUM (C. M.) Jaeg.
Hypnum comtifolium C. M. Linnaea 38: 653. 1874.
Valle de México: Desierto (Bro. Amable 1238).

BRACHYTHECIUM TROCHALOBASIS C. M. Bull. Herb. Boiss. 5: 238. 1897

Puebla: Esperanza (4729). Morelia: Cascade de Coincho (4713) ;
Carindapaz (7951) ; Santa Clara (4886).

BRACHYTHECIUM FLEXIVENTROSUM (C. M.) Jaeg.

Hypnum flexiventrosum C. M. Linnaea 38: 653. 1874.

Morelia: Cerro San Miguel (7546) ; Campanario (7940) ; Cerro
Azul (4532, 4541, 4554, 4788). Distr. Federal: Tlalpam (9498).
Valle de México (Bro. Amable) : Desierto (1222 p. p.) ; San Juanico
(1261)..

Several of these specimens oscillate between this species and the
preceding one. In their long and slender acumen and flexuose costa
they tend toward B. flexiventrosum; but the habit, the short nerve, and
the short pedicel (1 cm. or less) bring them nearer to B. trochalo-
basis. I am not far from believing that these two species should
be united into one.

36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

BRACHYTHECIUM SERICEOLUM Card. Rev. Bryol. 37: 66. 1910

Puebla: (4997) ; Hacienda Batan (4975).

BRACHYTHECIUM FLEXINERVE Card. Rev. Bryol. 37: 67. 1910

Puebla: (4862); Hacienda Santa Barbara (740). Tlaxcala:
(4855).

BRACHYTHECIUM ALBO-FLAVENS Card. Rev. Bryol. 37: 68. 1910

Puebla: Rancho Guadalupe (4614) ; Esperanza (4666). Morelia:
Campanario (7537, 7506); Cerro Azul (4530).

BRACHYTHECIUM ALBO-VIRIDE Besch. in Card. Rev. Bryol. 37: 69. 1910

Puebla: Boca del Monte (4674). Morelia: Campanario (7539).

I recognize in these specimens most of the characters attributed
to B. albo-viride: the green color of the tufts, the laxly foliate
branches, the lanceolate leaves strongly excavate at the base, long-
acuminate, with a costa reaching to two-thirds and even three-fourths
of the leaf ; but I have not noticed that the stems are more slender and
the branches more tenuous than in B. albo-favens.

BRACHYTHECIUM LAXIRETICULATUM Card. Rev. Bryol. 37: 67. 1910

Valle de México: Desierto (Bro. Amable 1412).

BRACHYTHECIUM ACUTUM (Mitt.) Sull. Icon. Musc. Suppl. 99. p/. 75. 1874

Pueblo: Rio San Francisco (5001 p. p.) ; sterile plant.

BRACHYTHECIUM SALEBROSUM (Hofim.) Bry. Eur.

Puebla: Esperanza (4515, 46064, 4690).

BRACHYTHECIUM SALEBROSUM var. POLYOICUM Thér., var. nov.

Synoicous and unisexual flowers, male and female, on the same
stem.

Puebla: Hacienda Batan (4937).

I combine this curious form with B. salebrosum on account of its
characters as a whole. It is much nearer to it than to the known
synoicous or polyoicous species B. acutum, B. conostomum, and

B. Mildeanum,

NO. 4 MEXICAN MOSSES—THERIOT 37

BRACHYTHECIUM INTEGRIFOLIUM Thér., sp. nov.
Distr. Federal: Tlalpam (9499).

Sterile. Caulis repens, radiculosus, sat regulariter pinnatus, ramis
inaequalibus, 3-4 mm., usque 10 mm. longis, patulis, attenuatis. Folia
caulina sicca et humida erecto-appressa, e basi decurrente latissime
cordato-ovalia in acumen longiusculum, patulum, acutum sat subito
constricta, haud plicata, marginibus planis, integerrimis, 1.6-1.7 mm.

Tey iC -

, ?

Fic. 19.—Brachythecium integrifolium Ther. 1, 2, 3, stem leaves; 4, median
cells; 5, basal and alar cells; 6, branch leaves; 7, acumen.

longa, 0.8-0.9 mm. lata, costa ad 3 evanida; rete pellucido, chloro-
phylloso, cellulis linearibus, par ean tenuibus, 35-45 «Xx 5-6», cellu-
lis basilaribus et alaribus laxioribus, breviter rectangularibus. Folia
ramea minora, secunda, subfalcata, 1 mm.xo.5 mm. Caetera ignota.

In size and habit like B. hylocomioides Card., but that species has
denticulate non-falciform leaves and looser areolation. It suggests
also B. refexum Starke, but is easily distinguished by its entire leaves,
with the nerve reaching only to the base of the acumen.

BRACHYTHECIUM CORBIEREI Card. Rev. Bryol. 38: 42. 1911

Valle de México (Bro, Amable) : Desierto (1222 p. p., 1441) ; Rio
Frio (1692, 1703, 1706) ; Contadero (1307).
The last number (1307) represents a form with long, flexuose

stems, irregularly ramose, with long, slender, almost flagelliform
branches.

BRACHYTHECIUM PLUMOSUM (Sw.) Bry. Eur.
Puebla: Huejotzingo (4856). Morelia: Loma Santa Maria
(4896). Valle de México (Bro. Amable): Desierto (1245, 1616,
1640) ; Contreras (1668). Hidalgo: El Chico (1586).

38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Of the material listed no. 1668 may be classified as var. sublaevipes
Card.,’ because the pedicel is scarcely papillose at the top; under a
strong magnification one can see only separated, depressed, low pa-
pillae. In this specimen a single costa is the exception; most of the
leaves have a double nerve of very variable length, sometimes very
short. The Bryologia Europaea indicates that this case is not of rare
occurrence.

BRACHYTHECIUM HASTIFOLIUM Card. Rev. Bryol. 37: 69. 1910

This species is not mentioned by Brotherus in his treatment of
Brachythecium in the second edition of Die naturlichen Pflanzen-
familien, but I am inclined to think that it is the one cited in the genus
Heterophyllum under the combination H. hastifoliwm (Card.) Fleisch.

Cardot says, “ Costa ad % evanida.’’ How can this character agree
with the genus Heterophyllum, which has “ Rippe sehr kurz oder
fehlend?”’ How could a moss which a bryologist of the standing of
Cardot affirms to belong to the genus Brachythecium have at the same
time the characters of the family Brachytheciaceae and those common
to the genus Heterophyllum of the family Sematophyllaceae ?

I have endeavored to solve this puzzle.. An examination of no.
10474 of Pringle’s exsiccata brought the answer to me. The speci-
men in my collection labelled Brachythecium hastifoliwm Card. 1s
not this species, but Heterophyllum affine (Hook.) Fleisch. Now if
one turns to the original description, where Cardot discusses Prin-
gle’s no. 10474, which he considers as a form of his Brachythecium
hastifolium, the inference is clear that Pringle distributed under
this same number (10474) two different species—the form just men-
tioned and Heterophyllum affine. 1 take no pride in this discovery,
but I cannot understand why such an expert and conscientious bryolo-
gist as Fleischer failed to find the clue and thus allowed himself to be
misled into giving full confidence to a specimen which did not agree
with the original description and was distributed by a collector who
was not a bryologist.

My conclusions are: First, that Heterophyllum hastifolium
(Card.) Fleisch. is a myth, and that this combination must disappear
from nomenclature ; secondly, that the binomial, Brachythecium hasti-
folium Card., which applies to one of the best characterized species
of the subgenus Salebrosium, ought to take again its place.

*Rev. Bryol. 37: 70. I9gT10.

NO. 4 MEXICAN MOSSES—THERIOT 39

RHYNCHOSTEGIUM SAINT-PIERREI Thér., sp. nov.

Valle de México: Contadero, on bark; leg. Marius Saint-Pierre
(Bro. Amable 1298).

6

x 200

|
\4

!
h
Fie, 20.—Rhynchostegium Saint-Pierrei Thér. 1, stem leaves; 2, apical cells;

3, upper cells; 4, median cells; 5, branch leaf; 6, marginal cells; 7, 8, perichaetial
leaves; 9, moist capsule.

j,?
| |

* 200

es /| / Va ;
PL)

Rh. leptomerocarpo (C. M.) sat simile, sed differt colore smaragdo-
viridi, caulibus gracilibus, laxe foliosis, foliis siccis valde patulis, duplo
angustioribus (1.7 mm.X0.5 mm.), tenuiter acuminatis, paulum de-
currentibus, rete densiore (cellulis mediis go-120 » x6»), foliis peri-
chaetialibus duplo majoribus, pedicellis longioribus (2 cm. longis).

RHYNCHOSTEGIUM HUITOMALCONUM (C. M.) Besch. Prodr. Bryol. Mex.
107. 1871

Hypnum huitomalconum C. M. Syn. 2: 248. 1850.

Morelia: Cascade de Coincho (4712a) ; Andameo (4827). Valle de
Mexico: Tlalpam (Bro. Amable).

RHYNCHOSTEGIUM LEPTOMEROCARPUM (C. M.) Besch. Prodr. Bryol.
Mex. 107. 1871

Hypnum leptomerocarpum C. M. Syn. 2: 354. 1850.

Puebla: Hacienda Alamos (586). Morelia: Loma Santa Maria
(4868, 4894, 5062). Distr. Federal: Tlalpam (9430a) ; Cuajimalpa
(9487, 9489). Valle de México (Bro. Amable): Santa Rosa (1515) ;
Contadero (1315).

40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

RHYNCHOSTEGIELLA ARSENEI Thér., sp. nov.

Puebla: Hacienda Santa Barbara, on sandy ground (739).

Sterile. Caespites lutescenti-virides, nitidi, Caulis repens, gracilis,
laxe foliosus, ramosus, parce radiculosus, paraphyllis raris; ramis
erectis, brevibus, 2-3 mm. longis, sat confertis, sat dense foliosis.
Folia caulina et ramea sicca erecto-patula, humida patentia, anguste
lanceolata-acuminata, acuta, decurrentia, marginibus planis, toto am-

3 2 i
x 200
x 200

x 200

Fic. 21—Rhynchostegiella Arsenei Thér. 1, stem leaves; 2, apical cells;
3, median cells; 4, marginal cells; 5, basal cells.

bitu minute denticulatis, 0.6-0.7 mm. x 0.2 mm. ; costa basi 30 m percur-
rente; rete opaco, cellulis linearibus, 36-4046 p, basilaribus sat
distinctis, marginalibus subquadratis (2-3 ser.), subhyalinis.

This species can be compared only with FR. Jacquini (Garov.)
Limpr. and Rk. Teesdalu (Sm.) Limpr. It is distinct from both by its
leaves denticulate all around and from the first species also by its
larger branch leaves, differentiated basal areolation, and more densely
foliate branches ; from the second species by its acute, decurrent leaves.

EURHYNCHIUM SUBSTRIATUM Thér., sp. nov.

Valle de México: Llano Grande, alt. 3,700 meters, on rocks (Bro.
Amable 1735).

I, striato (Schreb.) simillimum differt: statura graciliore, ramis
brevioribus, foliis minus profunde sulcatis, caulinis angustius de-
currentibus, rete basilari praecipue ad angulos densiore, foliis rameis
margine dentibus brevioribus,

NO. 4 MEXICAN MOSSES—THERIOT 4I

I segregate this plant from £. striatum, because, in addition to the
characters enumerated above, the European species 1s absolutely un-

Fic. 22.—Eurhynchium substriatum Thér. 1, stem leaf; 2, median and mar-
ginal cells; 3, basal cells; 4, 5, branch leaves; 6, cross-section of branch leaf;
7, apical cells; 8, median cells; 9, perichaetial leaf.
known throughout the American continent and it is therefore hardly
possible to consider the moss from Llano Grande as a local form.

EURHYNCHIUM STOKESII (Turn.) Bry. Eur.

Puebla: Boca del Monte (4738) ; a form with stems less densely
branched, elongate, and laxly foliate. Valle de México: Contreras
(Bro. Amable 1518).

ENTODONTACEAE (continued)

PTERIGYNANDRUM FILIFORME (Timm.) Hedw. var. MEXICANUM Ther.,
var. nov.

Folia valde secunda, latiora (0.5 mm.), cellulis apice parum pro-
minulis, costa gemella usque ad 4 folii producta.

Valle de México: Santa Rosa (Bro. Amable 1503). Hidalgo:
Mineral del Chico (Orcutt 6649).

The typical form of this species has not yet been found in Mexico.

ROZEA STRICTA Besch. Prodr. Bryol. Mex. too. 1871

Valle de México (Bro. Amable): Desierto (1418, 1425, 1444) ;
Llano Grande (1731).

ENTODON JAMESONII (Tayl.) Mitt. Musc. Austr. Amer. 525. 1869

Morelia: Cerro Azul (4779). Valle de México (Bro. Amable) :
Desierto (1245) ; Contadero (1311).

42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. &5

ENTODON ABBREVIATUS (Bry. Eur.) Jaeg.

Valle de México (Bro. Amable): Contadero (1302, 1305) ; San
Rafael (1280).

The pedicel is very variable in length. In the same tuft I have seen
pedicels 3 mm. long and others up to g mm. long.

ENTODON ABBREVIATUS (Bry. Eur.) Jaeg. var. TURGESCENS Thér.,
var. nov.

Caules et rami turgidi, folia 2.2 mm.xX1.3 mm., valde concava,
cochleariformia, rete laxiore, cellulis mediis 70-go px 9 p.

Valle de México: Contadero (Bro. Amable 1362).

The facies of this variety is very different from the usual forms
of E. abbreviatus. In its leaves and their areolation it comes close to
Pringle’s no. 15226, identified by Cardot as E. brevipes (Schimp.).

SEMATOPHYLEACE AE

RHAPHIDORRHYNCHIUM OBLIQUEROSTRATUM (Mitt.) Broth. in E. & P.
Nat. Pflanzenfam. ed. 2, 11: 428. 1925

Sematophyllum obliquerostratum Mitt. Musc. Austr. Amer. 472. 1860.
Morelia: Campanario (7926, 7941).
RHAPHIDORRHYNCHIUM DECUMBENS (Wils.) Broth. in E. & P. Nat.
Pflanzenfam. ed. 2, 11: 427. 1925

Hypnum decumbens Wils. (Ms.); Sematophyllum decumbens Mitt. Muse.
Austr. Amer. 488. 1860.

Valle de México: Desierto (Bro. Amable 1439).

SEMATOPHYLLUM CAESPITOSUM (Sw.) Mitt. Musc. Austr. Amer. 479.
1869

Morelia: (7893, 7909, 7912); Cerro Azul (4784) ; Campanario
(7515, 7516, 7520, 7537, 7559 7552; 7557, 7935)-
SEMATOPHYLLUM CAESPITOSUM (Sw.) Mitt. var. LATICUSPIDATUM

(Card.) Thér., comb. nov.

Rhaphidostegium caespitosum var. laticuspidatum Card. Rev. Bryol. 40: 39.
1913.

Morelia: (7890, 7916) ; Campanario (7642).

SEMATOPHYLLUM HAMPEI (Besch.) Broth. in E. & P. Nat. Pflanzenfam.
ed. 2, IL: 433. 1925
Rhynchostegium Hampei Besch. Prodr. Bryol, Mex. 105. 1871.

Morelia: Campanario (7518).

NO. 4 MEXICAN MOSSES—THERIOT 43

HY PNACEAE (continued)

STEREODON FALCATUS (Schimp.) Fleisch. in E. & P. Nat. Pflanzenfam.
ed. 2, I: 452. 1925

Stereodon subfalcatus (Schimp.) Fleisch. in E. & P. Nat. Pflanzenfam. ed. 2,
IIs 452. 1925.

Further observations have convinced me that in these two species
of Schimper’s there is only a single specific type.’

New localities: Valle de México (Bro. Amable): Desierto (1210,
1221) ; Acopilco (1201) ; Salazar (1236).
HYPNUM AMABILE (Mitt.) Broth. in E. & P. Nat. Pflanzenfam. ed. 2, 11:

454. 1925
Ectropothecium amabile Mitt. Musc. Austr. Amer. 513. 1860.

Puebla: (4945, 4947, 4948, 4949, 4951, 4953, 4956, 4959). Distr.
Federal: San Angel (9479).

ISOPTERYGIUM CYLINDRICARPUM Card. Rev. Bryol. 37: 56. 1910
Valle de México: Desierto (Bro. Amable 1247, 1623).
TAXIPHYLLUM PLANISSIMUM (Mitt.) Broth. in E. & P. Nat. Pflanzenfam.
ed. 2, 11: 462. 1925
Tsopterygium planissimum Mitt. Musc. Austr. Amer. 408. 1860.
Puebla: Hacienda Alamos (584). Distr. Federal: Tlalpam
(10999).

ISOPTERYGIUM PLANISSIMUM Mitt. var. LAXIRETE Thér., var. nov.
A forma typica differt: rete laxiore, cellulis diam. 8-9 p.

Morelia: Loma Santa Maria (4877).

VESICULARIA VESICULARIS (Schwaegr.) Broth. in E. & P. Nat.
Pflanzenfam, 1°: 1094. 1908

Hypnum vesiculare Schwaegr. Suppl. 2°: pl. 199. 1827.
Nuevo Leon: Monterrey (Bro. Abbon 10969).

MICROTHAMNIUM THELISTEGUM (C. M.) Mitt. Musc. Austr. Amer. 504.
1869

Hypnum thelistegum C. M. Syn. 2: 269. 1850.

Morelia: Campanario (7924).

Sterile, the determination uncertain. The cauline leaves are sharply
dentate and the branch leaves secund.

*See, Smithsonian Misc. Coll. 787: 28. 1926.

AA SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85
MICROTHAMNIUM SUBTHELISTEGUM (Card.) Broth. in E. & P. Nat.
Pflanzenfam. ed. 2, IL: 471. 1925

Mittenothamnium subthelistegum Card. Rev. Bryol. 37:55. I9Io.

Morelia: Jesus del Monte (7608a).

HYLOCOMIACEAE
LEPTOHYMENIUM EHRENBERGIANUM (C. M.) Fleisch. in sched.

Hypnum Ehrenbergianum C. M. Bot. Zeit. 14: 408. 1856.
Hylocomium Ehrenbergianum Besch. Prodr. Bryol. Mex. 111. 1871.

Vera Cruz: Jalapa (7998).

POLYTRICHACEAE, (continued)
POGONATUM BESCHERELLEI Hampe in Besch. Prodr. Bryol. Mex. 63. 1871

Valle de México: Salazar, alt. 3,100 meters (Bro. Amable 1715).

POLYTRICHUM ALPINIFORME Card. Rev. Bryol. 37: 6. 1910

Valle de México (Bro. Amable): Contreras (1667) ; Xoquiapan
(1750).

The last plant, which is in fruit, affords an opportunity to complete
the description:

Folia perichaetialia numerosa (12-15), remota, longe et late vagi-
nantia (vagina 4-5 mm. longa, 0.2 mm. lata), in acumen angustum
abrupte contracta, humida patulo-squarrosa. Pedicellus 20 mm. altus ;
capsula minuta, oblonga, laevis ; calyptra angusta, elongata, 6-7 mm.,
parce pilosa. Caetera ignota (capsulae immaturae).

SMITHSONIAN MISCELLANEOUS COLLECTIONS
VOLUME 85, NUMBER 4 (ADDENDUM)

INDEX TO PAPERS BY I. THERIOT ON MEXICAN MOSSES COL-
LECTED BY BROTHER ARSENE BROUARD PUBLISHED
BY THE SMITHSONIAN INSTITUTION’

INDEX

[Synonyms in italic. Page numbers of principal entries in heavy-faced type. ]

PNOMiaRGA CEO IT TO] atten. setae 8 staves ete Pes ettad. crcuoveteia ties dy crenata alep bie aly rage Siders III, 22
EMOTE Cll awe CALCU awe srottna Setey natetee Sica ahe oettrotssay2i-dcnye slate lari des, dvs ios exaudio iterates III, 22
EMD SteP Lat APSCHEL cid cel os eles ala dn es ein arn Go eth eda Boog Dele ders i iieee2
Frny sare Op Eye LUiTtameece says coe ha catvinis lol atbitouay shave, airi-d delat datleneusl ch actieie onavanees duct oe ee
Siuratzikan tists. sje rete eee A fete eal mies cereals deere eae el 2

COTgt nO Cle Cla trey ance aes vam aoa Bete o ees Sean hi Sh dias oh Stare Sh Sr toude coud Avoute ce, SM ottowarrd 6. LIT 32
HAG T Gell CMMs RVs 2 ae bis orate BL SEN RA NSB LA asl at sae afove cusriipleiasine dea tes Iii s2
Gla) GILG) aeteberene once oraheret oi seeney crore eterna cise Ga aaersie ice cree entice ere oie eee Lilien
SATU TTUN cc Geese hte treats ay af Shoes aya Mearns nseste APM G asnsnd sl Mau D aed adialaioea: Tht, 32

PN TES ETE Lessa ries cathe SVS AS re Ica oP Ra ns en eesine sone a BaP Ditne2

PMC OT AMITIRE ECGS alm ceeraics a eee a emis is csne Sieressiieseuese aie sucieloreus sioveen overstate eise oo aera I, 18
UTS CUO l Tel ot emer crerensaareueheee nen caaer erences ace teen oie a rene ore eure 1, 18

SELL lei reese acy enemas lesa rete teens ec ieiet reich at canter Oe ahs ets
SMDSESSil I Sme@merer araecre oat ieee ee er earn I, 19
BMIOECtalcilmy APICWAtUIM v.44p6s0 4 spanwise dine esoend che sd bdevar baeee ha Deas
HeCHINUp) A CULIT II sepemeraye ex aievensyevenstene lem gererevetciaie eaxceierstere reve ate ie cies re Sieiava chees eA
BONG elMSa tlimtin pantera ies sete eiyeee tre eusee wieresie oraniesic ese eae eidigen ale Ill, 5
SULCIN ORO acters teers ks sis iene iees ice inion Sernt neers eee es
mesa ATE AMAT Ie? "eres seas ay sco wiane eaicinta ae 4'a int eave Sc clase une cae are a seetesst aie tet Ill, 4
EMIGINODE YM ALiTOrmMe MEXICAN. .<.s 0.40 esgee vesee ven ceusveeeeseaes II, 1x
DISA hitin Meee eae eiiere reste Ticet erie een eee oe eee eee Le ex
VAMOMVOGOW TOCCOGE 5. 6.606scccecescarsncwscecoesacccancececceateae iW, 2
PES EEOC IA (DVECIPES 6 cuando G4 a4 velwn 658 Saareidae siuee a vas eck eaccead ac 3
Piet UCLA Came Senet ere Ine eaten rere neTeTE ere Te eee rere Setar ole oteretea sien seis ee
PRPIEUNCCTISIS: Sen enrebice ence tee See sie Sone Ao sve Sees ent ches Wanda nd Ds

Hilt cA CMe arty acea aS erEa Toya icteetsa tee terete a eee Ie ey ice eiecele he tue Sle Oalbaans's a I, 4

TST MA CL See sveneniercececircesee aes HTT eee oh ee aca altace nn i
SUMS 1 cals ewe sy aeryoe cece eee vercw ys reestal susyerau wer state vane Tete den Goae ear nci six eter cne ore aanns | oe
ESL ETE eS TS SOT Te 2
Atrichum contermanum ....00. ccc cece cece ccc eens cnecceeseevecevcccuves [ea2o
Mrrelleri *contermintm +..4-s00-.0e6.0. 4 ase ce lt cent ecole c I, 20; II, 26

TON GUY el UAT pee eee eae eee reo y Seay areyen UIE aT, Th ire MOREE Aedes sem vas 1, I, 20

* Part I, Smithsonian Misc. Coll., vol. 78, no. 2, June 15, 1926.
Part II, idem, vol. 81, no. 1, August 15, 1928.
Part III, idem, vol. 85, no. 4, August 25, 1931.

45

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Barbula Abbomity cekcccnea dees oe ere oe ees SHS Hee TREE RRS III, 20
AIEISStaY etic eis cvs ei Gla reo ees oe ICE aren eR or snepe ERE eienetonoeyetnenots ii rzeere

GIN PUD LAC CLL ” si esha cet ees Roe ecto hae Sao eS et RR Nach ntoeS coheed eo teceg= Ill, 23
Bescherelletie sian sacicearscryett tree torpeletay toecuntenteleterer ene ee oer Tench meene rate Tag,
CLASGCIMERV IQA oh SO rene ee ae ERR ee einer ere mien Ee taker Mia

SEEN OCAL PA % dra A syd sie ate a eee repairer ee ete eaten tere Lae
IByoyoirereaehich nogueongsoco pec Fi Naa PI Nis ry aN ch leg Say a OR III, 18

Cal Careat ecto 8 Sie cect ee ie AES Seo Bars ies rated eM Cen nee Iieezo
dialivérichoides s 4.2 o/s oe wre oictaieicverse terete eheicte-s etetehanchete RoteverrNeveie raha tate fetes fete III, 20
FEV ORD CH GONG Cie ules i wtarter a Sister tee aye ae eRoeneh A avete Mel) ieee toa eatche ez
Hhrenberoii mexicana meorcaerie seein irr reer III, 19, 20
POFVUGINED. waned na She ice hata nie hom ee ae elas eto De Seer III, 9
PEACUESCEMS. 5. in. aero Syapkl es a ese Ste ers o elas hac ree OS arson austohal pape Lil. 27s
PHACHIOLMIS: eee ciae oes siekaals arcisieshe aes chiara ora marae rereraie erie tiie Li r7ers
hypselostepia. esyris seawnrste aracee Sonia oa amin siete se anemia he aera eee III, 21
NATE TE CHALE: Ue ave ce tev mesos cotskg oua Soe aiie oe toce SRS Sea Sad Say oSne EN nen sone eters Eisen
OVIZADEMSTS: eacrex ees cisy so Soe cue iereter oh eelallele taverethlsh Pepateheto AMONG Caevoroen chee note treks Ill, 18
PUbTiCA ys, =z serstersafoide koe Shee Gua ae earch oleae te Lh nos Zo
SPIT ALES <a caiee cea wicca setae) te wicteadev ona loreal She area terete fole sfeomicpatater te aa nero III, 18, 19
SEETLOCH EGA ikon odie, Sua cra eee cls cnc aiwtal ai cn Sotelo ouanigee Sic re Mauey ores ehorenenterete Li om
Stubteretiuscula) sc.ccts 6 <a5 os cris a sroisin o-o,50e aucun uanvis es elsievn lorie euerepehtele III, 18

TO Caahbk cele len mena ao Samad neo mop ccaccuscmacnamdos odoudlD ono tc Ill, 18
ERICH O STOIMOLD ES ia ace, Hevseiere hia are ele Saree aie el Oe vole Oba cversy el srepeenstert Likes
Wasi ttl atten: say Metin ce wi atey hatese aie aeons oats oustay tetrsalct woke tcust ss Sistine tae IL a8
Bantramua thy plivila, eeeecaais oe emiayae ac rst xstoieerela cinecne ie Cera et eerste iene III, 30
En aan tL S Eats Ae here eee Te IN sr tie Rot aeceeteys ie Ore coke cen accuse tees chai fame Miteso
Bartramidulaimexicatias s.ceis/s osu siete eveuetteors seta erasers & alee siokeie rete) aerarsiotaeenenens III, 30
Brachymentunme barbacaimOmtise 2 7.) scien crcieneie syste eerie fae ako eee eter redone Dis
Capilla res nc elaine seeranai shane here snes aie tusr teva orate tay tsrover cite le Siatadl her sfegeenste a aaa II, 9
Chlorocanpitiniwe: mercies cee eset eborere ster usta tortor stoner keekersracnerater II, 10
COndensatutn sae ie Ce erste Oo eel cielo eit rer okey tel Revs atreionens II, 10
EMIGUUT cca as cone cle cae Se whe aise ctoyeilafe efeharon eighore oo eteyaoa ap Severe teats cher «tee keine II, 8

it 1CAtUtM Vora cys ses cies ei cis ciate ellevere esc letancnere siiaeye ei x steeere Totcoter Miah Melsnetelts Herel II, 9
TE ZAM ONT heel ee ees ieee oe a eas OOOO eros mete eae ll xo
AEC OLE Fars Penske Siete abe are Os, Oe ROIS RTE ECT CUAL Crate MeN We
TERI CATIUTI wea icts ctoetereceieere econ tees renewals listeners) sionerensyeneterelorekoteliheKcntonetare II, ro
INAitreric bitte eae ye ves rere rerericiess ace icioveie rene rcret ete telereree Velev eterno tener Melt oherehs tate II, 9
TEQUET LSS weoretcs csc eta ahaiieTes Sat eNes ay cod aT aOT ST Certo PMC H Ns cl oy sroueteowolotoitevo mete stsonerch seks II, 8
PLLVIGUITINT soleete OTR esr oe ene veme ere H A ene ic ieee ooh nereleers ter tenn ekeroRs II, ro
Saints PLERT OL sera ceercvsie Siero ero cree A ore a we ve re ferew clemency Rolcey eichaeoners Ill, 29
SQUATVUVOSUIN. Gcraratetciaveraidta ws 01h tie sieie) siels a'0 area lelavara/=imennve l= Waele sitet ere Lie

SYS tylitaini: shce ciara inet eons = ctas narrctors ees eiaiensile ete erate II, 10
Brachythecitum sActtuim' casa sos oo te ee cce esie eens cin inyeg eee eters Til, 36
BIBO= Aa eNSS Asc sc ae ae haste INOS OHS ww Blcveke ene cct ake Greteko MRCS emreene MI 36
ABO | VL a Rc ascaeca nese toler ee ca ol ele acre aes tone Renae ee poeeeemenontere III, 36
SUBS Ga rea 2.8, eae each oor tesco bate oro mene cas Pane SE ce eR ete RcTened emetoree III, 35
CLACOM EE OL rise een eae eel oe ae Coste elas eae cI tar ton ener Ill, 35
eres at Colt Lr Rae ee ee Morir ayteye cM Tale iss
GOTO SEGAL NN ee re ster cscs css ee edie ete eae ai ca NT Len tate oot one Iis6
COTBISL EL: ccs d eelece iG ecw ev ree eto eore lis Fe raed Pe ess ae aU redone oc Eos erst aes TIT 937

HEINER VE alts aise Dia teleoc rh oe eet TST Ue UT a SEE ere eerie hereon III, 36

NO. 4 MEXICAN MOSSES—THERIOT A7
Brachythecium—Continued.

HE MAVEMEUOSUIII Veeteve skate er, eee eit Crete cd Eile cvsy ot Aud Seuss Sane Meal aiacete. Seveienae tii 35

avs Gr T UTI Sees os Scent stent ater tse shayie Oat ksysiDaNe WOR ate «cies. III, 38

IyglG COMO LES dere, «5 fe se Be oa acoleia.c MEI SiS So aegietoy cs aie 37,

ATCC Casi LT CITT Weageu ee dyah Pat Sea sears Sait =. oa SA tpayednne els ee eeheews stl III, 37

HAM CEO LATO LIUITIN fis sce va. asroeseaklo ears aves ay laye 6 vnsieye O aya ers avdrmvagens ts yet Pisa ss

ated Gl Cig rer eave tey tio eka ve ee ome awe goatee he ak os hes cat yayzropenuseans sWevatene tev Ill, 35

Paes e tal CLUE CA EELEYN, 07 6: sy sear tevctiateuetete teeuscs Stat ye ralsl 6,6 Soren aie. a, cveNape, exact utbanecorvensare 66 III, 36

ANT Cea ttt rn aetna oo eon ¥ke Mey Sibs Schr os a Absyen saSs vale Gecitsou Sus teeeyee Spo ysebyES 8 aus III, 36

PUTA S CHT AMineyaysy nese ods ekavayev or Cayeyaranaietaparexe cieiers we scree eager whey aa ce eeeeye eter Les 7,

TET CORUII YM Mera consis cite oye abst av S00 oe os SV Syaos oelints 2) ote oe Be thysceis ) outyleee yeas Inn 37

Gall GPO Street tees es once ether ocr oee eres eretectere crereimesea cuentas: te Lie s6

TOO OL CUM area stun ceeveie ys ore reser ons ere eeserr eo eaele ete ercre eee ee ee lees

SETHI E OMT Avec ucita cre eecnau sun oe tap ve erteciete nia cies ei arauerseeravansrentiase erensuntelerc ran III, 36

PEIN UIULD © eV Coae wavereumasyfeusyeeey sxe apsye stores tetereietaney ery eerie ieicreeccinienriooeyes oheroreneictre I, 34

OG Hell OM asts ae seree peer enas ene eeray. eens aereeaiererrm everson III, 35

FEMA ZA MUP TEU 25 2 Bleisie cuscnisueia Se een e wo 6lm 6 Gyayts Suess Flom aed aloes Ge oe lvsd

Ce Tot | Tc we eeeyege eee een etane omens use Uae cereceretayeteticna e¥euole a cieieye cheretameices tei: Lez

lehiiantiiania ese eee eee cite. Sates ieee las

PUL Gevtay emaersrcmra ss ncpe mus eentriseisrers eee cise stele met eee naar eens den may ence ae Le2e3

GANIESCEM Sumer tae Ee oe erro eee l323

SE COUITI Cl cll ya cence emetic ee V Sec eer aaa ere iise ee tees fo Sotto nc ens teers e235) Dil ar

ENTICLGICLIXI Mee een acer ene eh tee nenereer ene raeye cities ees et ans I, 24

aSSINGL Siamese meister teeter eee ones ete ren Tiane Sie Rare cwi te salen enavoee 3 24

DEG Ga ee eee: Sicreee te rg ce crete Tee ON enc ohereecce ez

GLEN FIAGI (OS a eat ao ne Ea oe eee ie oto CL ee RU eC OE eee Le22

HSRC UILE Ice raTItGITMC Ici ercics iets scereieasie a tenoceneeo eens o.srei.s.s 61 are eecerctouc the sestha eo t.a? I-20

EC) TTT Call CCS ct ee a ee ON Petey ce ME eT, Non ain nye cert erny Ns ts cata) I, 20

Hes qy, OI 1) LI VUATIT THE CATIVITI eye eiene areve eterna ete ates slate core. 0s fay oi adavte neste e cue. .e- cyst Tez

FeTTo ATI DIWZOLE DIS: wereateasers.cevatst sects aise eneses cSt wo cua Shaye shu ay bvela toe ash nee Avesis. I, 16

UTC (|< eee a RENE STEM cia Al a oan. Br caste Ar Dens ya

AIGS GLUE TUTTI mmeeve grote naan aate terete her cieta ema ES viata lola days ay Stapnsnata eeen erst aie eet 7ien i eaeer

DEachy can pun Mea are teeta tees eae Ae ae rte Nace eR eae II, 11

EMMOPOCAU MIT amare ee kn eee lee «ced a tudes heen ar whales rt ae

COSLAICEIS Cua erent oe ee cerca cutee hoes oa eran fete Peete =e eer

LL CCITT penne a OIE OS 3, nes gee oes eM ane mc goaeg er re RNS fe cies tae Mey,

EEL cUY (1S gaa stn gntl eee Pha wed catoR ee a dhe ee, PINNARAL RAI ola @ NEE wih Sovh i

PAGS C11 GH wmeese cme te arch sapee eae T SOI cee ie Sete io ak arse Seat cathe etaeier eset et I, 16

IS OLLG Lol imenemenscerep siete eiisenk oo esae eas Geers es Sees Gee sys mh tube ionvha cae ayre sills ayes ure es III, 30

FS OUUG CCA) U1 1y Meee tee cee mee U Pre -cce ee tcataienns Meese tae ete oe aie vince tiers

Glia SEU reenere ene crsea te nmemetet ae cient tote, Meas. 6 stenc ste oos oavintye eve. eeie oxtail ee II, 12

SASS LITIMM ere ee Te enete ecient asics slates aca AR Pog ork toot ane ee oiSuatin'nva geod ROH I, 15

OU UL NMR esc emen ToMen tener swe wae tess cue o1act ee aye ou auevel Re) ae ire

EOD CITT meter neyetete tener ees iets ots ai ee cneccono iets Bo fe az. dod deeetvbs Benehtees.« II, 15

GidyMoOdontiunt oo... ccc cece ccc ec cc cecucuceencaceveusctecevevevees he

Ilirenbercianiiin eae. secs ses oe sda. oes ws aveeves sleces dccecsece- tet

DMO ICOT OUI bn aie ep ateseciniese SR so Seales Salad Sas Babs Dak ak he Shas II, 9

Ue eM UMN eae eee ee ne tye eerie Sec vs Sek asa tk cee eau oisee ae ist anche oe I, 18

POMCeGLOliuti ea ane cian eee ee ie Ue a Minow eden eda enki tl, 311

Weutalinn DAVCLiTmereatere ene ei eet te ier Sia eae sc ens TR eraios ad li rs

PB eraRMIITENIATACIINN 1p vseezssseiase ay Gun Cs eee eae erect Sewn arte ae aR wood tea ears I rE

48 SMITHSONIAN MISCELLANEOUS COLLECTIONS vor. 85

Bryum—Continued.

VUES OIA nese Gealecon Gt oh opehihe Teele) Sear aacuahis oeseed Neue AINE oper Veh enes Chern eeeNIS oe Seamer TT K@
{TAG HO DALAT heaters ee Slee cies acs ioeh eearaene Che eee Mee reReRe te eer ater Tule eres!
FULATDAE TAL TaLy eerie ce eh aes cycthe cot suntan lata cgRateG Neat aele eaten aaetace date ee ei epee eters I, 18
COSA tear eee Sac Bien a sie Wac ettre censuewonauuee eaiictlens Gieictiens euemaccieas oheeeme uel eee Das
SPECLUDALEN, Fixo% bynpec edn aioe oleae ert Ieee Maloferalete srelei 6 epee arene ILG
SCMAT EU OSIM ie-ciereys owicsonvs Ghee histones Bevae eben eee eImeIs s apaert en eee eres
ISD CHUN SEMA racine oe etic lehclciete ete ciel enir clever elei ces) cro iovectorehs ef rei he tetetcvers Ir
Stubelimbattn, 225. cs wih wons.cror a once ge deledvcistos sete tiersyele elo.cis na seattle Liang
Campy lium *chiry, Sop lay lenin sa. oo 2st tie es elaiwte eye elevelaeteie: = lle olereieve «re reretete er -tote ines
hispidulum: Sommienteltti cjasmeec sacise sale stlenclelecias eccleeteeieee 1. ann
Gampylopius, ame mstisal ters; ye vererssccchansvevoi crete stele evel ieleuate ecetyael iene cel eaerer erates hi
AOlMait els seiccdes ssi ers @ alo cepts oreyeoveen ieee bneeral iterate ctegen aslo Sperone Lie
DASES OTE: Sets a ceee ca lana Ors ah ece Ua oie tek UMN rakesta roe eeee ole TeL tote ouslanat Wnenatenner eer 125
AVS CEI: Face als eis is ote LAGS o Meu OLS OHS Osea NO BARS ENCES eue Te oMersy eed styebehel cre Reto 2
Ghristiarty scecetoats Gok cAea Ocean Oe eee EOUor er ierat re latene Teo ese
GESEUCENUS. 2:5, & cece. adverse ds aevcvoraherave esersie, OWS eLel eras Ss Gis mussel aera) w ahateeeheaiterens Te
LITER OPE SUIS! cis th srs shu Ses Sede ae ealsie iat che tla Paral custo ay witsbegetciren ou Stevslist so aoneriete tat eel ees
TTIATATENISIS® ances deo oh ctNauan avon cols de a tara ete ctioye Ye el cvaLeliellatansr-e/orelleyawe lavabeeroderamer acre III, 4
{TREK CALITIS * Vorsieearsuee devas, Sateeshraaay casa be tcue ate (itnavahorete Waa cs a ateieeoley sxe se toearaee Lone I,7

BP TEELE LE as gr tesco vere roar een oe tep a ra tad at ole Late aver™m ACR NSs eyelet Suche ere Tae s7eh a eee aenee Tit
PUUSHLTTTS, so, orace, scislels/ aoe are crercva cei emma aver Haver oye! oye oTnlare ede) oilers alaTehe te etetenate rs ttctode IG

TR OGUTT cctv Sea carey cies eR: ar ciate nat olerete ei Gie ee I ere ee ere I, 8

G aint PV eREEI/ lees eonapaeete i toera Davegoc deresaistua te (orators chee ol ote aveweeetro mipeirenotey sreyeenes hens eas
SLID ELIT RACES ees aces tie cecc hovel Neferieyc ellos fo onstel oy exerci oesicvatatsloretedaneuct foley eketareg rey tonne Te,
Geratodomm put purettse teres crests elevererarsje.ollelet\ste sie) eo iehlcie Mietelounverelobevovetey<l-tey fereere Ae
SEEMO CAT PUSi un. esl care eneyelcsterehastnt = spel aae elle el ctetetnt tetiar silage aiedctsy eneractors I, 23 tie
GOscinodomeATSEMe|ty ste crcicre see erete aravere le tersovellelajotals ietaTers aieraial"s\n/ote’= @' my efatel atenetetele II, 4
WYieiciintai. savarse erase seciapom Soule celatsvorney cates or Ate intel sisioueye tava eke arneorsy Retainers Li
Gry phaea apiculatar caterers ets eles oss te aaueyeray stene exayaneteteeetoransi ye 1 rareer eats II, 16
ACHE AL A eer ene ere eee ey aren eLeR STDC Te rere seN Ae raledeisteliosevedele oncaclcleveret te aane teen kenors II, 16
fiLT RO TALS doc eee ence rs ese arm soco ape oie lO aie daceeanoteoueyol over ote otulstaerch dereinnsecteterotaeke II, 16
OLIZAD AS cere, aaiete cere ee ee a rat eae one Ge eatin wteieicnan wie oe eos Meyer aoe ae Il, 26
PAESHISE carshate gale sncteere ciere'at'e whet ieicr oshoueler stale) she illest ler sYat=ioats) a avelietr icy e baleNeketaxel Thay
CE@CUIT CTS a nssseie se el cece oo chen eveves seh aiene ee relat orsiistel clays (eleietonevouapenerensires cciet TS ey,

SF LOLIE EL eet ere ere SUS ike FIST ae ret Terre vNaVON odsue ol taeelelsToreterhstchea cesta 1, 07,
Cyclodictyon! albicans © 27 ji. si. 20 -lefersye eee aie no eieos nee eterna ene ores lize
INTSCHOIME Ronee het ert ie Soe miei tehe rior eek recht er kre II, 20
Etre Cha Currant tetoes avo ae avee eG OGLE asie erceselchsi hate tototay Repenel ster Wen-tone II, 20
Tete lyrraatririd eco rhe avec etal ees Sector Tote feted clo shehoton oneal nel eel heNeomeretenexe ie ett: Tl ven
Dactylhymenium Pringlei .............- eee e eee e cece center renee eee Ill, 17
Dendropogonella rufescens ........-..ee eset cette tet teen ents TL a7,
Bicratellas varicie cceccassctaretetere eee tera toro 01c1 ole ite elereleley sxe 6 he Mics stiles rebalecete oOo rete I,
DicranOdOntuint COSEGTICENSE ccc do. ocise wi ciniern = wie nics oisles acleibia sosisie'e wicinielare sie ieee
DiCrAanUm Ce SETUCTLC ec eater oe eros) ores ie teree tect oink oustiotcks = erevalieder si(el skoteksi afer orerst ere talents iia
EFT OALII cece oc «5 oor eiacos eco rene Chetatae's tale #1 Serna te aie (otere inieiansls antes eee ES
Didymodon campylocarpus ...----- +e see e eee e eee e eee teense eee IT, x6
Giciplanmobasise ve 2s ict kpc certs asrelielt ate tal eee yarornwserelervtnsees orenaletacs memes III, 16

CAA COS VAITICIS eet aA rere tortie ler are eye cue a iekeiakey atc tenctebey clove eet shoberei we III, 16
INCraSSAtOlitmbatts) cic ce mteis uae oe ces oshetelere ce shencterehelesoleeltererenotonenstentorare III, 16

ATLCUCATIUIG esta ey ehe Ser een aria e feveenae ef okeneis eactey eno PeceteneNc totsielrerleiem eer choc ronan OT a7,

NO. 4 MEXICAN MOSSES—THERIOT AQ

Didymodon—C ontinued.

OEIC CLS ee ere rete eNc Pare re ever ce eeiers Petes stereos oss. eter oreeeh rete sae nse Suess III, 16

DS UCTUCUNO INTIS eee arava sesvevensleievcteierelcrciciets ole nic ieaeis ocho ora eee oedema acoso: Li x5
PUTS TT Seemann erecstet oemie reece renee tere arere etc nerc ek ere tomer siete eects cates oe III, 16
LOU MAS CON Steet aerer cere) otet cet ereh crates etsee tet cutee zea eee or aie aoe ae Cave oss Tis
MDSti cline Gaia Cetin merece ceeveretec stone cremate. cee Dee crete s eile ae orsr oe wht cueue ake Ill, x
Drepanocladus exannulatus meXiCanus ....... 6.60. e eee sense reese Mies
PSM ct Gee aargerieie, aes Greta ater 6 aesrcee cols geal ena Grd owe Ghai oe Seka Piss
PECIGOPOLNE CUMIN: AMADUE oo 5% slc.sis.cicceie:o.co.0 48 mcileseo asr010 acvle. ore alevniste ele shvie oes Ill, 43
TMUOC OM, SAD DGEVIALUS) se1ec susedsun cients od a sutsevalee viele ole rere [27 see S55) ci, Az
ELIF S'S COM SUMP ra eters etna riay ows enlarges tran aianstoten caine Re eee i as

BRC VAD CSM me ene ls eset tena Saat cccihh ua odie tended cee ahy aie, die tie III, 42

Clay LEO USI DLE VISCUUIS Et sir. tete ete cstte lore 6 ced sa've.cs saleuens a caee'e I, 26; Il, 2
AIEEE CT US ween tier Ae eee sates oa yeah vans Shay Ses ante tog acces tend i's audattsveees she I, 26
TILE XU CALAIS Meee vee te ey trey ester te yaa allay du ah ust ive tease n Gata feces eaves: os ezOse Lila

VitiTem Chitmmerrete oo ee eee ic teas ee nee GM CRS bin bad. odeottne bus’ Attonte a I, 26
AIFASTTT GS Oni eens sie Bh ya oe te cesta oteee e saceieé vaievs ccs parte at failed ops a Rpepees ee 1, Ax
PoMplery Sil MEKICAMUM |. hes vies dba w csaveulacaowes ers le Sis tile 27,
BOP IEC Ie Co Raters Seaeae ee na Nias. air o.< eats plate detiek OG aati O LR iiss
Brytnnrodontinm Cylndricaule’ ccc casas sees en ee eae see vs ce eee 45/5 IN 26)
METS (itrtmen ar teem tee ieee repre ona as Tacs apne Gye Re Varad eae esterase ern caine 27,
rEeysito WITT sya se easier tenner Maye clean noet ne aie sister eure een eer

CULE ENG ULE EEL! | Rae oP aE Pee Lee27
MOT AS CULT Ik mee Aedes erie Nee ena corse irne ria cian ere re aesie ie emiee Resta II, 26
ret] Oo Ot Mente yateet rarer atinceeveyeerneuenee eeatinicuss Airlie lenin sre aisieierere esos oe 2,
EOS Rca AR My irene Eales Choestatcj sits ey aveisae stage el usunra Sec verctey cera na eee ties
FetieMyMehi tim StOKeSIl . as agaidc me aienie sense ners aes dec cues un eee on TA
EMILE ahha lta c eed maha Pod Rhijsed beat G eee oon a estat oe ate III, 4o
ETUCLISS tilele UUTM Iaeyeuereisyanereremaceeacncrene ee meitiie oence eee eae Tee III, 40
EUSTICHUUNG MOTUCOICUM. .05ba<B5 es evncdlew asides cieeascuceecductsseesees 2
ao Orice CENtata panto oais.cise- vaineutrs Ware Avcusislaie crsisfs ee csess deci nev ermece ernest Jel sear
HA ATICTAVA'S oaks eas eyace a eetnera seenayeeseicrere siete sitesi srare serene ciate eee ez re lier
CGtO DIE Diarisen eae eae ia tee einai eee. ear
ALTE TIGA ae Crane pay ere oye yoo ian Meaney Te Ri tvante) oeroet eee earns eer
TENE Nel LNT cla eee eas eee ee Rea er eee een II, 21
PVECIIUIN OA ae ye ontva is Pe nace cesriciomers oi ecise titra emies ones 28 ssl 27
BIER ICAU «ate achat Aion ig Ae acre sonnei aee, sho aC eins RGAE les Gee Es I, 28
ISSTC EMS, ACCUALISS o o.c..0.5 bee trove oes stein oo olan aie ste oe eld) sisie a cle eas eecedaediees I, 10, Ir
PMTISEMEl y ereacis ain oncieeriesenToeieiciciel steerer riers MI ee ile dns Sam bles I, 8

AAS PUCIMLOLC ES o,e00 crc eens eieyeuste «ere eve 4, Atal e nEeoe te prepaa ite hudba sealed I, 14

ENT COUAT Cl lens, Sratstele vere emeiere tekinete cuss stir orerene series Smee oneal hicre sews irr; 13
Nha yas stays re vi Byer ee Seg gees cies wieises yas) yield Ges ARS he eoald b: aigetT ey
ARNO STG 5 gees Ne wre Gane cer stieae Geer eae or ortega os ade aie ln I, 14
emICraspedOnhy tS) dat acess veces es oases demas Save de beeioeead I, 12
Eeiat ya tcl las pe creepenetrcuae seen ee eee tee eeoreaeieass ete ee eee eee ee ce 4 I, 8
NOMCTCECUTHENS 4 asc esc ececisiee so secant eee ee feed ein eb dece ee I, 10
MINEO ACAI S: aewey vee ereinmeetere crtiseaiclevcle rey siecaiarercistans meee eras ees reed cncvy? aus Lar
PCA IOL SOU warraine ke cts eae aged chee ee ateicuveis eee once Sea ok ete cod aevie aloe ek [12
PR OPROVIMES) ec agate oes Se0 Sem sth aigte saree oe one a4 a oa cen ioe 6 cee I, 9

Fg MAGS Rares cee negara. n caies Chatee a Gyn coy be rerareton enue. ses Vn ted aie whine tend T8210

50 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Fissidens—Continued.

pseudo-exctlis y ona vsunc ote se Tatne conte Se ee Oe Ee eee 7 Hg

Ra vene hits xin ay-tarescsraiclene.« oc Seti ore ei eee eke eerealerenetiie hey chon etna p eters enya
POCIIMALUIUS. dcicitccrsiee «CeO Oe Aer steeO oo CNG OUT ETOC TC een eR Ctra I9
DHCD UROLIWS™ 2 een ne teak aaa neti tare te Reon ee secon eee cee cea ee Tee eerae I, 9

PORCUMTS eats Saye beceeietdon anavdlra atte Bioncrac cle eueeet ae, Neagle potate pees steno Met eee I, 9
brevifolivs. io. U eons e sociswe ine hab an oye see Moose eee aan 1)
Funatia. annulatay i.) siSoucaie ne iennin a nee etOn oon oe eS Ee Sn eT ae Re Ie
apiculatipilOSat Ach. sche eiows foe Seloers Os see om Ele seta ectae Ooi reece II, 5

Cal VESCEMS) 4<.34¥ tee aeyexeit © oe sae alee soul “tones apse eral ake easier o: se

CON VOliutay jeter veel soy cen See eee ee OE ener ee TG
EPID SH OSTE RIA. ro. ccacacee oss orise okt BaSrlA ole cut CO DLE MTR RTO ECT Veen reer ae ees

hy. orometriCay® os 4.<ccgioke s Ses Se eis a aloe eee oto a een ee sis eet ee ea LG
Calvescens: Owes veal ecient: Cele ear tae ee ne ee eee Il, 6

OLEH OPOMay us Hae Ree eseerese rele Te Sieees 1s a OT ee IE eee ees
SABEODIL vonseaveote cee ert yore ta tolerates eae Ronse tated ST OKIE CEE Ee ee TS
GrimmiaoArsenel:, octet scien saeiaeten ae Sues oo eine ciete s Seles ceo eee arenes 1
CalifOrnical We. tieo Reach oie Oe ere Oe aS Oiler eee I 53) Wiis
IMVOlUICEAtA 4 J. ter | .\besm re verucmaetare kore ereuntols te tale er eielonays orcreieds entrees erent Ill, 25
Oval Late sie k eae sega sci oiecss ae eiails ec arere ea eyetay age siete osevoteneteraisnater stern eee II, 5
practermissay 2: cei see yrmanue: mh rreseek Se eetotea ee aoe eae stoic ere oe tik 2s

PU Ba Gee oh cect See wis Soe aren Saver SR wit abet eel ora oe auc orca 2 ee 2s
Gymnostomunm calcareuniaa. amis ee erratic) ieieistoe eterna III, 6
INCUL TONS etre LOS OS TTA ee NRE eo ae Ti, 8
Gyroweisia- Obtusitolia® searentse cancaikrane os ener ten noe epee erator ere LLG
pPapillOSa Wassaeecie dels | fe eekti hi, Sreilor are ater ace “arte ouaredets avelene tego se Panes a beporeyev oncom Ill, 6
lekehoereleetebhonnal sonedojolanAlibhigls soqueccoq 4oancronaddgudondaccedacacsobagor II, 24
ledwiciaalbicansp swt cire act koe ore ete ee eee eee eee I, 21
Hedwisidiummeimberbes ameiccem peer comin ocean octet ism
ELerpetine wrong WOCCOA!! s.screc «)aciciayorel tone etic cetera store ouedetelotodtvey uals logerele einen 24
Eleterophyllumiiafiiie, Wace sere niente caterer reise sictel eo eus Late mora ene III, 38
Pasta Ole scalars See hse es Oe ROS OOO Sa ELSI a a Ill, 38
HolomitrivinisarbOreume versie fence or oe etek ere sella terse tear cn eee eA
PLOLMPEH UM, <hy.c 5% wsita eedeasls ewes Sees ee, Beate SO one heme eased Ne eee ees
SOUT CEMIN Mite cs ois caer ers ROBLES pe Bee SECC eT ee ee Le sim
Hiusnotiella: -Palamert ccs. cis crsmue see oo eie oe oeust rosie everapale eo slete Glebesneneicrs eraeyars Te;
TEVOUIEAL cy vileiahine totttare Moe, Mop Set olucic le euler a onan caay.c) MORI SEERA RE TSTE ee RCW LE Ora
CLE ee areas heehee a Bee eNOS IEG Pie TE Per ceed oreo eis roae Ee,
Peal rv Or thy cos oeesy fel denies tase creer ecole alan et ctereh telcnel Mepetetee re Ralerteret one Meare Lies
TOPQUESCENS? Cote fit 2 eel tin tesian ee Seer ear eee el eaten ae caer eats, eeestie seer aence ae petoae TES
Eby orohypntimispalusthe /sapre vtemicicctetor Srertestine se oeiedtetne eter ra ere iiss
Hylocomum Ehrenberguamum soa... otic os st cece essa sess cys se ne III, 44
HiymenostomumsimexXiCanuin > sic mitoras ssetvaseia terol ie onatotede ts toe cil atotescoiste eee 1,16
SMIUCIaphanutms co se cava sete ret ais iaclehe veel seer alee olteneor onto oes eetenew ene Ill, 5

Eby menostyliumCurgyinoStre tacsin clin rrr eke ce sitet re y-rteter rer eres Ill, 8
orb} 650) eee NS CI ico acaciocn cir cine arc omiGo ce atta salted on III, 8
INICUEVANS awe) racic oe Ge a Oe ee De ener es
Hyophilavangustiiolia ..s aay ce natant s n(n also cats ele esye anchotelala Simei a a oro ae III, 14
Bescherellei® ..dns4 Siesc.s dere oe Nee TO Ie Oo ee eee Tike

CLE MEAEA ieccar ag Saensna tA BzeiaSere fe ota mioueleie lee teceatiahe oi holes te ate oie Rete PT, 13

NO. 4 MEXICAN MOSSES—THERIOT 51
Hyophila—Continued.
Hats) all Sem Sees eevee gene ov aren tanetee tee veyetay cueite ome evens fe cosa romtinroe er coc tayias ah cerians wuioiay ait irs
AT Csxal (Uk) Asura Pe en eear Gee ee enN SUE err tee Neier err Gcn eect se @adale) stnkege keieuckarete ic Pil. rs
SUDAN PUIStINO lias epumeyse meeieiclterele meoteecaierei asics rere dee ees tee enol III, 14
elaygi OTL U (Tle CEL UG Cet Semleneperene nV oteaetene ei enehcter et seve eier dee: cls sk ciniaterst ores: siti ayn. ti eudssarbusicce Te20
ANNI <LI Caen ees utente tee were Sen aye I ICT rates ote oni pohyan ares ene e hd dh 4 III, 43
CHEN SO PI LLUGIN sarorerace oie cotnersiascstiavecs Saathsnscere © lelse ete rar'sroialete: olsnselavelle aie. axaltore at
GLOLOUEIL GOVE ET eS Tee TA Se Mises Bod ee eee IBA ei
GOPENO LMETH rete ere ae Tate RSet us Cites Ries asta leah Sas EERO ae LT 35
LECTIN CHS mre Ree tan oo eee skier aes nGBn Ca SHEN oto ANNA foe eee IAD
PAN CIUD CONAN UIM ct se specdeT lesbo lec ss fo se, Bae oetS Seta eoAeeT tse Misiare ub ae Oaoei III, 44
LGATUIGTILLOSWU THEM metanees orem coreMe dee aPSEE tego asl Seon a fe sT ea. wee Sakae Petes ectieha a 35
WarbOmalcOnutt oo. 0c cece cece ccccrsenscce tee bevettuteecccersess Les)
VOD EOUUCH OG OM DUT aeetee ee srasstden totisacusteies oid ates sors sie) sais Gis Sees F saNeNS 68s III, 30
FLOVGH.O PHL LLU MMER et siete peee soit eRe ele a MS ett coco fs Ge evaded ce alps < S aence SeETERENE sore II, 24
SLLIES GUO TTILC Met eadare, cia SaoN aa esr ashes eerdees SRS 4a. 16ad) Me dh aubys eis seners vised wee yea tees DU ee
LIUCUUSLG CULTIUMMMPRCCMNET terest AS sae fei ioca ashe "as areas QIAKOE ints ciate. ote oer III, 43
EISUCILLC CRM RR eg TES ARs 0 oh Peg cake Hata =, 6 oe eter tg PE Oh cea As
lsopterycium) cylindnicarpim, 2. ).2.-62 0s. csc eens eee meee ne es ss US
LCVOUSST VLU TN metero coe otk tee asta aL oc Sette nied ules chstens sess, Goede aciotewste teen Teas
ADA TEE CMM escalate coreg ses Iw avec he GO Ate Sire WA sass
WWeptobiry um) PInMrOnmNe e244) .ccuce ease sss cee eae tere -ses+ scans euessuas I, 15
PD ViLgLAO INIT Cemetery sacar rt ete ors Be ee aN anaes Caceres teeire siesee n sunray ene to 27
Leptodontium angustinerve ......... 0... cc ce cee cece ee cee eee eaes III, ro
PENSTSS TL Clie Wades osha Font Le PaMCNe UES A Cay cfs Sees IES cs ates ous, Span cea Tee a we cigtanenecensie os III, ro
NaS Cll Ul] fleets reenter emer reeteeteiterc terete eae ciercier aereeae eas Lenn
LCS CEN Summer aera emt c a tetor sere ier ce ec III, 10, 14
MelICOIdeSp eee ee eee eet eee eee eee I, 15
SU Llatgs@ SULIMLp ae ueneners sr tererrelcren terse ecineaey eee sitemeter recs acleereeentioreeiconen ie Ill, 11
TAL Gea liver eaten ep oy aueee tute cent a avr cue enene cies eierece etree COs ores a eaare eee eas
Leptohymenium Ehrenbergianum .............00ecee eee eens bees III, 44
Leptotrichii leptocarpum .....0ccc veces cccce seme ne ccsta nt seessessuvavs I, 8
BASIC OP ON PIQWOUL. oss ters aie sae etsiciers oe noe'e ere, sate ene, dae) sieve) renee G¥ op Biss bie scien Ill, 34
PILEALUG OGY pA eINrye erat ie rete ee eee aoe Ore ee oe rio II, 22
Et COUMOMMCHYPtOteGaman ame serie tee rieereeicieteieer te ere leis ere ee tie awe rz.
Say EUS US tnnmenam haute Rian wma es ee Geis a Patna s ewes sees «es. oe 24
Bima pencias mexicana, 2. ev. saueis situate ele aisieiei sis aie0sg slesscosu rece ere echomte, 2 ly 225,23
AGU IM Ate eke mrorneee erat te iceiceereeecee ieee eee hceiopeie es Il, 23
Opal el roe ver ee oie ce yey cn eyeraeueue ere ser ecticee aieneleetene teen enn Ie Tr el enti II, 23
Macromitrium flexuosum ......00c ccc vec cec cc cevcuceececcsesevevcettes II, 16
Rete SECO TIM secsyn as yeriiute eee sy asitee so ae en are orien tetas dla xe Ck es ans
IDO Cte paca erers ayaa riciraeee en erciere emitter Tae eiass cry, Sere omhectae thar Its
TIN Gel CEUNACT TTD shay chase eer ey onsts veeRye eee Vn ene eee ie ones ohare tan ce ex aiaes. Occasions sansa ohne II, 16
PYGMOp iy ltt rege seeeste ey epee trace nies sete eter ere noe a Mieneye ce cart. II, 16
ee ei os ee eres rena aera we eee are ree cash Sletee ssh es sha Seba MB yn Po II, 16
LOMEMOS tle meaennencvercrsverenerevensusinicrerceneearie areca cae eine ere ee tenet Il, 16
Miefeoriumillecebriutm <a cs 4 cferere so ee ee cel scle ae oa. conelewies soocusse. 6 II, 18
HELE LILO EI Cie serve ageiaiets epee earch crarde viele cise Moers eect ain Sraneoe cat II, 18
Wietzlerella scoStaniGensis: ¢..00c0.ces- cave suiees seecneecewsemicevivudecten Li 2
Neotel T3102 ary oy apse tan oreoueh i oensienea ay sreconrereetye hole terest che einer rere saaeie even eee

52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL.

Microthammnium. subthelistectuml \..oq-celece eo ceiice sceictcneelstenerecteuets Ill, 44
tel TSC meta a, < chs laleas specie Gi ck Metio-a) seca ayes emeenerone eiigatana itesar alters tag rewee ten TA
IMB Sa noone Seimei a goocokoonsoedonuiscn cu dasoddubooopoeDec TY 25
Mittenothamnium subthelistegum 22.0... ce cece c sees scons seeesencsees III, 44
Mintobrytim albicans.) oo c.csraie ici ios ceteis te baie icles sc orera ie terete
AMES SU Uta, Geog pises oes Chews ease eseacacel ett renee pedals ee ane astetene chase hatchet coher eee III, 28
Menivim walbrcans\ a: Aeacaditeersthee ree oe ee teas CIS: eect cine noe ene Toe ae
POS EATUITINE meters epee lars iavekakerenelayererelsiiev er cuoh aN ehencietetenaneyre ovetarel loch anc tator Nene Me ice i estetens
Milendoa -obtusifiolias .a2.. i422 eaten cos ocesee bate orc ee eer Ill 4
{NGLASSAtA. sais ool ere ers alee Rie aie che Oe Nee chen ere ee) olein: rete eee Lie |
Moriniasl hrenberoitamea cfc cti ste = oro crctars close segalotetaget ete fareleee nel teterctet Aletta Like |
PUIG SEOULOLLES™ xcs eed eadavs HORT INE eets to ore weotatoney aI ehedhe Ree tO ere Ti 22 |
Niecleeran chiloracatslis: caged snc cates sana ae crscn nore crolenciay Sstepene eicte layer sieyeinre eaehe II, 19 |
lornschtichtata <5 seis ces wisterstete etsy arenas = eielolsierietetenerereie omic overs lot teters II, 19 |
MIZE SBI eee oe Rae A Te ENS ele ote STG ales tel aoa etan eee hechctes Naan II, 18 |
LONGUS CEG he oa oxciereishe ote artes ie niet Meroe eoesn tat te sense Nel elo Coteaeh ee II, 26
OPDIGhVGNG 2 oda cine co aveees visinns Stine tle efaleler= @ wale le) m (ress olsie sey iav aioe II, 19
BETES AS eee OS TAS SR eee NaNO OTe CO aires Cee ae es Il, 25
FUP GOSCOMS © esc wichosaisisnyeverar8/ ey shalin ae) ese ei ese tele tie) «i ein ofore! oats tol aelaera esac TL
Nieocardotia cubmigra sc cco ace scrim telole + alaynininlajesat= alot siete Giot-poteregeuelers Lire
Oreas yy Niartiamayecs cicero cetecreie oral evar elciote lor crch atete clteneieneraeite te enn Narcrerse Jules
TOK ICAO. &cvecciecdasoarecrw arti ity lads ooied eps sien oiera mis oe emer eel oO stave oteuctotie as Ronstore ie
Oxrthodicranumpetlawellare: 426.4 weirs selore wlete ee ue cyelete, ow errpelene Miel=ts ache or eres ee eA
Orthotrichum’ diamham tne aco e cess eee iy lee eis UR eye exe recente Tos
TSO Z ATOM ere EAST onion le Facet on eveneNo rsa Pech ane DT, a6
Taal ACOp Mayline tere eters ete tetacad eos) ona ls oiete sellin over ceed ace ep eyetaelenece vel ero kenele Th oes
pYCTOp liyelltatray ertstee mre tsre wie rere vsl ee cae Osa er ole lye mileysi ofebereteranio tele a lotesenteenere ers
RECUEVAIISE Meee rae EE CI eon oe oe oreo Ncrenlotcycketeere Liens
Papiblaria appressa qo toca ce eeierede cic fete es crensh vue wuctate fafa tone orckaienecels terse a tate II, 18
DG ppeiy woot eae erate ess eas eer eto tatete io ence tncetote cleioteaeae ate a iea ven she II, 18
EAH Tite ree, on cetera Se ee a Te eee Nee ee toeneetet II, 18
TIUSTESCEMS. Ses eteieieiel cies cease Ma etesehae renr fe Reve reat eda eke: her teeter eee II, 18
STAD Ade OL teh er ee ae ee oe NI SNS Ios ed eiict nce g heey arsenic Cet Reena ee II, 18
Philonotis: amblyoblastas aajace otis otk liek trope Voteloo s oveneovercteren ete ole otenaTerayre rere I, £9
CUT Vat annie neice ote te ate aC ere ceener tenet eto choice tol atone on ean retest neni Remade I, 19
Clesamtuilgy ccs: erect ci aienneere teenie oparean ome erorare Gieleter ete ccefete lazer tier aero I, 19
STAMUMICOIA a aynins <a a e Cee ee tenet ane ke wld teaters oe telene oka teter ies tele [, ro
NA TTe CONS1S enone ork ae eee abayete eater a eee ens reteset NOI acto ceee sees a ctemeretee I, 20
TURE CEAL CA otitis erccvag cs ote ere eet eielcw oN oneTe oreo eC ekcpet eels orelionever cde tarsi ere eeey I, 19
TAAL CATS. Sea etars spac pense erate Oe Meira erate edsiote lene Leiria oe eens Lao
WETAGLALISY aves se iechresstc oe caverec Deeg aay eon leet oapst econ eko ears tees Steer tel eae feet cnet I, 21g
Pilotrichella cflescillish eesmersow ecient ioeioeteys aero terieter eet ey rneisr terete II, 18
FUP GOST OMS iid aa Shae ROTA Ia einer ee See Oe eee 118
Pylothichumutasciculatumls acre rie etote tere ren ri ereicne(naretereretsens Il 2
ATIC ICANT TI day ae rese ts tate ott enasa eleeie cea sien ese ORO ee OT Ieee II, 19
Platyeyriella shelicodontoidesie tac ccrec ers creates tcc stoke relator poleteheteoltehter eae

imbricatitolia neta eee eee CeCe eee Se ele

NO. 4 MEXICAN MOSSES—THERIOT 53
iatynypnidiain ODEUSHOMUAD 66s. cc cuss ose ve Go ccs o's « seeid vas eearw ates ie ss
SUG UAL MEET reso ere rere Srecetaraie cite nici east es ico non ears ace hor eau oes lees
Pettt1 es Ud eer tere che ae RII eo a aches acted Aes aise tetas Mood Lit, 33
SSNS LU SESLA COE Ig 1Ot els taper a Cyne ese e's a ee GE hes doar are es Wc a a sivas iiiess
lenrochaete Mitel vaca aac ein ssc fe oo bee arcla lees or Gtelecece so2ene'eereealy ecets'avv eave I, 14
TILELGUG OT] CL RAINE RO TO rhe tect dae T cA A a ent Be beh a Ne I,.14
SCL Tg OS mercenaria ie See Mvag e Mee S PNAS cas cake auction spt cesblac aye CeMaNen ioe feta ta ate este Pers
Pets HOM DIA] Se ae ean eeie aha deus Sia ais abiac ateiads enlerdiata comes ae hes DM 3%
Reo OMIALLITY yA ESI Oasis Gos, susiera tate a, aisye ¥ ACh 6: ee wd ase had aoasecdee oS wy Ctedordcg a weet ian
BS ESC HEE CIEL OMe prem cyan ls 2c: Mhe. eR Saha.) Spain So ese Una. oO a III, 44
ILL CED CULS Camm ee Re Sen ela, MEER tay GR py ae ar 5s / USE yesh nds PSUS HERS tae gous) oft enc et Mons 2
OVILO SULIT UMMC RSM Pepe eNe ary nes as Us ake honey cus ae eee dh diavel at TE Se tO ise eaeeS I, 20

GUS PIC Atlin cet oer mas Ooms ieee: Naked sei erSue Oo an eatin evateicey ores Ti 26

Gy ci CUrtip 2 cPepeicteke orcie oe shoe bes Ge leersiers e506 Hg ssi to sso wiscatee nadine oe aye arsé ane I, 20

Efe Sot SLO TTT alee reyes sae oud toe eaiyore oso da Payee thes fon sesh avandia chaser sc eerdite Bek Rdae tues I, 20

WO ZA OUR heer tin ok eto ae Sean ores ASHER Nee rae hs Gath atnad II, 26
IROZATIO Ihrer Ma tere tens, Aerie oer creme ie cians ciate Sieve eicleie e tiriaiere memes cre teeter: I, 20
EOIMICSTS Camere eter tena areyare acetone ots mucky ase vey cre ietgis ee vere fee ae RG osialareis ler

Fad (E10) 5) 01) <a ee aE ge ee ae

PTR AUCH LE CONG o5:a¥ ate his aihtiniatdigin as Bina eae GG dd idseee: $614 49 Hiaserecdeea me done i
olytrichtim calpintiorin€ ..c6--e..e00c oss sees cs senes sense we. le Tae Dl maea
ATLUTl eUTstd te eueverete eyelets siete arte ei sicimantens seco er oieiiel aie ierelcneretereieror tea reer eke J, 21
MLITIME TI MUTIN, eyetes aerpteeises eis ere esters sce erereree ie ves eretereuetere siaeiaiere [een leas G

He TeOTNG Gl OLY NeESETI Clap foycusaaieveveroievene sierevcte site sities errs crekaycas stein sir ietel es kare oterocicde I, 25
GiliciteiSumeterasta ier aceemerstierey cient icietete ec Steise eerie ciate ets ecto: I, 26
GIs Smee ge erer tay recreate cietceorsiseticn sae tare athe eee ees I, 24
VAEVAVISCUL ti Speers reeneien eres reterweneiene siete tstei eaters oy ctescu ec is 5 Ween Gee
LCG DOC IITs neem eros erence rece Skate ce eh dare nt I, 26
MITE CAMS, caper ee saetere einie eo etoncnaiers terete ears si aehae Cote eee ce witiccs oem 4
site a CHanererearsiscis reise cise teri neers earn Mea ee, 2 len
TASCUUOLES CMe SUUCULCHUIGLO: sere aero accins 2 ee Seine pede © veel Sieicus ie ieeteros « II, 24
Eterigynandrum filifiorme mexicanum ...... 555+ ieee .e0b bees e de dew'ws Ill, 41
EAeCLODIGVONSIS Illex! CANcdap ere emirates aiieiere seeinciione aarti ea esiare esc Ik 17
IESstelTd Oi tomes erases sialekeresversuetarersbens or ieierers camisetas ei eet sxere on ar hee Is Beal TE

Gy GHOM Ute pI OMiititItieerel ei olor wits eestor eet aecters-ceie <6 e e selsha (ous ocae x Ill, 30
Hesvaleuisi clinic CA tcl mir erereaeeiesates oreeseerre cities meter eens ats Senne aw eeeoats uate gad 4 I, 28
ITILEL ICC 1d Men ee eee ReCn ee Reena tii ter ieercrcien tennessee eleva sities c10 oe 1,28

SC IIMAMCEL. sae pad acabaw sede toe sajaietviadls oho Ladi eacteteie 2 OF Decachydatiaie I, 28

Sil biiell Gata mane Na ory airara iter aes tesie amen eet aeons usd. 6 Se tea wa nies I, 28
RATES GC ACCII 111 <1 Cl alee teee Neer terror areustecvereate te ise kaltc1cue sites giendveusetle aka ee cs I 24
erigcl dia eGchleatitGlia: S222 5.01 «oer gos suse ig Save acaete burg aeiwralaSated'eed II, 17
Maco pilin latistipulacetitmess smc desis cf ecin« eae = wesc «vice ase ev cerecfs one I, 29
EOUMELITOSTILIIN exeressererere ers cae re eee eter tere ee Caio ree ttrede cee St ed Sse chictits afisk Su, S88 I, 29
napoidoprhynchitim dectimbens ..2ecae. 2% oes sla sca ors weiss woe ses os lil 42
GME OScratlim meena eter evertererstere Corre ims rchecces ae hae ous Malo cgeiayaee < iia
Rhaphidostegium caespitosum laticuspidatum ..........: Me ers eee ITT, 42

incom focommtilt roLmlsieemeaeiersetneaicia sities cts ecco ecteietnctenreenie ook oc. e Li 22

54 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85
RihynehostesiellasArsener ataeecas eck roe eer eee eee ee cere eee lil, 40
VACQUI to widatee.c Aer lot oe Sel eae Oe SOE Ee OU eee III, 40
Meesdaliil ..cheseecsdeacvctas ie MESO AL ACEC CR eer eer Cio eer III, 40
Rhynchostesium selamper accep acc cee acie eit heir LL 42
ttitOmalComumay %.,., cise dea apoiice Gavtustsnn athens Wave ooo eaters eee eae III, 39
LEPLOMEROCAN DUM sce niekia cree serse eae Seas Ie OO eee eee ili 349
WUGLACOCTO DUAN. totat cis diac syaraue sd eustions ot aticone ash ekepeen ene EINE eR PISCE ieee
OOUSUFOLININ, oaioraierecdte sin Nafta eae ache Rese ee a oo Oe COO eer TIT; 38
PUG Le be ateieiaavateres 0 suai cD apeua tro el aeee a oe SO IO aoe ee Ss
Sammt=Prerneiy. Bais cs oye oss ecco aco Sis ere dora Olena tone tea ee It, 39
ROZea+ BOUTS Gata ver mec e icccuelsike ete tae tonee Ra eee ee i228
SELICEAL .< svsdete « Sesis vos Bicvoder grsveucutivel omairoraley sicko wesee the eral Shek teers cla alc ce ae ene LIT, 4a
Sematophyllumecaespitosum seer ceeeiece aie ein ee neiaron ner ane ne Ill, 42
laticuspidaturnin tose remot teen eee ee oe ee ee Ta yA2

DE CUMDEINS: ae Sika tks yd cn eee se NOE le OO ae Ee III, 42
Elaimpei” «i csuaeseie cer Gee eee COC ee Oe OER ECE eee Ere Dies
ODI GWer OSIM ATU 6 See Sacco Coe AES ee eee I 42
Stereodon ataleatusir meme sco oer eee eee aT ee ee IT, “43
SILO OLCOBUS) tevin fersjee See NI Cane ASIA oe Pics er ee 43
Syvanblephanise melicophylilagerecicsspencettscietien eile creer ie enteric eer ee ie}
siaxaphyllcimysplaniSsimumnenrsee cm cece iter ies teenie Ill 4s
NARIRETE Vis ak Ne eye SPR Cea OOO hae oO De eee LS 5
‘ayloriawtontelloidess saa. crac en eo eae Sore ete see eee III, 25
ihuicditimaimest Canim ete eee orem eerie Enmore re I; 24
MiTAC OFICIAL eee reer One cae Ion CCE 25

OV EOC OV PUNY racks seas SRST Es OO Eee Ee ees
HODUS HIT er epsesecs, ace sesso aioe Ble aie Sia outa orn) ene cars Maree Si eieraisie- oleae ee 25

S Chiluimmbe ne erie varsectssecalee Meer ee meee GP one ae ERG eee is
LOMLOVUSCUNALHE MINE UC ONIN ecotetsee leis eietele le eeclclel sire cieleieiolelcieieleiel = tle eter ti2s
Aber Clee limite epee eras ie arrest eee ic IS EN ree NIP ae ee ee ee Tia
AMA USTALAT 1. caters ress sreeyavoReuelfdeo tier c Suet govehceelece ions Revel re: cLevoten one aoe Ths
‘Mimmuicllapanomalauers erect eee eee eee ee III, 9
Sua mOrmal aye secceve wacko epane sooeease chan ere a areata ake ous see aa teen nese eve esis Recor Reger Ill, 9
Tortulacamphidiacea® 22. cere ee eee en er ee ee ence Ill, 23
GCOMPUSD (Foch x eee cs ge 6 oe ato ogee se Naialalat co Pete rotates aoe a oh ed eae oer Iii24
COMME CECWS Scotian Guntakomes atene WO, = taueilees Bic Mt ales uereua cle) element tae Mesa reHST LCOS TT, 2s

AA OAT Se mraye cre cote ei kiclyase Greteae koe ceo me Sierra ic a OTC IO CEs EMG eee Wi; 24
OBUASTSS ITA Lee. acd cct oe eacre see eesnssct ocevs eae cnt rtNCes ean us ete cio ea ee Ill, 24
COMME CHENS: sears Sac er oe roeice eon RCT eter ere tiers iis
Papillosae wroevsiec pees nets eae ale este ee aero nreme pero III, 24
parva latitoliay. 3,2 ot S-gcew ote eree was ste yciinve tacos sasere co eranctvarer tcccaansttoun epee Ill, “22
PAG LOU, sports in tnt site, een Lagan oo bog aa vers) eS. tere Oe ays Say ey syatte Po ararenaaa eee Vere eee gees III, 24
TUPI COlA Mra gy creuslshenead ues ens «, sto bateucopetsunyehtc vole mravel cleyaacin epee teen eee Til, 23
SUDO NUGT OL, oo escih abNe ooeie oi ste von eeUeiee SS Re anne TET ShG Creuse ere ae Saeeerrovorst eer ayers TIT. 2
TrichostomopsissCrispitolia, eee aaa ae ie oe Oe ee eee ee To
GCUASSIFE LISI. ccm aieclereies avs aston cor caens sence nite tk euchaiionemennee caste cocker ney areere III, 9
Mir1chostOmuUummncantpylOCA PUI) avers eysreleie aicratetcelotele) stoyeletaiere alehctenstieleraiereheoeers III, 16
chiorophytlum (brevitoliumy @.er o. nsqsacr seer cies oer TIES
(Nt mite head Sore etucts olga tower corsa ter cuvligc es sho cle oer Ne ere Tara hen eta Ill, 8

TELVO VIET Suc: tse ds ratece se eee eee EEE Ue ALLS Site BSE TOT tee en ene III, 8

NO. 4 MEXICAN MOSSES—THERIOT 55

Trichostomum—C ontinued.

DeMMDO UNE CAUITIUm teres is, eeeiere rs lercuer ster eset eye vs. cos airs csc. Meenas teatpmeseushara ese III, 8
LLCO UU amarante charsieiens 6 iskcierNaielis cist ele ren cael arc fe Aleianc ye ae eke eieal $s, OG couse ts I, 14
© ICEL ITEM OT ec earch lal oct ce oan ocoheaty setuthe dere cote III, 16
ee ECUMOOL ROPE os cease 1c eo eja eee pO nGeNee esd oY Sie EER eR Ee hea De III, 9
MeriG etre lar herEiUSiNed: Sarsac. aie creas ccs aidle lade .c0es sued as ee shece sundae Siwy vis III, 9
Bose Cll cata leey C1 Cillatise Perenere se cvaretars Geese sres oy5s Fin ivan sicaiebe ot oadecd ke tcmttetiay nical home III, 43
vee [espe COMTI TINA TicUt cameo ays eee: cece eee roe ose tee sage atau adestslay sea pe veh te sh bs taste Moma aie drese eee
fava lAT1G ta] Gaeta reais reret eSNG ls Miele eiaxs yo aera Aver hata eyaa ee CoA a haat ses Shoe Tea
RMLLIUCL IGG CE tare Sets Lopes Sie Sie RS RATE Silo BO yep eG Bu eeay Nea BUS esac take ce oe hoch ITI, 28
CGV IN OM ON tla weaes shstlerss es ca qe sb ohans 21, Sa as ivciole-wagtna ahaha ee eA ies le2o
RAVAGE mersrtepere eee etc ahs eat see uch ey eae RR Ay aith BAF ald Siobndystal ste vadiarolas Gleneoks III, 28
ISGUIG Om sail lei ie. ct tateeeterercics everarey cleus rose: s/S)ace-aaevens wieie colores aieleue arvane eee /ee III, 28
TAC GUL 01) LSM Ie yh Pearc AROA Se ssh esse OA aes) ata) a ss yauera syseynre es Tiegs Tih 28
ZAC AMECCAT IA alahetcs cc aes! a. oiee take staie' aft ate ere endls: se tsabehs-ocdia 0nd ape. ected aeons os II, 7
BMersIONSIs. SUCHOCAT DA, </s\os esis wis si eae oe ea cig ged ened syns ree wee See 10 ee
SHO MAO OM bal meverts sin aeevaieiarel tues a eusiek cis wisielsny ysneye 6 Goad’ gee esis maner swve.g es IIT, 15
PY COMON ODEUSITO]US Qencsscice nce sassw oe sass e gems sens eweessmaecvenes II, 14
PUM OI isis feta st Ase hr iStr een ne etotvig hts ave ipa ear bb aot onayeriyseeatease ee 2 II, 14
Seta OHO LITT Stata: systah are ores sHesayselateo es totais exaea le asec a1 sncweicl <n «, wuaregpnarecche were Il, 14
ERRATUM.

Part III, page 43 (Smiths. Misc. Coll. 85, no. 4)—For Isopterygium planis-
simum Mitt. var. laxirete Thér., var. nov., substitute Taxiphyllum planissimum
(Mitt.) Broth. var. laxirete Thér., var. nov.

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VOLUME 85, NUMBER 5

INFRA-RED ABSORPTION BANDS OF
HYDROGEN CYANIDE IN GAS
AND LIQUID

BY
F. S. BRACKETT
Chief, Division of Radiation and Organisms, Smithsonian Institution
AND
URNER LIDDEL
Fertilizer and Fixed Nitrogen Investigations

Bureau of Chemistry and Soils, U.S. Department of Agriculture

(PUBLICATION 3123)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
AUGUST 5, 1931

The Lord Baltimore Press

BALTIMORE, MD., U. 8. A.

INFRA-RED ABSORPTION BANDS OF HYDROGEN
CYANIDE IN GAS AND LIQUID

ByiS. BRACKET,

CHIEF, DIVISION OF RADIATION AND ORGANISMS, SMITHSONIAN INSTITUTION,

AND
URNER LIDDEL,

FERTILIZER AND FIXED NITROGEN INVESTIGATIONS

BUREAU OF CHEMISTRY AND SOILS, U. S. DEPARTMENT OF AGRICULTURE

INTRODUCTION

The absorption spectrum of hydrogen cyanide in gas phase in the
region from 3u to 15m was investigated originally by W. Burmeister,’
and more recently at higher resolution by E. F. Barker.” These investi-
gations have shown the presence of a strong doublet at 14, weaker
bands at 7p, 4.74, and 3.64, with another.very strong band at 3.04p.
Whereas the frequency relations supported the view earlier held that
the 7u, 4.74, and 3.6u bands were respectively second, third, and
fourth harmonics of a fundamental at 14, the intensity of the 4.7y
band led Barker to question this interpretation and to suggest that
very likely a new fundamental was present at approximately the same
wave-length as the third harmonic. The 3.6” band is thus more
likely to be a combination of the new fundamental at 4.7, with the
lower frequency vibration.

The band occurring at 3.04 is recognized as another fundamental.
The bands at 14m and 7p are clearly of the doublet character. Mol-
ecular moments of inertia are readily calculated from these data. From
Burmeister’s curves for the 14 band, yielding an apparent separation
of maxima of 37.5 cm.}, one calculates a moment of inertia of
33 X 10° *° g. cm.? If, however, the relatively large slit-width at which
this work was carried out is taken into account, estimates may be
made as to the degree of overlapping, and more probable positions of
the two components of the doublets may be plotted from the com-
posite curve. On this basis a larger separation is obtained, of the
order of 50 cm.'. The 7 band was investigated by Barker at suf-
ficiently high resolution so that no such correction needs to be made.

* Verh. Deutsche Phys. Ges., vol. 15, p. 589, 1913.
* Phys. Rev., vol. 23, p. 200, 1924.

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 5

2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Because of some uncertainty as to the exact intensity values, how-
ever, the form of the curve introduces some uncertainty in. the
determination of the separation of the maxima. In view of this, and
of the inadequate resolution of Burmeister’s apparatus, Barker’s cal-
culation of 13.2x10°*° g. cm.” for the moment of inertia, based on
the separation of 58 cm., is perhaps in as good agreement as could
be expected with the corrected value of Burmeister. As the band
occurring at 4.7m is certainly composite, no great significance can be
attached to calculations of moments of inertia involving data on this
band.

Recently, R. M. Badger and J. L. Binder* have carried out a
photographic investigation of the absorption spectrum in the near
infra-red of hydrogen cyanide vapor in a 280 cm. absorption cell.
They have observed two bands in this region occurring at A7QI2
and A8563, respectively. In these bands they have been able to resolve
the fine structure attributed to rotation in the molecules. On the
basis of their measurements, they interpret each of the bands as com-
posed of a P and & branch. From this rotational structure they are
able to compute an accurate moment of inertia of 18.79 x 10°*° g. cm.?
This calculation is in a reasonable agreement with the values based on
doublet separation for the 14m and 7p bands. The absence of a Q
branch is in harmony with the observations of a clearly doublet
character of both the latter bands. In regard to the apparent central
maxima in the bands occurring at 4.74 and 3.6u the question naturally
arises as to the possible presence of a Q branch. It should be borne
in mind, however, that these may readily be explained as due to
overlapping.

Assuming three fundamental frequencies corresponding to the
bands at I4pu, 4.74, and 3.04», which have been designated respec-
tively as 6, v2 and 1, Badger and Binder have interpreted the near
infra-red bands as 3v,+v. for the band at A8563 and as 4v, for the
band at A79g12. Because of the absence of a Q branch, they have
assumed a linear arrangement of atoms, and on the basis of three
fundamental frequencies, offered an interpretation of the three funda-
mental modes of vibration corresponding to these fundamental fre-
quencies. From an analysis of probable atomic distances of separation,
they have come to the conclusion that the molecule must be hydrogen
cyanide rather than hydrogen isonitrile (HNC).

* Phys. Rev., vol. 37, p. 800, 1931.

NO. 5 HYDROGEN CYANIDE—-BRACKETT AND LIDDEL 3

EXPERIMENTAL RESULTS

The results to be presented here were obtained with an automatic
recording apparatus yielding high resolution and possessing certain
novel features. An earlier self-recording instrument of high resolution
was set up at the University of California by F. S. Brackett, yieid-
ing on effective slit-width of 10 A. A similar instrument, but one
with considerably greater aperture, was constructed by E. D. McAlis-
ter at the University of Oregon, yielding an effective slit-width of
6 A. The instrument used in the present investigation at the Fixed
Nitrogen Laboratory is of approximately the same aperture, though
of considerably greater focal length, and yields the same effective
resolution.

Prisms & Litfrow Mirror

Collimating Mirror

=a bes en rae TES sega
Bie Sakae ei ee ee — = ri ource

Fic. 1—Diagrammatic sketch of spectrograph.

The instrument is of the Littrow form, wherein the light twice
traverses two 60° prisms. The aperture is limited by the smaller of
the two prisms, whose face is 20 cm. high and 15 cm. basal width.
The instrument is used at a focal length of 2 meters. The body,
a large casting, may be sufficiently evacuated to remove atmospheric
absorption. Wave length variation is accomplished by rotation of
the plane mirror, which is coupled with the motion of the photographic
plate by a lever system. The use of a mechanical lever system with a
variable pivot permits a wide range of variation of relative motion of
the plate carriage to the angular rotation of the mirror, giving prac-
tically any desired spread of spectrum. This improved mechanical
system, together with the use of photographic plates instead of film
or paper, gives a much greater reproducibility of spectrum and
accuracy of wave length than heretofore obtained. The calibration
was effected with mercury arc spectra and water vapor bands, the

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

observed wave lengths being consistent with grating measurements
within +2 A.

The thermocouple used is a modification of the type of single
junction vacuum thermocouple described by Brackett and McAlister.’
The source of continuous radiation is a tungsten ribbon filament,
using 16 amperes current at 6 volts and working at an approximate
temperature of 2900° K.

INVESTIGATION OF LIQUID HYDROGEN CYANIDE

In the present investigation, the absorption spectrum of hydrogen
cyanide in liquid phase has been studied with cell thicknesses of I mm.,
Icm., 5 cm., and 30 cm.

The liquid hydrogen cyanide was obtained through the courtesy of
W. B. Wood of the Plant Quarantine and Control Administration of
the Department of Agriculture. This product had quite a perceptible
odor of hydrogen sulfide. The original sample containing about
1500 cc. was distilled over P,O;, primarily to remove any water, but
a considerable quantity of sulfur was precipitated, as was expected
from the presence of hydrogen sulfide. The second 500 cc. fraction
was taken as an experimental sample. A drop of it did not affect
lead acetate paper. This purification was made possible through the
courtesy of Drs. G. E. Hilbert and L. B. Howard of the Fixed
Nitrogen Research Laboratory. ,

aN H
\
|
<p aon ae Oe Oe ee aE SL

Fig. 2.—Energy transmission curve showing absorption spectrum of 5 cm.
cell of liquid HCN at low dispersion. Slit width approx. 40A.

Figure 2 shows the absorption of a 5 cm. cell in the region from
7uto 2u. The steadiness of the thermocouple will be apparent from the
smoothness of the record. This illustration shows the instrument set
for a relatively narrow spread, covering the entire region, and gives
a general idea of the relative intensities. Actual observations of wave

Rev: Sei. Instr} vol; 1, p: 181, 1930.

NO. 5 HYDROGEN CYANIDE—BRACKETT AND LIDDEL 5
lengths, however, were made mostly on a much wider spread, in-
cluding simply the region from Ip to 2n. A typical plate at this spread
of the same cell length is shown in figure 3. Figure 4 shows an

! 1 : Ny — ~~ eee

10 Ll Le i 14 WED, 16 Ly 18

Fic, 3—Energy transmission curve showing absorption spectrum of 5 cm.
cell of liquid HCN at high dispersion. Slit width approx. 9A.

analysis of the bands in this region, the frequencies of the maxima
being plotted against percentage absorption. Table 1 gives the sum-
mary of the data obtained. The values of the fundamentals in vapor
are inserted for comparison since no liquid values have been obtained

SY+d,

2),+20,
t l
5000 6000 7000 8000 9000 10000 11000 12000 cm‘

Fre, 4—Diagrammatic representation of absorption maxima observed with
assignments of designations. Broken lines show vapor absorption. Fre-
quency is plotted against per cent absorption.

in that region. Intensity values, both as to percentage absorption and
absorption coefficient are only approximate. Generally the estimated
uncertainties in the frequency values indicated arise from the diffi-
culty of setting upon broad absorption maxima like those shown in
figure 3. Still less favorable are the conditions of the values for 41
and 371+ 12, which were obtained on a low spread plate such as that
shown in figure 1. In other cases the larger uncertainties indicated
arise from the proximity of strong absorption bands. It will be seen
that the observed values of Av lie well within the probable values to
be expected, taking into account the normal variation to be expected
in the successive differences, together with the probable uncertainty
of measurement. The agreement certainly excludes any uncertainty
as to identification. Not only do the wave lengths lead definitely to

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85-

the identification indicated, but the approximate intensities are con-
sistent with such an interpretation. Of the entire 15 bands observed,
only one has not been identified. This is almost immeasurably weak,
and occurs in a position slightly displaced from the frequency where
we should expect 3v,+6. This excellent agreement throughout leaves
little doubt as to the correctness of the choice of fundamentals pro-
posed by Badger and Binder.

TABLE I1—HCWN Bands in Liquid
Abs.

Notation Abs. coeff. r v Ap
% k A em.) cm.-1
4” 6.6 O14 8000 12500 + 30
3000
3" 57. U7, 10527 9500 = 10
3090
24 96.0 64 15600 6410 = 10
(1) vapor 3200
217+ 22 * 2. -004 9500 10527 = 20
2040
211 + 2 47. 213 11787 8487 += 10
2077
2” 6410 + 10
4v2 12. .024 12540 7974 = 10
IQIO
Bye (50) oe 16490 6064 + 10
(2v2)
(v2) vapor 2090
2r, + 26 14. .020 12760 7837 + 10
719
21+ 6 8&8. 42 14050 71S == TO
: 708
2V4 6410 + 10
(6) vapor 710
3r1 + v2 207) 008 8650 TTSOI ==\25
2061
3r1 g500 = 10
713
31 — 6 10. .O17 11380 8787 = 10
3r2 + 5 42. Ll 14690 6804 = I0
740
3¥2 6064 + 10
2r2 + 26 (20) ays 17900 5587 +15

vy + re 78. -30 19000 5203 22 15

NO. 5 HYDROGEN CYANIDE—BRACKETT AND LIDDEL 7

INVESTIGATION OF VAPOR

The absorption spectrum has been obtained of saturated vapor at
22.5°C., with a 30 cm. length of cell. The three bands observed in
the gas absorption, interpreted as 2v,, 3v1, and 2v;+ 2 all show clearly
a doublet structure. In the stronger bands, 2v; and 3, shown in
figure 5, separations of maxima are obtained of 47+2 cm. and
50+2 cm.“, which yield moments of inertia 21+2x10-* g. cm.’
and 18+2x107*° g. cm.” respectively. This is consistent, within the
order of the accuracy of the work, with the more accurate value ob-
tained by Badger. The combination band is too weak to obtain
separation values of significance. On the basis of this conclusive
identification of fundamentals, the clearly doublet character of the
gas absorption implying the absence of a Q branch, and the approxi-

Oil —~—

|

|
|
1034 tre 1.1340

ype

1.5280 BO ST

Fic. 5.—Energy curves showing absorption spectrum of 30 cm. Saturated
HCN vapor. Slit width approx. 9A.

mate values of moments of inertia, we had independently come to the
same conclusions regarding the arrangement of atoms, the approxi-
mate separations, and the probable modes of vibration before the
publication of the work of Badger and Binder.

The position of the 2y, and 37; bands, however, is not consistent
with the formula

Vn = 3333-70 — 43.70"
Assuming the formula
vV=Nwo — N-wo¥
our values indicate a variation in +. This is evident from table 2,
where the values of Av. or 2w)x have successive values 59, 107,

and 133, indicating values of wo, varying from 30 to 67, as against
Badger’s value of 43.7 based only upon v; and 41,. Third differences

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

suggest that a constant value of )1 may be approached equal to or
slightly greater than 67. On the basis of this value we may compute
a heat of dissociation corresponding to an absolute electron voltage
of 5.5 volts. This is in much better agreement with the value com-

TABLE 2.—HCN Bands in Vapor

Notation Abs. A v Ava I v(aver.) Av Av, yy va
% Mh cm.-—! cm.-1 Tot? cms! cmt cm: cm.-1 cm,.-t
(41) 12636 12500 + 30 136
2901

3n sSot2 18 9645 133 950010 145

3124
80 1.5280 6544 +2
Qn A 2 OST 107. 6410+10 III
75 1.5391 6497 +2
3231
(11) 3.04 3290 59
3200

bo

1.1610 8613 £6
2. + 44 10 8501 8487 =10 104
1.1670 8569 +6

bo

2070
(211) 6521
(72) 4.7 2090

puted from chemical data, 4.2 volts, than would be obtained from
Badger’s constant value of 43.7.

The displacements to lower frequencies in passing from vapor to
liquid show a marked increase for the higher harmonics, with the
exception of the 4v, value, for which it must be remembered that the
liquid value is only approximate.

SMITHSONIAN MISCELLANEOUS COLLECTIONS
VOLUME 85, NUMBER 6

MORPHOLOGY OF THE INSECT ABDOMEN

PART J. GENERAL STRUCTURE OF THE ABDOMEN
AND ITS APPENDAGES

BY
R. E. SNODGRASS

Bureau of Entomology,
U. S. Department of Agriculture

qoneeeeseg

(PuBLICATION 3124)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
NOVEMBER 6, 1931

Te Lord Baltimore Press

BALTIMORE, MD., U. S. A.

MORPHOLOGY OF THE INSECT ABDOMEN

PART I. GENERAL STRUCTURE OF THE ABDOMEN AND ITS

APPENDAGES

By R. E. SNODGRASS
BuREAU OF ENTOMOLOGY
U. S. DEPARTMENT OF AGRICULTURE

CONTENTS

PAGE
MTVETO@ CHUA CELO tlle sep eee epee tanta eee ie: < cas Meee dene renee tere ter eas, dakeke Susieus, sue ereie I

lee he -abdominallsclenOtizatiome. sriecee iciec se sic leper cosiesertsreiss arenaicn Slouaueveus oasis 6
Mipm ne mab comaitialieSeett ett aeeewtere irate cette <- c's lereter state ss « ies sielels otals cnusials 14
Abi eaviscetal USCESTMENtSs aa suse ete e a .o.8 eo oecaie eieiale aielwlaie @ sctiefere a cls.cesvaiave 16
iiveecenital mse otnenihs amentyem eres ated eerrtecrolcwet tore: o ersuebsrSoer ale cceloee 17
Mhemposteenitall Seo ments aes «= nels eto yetie cic lee ap vsiensie ecte.0 ale 19

ieee ire aba ominal emusctlatuce ers acai sessilis ts erete winters ear ceieis 28
General plan of the abdominal musculature....................... 31

The abdominal musculature of adult Pterygota.................4. 42

The abdominal musculature of endopterygote larvae............... 48

The abdominal musculature of Apterygota................ eee eee 56
INereliies abdominals appemdac ese m.aebimer es vrais sierra ene seit se naire 62
Bodyeappendacesmon a@hilopodasi sts arc ecta lee cus srnicerere sete ise cosine 5
Abdominal appendaresvor | Grustacediws... sauce ae eases cies ee tere 68

The abdominal appendages of Protura................eeeeeeecees 70
General structure of the abdominal appendages of insects........... 71

The abdominal appendages of Collembola................. 2.000. 2

The abdominal appendages of Thysanura.................--00000 74

The abdominal gills of ephemerid larvae..................0000005 77
Lateral abdominal appendages of sialid and coleopterous larvae.... 70

The abdominal legs of lepidopterous larvae..................... 83

“WER ONO DOC Sema ees se eeeke ness rer eis od So ais SoSNERAG ust onal ues esos Gadbenevditeos 88

PR ERCERCIE SCUITOWOUSI ie eeetrc srerhe enece ists ee ainieis acein vie ® eesteie te ciceterete is 2

The terminal appendages of endopterygote larvae................. 06
Terminal lobes of the paraprocts...0<52- leas sesso wise seecwe cess 107
Morphology of the abdominal appendages................ 0020000 108
Abbreviations used on the figures.............c.ceeccceeccccceuees 122
Tere RGU Semon eee acon lant eee Seti Nag ar nals Govan ye dart ais kek 641d act seens 123

INTRODUCTION

The incision of the insect into head, thorax, and abdomen is in
general more evident in the cervical region than at the thoracico-
abdominal line ; but anatomically the insect is more profoundly divided
between the thorax and the abdomen than it is between the head and

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, NO. 6

2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the thorax. The series of segmental appendages is continuous, and
uniform except for minor adaptational modifications, from the head
to the metathorax, inclusive ; on the abdomen the appendages abruptly
cease, or are greatly altered in form. The musculature of the abdomen,
consequently, is sharply differentiated from that of the thorax. Even
the respiratory mechanism attests that the thoracico-abdominal bound-
ary marks some deep-seated change in the body organization, since
the closing apparatus of the spiracles is almost invariably different
in one way or another before and behind the intersegmental fold
separating the thorax from the abdomen. The definitive head contains
three segments that have been, comparatively speaking, but recently
transferred from the body to the cephalic region; the waist line of
the insect has long been definitely established, and only in a single
order has an abdominal segment been given over to the thorax. The
abdomen is distinctly the visceral region of the body, and its major
active functions in adult insects are those of respiration, copulation,
and oviposition.

Yet, notwithstanding the functional and structural differences that
have come to separate the insect body into cephalothoracic and ab-
dominal regions, we can not avoid the assumption that modern in-
sects are derived from polypod ancestors, and that the abdominal
segments at some time in the past history of the Hexapoda had the
same essential structure as that of the primitive thoracic and gnathal
segments. In studying the morphology of the abdomen and its ap-
pendicular organs, therefore, we must attempt to find in the modern
structure a basic plan of organization the same as that we are led to
believe exists in the cephalic and thoracic regions from a’study of the
segments, the appendages, and the musculature in the preabdominal
parts of the body. Considering, then, the nature of the task, it is not
surprising that students of insect morphology find in the abdomen
problems far more difficult of solution than are those encountered in
the head or thorax, and that there are many fundamental problems
in the abdomen which are still unsolved.

To the systematist in entomology the study of the abdomen, or
particularly of the genital appendages, is becoming of ever increasing
importance, and specialists are coming to feel acutely the need of a
fundamental understanding of the organs that have been found in so
many cases to give the best characters by which species may be dis-
tinguished. Unfortunately, however, no investigator has yet discovered
a means for determining with certainty the homologies of the organs
most useful for diagnostic purposes. In truth, we may say that the

NO. 6 INSECT ABDOMEN—SNODGRASS a
morphology of the insect abdomen is a field in which no angel yet
has trod, and is, therefore, one in which the fool unhindered may
achieve his destiny. However, it is reputed that there is some merit
in knowing oneself to be a fool, and, if it is the wisdom of the wise
to enter only where the foolish have sprung the trap, the fool at least
has a mission to perform. Hence, the writer offers the generalizations
contained in this and a second paper to follow, not with the conviction
that they will prove infallible, but in the hope that others will try
to disprove them—and thereby establish their value.

The principal conclusions derived from the study of the abdomen
and its appendages to be given later in detail may be itemized in
advance as follows:

1. The insect abdomen consists primarily of 12 segments, the first
11 of which are true somites, while the last is the periproct, or telson.

2. The twelfth segment is practically obliterated in all the true
Insecta, except for possible remnants in larval Odonata.

3. The eleventh somite becomes the functional anal segment with

- the suppression of the twelfth segment. Its tergum forms the epiproct.
The venter of the eleventh segment is distinct in some Thysanura,
but it is usually reduced or obliterated, except for two terminal lobes,
which are the paraprocts. The appendages of the eleventh segment
pie tie Cercl,

4. The tenth segment is usually distinct in generalized insects, but
it is often combined with the eleventh to form a composite terminal
segment. The embryonic appendages of the tenth segment are sup-
pressed in postembryonic stages of the more generalized insects ; they
form the postpedes of holometabolous larvae, and perhaps the ap-
pendicular processes of the proctiger, or tenth segment, of some adult
Holometabola.

5. The eighth and ninth somites are the genital segments, which
bear the gonopods. The median gonopore of the female is located
typically behind the eighth sternum, that of the male behind the ninth.
Deviations from these positions are secondary; but the opening of
the paired oviducts of Ephemerida between the seventh and eighth
abdominal sterna probably represents a primitive condition, exhibited
also by the embryos of certain insects (see Heymons, 1892, and
Wheeler, 1893).

6. The abdominal appendages of adult and larval insects are serially
homologous with the legs and mouth parts. Each consists of a basis,
and of one or two movable appendicular processes. The basis appears
to comprise the coxal and subcoxal regions of a typical appendage,
the two parts being either distinct or united. The coxal part often

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

bears an eversible or retractile sac having various functions, repre-
sented by the vesiculae of Thysanura, the gill-bearing lobes of certain
neuropterous larvae, and the plantar lobes of caterpillars and sawfly
larvae. The basis of an abdominal limb usually takes the form of a
lobe or plate of the body wall, and in the pregenital segments the
limb bases are generally united with the sterna in adult insects. The
appendicular process of the basis more commonly present is the
stylus of the lower insects, or its derivatives, including the clasper-
like organs borne by the male gonopods of the higher insects. The
other limb process is the gonapophysis, which occurs only on the
gonopods. Both the styli and the gonapophyses may be movable by
muscles arising within the supporting basal lobes or plates, or on
segmental areas derived from the latter.

7. No positive evidence can be adduced from the known facts of
anatomy or embryology to establish the homology of either the stylus
or the gonapophysis. Many structural interrelations, however, sug-
gest that the stylus is the telopodite of the appendage, and that the
gonapophysis is an endite process of the basis.

8. The genital appendages, or gonopods, have the same essential
structure as the appendages of the pregenital segments. Their dis-
tinguishing feature is the presence of the gonapophyses arising
mesally from the bases. In the female, the gonapophyses of the two
pairs of gonopods form the first and second pairs of valvulae of the
ovipositor ; in the male the gonapophyses of the ninth segment be-
come the parameres. The styli of the gonopods are usually suppressed
in the female of pterygote insects; those of the ninth segment of
the male form the movable claspers, or harpes, of the copulatory
apparatus in the Endopterygota.

g. The bases of the gonopods in adult female insects become plates
supporting the first and second valvulae; those of the second genital
segment may form a third pair of valvulae. In the male the bases of
the single pair of gonopods often form distinct pleural plates of the
ninth segmental wall between the tergum and the sternum, or they
may fuse with either the tergum or the sternum, or with both; again
they may unite with each other to form a plate either coalesced with
the sternum or free and independently movable behind the latter.

10. The parameres of the male are associated with the median
penis in the lower insects, generally uniting with the latter except
in Thysanura; but the penis may be suppressed, and the parameres
then unite with each other+and inclose the terminal part of the
ejaculatory duct to form the more complex copulatory organ known
as the aedeagus. The parameres are to be identified throughout the

no. 6 INSECT ABDOMEN—SNODGRASS 5

orders by the muscles inserted on them, which take their origins in
the supporting basal plates.

11. The styli of the male gonopods become the movable claspers
known as the harpes in the copulatory apparatus of holometabolous
insects. They are to be identified by their muscles which arise in the
supporting basal plates. The harpes may be divided each into a pair
of movable claspers.

12. Numerous accessory appendicular lobes and processes may be
developed on all parts of the male genital segment and on segments
associated with it. These organs are secondary and are not necessarily
homologous in the several orders ; they are often flexible at their bases,
but are to be distinguished from the true harpes and from the para-
meres, with which they are associated, by their lack of muscles.

13. The postpedes, present in holometabolous larvae of several
orders, are the pygopods, or appendages of the tenth somite. The
postpedes are probably transformed into the appendicular processes
of adult males known as socii, found in adult Trichoptera and
Lepidoptera, or into the cercus-like appendages of adult chalasto-
gastrous Hymenoptera.

14. The cerci are the uropods, or the appendages of the eleventh
somite. Typically each is situated in a membranous area laterad otf the
base of the epiproct, and above the paraproct. Muscles that move
the cercus arise on the tenth tergum, or also on the epiproct, but these
muscles are not necessarily primitive muscles of the cercal appendages.
There is no intrinsic evidence that the cerci have any genetic relation
with the paraprocts. It is doubtful if true cerci occur in any holo-
metabolous insects, except possibly in females of Mecoptera.

It will be evident from statements given above, and more flagrantly
apparent in discussion to follow, that the writer still gives much
importance to the value of muscles as determinants of skeletal homol-
ogies—and this in the face of the edict against such practices recently
put forth by H. J. Hansen (1930). However, there surely can be
no question that in studying the insect skeleton we are dealing with
the passive elements of mechanisms, in which the active parts are the
muscles. The principal sclerotic areas of the body segments, and of
the limb segments, are always directly or indirectly associated with
muscle attachments or with mechanical strains resulting from muscle
actions, and there is every reason for believing that scierites have
been correlated with muscles in their evolution, if not necessarily in
their origin. It is, of course, true that, just as some sclerites are
secondary productions, so undoubtedly are some muscles. We must
admit that all kinds of deviations from a rule are possible ; but a few

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

exceptions do not discredit evidence supported by a long series of
uniform interrelations between muscles and skeletal parts. However,
a mere discussion of the matter is useless, and in a final analysis the
identities of muscles must be established by a study of the muscle
innervation. But, in the meantime, practical results may serve as a
basis of judgment. The results of the studies to be given in the fol-
lowing pages will appear principally in Part II of this paper, which
will attempt to analyze the organs of oviposition, and especially the
male organs of copulation, according to the light thrown on the homol-
ogies of their parts by an examination of their musculature. The
muscles furnish a means as yet but little used for identifying corre-
sponding structures in the male genital apparatus of the various orders,
and it will be found that they at least give something definite as a
working basis in a comparative study of the genitalia.

For most of the identified material on which the present paper is
based the writer is indebted to specialists in the United States Na-
tional Museum, including the entomologists of the Bureau of En-
tomology, Department of Agriculture, and Dr. Waldo L. Schmitt
and his associates, of the Museum’s Division of Marine Invertebrates.
Specimens of Heterojapyx and Nesomachilis, however, were obtained
through the interest of Dr. R. J. Tillyard of Australia. Furthermore,
much valuable criticism and information has been contributed by
Dr. A. G. Boving, Mr. Carl Heinrich, Dr. H. E. Ewing, and other
Museum entomologists of the Bureau of Entomology.

I. THE ABDOMINAL SCLEROTIZATION

For purposes of morphological description it is necessary to dis-
tinguish regions of the body wall from the sclerites that may partly
or entirely occupy the regional areas. Considering the body as a whoie,
there are two principal surface regions, one including the back and
sides above the limb bases, the other the under surface between the
limb bases. The first is the dorsum (fig. 1 A, D) ; the second is the
venter (V). Then, in a metameric animal, each somite is likewise
divided into a segmental dorsum and a segmental venter. Separating
the dorsum and venter of each segment are the latero-ventral limb
bases (LB, LB). The regions of the limb bases may be termed the
pleural areas of the segments. The free distal part of any limb, mova-
ble in a vertical plane on the basis, is the telopodite (Tlpd).

It is now well understood that the hardened areas, or sclerites, of the
body wall of insects, as well as of other arthropods, are not “ chitin-
ized” or “strongly chitinized” regions, but that they are areas of

0 ee ee

No. 6 INSECT ABDOMEN—SNODGRASS 7

the cuticula in which other substances than chitin predominate. It
has been shown by Campbell (1929), for example, that the exocuticula
of Periplaneta contains only about 22 per cent of chitin, while the
soft endocuticula contains about 60 per cent; and according to
Kunike (1926) the wing covers of a May beetle contain by weight
75 per cent of nonchitinous substances, and those of a grasshopper
as much as 80 per cent. The writer, therefore, follows the suggestions
of Ferris and Chamberlin (1928) in designating the sclerites as areas
of sclerotization rather than of “ chitinization.”

Sclerotization of the body wall usually produces definite piates in
the several segmental regions. According to the scheme of nomen-
clature adopted in this paper, a major segmental plate of the dorsum

Fic. 1—Diagrams illustrating the theoretical primitive structure of the ap-
pendages and their relation to the body wall.

A, cross section of a segment, showing the basis of each appendage (LB)
movable antero-posteriorly on the body segment by an axis (a-b) in a vertical
plane, and the telopodite (7/pd) movable dorso-ventrally at the coxo-trochan-
teral joint (ct).

B, the basis of each appendage subdivided into a subcoxa (Scx) and coxa
(Cx) by a secondary joint with a vertical axis (c-d); the upper part of the
subcoxa forming the “pleuron” of the body segment.

is a tergum (fig. 1B, T), a corresponding plate of the venter is a
sternum (Stn), and a single plate or group of plates in the pleural
region is a pleuron (Pl). Subdivisions or component elements of
these principal segmental plates then become tergites, sternites, and
pleurites, respectively, since the suffix ite grammatically can mean
only “a part of ” some larger unit designated by the stem of the word
to which it is appended.

The plan for distinguishing and naming the segmental regions, the
major sclerites, and the subdivisions of major sclerites given above
is not generally followed; but the writer has not found any nomen-
clatural scheme for these parts that is consistently applied, or that
adequately meets the situation. Some writers define the terms “ terga”
and “‘sterna”’ as the segmental dorsal and ventral regions, and then
designate the principal plates of these regions the “ tergites”’ and

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

“sternites.”’ This usage is without other objections than that it leaves
us no specific names for the subdivisions or component minor sclerites
of major areas of sclerotization. Writers that adopt it seldom follow
it consistently. On the other hand, many entomologists find it con-
venient for descriptive purposes to distinguish the segmental plates of
the abdomen from those of the thorax as “ tergites ”’ and “ sternites.”
The use of these terms in this manner, however, is not only arbitrary,
but it is ungrammatical, since it is clear that the terms ending in
ite may be applied to metameric units of any particular group of
somites only if the names ‘‘ tergum” and “sternum” are defined as
the entire dorsum and entire venter of this limited group of somites—
a usage which no one pretends to adopt. In general, nomenclatural
troubles arise not so much from a scarcity of suitable terms as from
a lack of consistent application of words in common use. The writer,
therefore, employs the terminology recommended above as the one
most adaptable to the needs of anatomical description. And yet, it will
be found that absolute consistency is not possible ; the insects are sure
to present some anatomical conditions that can not be made to fit
with any nomenclatural scheme that can be devised. Consistency
is said to be a jewel, but an excess of jewelry may become a burden.

Little is known concerning the nature of the sclerotic substances
in the cuticula of insects, or of the procedure by which a specific area
of the body wall becomes continuously sclerotized. We may believe,
however, that minor sclerites may have been produced phylogenetic-
ally by the secondary subdivision of major sclerites, though in the
ontogeny of the individual they may proceed from separate centers
of sclerotization. On the other hand, it is unquestionably true that
primarily distinct areas of sclerotization may unite, and give no trace
of their independent origin in the development of the embryo or pupa.
In the abdomen of most adult insects, for example, the pleural
sclerotizations derived from the limb bases are fused with the primi-
tive sterna, and each definitive “ sternal ” plate in such cases is a triple
structure, though it may lose all trace of its composite origin.

The spiracles are important landmarks in the study of the abdom-
inal segments. They never exceed eight pairs in number, and while
one or more of the posterior pairs may be absent, the presence of a
pair of spiracles is often better proof of the site of a primitive seg-
ment than is evidence derived from the sclerites. The primary posi-
tion of the spiracles is a matter on which opinion differs. There can
be no doubt that the spiracles are subject to migration, and that in
certain insects they have undergone an extreme displacement; but
in the more generalized segments of most insects the spiracles lie in

No. 6 INSECT ABDOMEN—SNODGRASS 9

the sides of the segments above the regions of the limb bases, and
therefore in the lateral parts of the dorsum (figs. 1, 2A, Sp).

the tergal sclerotization of a segment does not invade the spiracular
areas, the spiracles commonly lie in membranous lower parts of the
dorsum between the tergum and the limb bases (fig. 2 A), or between
the tergum and the definitive sternum, which has absorbed the limb
bases (B). The tergal plates of the abdomen, however, often extend

Stn

Fic. 2—Diagrams showing the sclerotization and the retractor mechanism of
the abdomen.

A-E, variations of sclerotization above and below the dorso-pleural line (a-a)
separating the dorsum from the region of the limb bases (LB). F, the retractor
mechanism as seen in vertical section, resulting from secondary segmentation
produced by the usual type of sclerotization in adult insects. G, two consecutive
segmental plates and their connecting muscles.

a-a, dorso-pleural fold; Ac, antecosta; acs, antecostal suture; DMcl, dorsal
longitudinal muscles; /sg, primary intersegmental fold; LB, limb basis; Mb,
secondary intersegmental membrane; MWcl, “longitudinal muscle; patg, parater-
gite; Pc, precosta; PJ, pleural plate formed of the limb basis ; cg definitive ster-
num including areas of limb bases; Sp, spiracle; Stn, primary sternal plate;
T, tergum.

so far downward on the sides of the dorsum as to include the spiracles
in their lateral parts (C). In some cases the spiracles occur in inde-
pendent lateral, or paratergal, sclerites of the dorsum (D). Finally,
the definitive ventral sclerotization is sometimes produced upward
on the sides of the abdominal segments, and the spiracles may then
be included in the lateral parts of the sternal plates (FE) ; but in such
cases it is to be suspected that the apparently sternal areas containing

IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the spiracles are really paratergal sclerotizations that have secondarily
united with the sterna.

In the larvae of holometabolous insects there is usually a more
or less distinct groove extending along each side of the abdomen
below the line of the spiracles (fig. 3 A, B, C, a-a), which is continued
upon the thoracic region above the regions of the pleural, or subcoxal,
sclerites, when the latter are present (B, C, Scx;). This groove,
therefore, evidently marks the division between the dorsal and pleural
areas of the abdominal segments, and may be termed the dorso-

Fic. 3—External structure of the abdomen in holometabolous larvae.

A, larva of Carpocapsa pomonella, showing the dorso-pleural fold (a-a) sepa-
rating the dorsum from the subcoxal areas in both the thorax and the abdomen.
B, larva of Silpha obscura, with series of paratergal sclerites (patg) above
dorso-pleural fold, and series of pleural subcoxal sclerites (Scx) below fold.
C, larva of Pteronidea ribesti, showing same structure as in B.

pleural groove. Sclerites or lobes of the body wall lying immediately
above this groove, then, belong to the dorsum, and may be called
paratergites (fig. 3 B, patg), but not “ pleurites”’ or “ epipleurites.”
Latero-ventral plates or lobes lying below the tergo-pleural groove
are properly termed abdominal pleura (B, Scx) if it is clear by their
position or by the presence of appendicular organs upon them that
they are the equivalents of the thoracic subcoxal pleura (Scx;). It
is convenient, however, to extend the term “ pleurites ” to any lateral
sclerites situated ventrad of the dorso-pleural groove, or below the
line of the spiracles, though such sclerites may be evidently secondary
sclerotizations of this region (fig. 24 D, rpl, 2pl, etc.). But if there is

No. 6 INSECT ABDOMEN—SNODGRASS TT

reason to believe that lateral sclerites are secondary subdivisions of
earlier formed pleurosternal plates, they may be given the non-
committal name of parasternites. When true pleural plates or lobes
of the abdomen are subdivided longitudinally, the upper and lower
parts may appropriately be termed epipleurites and hypopleurites,
respectively ; but such a division seldom occurs in the abdominal
pleura, and the term “ epipleurite ” is commonly misapplied by students
of insect larvae to paratergal lobes, or sclerites of the dorsum.
The fact that the dorso-pleural groove forms a conspicuous line
of infolding along the side of the abdomen in many insect larvae
(fig. 3, a-a) is probably the reason for its frequently having been
termed the “ pleural suture.” Hopkins (1909) thus named it in his
study of the larva of Dendroctonus, and he designated the iateral
lobes above the groove “ epipleural ” and those below it “‘ hypopleural.”
The former he believed represented the epimeron of a thoracic
pleuron, and the latter the episternum. No such homology as this,
however, is possible, since the pleural suture of a thoracic segment
is morphologically a vertical groove in the subcoxal sclerotization cf
the leg bases, taking only secondarily a horizontal position in the meta-
thorax of adult beetles. The so-called ‘‘ pleural suture ”’ of the larval
abdomen, moreover, as we have seen (fig. 3 A, B, C, a-a), extends
into the thorax above the subcoxal sclerotizations (Sc), and thus
throughout the body separates the dorsum from the true pleural
region. Lateral lobes or sclerites of the abdomen lying above the
dorso-pleural groove are, therefore, paratergal (fig. 3 B, patg), and
not “epipleural.’”’ Only the so-called “hypopleural”’ areas lying
ventrad of the dorso-pleural groove, that is, between it and the true
sternal region, are properly pleural in the sense that they correspond
with the subcoxal areas of the thorax (B, Scx;) containing the scler-
ites of the thoracic pleura. The ventro-lateral lobes or plates of the
larval abdomen may, then, be termed the abdominal pleura inasmuch
as they appear to represent the subcoxae of the thorax. The abdominal
pleura are never divided vertically in a way to suggest a true homology
with the division of a thoracic pleuron into episternum and epimeron.
The relation of the muscles to the lateral lobes of the abdomen
in the larvae of Coleoptera has been studied by Boving (1914) and
by Craighead (1916). Boving, here following Hopkins (1909), calls
the lateral groove of the abdomen the “ pleural suture,” but in all his
subsequent work he terms it the “ ventro-lateral suture.” Craighead
identifies the lateral areas of the abdomen with the corresponding

’

areas of the thorax in cerambycid larvae, but since he regards the

12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85
jateral groove of the abdomen as the “ sternopleural suture,” he
takes the paratergal areas to be the abdominal pleura.

The abdominal terga.—The dorsal sclerotizations of the abdomen
in general take the form of simple tergal plates characteristic of any
region of the body in which a secondary segmentation has been
established. Each tergum presents anteriorly a submarginal or often
marginal internal ridge, the antecosta (fig. 2F, G, Ac), on which
the principal longitudinal muscles usually have their attachments.
The antecostal suture (F, acs) is generally but faintly marked, and
the precosta (F, G, Pc) varies from a scarcely perceptible marginal
rim to a wide flange extending a considerable distance anterior to the
muscle attachments (G, Pc). Apodemal processes are frequently ex-
tended from the anterior margins of the terga. From the antecosta
of the first abdominal tergum there is commonly developed a pair of
phragmatal lobes, and the precostal part of this tergum, together
with the antecosta and the phragma, may be separated from the rest
of the tergal plate to form a so-called postnotal or postscutellar plate
of the metathorax. Otherwise the abdominal terga usually preserve
their structural unity. The postcostal areas of the abdominal terga are
seldom marked by sutural lines in adult or nymphal insects, and where
such lines do occur they can not be supposed to have any homology
with the sutures of a wing-bearing thoracic tergum, which adapt the
latter to its function in the wing mechanism.

The dorsal regions of the abdominal segments of soft-bodied holo-
metabolous larvae are usually divided transversely by impressed lines
or by strongly pronounced topographical features. The dorsal areas
thus formed are evidently mere adaptations to the contractile move-
ments of the larvae and have no morphological significance. That
the external body features of eruciform and vermiform larvae are
secondary larval characters is evident from the structure of the head,
which shows that such larvae are lateral derivatives from highly
evolved adult forms representing the immediate ancestors of the
order.

The abdominal sterna.—The definitive sternal plates of the abdomen
are in general similar to the tergal plates, each being a continuously
sclerotized area of the ventral integument of its segment, always
including the primary intersegmental area anterior to the somite,
corresponding with the intersternites, or spinisternites, of the thoracic
region. The antecostae may be coincident with the anterior margins
of the sternal plates, or set well back from the margins (fig. 2 F, Ac)
with distinct precostal regions before them. In the Japygidae a short
anterior division of each abdominal sternum is separated by a mem-

NO. 6 INSECT ABDOMEN—SNODGRASS 13

branous fold from the rest of the plate, forming a distinct sternal
apotome (fig. 24 D, Apt). Apodemal processes to give more effective
action to protractor and dilator muscles are commonly developed from
the anterior and lateral margins of the abdominal sterna.
Notwithstanding the apparent unity of structure in the abdominal
sterna, it is probable, as already stated, that the ventral plates of the
pregenital segments in most adult insects are triple structures, each
including in its composition the area of the true sternal sclerotization
of its segment, and the areas of the limb bases of the corresponding

Fic. 4—Ventral plates of abdomen of Nesomachilis maoricus.

I, II, VI, VIII, ventral plates of segments one, two, six, and eight in male.
LB, limb basis, or basal plate of appendage: rvs, muscles of retractile vesicle;
Stn, primitive sternal sclerite; Sty, stylus; ]"s, eversible and retractile vesicle.

segmental appendages. A comparatively generalized condition ‘is to be
seen in larvae of Ephemerida, in which lateral lobes of the abdomen
supporting the gills (fig. 34 A, B, LB), though forming a part of the
ventral wall of each segment, are distinct from the areas of the primary
sterna (Stn), and occupy the primitive position of limb bases on the
sides of the segments between the tergal and sternal sclerotizations.
A similar but less primitive condition is that occurring in some of the
Thysanura, as in the Machilidae (fig. 4), where each of the definitive
sterna in the pregenital region of the abdomen consists of a small,
median, true sternal sclerite (Stn) and of two, large, lateral stylus-
bearing plates (LB, LB) clearly representing the limb bases.

14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The frequent occurrence of styli on abdominal segments of insects
in which the sterna are undivided plates leaves little doubt that the
definitive abdominal sterna of insects in general are composite plates
including the limb bases as integral parts of their areas. Evidence
of the inclusion of the limb bases in the adult abdominal sterna might
be derived also from other sources, as in the Lepidoptera, where
the abdominal appendages of the larva at the time of pupation are
flattened out in the form of discs, and merge into the ventral areas
from which later the adult sterna are produced.

A definitive sternal plate that includes the primitive sternum and
the bases of the adjacent limbs is sometimes called a “ coxosternum,”
but, as will be shown later, there is a question as to whether the
abdominal limb bases represent the coxae or the subcoxae, or include
both of these usual basal elements of the appendages. A composite
sternal plate, therefore, is more appropriately distinguished from a
primitive sternal plate by the term sygosternum proposed by Prell
(1913). For the same reason the name “ coxite,’ often given to the
limb base element of the zygosternum, is objectionable as being more
specific in its meaning than is warranted by the known facts of the
origin of the part in question. Besides this, the suffix ite implies that
a structure so-named is “a part of” a coxa, and this implication 1s
clearly not intended.

II. THE ABDOMINAL SEGMENTS

Entomologists sometimes nominally distinguish the segments of
the insect abdomen frdm those of the thorax as “ urites,’ a term
perhaps recommended by its brevity, but one which, by inference,
reduces the entire abdomen to the status of a “ tail.”’, Consistent with
this usage, the abdominal appendages would all be “ uropods,” but
the custom of carcinologists in applying the latter term only to the
terminal pair of appendages has better anatomical sanction. (Lan-
kester, 1909; Sedgwick, 1909. )

From embryological evidence there appears to be little doubt that
the primitive number of abdominal segments in typical insects is at
least 12 (fig. 5 A). Twelve segments are actually present in adult
Protura (B), each having distinct tergal and sternal plates, but the
tenth and eleventh are said to be added by “ epigenesis,” that is, they
are developed during postembryonic growth. In many of the Aptery-
gota and in the lower Pterygota, 11 segments are present without
question, while in some forms there are possible rudiments of a
‘twelfth segment. The twelfth or primitive terminal segment is the

No. 6 INSECT ABDOMEN—SNODGRASS 15

periproct (fig. 5 A, Prpt), which carries the anus, but does not have
appendages. It is the terminal piece of the body anterior to which
the true somites are formed, segmentation of arthropods being intra-
somatic and not a process of budding. Some investigators claim that
a pair of coelomic sacs is formed in the periproct. The presence of not
more than six segments in the abdomen of Collembola is usually
taken to be the result of reductive specialization, or “ degeneration,”
in these insects, considering that the existence of even six segments
is obscure in some forms by a loss of intersegmental lines. Tillyard
(1930), however, sees in the Collembola a primitive “ protomorphic ”
condition in which only nine postcephalic somites have ever been de-
veloped. He points out that segmentation in the collembolan embryo
produces six abdominal segments and no more, and that we have,

fC aE ea IXApd ae

fe XApd

5 XApd
(Cer)

~Prpt
XI ae
ee ari xis B

Fic. 5.—Examples of the presence of twelve segments in the hexapod abdomen.

A, posterior end of abdomen of young embryo of Gryllotalpa (from Heymons,
1895), with large periproct (Prpt), or twelfth segment, behind the last true
somite (X/) bearing the cerci (Cer). B, end of abdomen of adult proturon,
Acerentulus confinis (from Berlese, 1910), showing twelve distinct segments.

therefore, no evidence of a greater number of somites ever having
been present in this group of arthropods, which he would not ally
closely with the other insects. A reduction in the number of abdominal
segments is the rule in both immature and adult insects generally.
As just stated, evident remnants of the periproct are rare except in
the Protura. While 11 segments are distinct in many of the more
generalized insects, in the higher orders, especially in the Holome-
tabola, not more than 10 segments are usually present, and sometimes
only nine are evident. In the more specialized insects there is a
tendency toward elimination of the first abdominal segment, but
generally reduction takes place at the posterior end of the body.
Since the periproct is commonly lacking, or reduced to a circumanal
membrane, the eleventh somite, which carries the last pair of seg-
mental appendages, becomes the definitive anal segment. The tenth

16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

segment is sometimes more or less united with the eleventh in orthop-
teroid insects; but in the Holometabola the eleventh segment 1s
generally suppressed, and the body ends with the tenth segment,
though possibly in certain holometabolous larvae the terminal segment
contains a remnant of the eleventh somite, while in coleopterous larvae
the abdomen ends in a distinct anal lobe, which appears to represent
the eleventh segment.

The association of the organs of copulation and oviposition with
the eighth and ninth segments of the abdomen is usually accompanied
by adaptive structural modifications in these segments that conspicu-
ously differentiate the latter as the genital segments, and separate the
rest of the abdomen into a pregenital region and a postgenital region.
Since the pregenital region contains most of the internal abdominal
organs, its segments may be termed the visceral segments. The seg-
ments beyond the ninth, which are usually more or less reduced and
united with each other, constitute the postgenital segments. It is not
possible, of course, in all cases to divide the abdomen consistently into
visceral and genital regions, since modifications adaptive to the major
functions of the eighth and ninth segments often affect one or several
of the preceding segments, but yet, for general descriptive purposes,
the term “ genital segments ” will have a specific meaning.

THE VISCERAL SEGMENTS

To describe here in full the visceral region of the abdomen would
be to repeat many well known facts without adding anything of 1m-
portance. The seven segments of this region are usually of simple
structure and differ but little from one another. In adult pterygote
insects they lack appendicular organs, and the definitive sterna prob-
ably always include the areas of the primitive limb bases. The first
segment is more subject to modifications than are any of the others.
In winged insects the antecosta of its tergal plate bears the third pair
of phragmatal lobes, and the precosta is usually much enlarged, form-
ing the so-called postnotal, or postscutellar, plate of the metathorax
(fig. 16, PN;), which, together with the base of the phragma, is
frequently removed from the tergal region of the first abdominal seg-
ment and more closely associated with that of the metathorax. The
rest of the first segment is often reduced, or fused with the second,
and the sternal sclerotization is sometimes obliterated. The first pair
of spiracles, however, are nearly always retained, and the spiracles
will generally furnish a key to the basal segmentation of the abdo-
men where the segmental limits are obscured. In the aculeate Hy-

No. 6 INSECT ABDOMEN—SNODGRASS 17
menoptera the entire first abdominal segment is so intimately united
with the metathorax that it forms anatomically a part of the thoracic
region of the body. Modification of the posterior visceral segments
will be noted in connection with a study of the genitalia. In females
of higher Diptera the functional visceral region is reduced to five
segments by the conversion of the posterior segments into a tubular
organ of oviposition.

THE GENITAL SEGMENTS

The lateral ducts of the reproductive organs in the majority of
the Hexapoda open into a common, median outlet tube. [Exceptions
to this rule occur in the males of Protura, in both males and females
of Ephemerida, and in males of some Dermaptera, the two lateral
ducts in these cases opening separately to the exterior. The position
of the genital apertures varies within the Hexapoda through an
extreme of six segments. In the Collembola the gonopores of botit
sexes are situated on the fifth abdominal segment, while in the Pro-
tura they occur between the eleventh and twelfth segments. In the
Ephemerida the paired oviducts open between the seventh and eighth
abdominal segments, and the vasa deferentia open on the penes be-
tween the ninth and the tenth segments. In all other Pterygota, except
Dermaptera, and in Thysanura and Dicellura, the single female aper-
ture lies between the eighth and ninth segments, and the male aperture
between the ninth and tenth segments. Apparent exceptions to this
rule occur where some of the terminal segments are fused, where
one or more of the pregenital segments have been obliterated, or
where, as in female Lepidoptera, the gonopore has evidently under-
gone a secondary change in position.

The genital apertures are described by some writers as being situated
on the segments, while others state that they occur between segments.
The gonopores, in truth, are probably located on the posterior parts of
the ventral surfaces of primary segmental areas, but since these parts
of the primitive somites become the intersegmental membranes of the
definitive segments, the gonopores of adult insects are anatomically
intersegmental. They lie behind the primary sterna of the segments
on which they are situated, and only rarely is a secondary sclerotization
formed behind them (male Odonata). The male gonopore is usually
carried outward on an evagination of the body wall forming a simple
penis, or it is situated on a more complex copulatory organ composed
of the penis and the parameres, or of the parameres alone, known
as the aedeagus.

18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Owing to the uniformity in the relation of the genital apertures to
the eighth and ninth abdominal segments, these segments in the
majority of insects become specifically the genital segments. Their
appendages form the principal organs of egg-laying and copulation,
and may therefore be designated gonopods. In some of the simpler
insects the gonopods are lacking and the genital segments have no
distinctive external features; but usually the segments show some
conspicuous structural adaptation to the functions of copulation or
oviposition.

The eighth segment——Modifications of the eighth segment (fig. 6,
VIII) occur principally in the female, since it is on the ventral part

VII VIII oe x XM Cer MiiGgey)
ppt

Fic. 6—Diagram illustrating the concept of the structure of the abdomen
adopted in this paper.

a-a, dorso-pleural line separating tergal region from pleuro-sternal region;
Cer, cercus; Eppt, epiproct; 1Gon, 2Gon, first and second gonapophyses; /a,
lamina subanalis; LB, limb basis; Papt, paraproct (lobe of eleventh sternum) ;
sa, lamina supra-analis; Sp, spiracle; Stn, primary sternum; Sty, stylus; T,
tergum; Tcl, telson (twelfth segment, greatly reduced or obliterated in insects).

of this segment that the first gonapophyses (1Gon), or genital proc-
esses of the eighth gonopods, are developed, and become the ventral
blades, or first valvulae, of the ovipositor in all species provided with
an ovipositor. The female genital opening is normally situated be-
tween the bases of the first gonapophyses in the membrane behind
the primitive eighth sternal plate, but the latter is frequently prolonged
beneath the base of the ovipositor, forming the subgenital plate of the
female. The bases of the gonopods of the eighth segment are never
united with the eighth sternum in female insects having an ovipositor.
In the Thysanura they are large, stylus-bearing plates or lobes which
retain the normal position of limb bases, but in pterygote insects they

No. 6 INSECT ABDOMEN

SNODGRASS 19
appear to form small suspensional sclerites of the first valvulae, known
as the valvifers, which always lack styli. Gonapophyses of the eighth
segment are known to be present in male insects only in some species
of Machilis, but the eighth segment of the male is frequently more or
less modified when associated with the ninth in the copulatory
mechanism.

The ninth segment.—The second genital segment (fig. 6, 7X) usu-
ally has less of the typical form than does the first. It is the somite of
the second gonapophyses (2Gon), or genital processes of the ninth
gonopods, which form the second valvulae of the ovipositor in the
female, and the usual parameres in the male. The sternum of the
ninth segment is generally reduced or rudimentary in the female, but
the bases of the gonopods are commonly retained, either in the form
of lobes, or as blade-like pieces of the ovipositor, the third valvulae.
In the male the ninth segment retains a generalized structure in the
Thysanura (fig. 33, B, C), but in the pterygote insects it is subject
to many modifications and takes on a great variety of forms. The
bases of the gonopods in male Pterygota either remain as distinct
lobes of the segment, or they unite with each other, or with the
sternum, or with both the sternum and the tergum. The styli, if re-
tained, usually take the form of movable clasping organs. Various im-
movable lobes also may be developed from the ninth segment of the
male, and sometimes from the eighth, which serve as accessory organs
in copulation.

The intromittent organ of the male arises in the Thysanura behind
the region of the ninth sternum and between the bases of the gono-
pods (fig. 33 B, C, Pen) ; but in insects having the gonopod bases
united with the sternum, it arises posterior to, or usually above, the
limb base area of the composite sternum. The membranous area
from which the organ arises is, furthermore, generally more or less
inflected between the ninth and tenth segments, forming a genital
chamber above the ninth sternum, and the ninth sternum is often
extended posteriorly as the male subgenital plate, or hypandrium.
The intromittent organ has the form of a simple, tubular penis in the
Thysanura, but in most pterygote insects it is a more complex struc-
ture, called the aedeagus, formed of the penis and parameres, or
of the parameres alone. The external genitalia will be fully described
in Part II of the present paper.

THE POSTGENITAL SEGMENTS

Beyond the second genital segment there are never more than three
segments represented in the hexapod abdomen (fig. 6, X, XJ, X//),

2

20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

and it is only in the Protura that the last of these segments is well
developed (fig. 5 B, X//). In some of the lower insects apparent
traces of the terminal segment are to be found, but in most of the
true Insecta there are only two postgenital segments present. A still
further reduction, however, has taken place in many insects, with the
result that but one segment is to be recognized beyond the ninth.
Where two postgenital segments are present there is little doubt that
they are the tenth and the eleventh; and where the number of post-
genital segments is reduced to one, it is usually to be assumed that the

Fic. 7—Terminal abdominal structures of Thysanura.

A, end of abdomen of male Nesomachilis maoricus. B, caudal filament and
cercus of same removed, showing their origin from small eleventh segment nor-
mally retracted into the tenth. C, dorsal view of same. D, ventral view of
eleventh segment, with bases of caudal filament and cerci. E, dorsal view of
eleventh segment and terminal appendages of Thermobia. F, ventral view of
same.

An, anus; Cer, cercus; cf, caudal filament; Eppt, epiproct; LB, limb basis;
Papt, paraproct; sa, possible rudiment of lamina supra-analis; Stn, primary
sternum; Sty, stylus.

single end-segment is the tenth somite, and that the eleventh has dis-
appeared, though it is probable that the tenth and the eleventh somites
are in some cases combined in the definitive terminal segment.

The tenth segment.—The tenth somite of the abdomen is clearly
retained as the numerical tenth segment in insects in which there are
two distinct postgenital segments. It is a complete annulus in the
Thysanura (fig. 7A, X), quite distinct from the eleventh segment
(B, X/), which is mostly concealed within it. In the Odonata the
tenth segment is a continuously sclerotized ring (fig. 12B, C, X),

NO. 6 INSECT ABDOMEN—SNODGRASS Zi

beyond which are the parts of the eleventh segment, and apparent
remnants of the twelfth (A). Likewise, in larvae of Plecoptera the
tenth segment 1s cylindrical in form (fig. 8 A, X), and the parts of the
eleventh segment (Eppt, Papt) are quite distinct from it. The struc-
ture in an adult plecopteron is essentially the same as that of the
larva, but the tenth segment is smaller. In the Ephemerida, both
larval and adult forms, the tenth somite is a well-developed segment,
which, because of the reduction of the eleventh somite, forms the
terminal segment of the body and appears to carry the long cerci.
Its tergal region is produced posteriorly in a median lobe, and thus
resembles the supra-anal plate of other insects, but the small true
epiproct of the eleventh segment lies beneath the lobe of the tenth
tergum and carries the median caudal filament. The venter of the
ephemerid tenth segment appears to contain the anal opening, but
it is evident that the anus is drawn forward and that the paraprocts are
united with the bases of the cerci.

Among orthopteroid insects the tenth segment is variable ; its ven-
tral region is usually membranous, and in some families its tergal plate
is fused with the eleventh tergum, or epiproct. In the Phasmidae,
however, the tenth segment is large and normally developed (fig.
8 E,G, X). In Diapheromera it has distinct tergal and sternal plates,
the tergum overlapping the edges of the sternum in the female (F),
though the two plates are ankylosed in the male to form a strong
support for the clasper-like cerci (2). The paraprocts are united
ventrally with the tenth sternum (F, Papt) and appear to be lobes of
the latter. The ventral region of the tenth segment 1s membranous
in most other Orthoptera (fig. 8 D), though the dorsum usually
contains a distinct plate (V7). In Acrididae the tenth tergum is a
narrow transverse sclerite fused laterally with the ninth tergum,
but it is separated from the epiproct by a complete suture. In Blat-
tidae (fig. 40 A), Tettigoniidae, and Gryllidae (fig. 8 B), however,
the tergum of the tenth segment (X), is generally fused with the
epiproct (Eppt), and since the paraprocts become continuous with
the membranous ventral wall of the segment, the tenth somite in
these families loses the status of an independent body segment.

It must be noted here that the principal muscles of the cerci arise
on the tergum of the tenth segment. The size of the tenth segment,
therefore, generally varies according to the size of the cercal muscles,
the segment being large in insects having long, mobile cerci (fig. 8 A,
XS), and strongly developed in insects in which the cerci have
a grasping function, as in Japyx (fig. 40 C),and in Diapheromera (fig.
8E, X). When the cercal musculature is weak the tenth segment

22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

is usually reduced, and its tergal plate becomes small, or unites with
the epiproct. In certain cases, however, the tenth segment is developed
quite independently of any relation to the cerci, as in some of the
Homoptera, in which the cerci are rudimentary or absent. In the
cicada (fig. 8 C) the tenth tergum is a strong plate produced down-
ward on the sides into a pair of hooked lobes (/) embracing the distal
end of the aedeagus.

XT Cer Eppt
4 eae

2 ae

~
Cc WUIS=
Fic. 8—Terminal abdominal structures of various insects.

A, larva of plecopteron, ventral view, showing cerci, paraprocts, and epiproct
as parts of eleventh segment. B, Gryllus assimilis, dorsal view, showing union
of epiproct with tenth tergum. C, Magicicada septendecim, with eleventh seg-
ment distinct from tenth. D, Scapteriscus didactylus, female. E, Diapheromera
femorata, male. F, female of same, showing paraprocts fused with tenth ster-
num. G, lateral view of genital and postgenital segments of female Diapheromera
femorata, showing ovipositor, and subgenital plate produced from eighth ster-
num. H, end of abdomen of female Panorpa consuctudinis.

True appendages are absent from the tenth segment in postem-
bryonic stages of all Apterygota and hemimetabolous Pterygota.
Rudiments of appendages, however, are well known to be present
on the tenth segment of many insect embryos (figs. 5 A,g A, XApd).
The idea that these appendages are developed in the female into the
third pair of valvulae of the ovipositor is now generally regarded as
erroneous, since it is clear that both the second and the third pairs
of valvulae are parts of the gonopods of the ninth segment. Berlese
(1906) records an anomaly found in an adult female of Locusta

No. 6 INSECT ABDOMEN—-SNODGRASS 23

viridissima, consisting of two pairs of appendicular processes on the
tenth abdominal segment closely resembling the two pairs of valvulae
on the ninth segment, the outer pair corresponding with the valvulae
formed of the basal plates, the inner pair with those formed of the
gonapophyses. It is scarcely to be supposed, however, that an abnor-
mality of this kind is a “‘ reversion” to an ancestral normality. The
embryonic limb rudiments of the tenth abdominal segment in all the
more generalized insects are normally suppressed before hatching.
In the Holometabola, on the other hand, appendicular structures fre-
quently occur on the tenth segment in postembryonic stages, and there
is little doubt that such organs on the larva, typically represented
by the postpedes of caterpillars, are true limb structures; in adults
they include the socii of Lepidoptera and Trichoptera, and the cercus-
like processes of Tenthredinidae, which appear to be derived from
the larval postpedes. The appendages of the tenth, or pygidial, seg-
ment of the abdomen may be generally designated the pygopods.

The eleventh segment (uro-segment).—The eleventh abdominal
segment represents the last true somite of the body, and is present in
the embryos of many insects as a well-developed ring bearing the
rudiments of the terminal pair of appendages (fig. 5 A, Cer). The
segment is present in adult Protura as a fully formed annulus with
tergal and sternal plates (B, X/), and in some of the lower Insecta
having 11 distinct segments in the abdomen the eleventh segment
is retained likewise as a complete annulus. This condition is well
shown in Nesomachilis (fig. 7B) where the eleventh segment, though
mostly concealed within the tenth (A), consists of continuous tergal
and sternal regions (C, D), and bears laterally the long, filamentous
cerci (Cer). The tergal region is produced into the median caudal
filament (cf), and the sternal bridge supports a pair of broad subanal
lobes, the paraprocts (D, Papt), separated by a median cleft. In
Thermobia the eleventh segment has a distinct tergal plate, or epiproct
(E, F, Eppt), but the sternal bridge is lost, and the sternal region
of the segment is represented only by the paraprocts (F, Papt), upon
which are borne the cerci (Cer). The median ventral region of the
eleventh segment is generally obliterated in pterygote insects that
have a well-developed tenth segment, but in some of the Homoptera,
as in the cicada (fig. 8 C), the venter of the eleventh segment is not
only present but it contains a distinct sclerotic sternal remnant (X/S).

The adult abdomen of most of the lower Pterygota ends with a
supra-anal plate (fig. 6, Eppt) which is in every way suggestive that it
corresponds with the epiproct, or tergum of the eleventh segment, in
the Thysanura. Some entomologists, however, basing their opinion on

24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Heymons’ (1895) assertion that the eleventh segment of the embryo
in Gryllotalpa and other Orthoptera is lost during development, regard
the supra-anal plate of pterygote insects as belonging to the twelfth
segment. On the other hand, Wheeler’s (1893) observations on the
development of Niphidium are fully convincing that the cercus-bear-
ing eleventh segment persists in the Orthoptera, and, though it be-
comes reduced, forms the terminal parts of the adult abdomen asso-
ciated with the cerci. Wheeler’s idea that the rudimentary appendages
of the tenth segment become the inner valvulae of the ovipositor 1n
the female does not conform with the evident facts of comparative
anatomy, but this detail of interpretation does not affect his exposition
of the segmentation.

When we compare the usual circumanal structures of pterygote in-
sects with the parts of the eleventh segment in the Odonata (fig. 12 A,

Aa
An JX
Fic, 9.—Posterior segments and appendage rudiments of embryos of N7phi-

dium. (Outlines from drawings by Wheeler, 1893, showing segmentation and
appendages, but with other details omitted. )

A, male embryo. B, female embryo, each with cercal appendages (Cer) on
eleventh segment. C, female embryo in later stage, showing retention of eleventh
segment structures (NX/).

Eppt, Cer, Papt), in which Heymons himself has shown that the
twelfth segment (Prpt) is present though rudimentary, we can
scarcely question the identity of the parts in all cases. In other words,
the epiproct, the cerci, and the paraprocts, which in larval Odonata
clearly belong to the eleventh segment, must be eleventh segment
structures in all Pterygota, as they are in Thysanura. Heymons’
(1904) claim that the appendages of the eleventh segment in the
Odonata are not true cerci, and that the latter are represented in the
apparent paraprocts finds no support in comparative anatomy, and
has been generally rejected on the ground that it would set the Odonata
apart from all other insects.

The writer would, therefore, agree with Crampton (1918) that the
epiproct is in all insects the tergum of the eleventh segment (fig. 6,

NO. 6 INSECT ABDOMEN—SNODGRASS ‘25

Eppt), but would dissent from Crampton’s opinion that the cerci
and paraprocts pertain to the tenth segment, since in such forms
as Nesomachilis (fig. 7 B) they clearly belong to the eleventh segment,
and embryologists agree that the cerci are appendages of the eleventh
segment (fig. 5 A). That the paraprocts at least belong to the same
segment as the epiproct is evident from their usual anatomical rela-
tions with the latter. Ford (1923), in her review of the musculature
of orthopteroid insects, says, “‘ from the musculature we find that the
supra-anal plate and paraprocts form a symmetrical group,” and fur-
ther she observes that “the transverse muscles between the supra-
anal plate and paraprocts show that all three belong to the same seg-
ment.” The segment represented by the epiproct and paraprocts Ford
believes is the twelfth, but she bases her opinion largely on Heymons’
statement that.the eleventh segment is suppressed in the adult.

Furthermore, the anatomical relations between the paraprocts and
the cerci do not support Crampton’s (1920, 1921) contention that
the paraprocts are the bases of the cercal appendages. The cerci may
be united with the paraprocts (figs. 7 F, 8A), but generally they
arise independently in membranous areas behind the tenth tergum
between the epiproct and paraprocts. The cerci never have muscles
arising in the paraprocts, and the ventral musculature of the para-
procts indicates that these plates are merely lobes of the eleventh
sternum (fig. 6, Papt), as they are actually in Nesomachilis (fig.
7D). With the usual suppression of the eleventh sternal area,
however, the paraprocts may appear to arise from the posterior margin
of the tenth sternum, and they are sometimes continuous with the
latter (fig. 8 F, Papt).

The cerci, as shown by their development, are the entire appendages
of the eleventh segment. Their primitive bases may be represented
by a small, ring-like segment at the root of each organ (Heymons,
1896, Walker, 1922), and, as noted above, they are sometimes united
with the sternal paraprocts, but the muscles of the cerci always have
a tergal origin. As already observed, most of the cercal muscles arise
on the tergum of the tenth segment. These anterior muscles of the
cerci, however, appear to be derived from the intersegmental, longi-
tudinal muscle fibers primitively extending between the tenth and
eleventh terga, which have secondarily become motors of the cerct.
In some Orthoptera, each cercus has a single muscle arising on the
epiproct.

The terminal structure of the generalized insect abdomen has a
certain resemblance to that of a malacostracan crustacean (fig. 10 A).
The twentieth body segment of the crustacean represents the eleventh

26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

segment of the insect abdomen, and its appendages, the uropods
(20Apd), evidently correspond with the cerci. The telson (Tel)
being suppressed in insects, the tergum of the eleventh abdominal
segment (twentieth somite) becomes the supra-anal plate, or epiproct.
The cerci, therefore, may be regarded as the uropods of insects.
The twelfth segment—Among adult Hexapoda a twelfth segment
of the abdomen is developed as a complete annulus with tergal and
sternal plates only in Protura (fig. 5 B, X/7). In the arthropods gen-
erally the terminal segment is the periproct, or end piece of the body
containing the anus, anterior to which the true appendage-bearing
somites are formed. In the malacostracan Crustacea the periproct

FrG. 10.—Terminal abdominal structures of a crustacean, Anaspides tasmaniae.

A, posterior part of abdomen, showing the uropods (20Apd) as terminal
appendages of penultimate segment. B, ventral view of telson, showing lamina
supra-analis (sa) and laminae sub-anales (/a) surrounding anus (An).

forms the telson (fig. 10 A, Tel), typically a broad terminal lobe of the
abdomen, having the anus situated in the basal part of its ventral
surface (B, An) between two lateral valve-like flaps (/a). A distinct,
anus-bearing, terminal lobe of the body is said to be present in the
embryos of some insects (fig. 5 A, Prpt), but in adult insects there
is never more than a vestige of the periproct, or rudiment of a seg-
ment beyond the cercus-bearing eleventh somite (fig. 6, XJ/).

The best example of the retention of a twelfth segment in insects
is furnished by the larvae of some Odonata, in which the anus is con-
tained in a small circular fold (fig. 12 A, Prpt) ordinarily concealed
between the bases of the epiproct (Eppt) and the paraprocts (Papt).
In the walls of this cirecumanal fold, as Heymons (1904) has shown,

NO. 6 INSECT ABDOMEN—SNODGRASS

to
N

/ \ Y = = ¥ Z x \
va ei /
Hypt Eppt Papt (© Hypt D
Fic. 11.—Posterior segments of a noctuid caterpillar.

A, dorsal view. B, lateral. C, ventral. D, posterior.
An, anus; Eppt, epiproct; Hypt, hypoproct; Papt, paraproct; Pp, postpedes
(pygopods ).

1Gor’ J
IXLB Sty

2Gon Papt paptl
Fic. 12.—Terminal abdominal structures of odonate larvae.

A, aeschnid larva, posterior view of end of abdomen, with epiproct and para-
procts spread out, exposing the periproct (Prpt) containing lamina supra-analis
(sa) and laminae sub-anales (/a) surrounding the anus (An). B, lateral view
of same with parts in normal position. C, end of abdomen of larva of Archilestes
grandis, showing gill plates formed of caudal filament (cf) of epiproct (Eppt),
and of lobes (paptl) of paraprocts (Papt).

28 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

there is a small dorsal sclerite, or lamina supra-analis (sa), and two
lateroventral sclerites, or laminae infra-anales (la). These sclerites
are lost in adult Odonata, but a small supra-anal lobe, possibly a rem-
nant of the lamina supra-analis, projects from beneath the epiproct.
A similar lobe occurs in Nesomachilis (fig. 7 D, sa), as well as in some
other Thysanura, and in larvae of Ephemerida. The supra-anal lobe
of these insects might be regarded, therefore, as a dorsal remnant of
the telson (fig. 6, Tel). In most insects, however, no trace of a twelfth
segment is to be found, and the periproct must be supposed to have
been reduced to the membranous area at the end of the eleventh
segment in which the anus is situated.

Ill. THE ABDOMINAL MUSCULATURE

We do not have as yet a sufficient knowledge of the comparative
myology of Arthropoda to furnish a basis for any theory as to the
nature of the primitive body musculature in this group of animals,
in which mobility of the body is a characteristic feature. Widely
different patterns of muscle arrangement are encountered 1n the several
arthropod classes, and even within a single class, while, among the
insects, extraordinary differences occur often between larval and
adult stages of the same species.

In the Insecta the abdominal musculature consists typically of dorsal
and ventral longitudinal fibers, dorsal and ventral transverse fibers,
and lateral dorsoventral fibers; but in none of these muscle groups
do all the fibers often retain their characteristic positions.

The development of the body muscles has been described by Cholod-
kowsky (1891), Heymons (1895), and Nelson (1915). The dorsal
and lateral muscles are formed from the lateral somatic plates of the
mesoderm; the ventral muscles arise from the median ventral parts
of the mesoderm where the somatic and splanchnic layers are united.
The muscle rudiments, or anlagen, according to Heymons, in insects
having open coelomic sacs (Blattidae, Gryllus, Acrididae), are formed
from sac-like evaginations of the mesodermal walls of the seg-
mental cavities, which are at first tubular, but sooner or later become
solid strands of cells. In the higher insects, however, in which the
coelomic sacs are small or but little developed, the muscles either are
formed by the proliferation of cells from the mesoderm segments, or
they arise directly from mesenchyme tissue at points corresponding
with the position of the coelomic sacs of lower insects.

Since the muscles are derived from the walls of the embryonic
coelomic sacs, or from the metameric divisions of the mesoderm, we

ee

No. 6 INSECT ABDOMEN—SNODGRASS 29

may assume that the primitive somatic fibers were all intra-segmental
in arrangement, as they are in the Annelida. With the acquisition
of secondary segmentation in arthropods, however, consequent upon
the development of sclerotic plates in the body wall, the principal
longitudinal fibers became functionally intersegmental (fig. 2 F). The
body of the animal can thus be shortened by a telescoping of its seg-
ments brought about by contraction of the longitudinal muscles, and
it can be compressed by contraction of the lateral dorsoventral muscies.
In most cases the opposite movements result either from the elasticity
of the body wall, or from pressure generated by contraction in one
part of the body transmitted to another through the medium of the
body liquid and the visceral organs; but in many insects a dilator ap-
paratus is developed in which certain muscles in both the longitudinal
and dorsoventral systems become antagonistic to the retractors and
compressors.

The abdominal musculature of adult insects is simpler than the
thoracic musculature because of the absence of leg muscles. There is
no evidence that the definitive lateral muscles of the abdomen have
been derived from the body muscles of the limbs. Muscles of the
movable parts of the abdominal appendages, as will be shown in the
next section, arise generally within areas of the body wall that may
be attributed to the limb bases (figs. 32 B, C, 34 B, 36D), except
the muscles of eversible or retractile sacs which in some cases have
evidently extended to the dorsum. The general segmental plan of
the abdominal musculature is usually repeated with only minor varia-
tions in each of the visceral segments ; in the genital and postgenital
segments it is more or less obscured by special modifications.

A rather simple scheme of abdominal muscle arrangement prevails
throughout all adult pterygote insects; but in the Apterygota and in
larval forms of holometabolous insects the musculature may be highly
complex. Some students regard the complex types of musculature as
representative of a primitive condition from which the simpler types
have been derived by elimination. There are reasons, however, for
taking the opposite view, as will later be shown.

Something is known of the abdominal musculature in most of the
principal orders of insects ; but the Odonata, Orthoptera, Coleoptera,
Hymenoptera, and the larvae of Lepidoptera and Diptera have re-
ceived special attention. Trichoptera and Neuroptera, on the other
hand, have been particularly neglected, and little has been done on
the abdominal musculature of Hemiptera, and of adult Lepidoptera

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

and Diptera. The following literature contains the principal descrip-
tions of the abdominal muscles of the various orders of insects known
to the writer.

PROTURA: Berlese (1910). COLLEMBOLA : Lubbock (1873).
THYSANURA: Grassi (1887), giving the characteristics of the
abdominal musculature of Campodea, Japyx, Machilis, and Lepisma.
ODONATA: Wallengren (1914), Whedon (1919), Ford (1923),
Steiner (1929), with descriptions of muscles of first three segments
in adult by Backhoff (1910) and Schmidt (1915), figures showing
larval musculature by Calvert (1911, 1915), and a tabulation of the
muscles by Tillyard (1917). EPHEMERIDA: Ditrken (1907).
ORTHOPTERA: Gryllus, Voss (1905), DuPorte (1920) ; Divip-
pus, first three abdominal segments, Jeziorski (1918) ; general com-
parative study of abdominal muscles of orthopteroid insects, Ford
(1923). HEMIPTERA: Aphis fabae, Weber (1928). ANO-
PLEURA: Haematopinus suis, Florence (1921). COLEOPTERA:
Melolontha, Straus-Dirckheim (1828); Hydrophilus, first three
abdominal segments, Berlese (1909) ; Dytiscus, Bauer (1910), Speyer
(1922), Korschelt (1924) ; larval musculature of other species, Ber-
lese (1909), Boving (1914), Craighead (1916), Paterson (1930).
LEPIDOPTERA: larval musculature, Lyonet (1762), Lubbock
(1859), Berlese (1909), Forbes (1914, 1916). HYMENOPTERA:
larval muscles of a chalcid, Tiegs (1922), of the honeybee, Nelson
(1924) ; adult musculature of pregenital segments of Vespa, Berlese
(1909), of Apis, Carlet (1890), Betts (1923), Snodgrass (1925) ;
full account of abdominal musculature of Apis, Morison (1927).
DIPTERA: larval musculature partly or briefly described or figured
in Syrphidae, Ktinckel d’Herculais (1875), in Chironomus, Miall
and Hammond (1900), in Anopheles, Imms (1908), in Musca,
Hewitt (1910, 1914), in Khagoletis, Snodgrass (1924) ; full account
of larval muscles of Psychoda, Dirkes (1928), of Culicidae, Samt-
leben (1929).

Less appears to have been done on the body musculature of other
Arthropoda than on that of insects. Descriptions of the abdominal
muscles of Crustacea will be found in the paper on the musculature
of Astacus fluviatilis by Schmidt (1915), in that on Pandalus danae
by Berkeley (1928), and in that on Copilia dana by Riester (1931).
A paper by Becker (1926) describes the dorsal body musculature
of Chilopoda.

ieee

eee

NO. 6 INSECT ABDOMEN—SNODGRASS 51

GENERAL PLAN OF THE ABDOMINAL MUSCULATURE

A review of the literature cited above gives a fairly comprehensive
survey of the abdominal musculature of insects in most of the principal
orders. There are notable blanks, however, since such important
orders as Neuroptera and Trichoptera are omitted entirely, and
adult Lepidoptera and Diptera have been given scant attention. On
the other hand it is gratifying to find that we have, as a basis for
a comparative study of insect myology, very full accounts of the body
musculature of the Odonata, Ephemerida, and Orthoptera. In the
Apterygota, we are indebted to Berlese for an excellent study of the
muscles in Protura, to Lubbock for a description of the collemboian
musculature, and to Grassi for brief descriptions of the characteristic
differences in the musculature of representative genera of Dicellura
and Thysanura, to which is added in this paper an account of the
abdominal muscles of Heterojapyx ; but a more complete study of the
musculature of Machilidae and Lepismatidae, and perhaps of Cam-
podea, is still to be desired. When we look to the papers treating
of holometabolous larvae, we find again satisfactory and in some cases
complete accounts of the body musculature in Coleoptera, Lepidoptera,
Hymenoptera, and Diptera, but note with regret a lack of information
on Neuroptera and Trichoptera.

To present here even a summary of the details known concern-
ing the abdominal muscles of insects would occupy an unwarranted
amount of space. A careful review of the facts to be obtained from
the works above listed, however, shows that we may with confidence
make certain broad generalizations concerning the fundamental plan
of the abdominal musculature of adult pterygote insects. The basic
plan is found to be simple; but, as so often occurs in insect mor-
phology, more difficulties are encountered in finding suitable terms to
express the facts than in discovering the facts themselves.

Voss (1905) classified the abdominal muscles as longitudinal mus-
cles, transverse muscles, and lateral muscles (Flankenmuskeln). This
classification is logical inasmuch as it probably conforms with the
primitive arrangement of the fibers. The muscles of the so-called
longitudinal groups, however, do not always preserve a lengthwise
arrangement; they are often strongly oblique, and some of them
frequently take a transverse position. The lateral muscles are desig-
nated “ dorsoventral’? muscles by many writers, but, though their
attachments are usually dorsal and ventral, some of their fibers com-
monly run in an oblique direction, The lateral muscles have also
been termed “ transverse’ muscles, but, as Samtleben (1929) points

32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

out, only the crosswise dorsal and ventral muscles are literally trans-
verse in position. Again, the body muscles are sometimes classed
as dorsal, ventral, and lateral muscles, the dorsals and ventrals com-
prising longitudinal, oblique, and transverse fibers, and the laterals
including dorsoventral and oblique fibers. This classification is evi-
dently the most nearly consistent one that can be devised, and it has
the added merit of being in conformity with the embryonic develop-
ment of the muscles. Unfortunately, however, in naming the secon-
dary muscle groups or individual muscles according to it, the plan
brings out such terms as “‘ median longitudinal dorsal muscles,” ‘‘ ex-
ternal median longitudinal dorsal muscles,” or “* second internal median

Fic. 13.—Diagrams of the more simple types of segmental musculature.

A, simple condition of musculature in right half of a segment, with dorsal
(d) and ventral (wv) fibers attached intersegmentally, external laterals (le)
intrasegmentally, and internal laterals (/i) on intersegmental folds.

B, upper ends of internal lateral muscles (/i) migrated posteriorly, separating
a paratergal muscle band (pf) from the rest of the dorsal muscles (d).

C, cross section of segment, showing relations of principal groups of muscles.
dl, lateral dorsal muscles; dm, median dorsals; Je, external laterals; li, internal
laterals ; p, paratergal muscle; td, dorsal transverse muscles; ftv, ventral trans-
verse muscles; v, ventrals.

longitudinal dorsal muscle.” Still more unwieldy do such terms be-
come when put into Latin form.

It is evident that strict anatomical and nomenclatural consistency in
dealing with the body musculature leads to impractical results. The
writer, therefore, has adopted a classification and nomenclature that
recognizes the anatomical arrangement of the muscles, but which,
in order to shorten the names, errs somewhat on the side of specificity
in terminology. Five principal groups of muscles are distinguished
and designated as follows: I. porsAL mMuscLes (fig. 13 A, d), the
fibers of which are typically longitudinal and primarily intersegmental
in their attachments. II. veNnrRAL MUSCLES (v), resembling the
dorsals in that their fibers are typically longitudinal and attached
primarily on the intersegmental lines. III. LATERAL mMuscLes (J),

NO. 6 INSECT ABDOMEN—SNODGRASS a3

typically dorso-ventral, their fibers both intersegmental and intraseg-
mental. IV. TRANSVERSE MUSCLEs (C, ft), lying internal to the longi-
tudinals, including dorsal transverse muscles (td), and ventral trans-
verse muscles (tv). V.SPIRACULAR MUSCLES (s), generally not more
than two connected with each spiracle, one an occlusor (os), the other
a dilator (dls).

All the body muscles are bilateral in their origin, and all of them
except the ventral transverse muscles remain so in the definitive state.
The fibers of the ventral transverse muscles, which primarily are
intersegmental, Heymons (1895) says are at first attached mesally on
a fold of the body wall between the nerve cords. Later the fold is sup-
pressed and the fibers from opposite sides become continuous across
the sternal region. The fibers of the longitudinal dorsal and ventral
muscles are always separated into symmetrical lateral groups along
the midline of the dorsum and venter, but the lateral sets of dorsal
transverse muscles come together on the ventral wall of the heart.

Each primary group of muscles may undergo an endless diversifica-
tion resulting both from a multiplication of fibers in the group, and
from a rearrangement of the fibers brought about by changes in the
points of attachment. With respect to the dorsal and ventral muscles,
the most general departure from the simple plan, in which the fibers
all lie in a single plane, consists of a differentiation of the fibers
in each group into internal muscles and external muscles. Thus it 1s
found in nearly all insects that the dorsal and ventral muscles com-
prise each two layers of fibers, namely, imternal dorsals (di) and
external dorsals (de), and internal ventrals (vi) and external ventrals
(ve). A second form of diversification affecting the dorsal and ven-
tral muscles is a more or less distinct grouping of the fibers into
median and lateral muscles. In most insects, therefore, we may distin-
guish four sets of dorsal fibers, and correspondingly four sets of
ventral fibers. The several resulting muscles may then be distinguished
as median and lateral internal dorsals (fig. 14 A, B, dim, dil), median
and lateral external dorsals (dem, del), median and lateral internal
ventrals (vim, vil), and median and lateral external ventrals (vei,
vel).

The lateral muscles are more subject to irregularities than are the
dorsal and ventral muscles, but they likewise are often divided into
internal laterals (fig. 13 B, C, lv) and external laterals (le).

Associated with the dorsoventral lateral muscles there is sometimes
present a longitudinal muscle or group of longitudinal fibers lying
external to the upper ends of the internal laterals (fig. 13, p). This
muscle is called the ‘“ epipleural”” muscle by Ford (1923), but since

34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

it occurs on the region of the dorsum, being situated above the line of
the spiracles, and therefore not on the true pleural region, the writer
would term it the paratergal muscle (p). Since this muscle belongs
to the dorsum it should be classed as a dorsal muscle.

The fibers of the transverse muscles are never differentiated into dis-
tinct layers, but they may be variously grouped in both the dorsal
system (fig. 13 C, td) and the ventral (tv).

The spiracular muscles comprise usually not more than two mus-
cles associated with each spiracle. One is an occlusor of the spiracle,
the other a dilator. The occlusor is seldom lacking ; the dilator is less
constant.

To express more concisely the principal groups of abdominal mus-
cles and their subdivisions, we may tabulate the muscles in the follow-
ing manner :

I. Muscutti porsaLes (d).

1. M. dorsales interni (dz).
a. M. dorsales interni mediales (dim).
b. M. dorsales interni laterales (dil).
2. M. dorsales externi (de).
a. M. dorsales externi mediales (dem).
b. M. dorsales externi laterales (del).
3. M. paratergales (p).
II. MuscuLt VENTRALES (v).
I. M. ventrales interni (v1).
a. M. ventrales interni mediales (vim).
b. M. ventrales interni laterales (vil).
2. M. ventrales externi (ve).
a. M. ventrales externi mediales (vem).
b. M. ventrales externi laterales (vel).
III. MuscuLt LaTERALEs (J).
1. M. laterales interni (Jz).
2. M. laterales externi (Je).
IV. MuSCULI TRANSVERSALES (ft).
1. M. transversales dorsales (td).
2. M. transversales ventrales (tv).
V. MUSCULI SPIRACULORUM (s).
1. M. occlusores spiraculorum (os).
2. M. dilatores spiraculorum (dls).

Each secondary group of muscles is often again subdivided into
several bundles of fibers. These ultimate individual muscles may be
given numerical designations, beginning medially in the case of the
longitudinal muscles and anteriorly with the lateral muscles. Thus

No. 6 INSECT ABDOMEN—SNODGRASS 35

the individual muscles of the median internal dorsals may be
specifically indicated rdim, 2dim, 3dim, etc., the external laterals rle,
2le, 3le, etc., and the muscles of the other groups in like manner (fig.
15 A). If it is desired to show that a muscle belongs to a particular
segment, this may be expressed by adding to its symbol a Roman
numeral designating the number of the segment, thus rdimI1, 3vimlV,
aleV I, etc. In describing the complete musculature of a species, how-
ever, the writer has found it more practical to number the muscles
with Arabic numerals, rather than to attempt to follow any system
of lettering that pretends to identify homologous muscles in con-
secutive segments.

The dorsal muscles—The muscles of the dorsum are primarily
composed of longitudinal fibers of segmental length attached on the
intersegmental folds ; in many larvae the principal dorsal fibers retain
this primitive condition. Wherever the dorsum, however, contains
fully-developed sclerotic terga, a secondary segmentation is estab-
lished, and the folds on which the dorsal muscles are attached become
the antecostae of the definitive tergal plates (fig. 14C, Ac). The
longitudinal, primitively intrasegmental muscles thus become func-
tionally intersegmental, and serve to contract the abdomen in a length-
wise direction by retracting each tergum into the posterior end of
the segment preceding, as far as the intersegmental membrane will
allow. The anterior end of a longitudinal abdominal muscle, there-
fore, may be termed the origin, and the posterior end its isertion.

The differentiation of the dorsal fibers into internal and external
muscles is the rule in both adult and larval stages of pterygote in-
sects. The internal dorsals commonly retain their longitudinal posi-
tions, their segmental lengths, and their attachments on the antecostae ;
but there are many departures from this generalized condition. Fre-
quently the fibers take an oblique position, and sometimes they become
shorter than segmental length by a migration of their origins to the
postcostal area of the tergum, or of their insertions to the precostal
area, The external dorsals seldom retain a segmental length ; typically
they are short muscles lying in the posterior parts of the segments
(fig. 14 C, de), and often they become strongly oblique, sometimes
actually transverse, giving a movement of torsion between the two
segments they connect. Finally, the external dorsals may become
completely reversed in position (D, de), their origins being so far
back on each tergum that they lie posterior to the points of insertion
on the anterior edge of the precostal rim of the following tergum.
In such cases, the external dorsals become antagonistic to the internal

3

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

dorsals (di), and function as abdominal protractors, since their con-
traction lengthens the abdomen by decreasing the overlap of the
segments.

dem

Se

Fic. 14.—Diagrams illustrating more evolved types of musculature, and seg-
mental mechanisms.

A, dorsal muscles differentiated into internal and external median dorsals
(dim, dem), and internal and external lateral dorsals (dil, del) ; lateral muscles
(1) differentiated into tergo-sternal (t-s), tergo-pleural (t-p), and sterno-
pleural (s-p) groups; ventral muscles differentiated into internal and external
median ventrals (vim, vem), and into internal and external lateral ventrals
(vil, vel).

B, illustrating the compressor and dilator mechanism of an abdominal seg-
ment, in which some of the lateral muscles function as compressors (cp), and
others, attached ventrally on edge of tergum, become dilators (dlr),

C, usual arrangement of dorsal muscles as seen in longitudinal section, with
internal dorsals (di) attached intersegmentally on antecostae (4c), and external
dorsals (de) arising on posterior parts of terga, both sets acting as retractors
of the terga.

D, modification by which external dorsals (de), attached in posterior folds
of terga, become antagonistic to internal dorsals (di) and act as protractors of
the terga.

The division of the dorsal muscles into median and lateral groups
of fibers affects both the inner and the outer layers (fig. 14 A, B, dim,
dil, and dem, del), but it is not always apparent, and the lateral dorsals
are sometimes absent. The line of division of the inner dorsals into

No. 6 INSECT ABDOMEN—SNODGRASS By

median and lateral muscles is sometimes marked by the points of
origin of the dorsal transverse muscles on the tergal wall (figs. 13 C,
15 B, td).

The paratergal muscle of the dorsum (figs. 13 B, C, 14 B, p) is not
commonly present in adult insects, or, at least, its fibers are not
generally separated from those of the other lateral dorsal muscles. It
is well developed in the Acrididae (fig. 15 B, 169), where it is repre-
sented in each of the pregenital segments but the first (fig. 16) by
a band of intertergal fibers lying above the line of the spiracles ex-
ternal to the upper ends of the internal lateral muscles (fig. 15 B,
175, 176). According to Ford (1923) it is also present in the
Plecoptera (“‘epipleural muscle’’). The paratergal muscle occurs
more commonly in larval insects. In the abdomen of the larva of
Dytiscus it is represented by a lateral band of fibers (fig. 18, p) cut
off from the other dorsal muscles (di/, dill) by the upper ends of the
internal lateral muscles (JiJJ). An apparently corresponding mus-
cle, consisting of a pair of parallel fibers extending lengthwise on
each side of the body above the line of the spiracles, is characteristic
of the caterpillars (figs. 20, 21, p). In the larva of Tipula there is a
broad band of paratergal fibers on each side of the body attached on
the middle of the laterodorsal areas of successive segments (fig.
22,\p).

The ventral muscles——TVhe ventral abdominal muscles undergo an
evolution parallel in most respects with that of the dorsal muscles.
Their fibers are nearly always differentiated into internal and external
layers, and those of both groups are commonly separated into
median and lateral groups. The fibers of the internal layer are typically
intersegmental wherever complete sternal plates are developed, and
serve as retractors of the ventral arcs of the segments. The external
ventrals are usually short and take their origins on the posterior parts
of the sterna. Frequently they become sternal protractors by a re-
versal of their position, owing to the carrying forward of their points
of insertion on apodemal arms of the anterior margins of the sterna
until their morphologically posterior ends lie anterior to their points
of origin on the preceding sterna. The length of the sternal apodemes
commonly gives to the ventral protractor mechanism of the abdomen
a greater effectiveness than has that of the dorsum.

The lateral muscles—It.is difficult to make satisfactory general-
izations concerning the lateral muscles of the abdomen, because these
muscles are subject to more variations in position and attachments
than are either the dorsals or the ventrals. Most commonly the lateral
muscles are tergo-sternal in their attachments (fig. 13 C, li, le), and

38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

typically they are vertical in position; but they may comprise tergo-
pleural and sterno-pleural fibers (fig. 14 A, t-p, s-p), as well as
tergo-sternal fibers (t-s), and generally some of them are oblique.
A division of the lateral muscles into internal laterals and external
laterals (fig. 13 B, C, li, le) is not always apparent, often because
of the absence of the internal group, but it is of common occurrence.

The internal lateral muscles, when present, are longer than the
external laterals because their upper attachments are at a higher
level on the dorsum than are those of the external muscles (figs.
13. C, li, 15 B, 175, 176). The position of the internal laterals along
the sides of a segment is variable. The muscles are usually situated
in the middle or anterior parts of the segments (fig. 16), but in
some cases they are limited to the extreme anterior regions, and in
certain holometabolous larvae they lie on the intersegmental folds.
The internal lateral muscles, however, do not in all cases constitute
a homogeneous group of muscles ; one or more sets of anterior fibers,
such as those forming the first internal lateral muscle of Dissosteira
(fig. 15 B, 175), lie internal to the lateral tracheal trunk, while the
more posterior fibers, as the second internal lateral of Dissosteira
(176), may lie external to the tracheal trunk. In some insects, on
the other hand, the entire series of internal lateral fibers are internal
to the lateral tracheal trunk (fig. 22).

An example of the limitation of the internal lateral muscles to the
intersegmental regions is well shown in the larva of Rhagoletis po-
monella (fig. 23), a trypetid fly, in which the muscles consist of
slender hands of fibers (7) lying laterally on the intersegmental folds
in both the abdomen and the thorax. Similar intersegmental muscles,
comprising each three groups of fibers, are described by Samtleben
(1929) in the larvae of Culicidae as “‘ musculi dorsoventrales medi-
ales,” the upper attachments of which are between the ends of the
dorsal longitudinal muscles, and the lower attachments between the
ends of the ventral muscles. In the larva of Tipula (fig. 22) the
internal lateral muscles consist of a series of approximately vertical
fibers (li) occupying the anterior half of the lateral wall of each seg-
ment, but the anterior fibers in each segmental group are attached
on, or close to, the intersegmental fold. All of these fibers lie internal
to the ventrolateral tracheal trunk (L7ra) and a broad band of lon-
gitudinal paratergal fibers (p). In the caterpillars a group of several
internal lateral fibers (figs. 20, 21, li) arise from the lateral extremity
of each ventral intersegmental fold and diverge posteriorly to the
dorsum, going internal to the lateral tracheal trunk (Tra) and the
paratergal muscles (p/p).

No. 6 INSECT ABDOMEN—SNODGRASS 39

Similar groups of internal lateral muscles occur in both the abdomen
and the thorax of the larva of Dytiscus. According to Speyer (1922),
a two-branched internal lateral muscle, ‘‘ musculus dorsoventralis
abdominis a,” occurs in the anterior part of each of the first five seg-
ments of the abdomen in the larva of Dytiscus marginalis (fg. 18,
lill). The lower ends of these muscles are inserted on the interseg-
mental folds, but their upper ends are attached in the anterior parts of
the segments following on the tergal plates between median and
lateral groups of fibers of the dorsal longitudinal muscles (di, /).
In the thorax, the upper ends of the corresponding muscles (/i) are
more nearly intersegmental ; their lower attachments are on the inter-
segmental “ furcillae ’’ and on the sternal apophyses. These muscles
of the larval thorax possibly correspond with muscles of the adult
described by Bauer (1910), “ musculi levatores prothoracis and meso-
thoracis,” extending from the sternal apophysis of the prothorax
and mesothorax to the first and second phragmata, respectively. Sim-
ilar muscles are sometimes present in the thorax of other adult in.
sects ; one such occurs in Acrididae attached ventrally on the prosternal
apophysis and dorsally on the intersegmental fold in front of the
mesepisternum (see Snodgrass, 1920, figs. 32, 34, 50).

The fragmentary review of the position of the internal lateral mus-
cles given above suggests that the anterior fibers at least of each
segmental group represent lateral dorsoventral muscles that are
primarily intersegmental (fig. 13 A, li). The fibers have a tendency
in the abdomen to migrate posteriorly, especially on the dorsum (B),
though they may extend backward along the lateral edges of the
sternum also (fig. 22). Their upper ends thus cut off a lateral group
of fibers (fig. 13 B, fp) from the longitudinal dorsals, which become
the paratergal muscles. These primarily intersegmental internal lateral
muscles run mesad to the lateral tracheal trunks in some insects
(Acrididae, lepidopterous larvae, tipulid larvae), and their homologues
presumably should do so in all insects, but this point has not been
determined. Other internal lateral muscles of the abdomen, lying
external to the lateral tracheae, are probably intrasegmental in their
origin.

The external lateral muscles are typically dorsoventral and intra-
segmental (fig. 13 A, B, le). Some of them, however, are frequently
oblique (fig. 15 B, 775, 770), and the latter may include an interseg-
mental muscle (fig. 15 A, 4/1). The dorsoventral fibers are sometimes
attached on the pleural membrane or on “ pleural’’ sclerites, form-
ing thus tergo-pleural and sterno-pleural muscles (fig. 14 A, t-/~, s-p).

40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

While most of the lateral muscles are compressors of the abdomen
(fig. 14. B, cp), since they serve to approximate the sternum to the
tergum in each segment, some of them, at least in insects that make
active respiratory movements, serve as dilators (dir). The lateral
dilators become mechanically antagonistic to the compressors by rea-
son of the fact that their points of origin are on the lower edges
of the terga ventral to their insertions on the overlapped edges of
the sterna. As in the case of the sternal protractor muscles of the
abdomen, the effectiveness of the dilators is commonly increased
by the dorsal extension of their points of insertion on apodemes of the
sterna.

There is no evidence to suggest that any of the lateral abdominal
muscles of adult pterygote insects are derived from the primitive body
muscles of the lost appendages. In larval forms that retain appendage
rudiments on the abdomen, the lateral muscles lie mesad of the limb
bases (figs. 34 A, 36 C, 1), attached above on the tergum and below
on the sternum. The persisting muscles of the abdominal appendages
pertain to the distal movable parts of the organs, and these muscles
take their origins within the limb bases (figs. 34 A, 36D). Exceptions
to this occur in the case of the muscles of retractile vesicles of holo-
metabolous larvae (fig. 36 C, D, rvs), which take their origin on the
dorsum, but these muscles are not retained in the adult.. The branchial
muscles of ephemerid larvae (fig. 15 A, bmcls) are said to persist in
the adult stage, but they do not appear to correspond with any of
the lateral muscles in other pterygote insects.

The transverse muscles——The transverse muscles of the abdomen
are best known as the muscles of the dorsal and ventral diaphragms
(fig. 13 C, td, tv). It seems probable that primitively these muscles
were intersegmental in position, their fibers being attached on the
intersegmental folds, one set being dorsal, the other ventral.

The fibers of the dorsal transverse muscles arise typically in groups
on the anterior edges of the lateral parts of the abdominal terga, and
spread mesally to their insertions along the ventral wall of the heart.
Only in a few insects are they evenly distributed along the entire
length of the tergum, or collected into anterior and posterior groups.
The usual anterior origin of the fibers, therefore, suggests that the
dorsal transverse muscles are primarily intersegmental. In the cater-
pillars (fig. 21 A, td) they practically have this position, except that the
diverging inner ends of the fibers spread into the anterior and posterior
parts of the adjoining segments. Usually the muscles of the dorsal
diaphragm extend from the second to the eighth or ninth abdominal
segment, but in the Blattidae they are said to occur not only in the

NO. 6 INSECT ABDOMEN—SNODGRASS 41

first abdominal segment, but also in the mesothorax and the meta-
thorax (Brocher, 1922). In the larvae of anisopterous Odonata,
according to Whedon (1919), a muscular dorsal diaphragm is present
only in the fourth or the fifth segment of the abdomen.

The ventral transverse muscles in some of the Orthoptera (‘Tet-
tigoniidae and most of the Gryllidae) take the form of widely sepa-
rated compact bundles of fibers crossing the anterior parts of the
segmental sterna. In others, as in the Acrididae, the origins of the
fibers are distributed along the sides of the sterna, and the muscles
form a typical ventral diaphragm occupying most of the length of the
abdomen. Ford (1923) thinks that the compact type, that is, the one
in which the fibers form individual transverse muscles segmentally
arranged, is the primitive type, and that it has been derived from a
diffuse or web-like type. The writer, however, believes that the
relations may be the reverse, especially considering Heymons’
(1895) statement that the transverse muscles of Orthoptera are
formed in the embryo along the intersegmental folds. In the larvae
of anisopterous Odonata, according to Whedon (1919), there is
in the abdomen only a single, large, spindle-shaped, somewhat flat-
tened ventral transverse muscle lying in the extreme anterior part
of the sixth segment, attached laterally on the intersegmental fold.

In the higher insects in which ventral transverse muscles are present
in the adult, as in Hymenoptera, the muscles form a continuous sheet
of tissue over the ventral sinus. Ventral transverse muscles are usu-
ally absent in holometabolous larvae. In the honeybee larva, Nelson
(1924) says, there is present in the newly hatched larva a well-
developed ventral diaphragm consisting of a continuous sheet of trans-
verse fibers, but in older larvae it becomes a vestigial structure formed
of more or less isolated fibers entirely too few in numbers to con-
stitute more than a loose and insignificant meshwork.

The spiracular muscles—The musculature of the abdominal spira-
cles includes one or two muscles associated with each spiracle. The
muscle most generally present is an occlusor. This is a short muscle
usually attached at both ends on the base of the spiracular atrium,
where its contraction compresses the inner end of the atrium and so
closes the entrance to the trachea. In the Acrididae the occlusor
muscle arises dorsally on the tergal wall close behind the spiracle. A
dilator, or opening muscle of the spiracle, antagonistic to the occlusor,
occurs at least in most of the Orthoptera, Lepidoptera, and Hymenop-
tera, but it is absent in Odonata, some Orthoptera, and Coleoptera.
The dilator commonly takes its origin on the tergum or on the lateral
margin of the sternum of the segment in which the spiracle is situated.

42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The musculature of the thoracic spiracles is usually different from
that of the abdominal spiracles, as is the structure of the spiracles
themselves. The spiracles of Apterygota and Ephemerida are said
to have no musculature.

THE ABDOMINAL MUSCULATURE OF ADULT PTERYGOTA

The musculature of the visceral segments of the abdomen in
pterygote insects adheres closely to the generalized plan of structure,
though there are usually slight aberrations in the first segment or
first and second segments. The musculature of the genital and post-
genital segments is often highly specialized or reduced, but it is un-
doubtedly derived from the same muscle pattern as that prevailing
in the less modified segments. The usual departures from the gener-
alized musculature in the visceral region of the abdomen consist
principally of a reduction in the number of muscles, a shortening in
the length of some of them, and a shifting of the points of attach-
ment, bringing about simple changes in the position of certain mus-
cles. A brief examination of the orders in which the abdominal mus-
culature is best known will serve to show the extent and nature of the
modifications that take place in the visceral segments. The more ex-
tensive modifications in the specialized genital and postgenital seg-
ments need not concern us here.

E phemerida.—The most generalized abdominal musculature of the
adult pterygote type occurs in the Ephemerida, and the muscle pattern
is here essentially the same in both adult and larval stages. We may,
therefore, follow Dutrken’s account of the larval musculature of
Ephemerella ignita, which can easily be verified in any ephemerid
species. Most of the abdominal muscles (fig.15 A), except those 1n-
serted on the gill bases (bmcls), lie in a single plane against the body
wall, and are comprised in dorsal (d), lateral (/), and ventral (v)
groups. The first two dorsals (rd, 2d), counting outward from the
median line, and the second and third ventrals (2v, 3v) in most of
the segments are typical intersegmental, longitudinal muscles attached
on the anterior margins of successive segmental plates. The third
dorsal (3d), however, is atypical in that most of its fibers take their
origin on the middle of the tergum and cross the following segment
to be inserted on the anterior margin of the second tergum follow-
ing. This muscle Durken calls a “ compound intersegmental ” mus-
cle. The fourth dorsal (4d) and the first and fourth ventrals (7v,
jv) are short muscles arising on the posterior parts of the segmental
plates before those of their insertions. These muscles thus appear

NO. 6 INSECT ABDOMEN—SNODGRASS 43

to correspond with the external dorsal and ventral muscles usually
more definitely differentiated from the internal muscles in the majority
of the Pterygota.

The true lateral body muscles of Ephemerella include only the
intrasegmental, vertical, tergo-sternal muscles (fig. 15 A, zl, 2l, 31),
and the intersegmental, oblique tergo-sternal muscle (4/). The
groups of branchial muscles (bimcls), inserted in the larva on the
bases of the gills, are described and figured by Durken as arising
on the lateral parts of the sterna. The areas on which these muscles

NIT
/

\ /
Ac ms RB Ivs

Fic. 15—Examples of abdominal musculature.

A, musculature of left half of seventh and eighth segments (seen externally)
of abdomen of larva of Ephemerella ignita (from Diurken, 1907) ; bmcls, bran-
chial muscles, shown here as originating on lateral part of sternum, which is
probably the ventral area of limb basis (see fig. 34 B).

B, muscles of right half of third abdominal segment (seen internally) of an
acridid, Dissosteira carolina, 167, internal median dorsals; 168, internal lateral
dorsal; 769, paratergal muscle; 170, external median dorsal; 177, external
lateral dorsal; 172, median ventral; 173, internal lateral ventral; 774, external
lateral ventral; 775, 176, internal laterals; 177, 178, 179, external laterals; td,
insertion point of dorsal transverse (cardiac) muscles.

arise, however, are clearly distinct from the true sterna (fig. 34 A,
B, Stn), and very evidently represent the bases of the abdominal
limbs (LB), of which the gills (Brn) are the distal movable parts.
The branchial muscles, therefore, are not body muscles, but are in-
trinsic muscles of the appendages, and take their origins within the
limb bases. The true lateral body muscles (fig. 34 A, J) are tergo-
sternal in their attachments and lie mesad of the lobes (LB, LB)
supporting the gills. The gill muscles of the ephemerid larva, Diirken
says, are retained without change in the adult. They do not appear

44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to have representatives in the pregenital segments of any other adult
pterygote insect.

Odonata.—The muscles of the first three abdominal segments of
adult Odonata are described by Backhoff (1910) and by Schmidt
(1915) in connection with a study of the male genital organs, and
some of the abdominal muscles of odonate larvae are figured by Cal-
vert (I9gII, 1915). A more complete description of the larval muscu-
lature as a part of the respiratory mechanism is given by Wallengren
(1914), and of that of the adult by Steiner (1929), while a full
account of both the adult and larval muscles in Zygoptera and Anisop-
tera will be found in the paper by Whedon (1919) on the morphology
of the odonate abdomen, a few errors in which are corrected by
Steiner (1929).

In the abdominal musculature of the Odonata there is nothing
to suggest a type of structure more primitive than that of other
Pterygota ; the fundamental plan of the muscle arrangement is that of
pterygote insects in general, and has little to distinguish it from the
muscle pattern of orthopteroid insects. The generalized plan of
musculature is best retained in the larvae of Zygoptera. The muscles
here comprise internal and external longitudinal dorsals, internal and
external longitudinal ventrals, and dorsoventral and oblique lateral
muscles. The internal dorsal and ventral fibers are of segmental
length ; but the externals in each set are short, taking their origins
on the posterior parts of the segments. In the Anisoptera the larval
muscles are more strongly developed than in the Zygoptera, evidently
as an accommodation to the respiratory and locomotor functions
of the rectum, and the broad internal dorsal and ventral bands of
fibers take on oblique direction. The adult musculature is much re-
duced in the abdomen, and most of the muscles are very short, but
the arrangement of the muscles shows no radical departure from
the fundamental pterygote pattern better preserved in the larva.

Orthoptera—The comparative myology of the abdomen is better
known in the Orthoptera than in any other of the larger orders of
insects owing to the comprehensive review by Ford (1923) of the
abdominal musculature of orthopteroid insects. Then, too, Voss
(1905) in his thorough study of the thorax of Gryllus includes an
account of the muscles of the anterior abdominal segments, and
Du Porte (1920) describes the entire musculature of the abdomen
in the same genus.

The abdominal musculature of the Orthoptera and related orders
shows in all groups a differentiation of the dorsal and ventral muscles

NO. 6 INSECT ABDOMEN—SNODGRASS 45
into internal and external layers of fibers, and in most cases a well
marked separation between median and lateral fibers in each of these
groups. In a general statement on the abdominal musculature of
orthopteroid insects, Ford (1923) says: ‘In the common ancestors
of the orthopteroid insects the tergal musculature probably consisted
of two broad layers, an internal longitudinal and an external oblique,
with the inner layer approximately equaling the length of the tergum,
and the outer layer much shorter. Of the present-day orders the
Blattaria approach closely this hypothetical type.” Of the ventral

PNz I 140 14:9
! Il/
197
ae va |
lay — Vil
—— font
rs = AA
=
= LAA
i a Lares
LS] Sls & | i
Se =
DEEL [7A

7. Yi [aN \ \ \
Pe 145 143 IS 154 ie 187 202
Fic. 16.—Musculature of the right half of the first five segments of the ab-
domen of Dissosteira carolina, together with dorsal muscles (7112) of meta-
thorax. (Compare with fig. 15 B.)

140-197, median internal dorsals; 150-198, lateral internal dorsals; 143-202,
median ventrals; 744, first lateral internal ventral; 145, first lateral external

ventral.

musculature Ford says: “The hypothetical type of sternal muscu-
lature is similar to the tergal, having the two-layered arrangement
of longitudinal ental and oblique ectal muscles. The Blattaria again,
resemble the hypothetical type.”

Since the external muscles of the dorsum and venter are not always
oblique, being often parallel with the internal muscles, and since,
furthermore, the dorsal and ventral muscles are not always differ-
entiated into external and internal layers (fig. 15 A), it would seem to
the writer more probable that all the dorsal and ventral fibers were

46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

primarily longitudinal and of segmental length, and that they had
this arrangement in pre-orthopteroid insects. The internal dorsals and
ventrals are likewise often oblique. Obliquity, therefore, would ap-
pear to be secondary also in the external muscles, in which it may
be so accentuated that the muscles lie in a transverse direction, or are
even reversed in position,

The lateral muscles of the Orthoptera are variable in their positions
and in their attachments. They include typical vertical and oblique
intrasegmental tergo-sternal muscles, oblique intersegmental tergo-
sternal muscles, and in some cases muscles that may be termed “‘ tergo-
pleural” and “ sterno-pleural,” since they are inserted on the lateral
membranes or on sclerites below the line of the spiracles. The so-
called pleural areas on which these last named muscles are attached,
however, probably really belong either to the dorsum or to the venter
of the segment, and, if so, none of the lateral muscles is properly a
* pleural” muscle.

In the Acrididae the internal dorsal muscles are distinctly separated
into median and lateral groups of fibers (figs. 15 B, 16, 167, 168)
by the points of attachment of the dorsal transverse fibers on the
tergum (fig. 15 B, td). The external dorsals assume very oblique
or transverse positions (170, 171). The ventral muscles are well
differentiated into median and lateral groups of internal fibers (172,
173) and into lateral external muscles (174). The external ventrals
(174) are sternal protractors by a complete reversal in the relation
between their points of attachment. The lateral muscles in the third
and succeeding segments (figs. 15 B, 16) comprise two internal dor-
soventral laterals (fig. 15 B, 175, 176), and three external laterals
(177, 178, 179), of which the first (777) is an abdominal dilator by
reason of its sternal attachment being on the upper end of a large
lateral sternal apodeme (/A4p). The upper ends of the internal
laterals (175, 176) are attached on the tergum between the lateral
internal dorsals (168) and a broad paratergal dorsal muscle (1769).
This last muscle is the “epipleural’’ muscle of Ford (1923), who
says a similar muscle also is present in the Plecoptera.

Coleoptera——The abdominal musculature of adult Coleoptera is
known principally from the description of Melolontha vulgaris by
Straus-Durckheim (1828), and of Dytiscus marginalis by Bauer
(1910) and Korschelt (1924).

The adult musculature of the abdomen of Dytiscus is relatively sim-
ple. As described by Bauer (1910) it consists of dorsal longitudinal
muscles, ventral longitudinal muscles, and lateral muscles, to which
list should be added the transverse muscles of the dorsal diaphragm.

No. 6 INSECT ABDOMEN—SNODGRASS 47

Bauer terms the lateral muscles ‘“‘ musculi transversales abdom-
inis,” but, as pointed out by Samtleben (1929), the lateral muscles are
dorsoventral and should not be termed “ transverse.” The largest mus-
cles of the Dytiscus abdomen are the dorsal muscles. These consist of
broad bands of fibers forming a wide sheet of muscles against the ter-
gal region in each of the first six segments. The ventral muscles are
present only in segments III, 1V, and V. They include large median
ventrals and small lateral ventrals. The median ventrals form three
pairs of muscle sheets occupying the median sternal region of the seg-
ments, the fibers of the opposite groups in each pair converging pos-
teriorly. The lateral ventrals (“ musculi ventrales externi’”’ of Bauer)
are very small, each arising on the posterior lateral angle of the
sternum of its segment, and being inserted on the anterior margin of
the sternum following. The lateral muscles (‘‘ musculi transversales ”
of Bauer) comprise a pair of small, oblique tergo-sternal muscles
crossing each other in the form of an X in each side of segments I]
to V inclusive, and a single oblique muscle in segment VI.
Hymenoptera—The honeybee furnishes the principal information
that we have on the abdominal musculature of Hymenoptera. The
muscles of a typical abdominal segment of the honeybee have been
described by Carlet (1890), Betts (1923), and Snodgrass (1925) ;
the complete abdominal musculature is given by Morison (1927).
The muscles characteristic of the part of the abdomen involved in
respiration are well shown in the third and fourth segments (fig. 17).
The dorsal muscles consist of three sets of fibers in each half of the
seginent, two of which are internal and one external. The internals
form a broad median band of fibers (dim) slanting ‘from in front
posteriorly and medially, and a slenderer lateral muscle (dil) ex-
tending from in front posteriorly and laterally. The external dorsal
isa short muscle (del) arising laterally on the posterior margin of the
tergum and extending forward to its insertion on the tip of a lateral
tergal apodeme of the following segment. The two sets of dorsals
are thus antagonistic, the internal fibers being tergal retractors, and
the external fibers tergal protractors. The ventral musculature com-
prises internal and external muscles, which are likewise antagonistic.
The internal fibers form an oblique median internal ventral (wi)
on each side of the sternum, the two converging mesally in the form of
a V, anda slenderer lateral muscle (vil) oblique in the opposite direc-
tion. The external ventrals consist of a single small, fan-shaped lateral
muscle on each side (vel), arising laterally on the posterior part of
the sternum and inserted anteriorly on the lateral anterior apodeme

48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL 85

of the following sternum. The lateral muscles comprise three tergo- —

sternal muscles in each side of the segment. .The first (71) is a dilator
of the abdomen, since it arises ventrally on the lateral part of the ter-
gum and is inserted dorsally on the tip of the lateral apodeme of
the sternum; the second and third laterals (2/, 31) are oblique tergo-
sternal compressors of the abdomen.

i i
If

li

Fic. 17—Musculature of two consecutive abdominal segments of adult Apis
mellifica, showing the muscles on the left side.

Ac, antecosta; del, lateral external dorsal muscle; dil, lateral internal dorsal;
dim, median internal dorsal; zl, 2/, 3/1, first, second, and third laterals; dls,
dilator muscle of spiracle; Sp, spiracle; vel, lateral external ventral muscle;
vil, lateral internal ventral; vim, median internal ventrals.

The first lateral muscles (z/) are dilators of the abdomen, the second and
third laterals (2/, 31) are compressors; the internal dorsal and ventral muscles
(dim, dil, vim, vil) are retractors of the segments, the external dorsals and ven-
trals (del, vel) are protractors.

THE ABDOMINAL MUSCULATURE OF ENDOPTERYGOTE LARVAE

The body musculature of endopterygote, or holometabolous, larvae,
in its*higher forms of development, attains an extreme degree of com-
plexity ; in its simpler forms it differs but little from the body muscu-
lature typical of all adult Pterygota. It appears, therefore, that the
complex types of larval musculature represent specialized conditions
adapting the larvae to their individual ways of living, and are not to
be interpreted as meaning that insects are derived from ancestral
worm-like forms having an intricate body musculature. The changes
in the musculature that occur during the pupal metamorphosis are to
be regarded as alterations necessitated by the restoration of the
normal adult musculature, which involve varying degrees of destruc-
tion or reconstruction in the special, temporary larval musculature.

Coleoptera.—The larval musculature of Trichoptera and Neurop-
tera has not been fully studied, nor do we have any comparative work

a

No. 6 INSECT ABDOMEN—SNODGRASS 49

on the larval muscles of the Coleoptera. The complete account of the
muscles of the Dytiscus larva given by Speyer (1922) and by Kor-
schelt (1924), however, furnishes a basis for an understanding of
the relation between the larval musculature and the musculature of
adult insects. The structural changes which take place in the trans-
formation from the larval to the imaginal musculature have been
described by Breed (1903) in a trogositid, Thymalus marginicoliis.

The abdominal musculature of the larva of Dytiscus marginalis,
as described by Speyer, consists of four primary groups of fiber
bundles, namely, dorsal muscles, ventral muscles, and, on each side,
a set of lateral (dorsoventral) muscles. In the region of the first seven
abdominal segments, the dorsal muscles comprise an internal set of -

Fic. 18.—Body muscles in right half of thorax and first two abdominal seg-
ments of Dytiscus marginalis larva. (Outline from figure by Speyer, 1922, re-
lettered in accord with muscle nomenclature adopted in this paper.)

di, internal dorsal muscles; /’, internal lateral muscles, upper ends of which
cut off paratergal muscles (p) from the other dorsals; wi, internal ventrals.

intersegmental longitudinal fibers of segmental length (fig. 18, dil,
dill), and outer sets of short fibers extending from the posterior
parts of the terga to the following intersegmental folds. The ventral
muscles consist likewise of internal (vi/, vill) and external sets of
intersegmental fibers. In each side of the first five abdominal segments
Speyer distinguishes six lateral (dorsoventral) muscles. Five of
these are external laterals, three of which are tergo-sternal and two
tergo-pleural in their attachments. The other lateral muscle is an
internal lateral and consists of two bundles of fibers (liJJ), which
arise ventrally by a common base on the intersegmental fold. Dorsally
the two branches are inserted on the anterior lateral part of the
tergum between median and lateral sets of the dorsal longitudinal
fibers (di and p). In the thorax the muscles (li) corresponding with

50 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the internal laterals of the abdomen occur between the prothorax
and mesothorax, between the mesothorax and metathorax, and be-
tween the metathorax and first abdominal segment. They are attached
above more nearly on the anterior margins of the terga, and are in-
serted ventrally on the intersegmental “ furcillae ”’ and on the sternal
apophyses. The posterior migration of the upper ends of the internal
laterals in the metathorax and abdomen cuts off a lateral group of
fibers (fp) from the longitudinal dorsals that evidently corresponds
with the paratergal muscle of Dissosteira (fig.-15 B, 160).

It is clear that the abdominal musculature of the Dytiscus larva
differs in no essential respect from that characteristic of adult insects
generally. It presents a more primitive condition in that the internal
lateral muscles retain ventrally their intersegmental attachments,
whereas in most adult insects, when present, their ventral ends have
migrated posteriorly along the edges of the sterna (fig. 15 B, 175,
170, tie. iO): |

The figure given by Berlese (1909) of the muscles in the first three
abdominal segments of the larva of Pentodon, and the studies of
Boving (1914) and of Craighead (1916) on the abdominal muscula-
ture of coleopterous larvae, including species of Cleridae, Trogositidae,
Elateridae, and Scarabaeidae, suggest that the chief deviation’ from the
Dytiscus larval muscle pattern consists only of a greater diversification
in the position of the muscles, and of an increase in the number of
muscles or individual fibers in each group. In any case it is clear that
the larval musculature in the Coleoptera presents at most but a small
increase in complexity beyond the minimum characteristic of adult
pterygote insects. Proceeding from this condition found in the
Coleoptera, therefore, we may expect to find that the more complex
musculature of other holometabolous larvae represents only a more
highly specialized condition.

Hymenoptera.—In the larvae of Hymenoptera the body muscula-
ture also retains a relative simplicity. The pattern of the abdominal
muscles of the honeybee larva (fig. 19), as described by Nelson
(1924), departs but little from the basic plan of the general adult
pterygote musculature, though it is somewhat more complex than the
abdominal musculature of the adult honeybee (fig. 17), and is not at all
like the latter in detail. The dorsal muscles of a typical abdominal
segment of the larva (fig. 19) consist of broad bands of internal
longitudinal fibers (di) of segmental length, and of shorter, oblique
external fibers (de). Some of the external fibers, by a transposition
of their posterior attachments on the intersegmental fold, have come

no. 6 INSECT ABDOMEN—SNODGRASS 51

to overlap internally the internal dorsals. The ventral muscles include
internal oblique (vi) and external oblique muscles (ve), all of seg-
mental length. The lateral muscles comprise dorsoventral and oblique
external laterals (/e), and a strong, oblique, internal sterno-tergal
muscle (Ji) attached on the consecutive intersegmental folds. It may
be questioned whether this last muscle represents the internal laterals
of the Dytiscus larva (fig. 18, li), but the only difference between
the two is that the upper end of the muscle in the bee larva is attached
on the intersegmental fold following that of its ventral attachment,
a change that might have come about by a posterior migration of its
dorsal end.

wit\

WF.
Wf

Fic. 19—Musculature of right half of two consecutive abdominal segments
of honeybee larva. (Figure from Nelson, 1924, but relettered in accord with
muscle nomenclature adopted in this paper.)

Con, ganglionic connectives; de, external dorsal muscles, some of them secon-
darily internal at posterior ends; di, internal dorsals; Gung, segmental ganglion;
Ht, heart; le, external lateral muscles; //, internal lateral muscle; Sp, spiracle;
ve, external ventral muscles; vi, internal ventrals.

Lepidoptera.—the larvae of Lepidoptera have long been noted for
the great number of muscles that lie against the body wall, and for the
extreme complexity in the arrangement of the fibers. Fully 150 mus-
cles, mostly individual fibers, may be counted in a typical abdominal
segment of any caterpillar (figs. 20, 21). The principal muscles of
the innermost layer (figs. 20, 21 A) are definite bands of parallel
longitudinal fibers having segmental lengths and attached on the inter-
segmental folds. In the ventral region there are also strong external
muscles of segmental length having an oblique position. Most of
the external fibers, however, are of various lengths and are disposed
in all directions against the body wall (fig. 21 B). On each side of
the body, between the principal dorsal and ventral groups of muscles

4

52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

there is a pair of slender, longitudinal paratergal muscles (figs. 20,
21 A, B, p) lying just above the line of the spiracles and the lateral
tracheal trunks (7va). Anteriorly in each segment the paratergal
muscles are crossed internally by a group of internal lateral fibers
(li) arising ventrally on the intersegmental fold, and diverging dor-
sally and posteriorly to their attachments on the dorsum. These in-
ternal lateral muscles of the caterpillar lie internal to the lateral tra-
cheal trunks. The dorsal transverse muscles of the caterpillar (figs.
20, 21 A, td) arise in groups immediately dorsad of the paratergal
muscles from the posterior margins of the intersegmental folds.

P Tra 2S et ee VNC vi Tra Zp

Fic. 20.—Ventral musculature of fourth abdominal segment of a caterpillar,
Estigmene acraea.

li, internal lateral muscles; p, paratergal muscle; fd, origins of dorsal trans-
verse (cardiac) muscles; Tyra, lateral tracheal trunk; vi, internal ventral mus-
cles; NC, ventral nerve cord.

The complexity of the body musculature of the caterpillar appears
to demonstrate that the muscle system of insects has no limits imposed
on its possibility of diversification both by multiplication and by re-
arrangement of its fibers, since there is no reason to believe that the
intricate pattern of the caterpillar muscles represents in any way the
primitive plan of insect musculature. In the other organization of the
lepidopterous larva there is little to suggest a primitive condition. The
head and mouth parts present the typical fundamental structure of
these organs that has been developed in adult Pterygota, and on this
basic structure have been built up the many special features of the

NO. 6 INSECT ABDOMEN—SNODGRASS 53

caterpillar head and mouth parts adapted to the needs of the larva.
The alimentary canal of the caterpillar is highly specialized in its
musculature. The ae of the nervous and tracheal systems

eS

CLL

Fic. 21.—Abdominal muscles of a caterpillar, Estigmene acraca, seen by re-
moval of inner muscles shown in figure 20.

A, muscles of right half of fourth segment (//”) after removal of internal
ventrals (vi). B, outermost muscles in right half of third segment, showing
particularly the leg muscles; cross-hatched areas represent bases of hair-bearing
tubercles.

Cx, basal rim of leg; di, internal dorsal muscles; Ht, heart; /i, internal lateral
muscles; p, paratergal muscles; td, dorsal transverse (cardiac) muscles ; Tra,
lateral tracheal trunk; vi, internal ventral muscles.

is not necessarily an indication of a primitive state; it is merely the
retention of a generalized structure in these organs accompanying ¢
high specialization in others. The presence of appendages on the pre-

54 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

genital segments of the abdomen likewise signifies nothing more than
the retention of organs useful in the larval stage. In short, the worm-
like form of the caterpillar and of other holometabolous larvae has
no phylogenetic significance. It is a secondary adaptation, derived
from the normal adult pterygote structure, accompanied by numerous
specializations peculiar to the larva, and later discarded. The com-
plex musculature of the caterpillar is only one of the features in the
larval organization that have been specially evolved from the gener-
alized adult structures of the immediate ancestors of the Lepidoptera
to enable the caterpillar to perform more efficiently the duties that have
devolved upon it from the apportionment of the life processes between
the immature and adult stages of the individual.

Diptera——The musculature of the larvae of Diptera shows a
unique type of specialization in its highest development, but at the
other extreme it has a pattern corresponding entirely with that of the
generalized plan of abdominal musculature in adult Pterygota.

The simpler forms of dipterous larval musculature, known in the
Tipulidae, Psychodidae, Chironomidae, Culicidae, and Tabanidae, con-
sist of dorsal and ventral bands of longitudinal fibers, and of lateral
dorsoventral muscies. A primitive type of musculature occurs in the
Psychodidae, where, as described by Dirkes (1928) for Psychoda
alternata, the dorsal and ventral muscles are mostly longitudinal and
attached on the intersegmental folds, though a few in each set are
shorter than segmental length. In the first abdominal segment there
are five dorsoventral laterals and two oblique laterals on each side.
The first of the dorsoventral muscles is attached on the intersegmental
fold between metathorax and abdomen, the others follow along the
side of the segment. A similar condition exists in the Culicidae, as
described by Samtleben (1929), except that here some of the inner
muscles of the dorsal and ventral series in each segment cross ob-
liquely over the outer muscles, and the inner lateral muscles are con-
fined to the anterior parts of the segments, where they are attached on
the intersegmental folds between the ends of the dorsal and ventral
muscles. In the larva of Tipula (fig. 22) the musculature is compli-
cated by a great increase in the number of fibers in all the principal
groups, and by a diversification in their points of attachment, but
there are few fibers taking an oblique course. In both the dorsal and
ventral groups certain sets of fibers are attached regularly on the
intrasegmental transverse folds of the body wall, and some of the
median ventral fibers form somewhat oblique interlacing bundles. The
internal lateral muscles (Ji) comprise a series of dorsoventral fibers

5

Cn

No. 6 INSECT ABDOMEN—SNODGRASS

lying in the anterior half of each segment internal to the laterai
tracheal trunk (LTra). Outside of these muscles, and external to the
tracheal trunk, is a wide band of longitudinal paratergal fibers (/) of

i

ABLE
SEW ue Bl

mM Ho
FY best
J TG
(ES

=
Fy
|
H
a
HI

Fic. 22.—Musculature of right half of two consecutive abdominal segments
of Tipula abdominalis larva.

di, internal dorsal muscles; /i, internal lateral muscles distributed in anterior
half of segment internal to lateral tracheal trunk (LT7ra) ; p, band of paratergal
fibers; vi, internal ventral muscles.

Fic. 23.—Musculature of the body wall of the larva of a cyclorrhaphous
dipteron, Rhagoletis pomonella.

A, musculature of the thoracic and first abdominal segments, seen from above,
with pharynx and connected parts turned forward. B, musculature of right

half of two consecutive abdominal segments.
_ASp, anterior spiracle; fs, frontal sacs; Ht, heart; /e, external lateral muscles ;
li, internal lateral muscles; Phy, pharynx, turned forward; wi, longitudinal

bands of internal ventral muscles.

segmental length, but attached on the median intrasegmental folds of
the dorsum. The numerous external laterals lie against the body wall
external to the paratergal fibers.

56 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

In the larvae of the higher Diptera the dorsal and ventral muscu-
lature appears to be merged into a double series of oblique muscles
regularly crossing one another to form a network pattern repeated
throughout the thorax and the abdomen (fig. 23). The only muscles
that preserve a longitudinal direction are two bands of ventral fibers
continued from the prothorax to the end of the abdomen (v7).
There can be little doubt that this type of musculature represents a
highly specialized condition, correlated with the great specialization
which the maggot shows in nearly all other parts of its body organ-
ization. The lateral muscles, on the other hand, appear to retain
a primitive condition. The internal laterals consist of slender fibers
lying on the intersegmental folds (Ji) along the sides of the body.
The external laterals (Je) comprise a small group of fibers in the
side of each segment against the body wall external to the network
of oblique muscles.

THE ABDOMINAL MUSCULATURE OF APTERYGOTA

The body musculature of apterygote hexapods is not well known
in all the major apterygote groups; it has been carefully studied in
representatives of Protura, Collembola, and Dicellura, but only casu-
ally examined in Thysanura. Particularly desirable, therefore, would
be a complete account of the body muscles of Machilidae and Le-
pismatidae.

Protura—The abdominal musculature of the Protura is fully
described by Berlese (1910) in his monograph on the “ Myriento-
mata.” In this group of hexapods, Berlese says, “ the musculature
is extraordinarily complex by reason of the great multiplicity of
fibers extending in all directions, very much as in the larvae of meta-
bolic insects.” The muscle pattern of the proturan abdomen as shown
by Berlese, however, is not complex by comparison with that of a
caterpillar or of a muscoid maggot, and the proturan body muscles
clearly fall into the three usual categories of insect muscles, namely,
dorsal muscles, ventral muscles, and lateral muscles, to which are to
be added the body muscles of the appendages.

The dorsal abdominal muscles of Protura are divided into external
dorsals and internal dorsals. The internal dorsals (muscles of the
second stratum of Berlese) consist of broad bands of fibers in the
Acerentomidae attached on the successive tergal antecostae. In the
FEosentomidae they are differentiated into median and lateral groups
of fibers. The external dorsals (muscles of the third stratum of

serlese) include two large oblique muscles on each side of each seg-

NO. O INSECT ABDOMEN—SNODGRASS 57

ment; one (the intersegmental tergal muscle of Lerlese) arising
medially on the anterior part of the tergum and inserted laterally on
the antecosta of the following tergum, the other (the intersegmental
tergopleural muscle of Berlese) arising anteriorly on the tergal ante-
costa and inserted posteriorly on the “pleuron” of the following
segment. The so-called ‘ pleuron,” however, is probably to be re-
garded as a paratergal sclerite.

The ventral muscles of the abdomen include likewise external
ventrals and internal ventrals. The internal ventrals consist of paired
bands of longitudinal fibers extending throughout the length of the
abdomen. Each muscle band is divided into a median group of fibers
attached on the sternal antecosta of each segment, and into longer
lateral muscles in the first five segments attached on alternate sterna.
The external ventrals occur only in the first three abdominal seg-
ments of Acerentomidae. Those of the first segment extend from the
center of the sternum laterally to the bases of the appendages of
this segment, and serve as adductors of the appendages. In the sec-
ond and third segments corresponding pairs of muscles arising on
the anterior median part of each sternum diverge posteriorly to the
antecosta of the following sternum.

The lateral musculature of the abdomen in Protura has a very sim-
ple pattern. The lateral muscles comprise intrasegmental vertical
tergo-sternal muscles, and intersegmental oblique tergo-sternal mus-
cles. The intrasegmental laterals include in each of the first three
abdominal segments of EKosentomidae, and in the first segment only of
Acerentomidae, a pair of tergal muscles inserted ventrally on the base
of the appendage, and in the following segments of Acerentomidae
a single lateral tergo-sternal muscle. The intersegmental laterals con-
sist of two slender muscles arising laterally at each end of the tergal
antecosta in each segment, one of which goes to the anterior margin
of the sternum of the preceding segment, the other to a corresponding
point on the sternum of the following segment. From the first ab-
dominal tergum a muscle extends downward to the posterior edge of
the metathoracic sternum, and another goes forward to the posterior
edge of the mesothoracic sternum.

An analysis of the proturan musculature, as described by Berlese,
thus shows that the Protura suggest nothing different as to the pattern
of the primitive body musculature of the Hexapoda from the idea
to be derived from a study of the muscles of adult Pterygota.
Since the lateral musculature in the Protura does not match with that
of any pterygote insect, it does not appear to be the prototype of
the lateral musculature characteristic of the Pterygota, and therefore,

58 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

probably represents a special development. The Protura have no
transverse muscles. Berlese describes a dorsal septum above the
alimentary canal, but, he says, it is composed entirely of a connective
tissue membrane and contains no muscle fibers. A closed dorsal vessel
is likewise absent.

Collembola.—The account of the body musculature of the Collem-
bola given by Lubbock (1873) is so complete and so convincing in
its detail that no doubt can be entertained of its accuracy, though
apparently no subsequent investigator has verified it, or given any
attention to the musculature, other than that of the appendages, in
this interesting group of insects. Lubbock describes the muscles of
Lomocerus as an example of the musculature of a “ linear ” species,
and those of Smynthurus to illustrate the musculature of a “ globular ”
species. It is clear that the muscle pattern in the abdomen of the
former is more generalized than in that of the latter, but that in both
forms the musculature is modified in adaptation to the specialized
functions of the abdominal appendages.

The abdominal musculature of Tomocerus is highly developed,
consisting of strong bands of longitudinal dorsal and ventral muscles
differentiated into internal and external groups of fibers, and of
vertical and oblique dorsoventral lateral muscles. In the first segment
two strong muscles arising on the tergum are inserted on the eversible
vesicle of the collophore. In the third segment groups of lateral
muscles are attached ventrally on the sternal region in the neighbor-
hood of the tenaculum, but they do not appear necessarily to be
primarily muscles of the pair of appendages presumably combined
in this organ. The muscles of the furcula, or spring supported on the
fifth segment, take their origins in the fourth and third segments,
but they appear to be parts of the system of longitudinal body muscles
rather than specific muscles of the leaping appendage. In Smynthurus
the abdominal musculature is highly modified. The longitudinal mus-
cles appear to be reduced and are mostly absent in typical form. On
the other hand, there is a great development of vertical and oblique
dorsoventral muscles associated with the base of the furcula, taking
their origins in the posterior and middle parts of the abdomen.

In no respect can the collembolan musculature be said to be primi-
tive ; but it is evident that it may be derived from the same generalized
plan of muscle arrangement that underlies the abdominal musculature
of adult pterygote insects.

Dicellura.—lIt is most interesting to aad in Grassi’s description of
the muscles of Campodea that the pattern of the abdominal muscula-
ture of this primitive apterygote insect conforms closely with the

No. 6 INSECT ABDOMEN—SNODGRASS 59

fundamental plan of the abdominal musculature of pterygote insects.
According to Grassi, the musculature of an abdominal segment of
Campodea comprises longitudinal dorsal and ventral muscles, oblique
dorsal and ventral muscles, and dorsoventral lateral muscles. The
longitudinal muscles are clearly the internal dorsals, and internal
ventrals. The oblique dorsal muscles are the external dorsals. The
external ventrals are represented by a pair of muscles convergent from
the posterior margin of the segment to the mid-sternal region below
the ganglion. These muscles Grassi terms musculi subganglionares.
In addition to these there are also small oblique and transverse lateral
ventral muscles. The true lateral muscles include several small tergo-
sternal fibers on the sides of each segment. Finally there are the
muscles of the styli and eversible vesicles.

By comparison with Campodea, or with almost any other insect,
the body musculature in the Japygidae is extremely intricate, being
highly complicated by the presence of numerous muscles that appear
to have no relation to muscles in a simple type of musculature. The
following account of the abdominal musculature of a member of this
group is based on a study of specimens of the Australian Heterojapyx
gallardi, for which the writer is indebted to Dr. R. J. Tillyard.
Females of this huge japygid reach a length of 40 millimeters, and
a dissection of the muscles in well preserved specimens is not a
particularly difficult task.

The entire body musculature of Heterojapyx anterior to the ninth
abdominal segment is highly complex, there being in each of the first
eight segments of the abdomen at least 4o pairs of muscles, the
arrangement of which makes a most intricate pattern against the body
wall (fig. 24). In the mesothorax and metathorax the musculature is
quite as complex, and in many details quite different from that of the
abdomen, and is more diversified by the presence of the leg muscles.
In the ninth abdominal segment the musculature is simplified. In the
tenth it consists of a single pair of fiber bundles, but these constitute
two great lateral muscles, almost completely occupying the segment,
which act as adductors of the cercal forceps.

The 40 muscles in either half of a typical abdominal segment of
Heterojapyx, shown in figure 24 representing segment VI, are com-
prised in the following groups:

I. Dorsat MuscLes.—A median band of inner longitudinal inter-
segmental dorsals (A, ta, 1b, Ic); two medio-lateral oblique inter-
segmental dorsals (A, B, 2, 3) ; two latero-median oblique interseg-
mental dorsals (B, 4, 5); and an outer longitudinal intersegmental
dorsal (B, 6).

60 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

| /
/ 7),
VUS B
Fic. 24 A, B.—Musculature of sixth abdominal segment of Heterojapy.+
gallardi.
A, muscles of ventral region and right half of sixth segment, and anterior

part of seventh segment. B, same view with muscles 7, 8, 9, 17, 10, 20, 21 partly
or entirely removed.

No. 6 INSECT ABDOMEN—SNODGRASS 61

Fic, 24 C, D.—Outermost muscles of right half of sixth abdominal segment
of Heterojapyx gallardi.

C, posterior part of right half of sixth segment and anterior part of seventh
sternum. D, right half of sixth segment, showing segmental sclerites.

Ac, antecosta; Apt, sternal apotome; /t, laterotergal lobe; Wb, intersegmental
membrane; Pc, precosta; rpl, 2pl, 3pl, 4pl, pleural sclerites; Sp, spiracle; Sty,
stylus; V/S, VIS, sixth and seventh abdominal sterna; ’ JT, V/IT, sixth and
seventh abdominal terga; 7-40, muscles of sixth segment (see text, pages 59, 62).

62 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL: 85

IJ. VENTRAL MUSCLES.—A group of internal transverse interseg-
mental sterno-pleural fibers (A, 7) ; an oblique intersegmental sterno-
sternal muscle (A, 8); two imner longitudinal intersegmental ven-
trals (A, 9, 10); five outer oblique intersegmental ventrals, four
of which are median (B, 77, 72, 13, 14), and one lateral (75).

II. LaTerRAL MUSCLES.—An anterior intrasegmental tergo-pleural
muscle (A, B, 16); an oblique intersegmental sterno-tergal muscle
(A, 17) ; an oblique intrasegmental tergo-sternal muscle (A, B, 18) ;
a series of five intrasegmental transverse tergo-sternal muscles (A,
19, 20, 21, 22, 23); two short lateral intra-tergal muscles (D, 24,
25); two small anterior intrasegmental tergo-pleural muscles (D,
26, 27); two median intrasegmental pleuro-sternal muscles (C, D,
28, 29) ; an oblique intersegmental pleuro-sternal muscle (A, B, C,
30) ; a group of small external posterior lateral intrasegmental mus-
cles (C, 32, 32, 33, 34, 35; D, 38) ; and two small posterior interseg-
mental muscles, one sterno-pleural (C, 36), the other pleuro-sternal
(vas

IV. Muscles OF THE STYLUS.—Two small muscles (D, 39, 40)
arising in the posterior lateral lobe of the sternum, inserted on the base
of the stylus (Sty).

This complex and strongly developed musculature of Heterojapy.,
which presumably is characteristic at least of the Japygidae, contains
nothing to suggest that it represents the primitive plan of the body
‘musculature of insects. It indicates, on the other hand, a highly
specialized condition giving to these very small creatures a strength
out of proportion to their size, which might enable them to burrow
into hard soil or to insinuate their bodies into minute irregular spaces.
In the multiplicity of individual muscles and in the diversity of
their attachments, the body musculature of H eterojapyx resembles
that of a caterpillar, but there is not the remotest likeness in detail,
showing that the complexity of the muscle pattern in each case is but
the result of a high degree of specialization adaptive to demands for
dexterity of body movements. Both the caterpillar and Heterojapyx
demonstrate the limitless potentiality of the insect muscular system,
and make it all the more surprising that there are so few departures
from the fundamental plan of muscle arrangement.

IV. THE ABDOMINAL APPENDAGES

There is no more vexing subject in the whole field of insect mor-
phology than that of the homologies of the appendicular organs of
the abdomen. Embryology shows at most that these organs are de-

No. 6 INSECT ABDOMEN—SNODGRASS 63

rivatives of the segmental appendages ; it gives no positive evidence as
to what part of a primitive limb may be preserved in the definitive
rudiment, since the latter, whatever it may be, develops directly from
the embryonic rudiment, instead of following what we should suppose
would be the course of the phylogenetic evolution of the organ.
Comparative anatomy is more likely to foster illusions than to lead
to definite results, for while certain categories of facts may seem
to align themselves satisfactorily in some limited scheme of suggested
homology, the plan invariably breaks down when wider generaliza-
tions are attempted. The writer, therefore, can offer nothing new on
the fundamental morphology of the abdominal appendages of insects
that is likely to be generally accepted. Even so, however, it will
be sufficiently worth while to bring together the principal facts at
present known concerning the anatomy of the various appendicular
structures.

Though the appendages of the insect abdomen are rudimentary in
the sense that they do not in any case represent a fully-developed
limb, they are in all cases specialized by a structural adaptation to
some particular use. The abdominal appendages of most interest to
entomologists are those of the genital segments, and if we can dis-
cover a means of identifying these organs in the various insect orders,
this discovery alone will be of much practical value, and it then be-
comes a less consequential matter if we can not fully decide the exact
morphological nature of the organs themselves.

It is not possible, however, to study with profit any modified or
specialized appendicular organ without having some concept of the
nature of the primitive limb structure from which it has been derived.
Since there are current several different ideas concerning the funda-
mental structure of a primitive arthropod limb, it is therefore neces-
sary for a writer to make clear at the outset of a discussion the par-
ticular theory from which he proceeds. The following sketch will
give briefly the view on this subject here taken, and a more extensive
discussion at the conclusion of this section will examine the possibil-
ities of interpreting the structures of the abdominal appendages of
insects according to the terms of the theory adopted, which is essen-
tially that of Borner (1921), though with differences in special appli-
cations.

A comparative study of arthropod appendages soon shows that the
number of segments in the limbs, the relative size of the segments,
and even the segmental musculature are so variable in different
arthropod groups that none of these features can be used as a guide

64 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

for establishing the homologies of the segments or parts of the limb
in any specific case. There are two joints of the limb, however, that
recur in the same form in such a large number of appendages in
the various arthropod groups as to suggest that they represent two
primary points of flexure in the primitive ambulatory appendages,
and that they may, therefore, be accepted as “‘ constants ” in the limb
structure. These joints in a thoracic leg of an insect are the coxo-
trochanteral joint, and the femoro-tibial joint (the Hiiftgelenk and
the Kniegelenk of Borner, 1921). The first (fig. 25 A, f-g) divides
the appendages into a basal region, or limb basis (LB), and a distal
shaft, or telopodite (Tlpd), which is movable on the basis in a vertical
plane by a horizontal, dicondylic hinge (f-g). The limb basis, in the

b B Pa. \Ptar

Fic, 25.—Diagrams showing the relation of the base of a tee te to the body, and
the theoretical progress of segmentation in the limb.

A, theoretically primitive appendage divided into a basis (LB) and a telopodite
(Tlpd); the first movable antero-posteriorly on the body by a vertical axis
(a-b) between tergum and sternum, the second movable on the basis in a ver-
tical plane by a dicondylic, horizontal hinge (f-g) with levator and depressor
muscles (O, Q) arising in the basis.

B, the fully segmented appendage: the basis divided into coxa (Cx) and sub-
coxa (Scv), the latter becoming the pleuron; the telopodite divided into the
usual segments of an arthropod leg beyond the coxa.

sense here understood, includes the potential coxa and subcoxa, which
in some arthropods are differentiated as distinct parts of the basis (B,
Cx, Scx), the coxa then becoming the functional or movable base of
the appendage, while the subcoxa becomes a part of the lateral and
ventral walls of the supporting body segment. The second funda-
mental joint of the limb forms the “knee” (A, ft), and divides the
telopodite into a proximal trochantero-femoral piece, and a distal
tibio-tarso-praetarsal piece, the two movable on each other in a vertical
plane by an articulation which is either monocondylic, or dicondylic.

If we conceive, thus, that the primitive arthropod limb is divided
primarily into a basis and a telopodite, we should expect the baso-
telopodite joint to be the point of flexure most generally preserved,

No. 6 INSECT ABDOMEN—SNODGRASS 65

and, as above noted, a joint does occur in the proximal part of
practically all fully-developed arthropod appendages that is evidently,
from its structure and musculature, to be identified as the joint be-
tween the primitive basis and the telopodite. This joint is the coxo-
trochanteral joint of an insect’s leg. It is then reasonable to assume
that the same joint 1s retained in reduced appendages, and that, finally,
in an unsegmented limb rudiment it is the telopodite that has been
lost, and that the part which remains is the basis.

The appendages of arthropods are prone to develop appendicular
processes on the limb segments. Such processes may be either endites
or exites, or both forms may occur on the same segment. [Endites are
developed particularly on the basis, serving as masticatory lobes on
the gnathal appendages. In the Crustacea, exites of the basis are often
gill-bearing organs, and an exite of the proximal segment of the
telopodite commonly forms an outer branch of the appendage known
as the exopodite. The study of rudimentary appendages becomes
complicated by the fact that it is often difficult or impossible to
determine whether a persisting part represents the main shaft of the
limb, or an appendicular process of the latter.

BODY APPENDAGES OF CHILOPODA

The centipedes furnish a good example of arthropods that have
retained a long series of body appendages preserving the form and

Ftc. 26.—Somewhat diagrammatic cross section of a body segment of Litho-
bius, showing the relation of the subcoxa (Scr) and coxa (Ca) to the body
and to the telopodite (7/pd).

function of ambulatory limbs. Each appendage of the pregenital re-
gion of the body is implanted in a membranous pleural area of its
supporting body segment between distinct tergal and sternal plates
(figs. 26, 27 A). The movable basal piece of a typical chilopod leg is
a small segment generally termed the coxa, or coxopodite (Cx). The

66 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

coxa supports the six-segmented telopodite (fig. 26, T/pd), the proxi-
mal segment of which, or first trochanter (177), is articulated to the
coxa by a typical coxo-trochanteral hinge (fig. 25 B, f-g). Surround-
ing or partly surrounding the base of the coxa, in most of the pre-
genital segments but the last, is an area of the body wall containing
one or several small sclerites (fig. 27 A, Scx). These sclerites appear
to belong to the subcoxal region of the primitive limb basis, since
upon this region are inserted the tergal muscles of the appendage, and
within it arise muscles of the coxa. The coxa turns antero-posteriorly
upon the subcoxa by an approximately dorso-ventral axis (fig. 25 B,
c-d).

The large terminal pair of legs of a chilopod borne by the last pre-
genital segment (fig. 27 A, T/pd) are supported each upon a single
large plate in the lateral segmental wall (LB). The basal joint of each
of these legs clearly corresponds with the coxo-trochanteral joints of
the preceding appendages, and a comparison of the leg-bearing plate
of this segment (LB) with the coxal and subcoxal sclerites of the
segments immediately anterior to it leaves little doubt that the single
‘“pleural”’ plate of the last segment represents both the coxa and the
subcoxa of the preceding segments (Ca, Scv). In other words, the
large pleural plates supporting the legs of the last pregenital segment
are the limb bases (LB) undivided into coxal and subcoxal parts as in
the other segments. The condition here, of course, may be the result
of a secondary union of the subcoxal sclerotizations with the coxa,
but it gives a convincing demonstration of the potential unity of the
coxal and subcoxal regions of the limb basis, and at least suggests
a primitive condition in which the limb basis occupied the lateral walls
of the body segment between the tergal and sternal plates (A, C, IT,
IStn). The levator and depressor muscles of the telopodite of the
last pair of legs arise on the plate of the limb base and on the
sternum (D, O, Q), and have their insertions on the first trochanter
(rey

The basal structure of the last pair of legs in the chilopoda is
paralleled exactly in that of the legs of more generalized Arachnida
as in the Phalangidae (fig. 46 A), in which the free part of each leg
is supported on a large basal plate (1B) implanted in the lateral wall
of the body. Borner (1904) regards the single basal plate of the
terminal pair of chilopod legs as the united coxa and subcoxa; but in
the Arachnida, he concludes (1921) that subcoxae are absent and that
the plates supporting the telopodites are the coxae alone. It is not
clear why structures so evidently similar should be differently inter-
preted.

No. 6 INSECT ABDOMEN—SNODGRASS 67

In the Diplopoda the free basal segment of the leg, judging from
its structure and the nature of its articulation with the next segment,
would appear to be the coxa, and since the sternal plates of the Diplop-
oda surround the bases of the legs, we may conclude with Borner
(1921) that the definitive sterna include the subcoxae. Silvestri
(1903), however, regards the free basal segment of the diplopod leg
as the subcoxa, and the next segment as the coxa, though the latter

Fic. 27.—Appendages and terminal body structures of Chilopoda.

A, posterior part of body of Scolopocryptops sexspinosa, telopodites removed
except from last segment, showing union of subcoxal sclerites with coxa in large
pleural plate (LB) on last pregenital segment. B, ventral view of genital and
pregenital segments of Lithobius. C, lateral view of same. D, base of pregenital
appendage of Lithobius. E, telopodite of right gonopod of Lithobius, mesal
surface.

An, anus; Cx, coxa; Crpd, coxopodite; GO, gonopore; Gp, gonopod; gS,
sternum of genital segment; GSeg, genital segment; /st, intersternal sclerite;
LB, limb basis; ]Stn, sternum of pregenital segment; /7, tergum of pregenital
segment; O, levator of telopodite; Prpt, periproct; Q, depressor of telopodite;
Scx, subcoxa; Stn, segmental sternal plate; T/pd, telopodite; rT yr, first trochan-
ter; 277, second trochanter (praefemur ).

segment has all the structural features and usual relations of a
first trochanter.

Following the last leg-bearing segment in the Chilopoda comes
the definitive genital segment (fig. 27 A, GSeg), beyond which is the
periproct (Prpt), or anal segment. According to Heymons (1gor) the
genital region of the body contains two small somites in the embryo,
parts of both of which are sometimes retained in the adult stage. The

5

68 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

apparently single genital segment of the adult may be a mere mem-
branous ring (A, GSeg), or it may be a well-developed segment with
dorsal and ventral plates (C, gT, gS). In Lithobius the ventral plate
of the genital segment bears laterally on its posterior margin a pair of
small, three-segmented appendages, the gonopods (B, C, Gp, E), the
basal muscles of which (E, O, Q) arise on the sternal plate of the
segment. The definitive sternum of the genital segment, thegefore, is
clearly a composite plate which includes the true bases of the genital
appendages united with the primitive segmental sternum. The free
genital appendages, then, are not the entire gonopods, but are the
telopodites of the latter, and their muscles (E, O, Q) are the levators
and depressors of the first trochanter (Tr).

In the males of many insects of the higher orders the structure of
the second genital segment and its clasperlike appendages (the harpes)
closely resembles the condition in Lithobius. Though the claspers are
but one-segmented, they are movable by muscles arising in the gonopod
bases, and the latter are generally more or less united with the sternum.
In the insects, however, it is not so clear that the claspers are the true
telopodites of the gonopods, since there is evidence to suggest that
they may be other appendicular processes of the bases of the genital
appendages.

ABDOMINAL APPENDAGES OF CRUSTACEA

All the body segments of the Crustacea anterior to the telson are
usually provided with well-developed appendages. In the lower crus-
tacean groups, the appendages of the entire body series, as in Apus
(fig. 28 A), are fundamentally uniramous in form, though the various
segments may be provided with endite and exite lobes. Each limb
consists of a basis (LB), called the coxopodite, and of a telopodite
(Tlpd). The frequent biramous form of crustacean appendages (C)
is evidently the result of the hyper-development of an exite of the
basal segment of the telopodite (the first trochanter, or basipodite,
Bspd). The shaft of the telopodite beyond the basipodite then be-
comes the endopodite (Endpd). The exite lobes are movable by mus-
cles arising in the limb segments that support them.

The abdominal appendages of the Malacostraca are typically bira-
mous limbs (fig. 28 C) in which the endopodite (Endpd) is usually
reduced to the size of the exopodite (Expd). The basis, or coxopodite
(Capd), and the basipodite (Bspd) may be distinct segments, but in
some forms (B) they are united in a single protopodite (Prtpd).
In certain cases the abdominal appendages become practically unira-

No. 6 INSECT ABDOMEN—SNODGRASS 69

mous by a suppression of the endopodite (B, Endpd), or by its con-
version into a genital process. In such cases the functional or loco-
motory shaft of the appendage is the exopodite (Ewpd). A crustacean
limb of this type of structure furnishes an analogy with the abdominal
limbs of Thysanura on the assumption that the stylus of the latter
(fig. 31 A, Sty) is the exopodite, and that the endopodite has been
entirely suppressed, or preserved only in the gonapophyses of the
genital appendages (B, Gon).

Fic. 28.—Abdominal appendages of Crustacea.

A, Apus longicaudata, appendage from posterior part of body, left, anterior
surface. B, Anaspides tasmaniae, third abdominal appendage of left side, anterior
surface. C, Spirontocaris groenlandicus, abdominal appendage of left side,
anterior view.

Bnd, basendite; Bspd, basipodite; Capd, coxopodite, or limb basis; Endpd,
endopodite; Expd, exopodite; LB, limb basis, or coxopodite; Prtpd, protopodite
(united basis and basipodite) ; 7/pd, telopodite.

The genital claspers of the higher insects, which are clearly homo-
logues of the abdominal styli of the Thysanura, may thus be
likened either to the gonopods of the Chilopoda (fig. 27 C, Gp), if we
assume that they are the main shafts of the telopodites, or to the ab-
dominal appendages of such crustaceans as Anaspides (fig. 28B),
if we assume that they are exites of the appendages. As will later be
shown, however, it is difficult to obtain positive evidence as to the
nature of the insect abdominal styli; whether they are likened to the
main shaft of the telopodite or to an exopodite branch depends largely
on the student’s bias toward a myriapodan or a crustacean ancestry
for the insects.

70 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

THE ABDOMINAL APPENDAGES OF PROTURA

A pair of short, cylindrical appendages is present on each of the
first three abdominal segments of all adult Protura. These appendages
arise from the membranous parts of these segments between the pos-
terior angles of the tergal and sternal plates. They are best developed
in Eosentomidae, where the three pairs are alike in size and structure,
and each organ (fig. 29 B) consists of two segments and a small
terminal vesicle (v) which is eversible and retractile. In Acerento-
midae (A) the appendages of the first pair are like those of the
Eosentomidae, but the second and third pairs are simple, tuberculiform

2 oex

—Cx

Fic. 29.—Abdominal appendages of Protura.

A, Eosentomon germanicum, abdominal leg (from Prell, 1913). B, Aceren-
tomon doderoi, first abdominal leg (from Berlese, 1910).

Cx, coxa; I, promotor muscle of limb base; J, remotor muscle; rv, retractor
of vesicle; Scr, subcoxa; T/pd, rudiment of telopodite; v, terminal vesicle.

protuberances, unsegmented and lacking the terminal vesicle. Each
appendage of the larger type in the two families, as described by
Berlese (1910), is movable by two tergal muscles (B, /, J) inserted
on the basal segment, one anteriorly, the other posteriorly. The sec-
ond segment is provided likewise with two muscles, one arising
anteriorly, the other posteriorly in the proximal segment, the two
crossing each other axially to be inserted on opposite sides of the base
of the distal segment. The terminal vesicle is retracted by a single
large muscle (rv), which takes its origin mesally on the base of the
first segment of the appendage, and is inserted on a central depression
of the ventral face of the vesicle. The extrusion of the vesicle is
evidently brought about by blood pressure from within the body.

No. 6 INSECT ABDOMEN—SNODGRASS 71

There appears to be no reason to doubt that these abdominal
appendages of the Protura are remnants of true post-thoracic limbs.
They have, as Berlese points out, a certain resemblance to the ab-
dominal legs of lepidopterous larvae; but a closer comparison shows
differences in the segmentation and musculature which makes it seem
probable that there is no close genetic relation between the two sets of
organs. Prell (1913), in his study of Eosentomon germanicum, finds
at the base of each abdominal leg two small sclerotizations which
he regards as remnants of the subcoxa (fig. 29 A, Scv). The large
basal segment he believes is the coxa (C2) and the smaller distal seg-
ment the rudimentary telopodite (7/pd). The homology of the
terminal vesicle (v) is doubtful. The organ does not appear to
represent the eversible sacs of Thysanura, since the latter are borne
by the limb bases (fig. 4, Vs) ; it might be, however, as Prell sug-
gests, the praetarsus, since it has a certain resemblance to the vesicular
praetarsus of Thysanoptera. The most likely homologue of the
proturan leg vesicles is to be found in the eversible sac on the
collophore of Collembola (fig. 30 B, v), which probably represents the
united vesicles of a pair of fused appendages.

GENERAL STRUCTURE OF THE ABDOMINAL APPENDAGES OF INSECTS

Most of the appendicular organs found on the abdominal region of
insects fall into two quite distinct categories distinguished by the
insertion points of their muscles. In those of one group the muscles
are inserted on the base of the organ; in those of the other the
muscles traverse the organ and are inserted within its distal extremity.
Appendicular structures of the first class are typically stylus-like in
form, though they take on various other shapes. They include such
organs as the abdominal styli of the Thysanura and the more general-
ized Pterygota, the furcula of Collembola, the gills of ephemerid
larvae, the terminal claws of trichopterous larvae, the lateral abdominal
appendages of larvae of Sialidae, the gonapophyses, the movable
claspers of male pterygote insects, and the cerci. Organs of the sec-
ond class are sac-like or tubular in form, and are usually retractile and

-eversible. They include the collophore of Collembola, the eversible

vesicles of Thysanura, the gill-bearing tubercles of some sialid larvae,
and the plantar lobes of the abdominal legs of larvae of Lepidoptera
and chalastogastrous Hymenoptera.

If we could accept the two categories of abdominal appendicular
structures, distinguished by the muscle insertions, as morphological
groups of organs, the study of the abdominal appendages of insects

72 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

would be much simplified. But, unfortunately, there are other fea-
tures than the muscle insertions to be considered, such as the origins
of the muscles and the position of the organs on the body, that make
it doubtful if either constitutes a homogeneous group. Furthermore,
organs of each type frequently occur together supported on a common
basal structure, and the latter must then be reckoned as an essential
part of the primitive appendage, of which the free, movable parts are
but accessory structures of one kind or another. Any attempt to make
theoretical generalizations on the primitive form or on the homologies
of the abdominal appendages of insects will be premature until we have
more closely examined the structure of the principal types of such
organs as occur in both larval and adult stages of the various insect
orders.

The abdominal appendages of insects fall into three groups corre-
sponding with the subdivision of the abdomen into pregenital, genital,
and postgenital regions. The appendages of the pregenital segments
may be designated the pregenital appendages; those of the genital
segment are the gonopods; those of the first and second postgenital
segments are conveniently termed the pygopods and uropods, respec-
tively. The pygopods are the anal legs, or postpedes, of holometab-
olous larvae, and probably the socii and so-called “cerci” of holo-
metabolous adults. The uropods are the true cerci.

THE ABDOMINAL APPENDAGES OF COLLEMBOLA

The well known appendicular organs on the abdomen of Collembola
include organs of each type of structure as defined above according
to the muscle insertions.

The collophore (fig. 30 A, Col) is a large, thick, tubular pouch of
the body wall projecting from the sternal region of the first abdominal
segment. In most species it ends in a bilobed terminal vesicle (B, v),
which is ordinarily retracted but is capable of being protruded by
blood pressure. The entire collophore is traversed by a pair of large
muscles (rv) arising on the tergal region of the body and inserted on
the lobes of the terminal vesicle. The structure of the collembolan
collophore thus suggests that it is formed by the union of a pair
of abdominal appendages resembling those of the Protura (fig. 29),
though in the latter the retractor muscle (B, rv) arises in the base
of the appendage, and the appendage itself is movable by two muscles
(J, J) arising in the body and inserted on its base. In some of the
Collembola each lobe of the vesicle is produced into a long eversible
tube.

No. 6 INSECT ABDOMEN—SNODGRASS 73

The typical spring, or leaping organ of the Collembola, known as
the furcula (fig. 30 D), has quite a different type of structure from
that of the collophore. It consists of a large median base, the manu-
brium (mn), and of two slender arms, each of which is subdivided
into a long proximal segment, the dens (d), and a short terminal
segment, or mucro (m). On the base of the manubrium are inserted
flexor and extensor muscles arising in the fourth and third abdominal
segments, but, as already observed, these muscles apparently belong
to the system of longitudinal dorsal and ventral body muscles, and
are not specifically muscles of the spring. In Tomocerus vulgaris

Fic. 30.—Abdominal appendages of Collembola. Tomocerus vulgaris.

A, lateral view of insect. B, anterior view of collophore. C, tenaculum. D
furcula.

a, aperture between bases of furcular arms receiving prongs of tenaculum;
ab, abductor muscle; ad, adductor muscle; Col, collophore; d, dens; Fur, fur-
cula; m, mucro; mn, manubrium; rv, retractor muscles of vesicle; v, terminal
vesicle of collophore.

(fig. 30) each of the arms of the furcula is provided with an abductor
muscle (D, ab) and an adductor muscle (ad) having their origins in
the manubrium. According to Quiel (1915) adductor muscles are
absent in Orchesella cincta, though he says a few obliquely transverse
fibers are present in the manubrium. It is possible that Quiel did
not observe in studying sections that these transverse fibers are
attached on each side to a slender adductor tendon of the dens. The
structure of the furcula readily suggests that it is composed of a
pair of segmental appendages united by a fusion of the coxae, which
become the manubrium, while the reduced telopodites become the
arms.

74 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The minute tenaculum of the third abdominal segment (fig. 30 C)
looks like a miniature furcula, and likewise suggests that it has been
produced in the same manner by the union of the bases of a pair of
appendages.

THE ABDOMINAL APPENDAGES OF THYSANURA

It is in the Thysanura that the abdominal appendages best preserve
the fundamental structure characteristic of the abdominal appendages
of adult Pterygota, as shown in the gonopods of the latter group ; but,
as will be seen later, it appears that the abdominal appendages have
a more primitive form in the larvae of Ephemerida and in the larvae

Fic. 31.—Diagrams of structure of abdominal appendages of Thysanura and
Pterygota.

A, a typical pregenital appendage. B, a gonopod, or genital appendage.

gmcls, muscles of gonapophysis; Gon, gonapophysis; LB, limb basis, usually
a lobe or plate of body wall; rvs, retractor muscles of eversible vesicle; smcls,
muscles of stylus; Stn, sternum; Sty, stylus; /’s, eversible and retractile vesicle.

of certain holometabolous insects: The thysanuran appendages retain
most completely their independence in the Machilidae.

The appendages of the pregenital region of the abdomen are
typically developed on each of the pregenital segments except
the first in Machilidae (fig. 4, II, VI), and those of each pair are
distinct from the small median sternal plate (Sin), though their
bases (LB, LB) are ankylosed with the latter, and are united medi-
ally with each other. Each appendage consists of a broad basal plaie
(fig. 31 A, LB), of a stylus (Sty) borne by the distal free margin of
the basal plate, and of an eversible sac, or vesicle (V's), lying mesad of
the stylus and retractile into the basal plate (fig. 4, II, Vs). The
posterior part of the basal plate projects from the ventral wall of the

NO. 6 INSECT ABDOMEN—SNODGRASS 75

abdomen as a free flap with a membranous dorsal wall (fig. 32 B).
Both the stylus and the eversible sac are provided with muscles arising
within the basal plate (figs. 31 A, 32 B, smcls, rvs). The muscles of
the stylus are inserted on the base of the stylus; the muscles of the
vesicle traverse the latter, when the vesicle is everted, to be inserted
within its distal extremity.

The appendages of the pregenital segments are never developed
into any other form in the Thysanura than that which they have
in the Machilidae, but they may be variously reduced, or united with
the sternum. The styli and eversible sacs are sometimes absent, or
either organ may occur alone (fig. 4, I, VIII). A pair of vesicles
is frequently present on each basal plate, but the stylus never occurs

Fic. 32.—Appendages of Thysanura and Dicellura.

A, base of metathoracic leg of Nesomachilis maoricus, showing styliform spur
on coxa. B, typical structure of a pregenital appendage of same, dorsal view.
C, posterior lateral part of sternum of Heterojapyx gallardi with united limb
basis bearing the stylus.

in duplicate. In Lepismatidae the basal plates of the appendages in
each segment are fused with the primary sternum to form a large
zygosternum, which is the definitive sternal plate of the segment.
The same is true in the Dicellura, though the regions of the limb
bases may remain partially separated from the region of the primary
sternum (fig. 32 C).

The basal plates of the thysanuran appendages are commonly called
“coxae”’ (or “coxites”) by American entomologists, while certain
European entomologists call them “ subcoxae.’’ The idea that the
plates are coxae is based chiefly on the fact that in the Machilidae
stylus-like spurs occur on the coxae of the second and third pairs of
thoracic legs (fig. 32 A, Sty?), which appear to be homologues of
the abdominal styli. The question of the possible identity of the

76 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

thoracic and abdominal styli will be discussed in the concluding part
of this Section, but in any case the term “ coxite” should not be used
to designate the stylus-bearing plates, because the word can properly
mean only “a part of a coxa.” Since the writer is inclined to believe
that the abdominal plates in question represent both the coxae and
the subcoxae of the thoracic appendages, they are here termed simply
the limb bases, or basal plates.

The gonopods, or appendages of the genital segments, have the
same structure as the pregenital limbs in Thysanura, with the excep-
tions that they always lack eversible vesicles, and that typically each
bears a median genital process, or gonapophysis (fig. 31 B, Gon).
Two pairs of gonapophyses are regularly present in the females of

Fic. 33.—Genital and postgenital segments of Machilidae.

A, Machilis variabilis, male, ventral plates of eee abdominal segment, dor-
sal view, with small first gonapophyses (1Gon). B, ventral plates of ninth seg-
ment of same, dorsal view, showing second pair of gonapophyses (2Gon) at
sides of median penis (Pen). CG, Nesomachilis maoricus, male, posterior part of
abdomen, ventral view, showing absence of gonapophyses, and ventral structure
a taleil segment, bearing paraprocts (Papt), caudal filament (cf), and cerci
(Cer

Thysanura, one pair borne by the gonopods of the eighth segment, the
other by the gonopods of the ninth segment. In the male, gonapophy-
ses are known to occur on the gonopods of the eighth segment only
in certain forms of Machilis (fig. 33 A, 1Gon); they are usually
present on the gonopods of the ninth segment (B, 2Gon), but they
may be absent from both genital segments (C). Each gonapophysis,
when present, arises from the median, basal angle of the free dorsal
surface of the stylus-bearing plate, and is provided with short muscles
(fig. 31 B, gmcls) arising within the supporting plate and inserted
on its base.

Between the bases of the gonopods of the ninth segment in the male
is a short membranous penis (fig. 33 B, C, Pen), a tubular evagination

No. 6 INSECT ABDOMEN—SNODGRASS 77

of the body wall from behind the region of the ninth sternum, having
the opening of the ejaculatory duct at its extremity.

The thysanuran gonopods contain, in their simplest form, the
fundamental elements of the organs of copulation and oviposition of
pterygote insects. In the male the gonapophyses of the second gono-
pods become the so-called parameres of the copulatory apparatus ; in
the female, the first and second gonapophyses become the first and
second valvulae of the ovipositor, and the basal plates of the second
gonopods form the third valvulae, when the last are present.

The uropods, or cerci, of the Thysanura are typically long, multi-
articulate filaments (fig. 7 A, Cer) borne by the eleventh segment
(B, XJ). In the Dicellura the abdomen contains only 10 segments,
and the uropods, therefore, in this group appear to belong to the
tenth segment. They are filamentous in Campodea, styliform in
Projapygidae, and take the shape of large pinchers in Japygidae (fig.
40 C, Cer). The uropods differ from the preceding appendages in
that they are not differentiated into a basal plate and a stylus, and
they bear neither eversible sacs nor processes corresponding with
the gonapophyses.

THE ABDOMINAL GILLS OF EPHEMERID LARVAE

The abdominal gills of ephemerid larvae, together with the lateral
lobes of the body wall supporting them (fig. 34 A), appear to be ap-
pendages of a more primitive form than the abdominal appendages
of the Thysanura. Their structure, with certain modifications, is
repeated in the abdominal appendages of several groups of holometab-
olous larvae; but it does not furnish the basis of the structure of
the gonopods in adult pterygote insects, which, as already stated,
is to be derived from that of the gonopods of the Thysanura.

The ephemerid larval gills have various shapes, some being taper-
ing stalks, either single or double, fringed with filaments (fig. 34 A, B,
Brn), while others are expanded into broad plates ; but, whatever the
form, each organ is movably attached by its base to a large lateral
lobe of the body wall (LB). The gill is movable by muscles arising
in the ventral part of the supporting lobe (A, B, bmcls).

The gill-bearing lobes are not movable, since there are no body
muscles inserted upon them. The vertical lateral muscles of the
abdominal segments extend from the tergum to the edge of the ster-
num mesad of the gill lobes (fig. 15 A, 7/, 21, 3l, fig. 34 A, 1). The gill-
bearing lobes therefore have the character of limb bases implanted
in the pleural areas of the segments between the tergal and sternal

78 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

plates (compare LB of fig. 34 A with LB of fig. 1 A). The lateral
and ventral surfaces of each lobe are often separated by a sharp fold
or ridge (fig. 34 B, LB), and the ventral part of the lobe may be
more or less united with the sternum of its segment (Stn). The gill
(Brn) arises from the posterior end of the lateral surface of the
supporting lobe, and is usually provided with four muscles inserted
on its base (C). The branchial muscles, as above noted, take their
origins on the ventral plate of the supporting lobe (B), and therefore

Fic. 34—The abdominal gill-bearing appendages of ephemerid larvae.

A, diagrammatic cross section of abdominal segment, showing limb base lobes
(LB) bearing gill appendages (Brn), which are evidently the styli, each mov-
able by muscles (bmcls) arising in the bases; the lateral body muscles (/)
extend from tergum to sternum.

B, external view of a gill and its supporting lobe (LB), in which arise the
gill muscles (bmcls).

C, musculature and tracheation of a gill, lateral view.

not on the sternum, as stated by Dtirken (1907), who did not dis-
tinguish the ventral plate of the gill basis from the true segmental
sternum.

The old idea that the ephemerid larval gills are appendages of the
dorsum is no longer tenable. The organs are very evidently rudiments
of segmental appendages, as claimed by Heymons (18 96a, 1896)
and by Borner (1909), each being composed of a basis (LB), and
of a terminal appendicular part (Brn) movable on the basis by mus-

NO. 6 INSECT ABDOMEN—SNODGRASS 79

cles arising in the latter. The gill bases are interpreted by Borner as
the subcoxae of the abdominal appendages, since they follow exactly
in line with the subcoxal, or “ pleural,” plates of the thorax. The
gills, therefore, Borner contends, are the equivalents of the legs,
and the proximal end of each represents the coxa. On the other
hand, we might assume that the gill basis includes the equivalents
of both the subcoxa and the coxa of a thoracic leg, in which case the
gill shaft or plate might be supposed to be the telopodite with its
proximal end representing the first trochanter. The presence of the gill
muscles arising in the gill basis suggests this homology, since the
muscles are comparable with the trochanteral muscles of a leg (fig.
26, O, Q). But again, the gill resembles the stylus of a thysanuran
appendage, and there is doubt as to whether the abdominal styli
are true telopodites or secondary appendicular processes of the coxae.

Leaving aside, for the present, the question of homologies between
the parts of the abdominal appendages and those of the thoracic
appendages, a comparison of the abdominal appendages of the ephem-
erid larva with the abdominal appendages of Thysanura leaves
little doubt that the gill-supporting lobes of the former are the
equivalents of the stylus-bearing plates of the latter, and that the
gill stalks or plates are the homologues of the styli. Neither the
abdominal gills of ephemerid larvae nor the styli of Lepisma are
present on the early postembryonic stage of the insect. In the newly
hatched larva of Ephemerella vulgata, according to Heymons (1896),
there are lateral protuberances of the abdominal segments, especially
prominent on segments II and VII, which are derived during em-
bryonic development from the embryonic limb rudiments. The giils
first appear as outgrowths from these abdominal lobes about four
days after hatching. The styli of Lepisma saccharina, Heymons
(1897) says, appear likewise a considerable time after hatching, and
arise from the parts of the ventral plates of the eighth and ninth
abdominal segments that are derived from the “ Anlagen” of the
embryonic appendages.

LATERAL ABDOMINAL APPENDAGES OF SIALID AND COLEOPTEROUS
LARVAE

The larva of Sialis (fig. 35 A) presents at least an excellent imita-
tion of an insect that has carried the primitive polypod condition
into a postembryonic stage. The long, tapering, segmented, ap-
pendicular organs, usually termed “ gills,’ projecting from the sides
of the first seven abdominal segments have a striking resemblance to

8o SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

legs. Each appendage (C) is a hollow process of the body wall, dis-
tinctly jointed, and composed of six segments, of which the proximal
three are relatively thick, while the region of the distal three is slender
and rapidly tapering to the apex.

Each appendage is supported on a lateral lobe of the body segment
(fig. 35 C, LB). The series of lobes appears at first glance to belong
to the dorsum of the abdomen, but each one, though so closely amal-
gamated with the tergum that the spiracle appears to be situated on
its dorsal part, really occupies a pleural position between the tergum

Fic. 35——Abdominal appendages of sialid larvae.

A, larva of Sialis, showing jointed appendages of abdomen. B, metathoracic
leg of Sialis larva, with subcoxal sclerites at its base. C, abdominal appendage of
Sialis larva, showing division into basis (LB) and jointed stylus (Sty). D,
posterior end of larva of Chauliodes, dorsal view, showing pygopods (Pp) of
terminal segment. E, the same, ventral view.

and the sternum of its segment. The lumen of each lobe is separated
from the general body cavity by a vertical sheet of tergo-sternal
lateral muscles, and within the lobe arise anteriorly and posteriorly
muscles inserted on the base of the movable shaft of the appendage.
It is clear, therefore, that we have here an organ corresponding
in every respect with a gill-bearing appendage of an ephemerid larva
(fig. 34 B), and that in both of these structures the basal lobe repre-
sents the stylus-bearing plate of Machilis, and the movable distal
appendage the stylus.

no. 6 INSECT ABDOMEN—SNODGRASS 81

The distinct segmentation of the abdominal appendages of the
Sialis larva almost unavoidably gives the impression that these organs
are the true telopodites of the abdominal limbs. The impression,
moreover, comes close to a conviction when it is discovered that each
appendage is provided with internal muscles in addition to those in-
serted on its base. The presence of these muscles was first mentioned
by Heymons (1896a) ; and it can be demonstrated by dissection and
staining of specimens preserved in alcohol that bundles of muscle
fibers are present in at least each of the first three segments, inserted
on the bases of the second, third, and fourth segments, but it is diffi-
cult to make an exact study of them without properly prepared ma-
terial. The slender distal part of the shaft beyond the third segment
is penetrated by a branched trachea, and this part of the appendage
might serve as a tracheal gill ; but the strong musculature of the proxi-
mal part of the organ, and the long hairs that fringe the segments,
suggest that the abdominal appendages of the Sialis larva have an
important locomotory function.

In the sialid genera Chauliodes and Corydalus the larvae are like-
wise provided with long, lateral abdominal appendages, a pair being
present on each of the first eight segments, and a terminal pair on the
last segment (fig. 36B). In these genera, however, the appendages
are simple, tapering, hollow processes of the integument, unsegmented,
and containing no muscles. Each is supported on a lateral lobe of the
body wall (LB).

The basal lobes of the lateral appendages of the Corydalus larva
are large and prominent as seen in a transverse section of an abdominal
segment (fig. 30C, LB, LB). Each projects laterally beyond the at-
tachments of a set of strong tergo-sternal lateral body muscles (/) ; and
on the inner margin of the ventral wall of the lobe arise three muscles
(D, smcls), one anteriorly and two posteriorly, which are inserted on
the base of the distal appendicular process. Here again, therefore, we
find repeated the same structures that occur in the ephemerid larval
gills and in the thysanuran abdominal appendages. In the Corydalus
larva the appendage-bearing lobes of the abdomen fall in line with
the subcoxal lobes of the thoracic segments (fig. 36 A, Scv), rather
than with the long coxae (Cx); but the muscles of the abdominal
appendages (D, smecls), taking their origins in the supporting lobes,
can be compared only with the basal muscles of the leg telopodite
(A, O, Q) inserted on the trochanter, The abdominal lobes, there-
fore, would appear to contain both the subcoxal and the coxal parts
of the limb bases.

82 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The Corydalus larva differs from the Sialis and Chauliodes larvae
in that the basal lobes of the first seven abdominal appendages bear
each a large ventral tubercle supporting a circle of respiratory fila-
ments (fig. 36 B, Vs). When the filaments are removed it is seen that
each tubercle is subdivided distally into three terminal lobes (C, Vs),
and dissection reveals the fact that the tubercle is provided with a large

Sty

Frc. 36.—Thoracic and abdominal appendages of the larva of Corydalus
cornutus.

A, transverse section through anterior end of metathorax, showing subcoxal
lobes (Scx) above bases of coxae (Cx), and levator and depressor muscles of
trochanter (O, Q) arising in coxa.

B, last four segments of abdomen, showing segmental appendages.

C, transverse section through posterior part of an abdominal segment, seen
from behind, showing limb base lobes (LB) supporting each a stylus (Sty),
and a retractile gill-bearing tubercle (I’s).

D, section through base of right abdominal appendage and gill tubercle, seen
from behind, showing muscles of stylus and gill tubercle.

FE, antero-mesal view of right appendage (pygopod) of tenth abdominal
segment.

F, section of terminal appendage, showing insertion of retractor muscles
behind bases of claws.

Cx, coxa; d, claws (crochets); e¢, insertion point of retractor muscles of
claw-bearing tubercle (planta); /, lateral body muscles; LB, limb basis; O,
QO, muscles of trochanter; rvs, retractor muscle of gill-bearing or claw-
bearing lobe of appendage; Sc, subcoxa; smcls, muscles of stylus; Sp, spiracle;
Stn, sternum; Sty, stylus; 7, tergum; /’s, retractile lobe of appendage (retrac-
tile vesicle).

retractor muscle (C, D, rvs) arising on the dorsum of the segment and
inserted by three diverging branches on the distal surfaces of the three
terminal lobes of the tubercle. The gill tubercles are thus highly
suggestive of the eversible sacs of the thysanuran appendages, except
for the difference that their retractor muscles arise on the tergal

CO

no. 6 INSECT ABDOMEN—SNODGRASS 3
region of the body segment, instead of in the bases of the appendages,
as in the Thysanura (fig. 32 B, rvs). On the other hand, as we shail
presently see, the gill tubercles of the Corydalus larvae are almost
identical in structure with the abdominal feet of lepidopterous larvae.

The appendages of the last abdominal segment in the larva of
Chauliodes and Corydalus are remarkable structures in that they com-
bine the characters of the preceding appendages of Corydalus with the
features of an abdominal leg of a caterpillar. Each of these terminal
appendages (figs. 35 D, Pp, 36 B, E) consists of a large, hollow,
somewhat cylindrical lobe of the body wall. The basal part of the
organ (fig. 30B, E, LB) bears laterally a tapering process (Sty)
similar to the lateral processes of the preceding appendages, and ends
distally in a thick tubercle (Vs), which lacks gill filaments, but is
armed on its terminal surface with a pair of large, curved claws (d).
The appendage is traversed by a strong retractor muscle (F, rvs)
taking its origin on the dorsum of the tenth segment, and having
its insertion on the distal wall of the tubercle at the posterior ends
of the long bases of the claws (e). The resemblance in structure
and mechanism of these appendages to the “ anal” legs of caterpillars
is so striking that it is difficult to believe the likeness is fortuitous.
The terminal body segment is better developed in the Corydalus larva
(fig. 36B, X) than in Chauliodes (fig. 35 E), in which its dorsal
part is rudimentary.

Lateral appendicular processes of the abdomen, similar in every
respect to those of the sialid larvae, are present also on certain aquatic
coleopterous larvae, especially in the families Dytiscidae and Gyrin-
idae. In the gyrinid Dineutes, for example, the larva is provided with
long, tapering processes arising from lateral lobes of the body wali
on each side of each of the first eight segments, and with a pair of
two-branched processes on the ninth segment. Each process is pene-
trated by a trachea from the lateral respiratory trunk, and is fur-
nished with two short, antagonistic muscles arising in the supporting
lobe of the body and inserted on its base. The larva of the dytiscid
Coptotomus has the same equipment of lateral processes, but the writer
did not find muscles connected with them in a specimen examined.

THE ABDOMINAL LEGS OF LEPIDOPTEROUS LARVAE

A typical abdominal leg of a caterpillar consists of three parts
(fig. 37 A). At the base is a ring of flexible integument (mb) ; be-
yond this is a longer, cylindrical section (Cx) forming the greater part
of the appendage, and frequently having a sclerotic plate in its outer

6

7

84 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

wall often marked by distinctive groups of setae (fig. 3 A) ; distally
the leg ends in a retractile lobe (Vs), called the planta, which bears
the claws, or crochets (d).

Functionally the planta is the most important part of the leg, and
structurally it is the most variable. The proximal parts of the ap-
pendage differ principally in relative size in different species. The
planta in its more generalized condition is a short cylindrical pad with
a circular distal surface, on the center of which is inserted a group

Fic. 37.—Structure of an abdominal leg of a caterpillar. Malacosoma
americana.

A, posterior view of a pair of abdominal legs. B, ventral view of a left ab-
dominal leg, showing crochets (d) turned outward, and insertion of retractor
muscles (¢) at inner margin of plantar lobe (Vs). C, dorsal view of leg mus-
culature, right side. D, posterior view of musculature of a right leg.

a, dorso-pleural groove; Cx, coxa; d, crochets; e, insertion point of retractor
muscles of planta; mb, membranous area between subcoxa and coxa; rvs,
retractor muscles of planta; Scx, subcoxa; Vs, planta (retractile vesicle).

of retractor muscle fibers. In such cases the crochets may be arranged
in a complete circle around the distal plantar surface, with their re-
curved points turned peripherally and upward. With most cater-
pillars, however, the claws are limited to a semicircle or a small arc
usually on the inner margin of the planta (fig. 37 B), and in such cases
the planta itself (Vs) generally becomes asymmetrical by a reduction
or obliteration of its outer half. The planta then assumes the form
of a lobe projecting to the mesal side of the axis of the limb, marked
by the insertion point of the retractor muscle (e), and its claws (d)

No. 6 INSECT ABDOMEN—SNODGRASS 85

curve mesally and upward when the planta is protracted in the usual
position. The various types of foot structures resulting from modi-
fications in the form of the planta and in the arrangement of tie
crochets characteristic of the different groups of caterpillars have
been described by Fracker (1915).

Immediately above each abdominal leg there is usually a prominent
lobe or swelling of the body wall (fig. 37 A, Scv), separated from the
latero-dorsal area of the segment by a distinct groove (a). Corre-
sponding lobes are present also on the legless abdominal segments, and
likewise on the metathorax and mesothorax (fig. 3 A, Scr). The
serial identity of these suprapedal lobes of the abdomen and thorax
is demonstrated by their uniform position relative to the appendages,
and by the fact that in many species they bear similar or identical
marks or groups of setae. In the anatomy of the caterpillar, therefore,
the abdominal and thoracic appendages appear to be homodynamous
structures. Eastham (1930), in his study of the embryology of
Pieris rapae, says: “The prolegs which are retained on their seg-
ments must be regarded as true appendages. They develop in the same
manner as those of the head and thorax, have the same relation to their
own somites, and a musculature develops in connection with each com-
parable to that of the thoracic limbs though of a weaker order.”

The suprapedal lobes of the caterpillar (fig. 3 A, Scx) are clearly
the subcoxal areas of the appendages, since those of the thorax are
identical with the areas which in certain other holometabolous larvae
contain the pleural sclerites of the thorax (B, C, Scv3). The free
part of the abdominal appendage in the caterpillar is, therefore, ap-
parently the coxa (fig. 37 A, Cx). The planta (Vs), then, is either
a rudiment of the telopodite, or a highly specialized retractile vesicle
of the coxa. Further light on the morphology of the caterpillar proleg
may be obtained from a study of the musculature.

The musculature of an abdominal leg of a caterpillar is compara-
tively simple. It comprises two sets of muscles, those of one set
being inserted on the base of the principal part of the leg (fig. 37 D,
Cx), those of the other on the planta (Vs). The muscles inserted on
the proximal rim of the leg include three groups of fibers represented
in Malacosoma americana and Estigmene acrea as follows: (1) a
series of median fibers (fig. 21 B, 37 C, D, r) arising on the midline
of the venter, or also on the mesal parts of the anterior and posterior
intersegmental folds, and converging to the mesal rim of the base of
the principal segment (Cx) of the leg; (2) a group of two fibers (fig.
37 C, D, 2) arising on the groove (a) above the suprapedal lobe of
the body wall, and inserted on the mesal rim of the leg base posteriorly

86 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

just ventrad of the insertions of the median muscles; (3) a group
of two or three fibers (C, D, 3) taking their origins on the middle of
the lateral wall of the segment posterior to and a little above the level
of the spiracle (fig. 21 B), and having their insertions on the outer side
of the proximal rim of the leg (figs. 21 B, 37 C, D, 3).

The muscles of the planta take their origins outside the leg from
two widely separated parts of the body wall. The plantar muscles
of Malacosoma americana (fig. 37 C, D) comprise four fibers. The
principal one is a long fiber (4) arising on the middle of the lateral
wall of the segment, close to the origins of the lateral muscles of
the leg (3), from which point it curves downward into the leg to be
inserted on the inner face of the planta. The other plantar muscles
arise on the dorso-pleural groove (a) above the suprapedal lobe (fig.
37 D, Scx). One consists of a single fiber (C, D, 5) arising pos-
teriorly just before 2, and entering the leg with 4. The other in-
cludes two fibers (6) in Malacosoma, represented by a single fiber
in Estigmene (fig. 21 B, 6), arising anteriorly on the dorso-pleural
groove, and curving posteriorly and downward into the leg to join
with 4 and 5. In the distal part of the leg (fig. 37 D) all the fibers
of the plantar group unite to form a common stalk which is inserted
on the inner surface of the ventral wall of the planta. In caterpillars
having a disk-shaped planta, the muscle insertion is at or near the
center of the latter, but with species in which the planta has the form
of a mesal lobe, the muscle attachment is at the outer side of the
plantar lobe (fig. 37 B, D, e).

On comparing, in the caterpillar, the musculature of an abdominal
leg with that of a thoracic leg, it is found that though there is ne
exact correspondence in the number and arrangement of the fibers,
there is a general similarity in the disposition of the muscles sufficient
to suggest a derivation of the muscles in the two cases from one
fundamental plan of musculature. Thus, in the metathorax of Mala-
cosoma (fig. 38 A, B) there is a set of sternal fibers (a) arising an-
teriorly on the intersegmental fold, and inserted mesally on the rim of
the coxa (Cx), which correspond with the median muscles of an ab-
dominal leg (fig. 37 C, D, 1). Likewise, there are muscles from the |
lateral wall of the thoracic segment inserted on the outer rim of the |
coxa (fig. 38 B, C, b), having thus the same relation to the appendage |
as the fibers of muscle 3 in the abdomen (fig. 37C, D). In the |
thorax there are several subcoxo-coxal muscles (fig. 38 B, c) which |
have no exact counterparts in the abdomen, though in the latter there |
is a muscle from the groove above the subcoxal lobe (fig. 37 C, D, 2)
to the inner margin of the apparent coxal segment of the leg. The —

SNODGRASS 87

No. 6 INSECT ABDOMEN 7

other muscles in the base of a thoracic leg (fig. 38 B, C, d, e, f, g)
are coxo-trochanteral and coxo-femoral muscles, representatives 0}
which are entirely absent in the abdominal legs. On the other hand,
the plantar muscles of the abdomen have no evident counterparts in
the thorax.

The general parallelism between the muscles of the abdominal ap-
pendages and those of the thoracic legs shows that the musculature
of the prolegs in the mature caterpillar is, as Eastham says of the
comparable to that of the thoracic limbs
Moreover, if the musculature has any

ce

musculature in the embrvo,

though of a weaker order.”

c

Wy

1 CW ee ace
¢ SS” mM i :
; ; a G/,

}
[
ot —

q caine % | Sap

Fic. 38.—Body muscles of a thoracic leg of a caterpillar.

A, external body muscles and leg muscles of ventral area and right half of
metathorax of Malacosoma americana. B, basal muscles of right metathoracic
leg of same. C, coxal and coxo-trochanteral muscles of right mesothoracic leg
of Estigmene acraea.

bearing on homologies in the segments of the appendages, it shows
that the suprapedal lobes of the body wall above the abdominal limbs
are the subcoxae (fig. 37 A, D, Scv), and that the principal parts
of the legs are the coxae (Cx). The plantae of the abdominal ap-
pendages, however, have no evident homologues in the thoracic legs,
nor do their muscles correspond with any of the muscles of the legs
in the thorax.

The nature of the planta becomes clear when we compare an ab-
dominal limb of a caterpillar (fig. 37 D) with an abdominal appendage
of the Corydalus larva (fig. 36C, D). A striking resemblance is

88 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

seen at once between the retractile planta of the former and the gill-
bearing tubercle (Vs) of the latter. Moreover, the likeness amounts
almost to a structural identity when the planta is compared with the
claw-bearing lobe of one of the terminal appendages of the Cory-
dalus larva (fig. 36 E, Vs). In each case the organ is provided with
strong retractor muscles arising on the dorsum of the body segment.
It is a mere detail that the muscles in Corydalus (fig. 36 D, rvs) branch
to the several lobes of the gill tubercle; in the terminal appendages
the bundle of retractor fibers (F, rvs) tapers to a narrow stalk inserted
at the bases of the claws exactly as in the foot of the caterpillar (fig.
37 D). The Corydalus larva lacks only the accessory muscles of the
planta arising in the base of the subcoxa. We can, therefore, scarcely
avoid the conclusion that the planta of the caterpillar’s abdominal
leg is an eversible vesicle of the limb basis, here borne by the coxa,
which is not consolidated with the subcoxa. The representatives of
the styli, preserved in the movable distal processes of the sialid ap-
pendages (fig. 36 C, D, E, Sty), have been lost from the abdominal
limbs of all lepidopterous larvae.

The anal legs, or appendages of the last abdominal segment of the
caterpillar, differ from the legs of the preceding abdominal segments
only in details of their musculature. Their structure will be described _
in the subsequent discussion of the terminal appendages of holometab-
olous larvae.

The abdominal appendages of chalastogastrous hymenopterous
larvae have the same essential structure as the abdominal limbs of
caterpillars, though they are not so highly organized, and the plantar
lobes are reduced to terminal disks of the coxal segments (fig. 3 C).

THE GONOPODS

The appendages of the eighth and ninth segments of the abdomen
are potentially gonopods because of the association of the openings
of the genital ducts with these segments. They are, however, not
necessarily modified for reproductive purposes, as in the males of cer-
tain Thysanura in which the appendages of the eighth and ninth
segments (fig. 33 C) do not differ structurally from those of the
preceding segments.

Typically a gonopod is distinguished from the pregenital and post-
genital appendages by the development of a median process from the
proximal part of its base. A complete gonopod, therefore, consists
of a basal lobe or plate (fig. 31 B, LB), of a lateral, distal stylus (Sty),
and of a median, proximal gonapophysis (Gon). Both the stylus and

no. 6 INSECT ABDOMEN—SNODGRASS 89
the gonapophysis may be movable on the basis by muscles arising in
the latter and inserted on their bases (smcls, gmcls). So far as has
been ascertained neither of the appendicular parts of a gonopod ever
contains intrinsic muscles. In the males of some Ephemerida the
genital styli are distinctly jointed and thus separated into apparent
segments, though the latter are not independently movable. In the
Thysanura the gonapophyses are marked by circular constrictions,
varying from a few to many according to the length of the organs
(fig. 33 A, B), but the resulting subdivisions have none of the charac-
ters of true segments, and are entirely comparable with the annulations
of the caudal filament and cerci (C; cf, Cer).

In female insects the gonopods form the ovipositor, when this organ
is present, and both pairs of appendages enter into its composition.
In the Thysanura the gonapophyses only are involved in the ovipositor,
the basal plates and the styli retaining the structure typical of these
parts in the pregenital segments of Machilidae. Evidently the condi-
tion here represents a primitive stage, in which two pairs of median
apophyses of the appendages of the eighth and ninth abdominal seg-
ments form a simple egg-laying organ. In female Pterygota the styli
of the gonopods are usually lost, those of the first pair being always
absent, and the basal plates are transformed into a suspensory appara-
tus for the gonapophyses. The basal plates of the first gonopods (fig.
39 A) evidently become the small sclerites known as the valvifers
(B, Vif), which support the first gonapophyses (1//1), though there
is a difference of opinion on this point. Those of the second gonopods
form lobes (JX LB) supporting the second gonapophyses (2V1), or
they are drawn out into long processes that become a third pair of
blades in the ovipositor (C, 37). The component blades of the adult
ovipositor are commonly called valvulae. It is to be observed that only
the first and second pairs of valvulae (B, C, rVl, 2V1) represent the
gonapophyses (A, 1Gon, 2Gon), those of the third pair (C, 3/1)
being derived directly from the basal plates of the second gonopods,
the styli of which are apparently lost.

The ovipositor is absent or rudimentary in many groups of insects,
but its wide distribution throughout the orders leaves little doubt
of its being a primitive structure of the Insecta. It has no homo-
logue in other Arthropoda, and it is doubtful if the rami of the gono-
pods in the Crustacea are homodynamous with the gonapophyses
of insects. The various theories concerning the possible homologies
of the genital processes of the gonopods in insects will be considered
in the closing discussion of this paper.

F

go SMITHSONIAN MISCELLANEOUS COLLECTIONS — VOL. 85

In male insects the history of the gonopods is much more involved
than in the female, and the evolution of the genital appendages into
organs of copulation has produced many different kinds of struc-
tures. It is only in certain species of Machilis that both pairs of gono-

VIII LB (VIF) XLB Sty D
\

Vig.

N \
G iVI 2V1 3V1

Fic. 39.—Diagrams showing the morphology of the ovipositor, and the exter-
nal genitalia of the male.

A, female genital segments and generalized structure of the gonopods. B, an
ovipositor with two pairs of valvulae formed of the gonapophyses. C, an ovi-
positor with three pairs of valvulae, the third pair (3/1) formed of the basal
plates of second pair of gonopods (A, LB).

D, generalized structure of ventral parts of ninth abdominal segment of male,
showing sternum (Stn), gonopods (Gp, Gp), and penis (Pen). E, specialized
structure of male genitalia; basal plates of gonopods (LB, LB) united by pons
basalis (pnb), parameres united with penis to form an aedeagus (ded), and
styli (Sty) transformed into clasping lobes.

Aed, aedeagus; Gon, gonapophysis; 1Gon, 2Gon, gonapophyses of first and
second gonopods; LB, basal plate of gonopod; Ovd, oviduct; Pm, parameres
(male gonapophyses) ; pub, pons basalis; smcls, muscles of stylus; Stn, ster-
num; Sty, stylus; 7, tergum, rV1, 2V1, 3V1, first, second and third valvulae;
Vif, valvifer.

pods bear gonapophyses (fig. 33 A, B, 1Gon, 2Gon). In all male
pterygote insects gonapophyses are lacking on the eighth segment, and
the gonopods of the ninth segment only enter into the copulatory
apparatus, though accessory structures of the eighth and the tenth
segment may be included.

No. 6 INSECT ABDOMEN—SNODGRASS gI

The basal plates of the second gonopods in the male may unite with
each other and with the primary sternum of their segment, as do those
of the preceding segments, to form a composite zygosternum, and in
such cases the styli either retain the typical shape of styliform organs,
or they are lost. The gonapophyses, usually termed the parameres,
however, in the more generalized insects, are associated with a median
intromittent organ, or penis (figs. 33 B, 39 D, Pen), which is a tubular
evagination of the segmental wall behind the ninth sternum, bearing
the opening of the ejaculatory duct at its extremity. In the higher
insects the primitive penis becomes partly or entirely suppressed, and
the parameres unite with it or with each other to form the secondary
and often more complicated intromittent organ usually termed the
aedeagus (fig. 39 E, Aed), which incloses the ejaculatory duct and
bears the gonopore.

The basal plates of the gonopods of the ninth segment in the male,
if not completely amalgamated with the sternum, may form free lobes
of the ninth segment, or they may unite with each other, with the
sternum, or with the sternum and the tergum of the ninth segment.
In this way the genital segment of the male, especially in holometa-
bolous insects, assumes a great diversity of structure, and it is often
reduced to a simple continuously sclerotized annulus. The ninth seg-
ment, however, regardless of its form, always bears the aedeagus,
which may be partly or wholly concealed in a genital chamber of its
ventral part, and it generally carries clasping organs of various forms
on its posterior margin. Usually, among the clasping organs of the
ninth segment, or often the only structures having a clasping func-
tion, is a pair of lobes flexible at their bases and independently mov-
able by muscles taking their origins in the basal plates of the gonopods,
or in the regions of the ninth annulus derived from the gonopod bases.
These movable claspers, designated the harpes by students of Lepi-
doptera, are evidently the homologues of the styli of the more general-
ized insects (fig. 39 D, E, Sty).

It is most important, now, to observe that in the fundamental or-
ganization of the gonopods there are only two sets of appendicular
structures that are independently movable by muscles inserted directly
on their bases. These structures are the styli and the gonapophyses.
Therefore, in the ninth segment complex of the male genitalia, there
will generally be two sets of appendicular structures, the harpes and
the parameres, provided with muscles arising in the basal plates of
the gonopods, or in the parts of the ninth segmental ring derived
from the latter. By a study of the genital musculature, then, thesc
two structures can be identified with certainty in almost all cases

92 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

throughout the orders, unless one or the other or both are absent.
In addition to these fundamental, musculated processes, however,
there are innumerable other secondary genital processes having no
necessary homology in the different orders, which may be developed on
the ninth segment, on the aedeagus, or on the segments preceding
and following the ninth. These structures, except in rare cases, have
no muscles of their own, and are not independently movable, though
some of them may be moved incidentally by the usual segmental mus-
cles attached at their bases. The movable claspers derived from the
styli, however, are sometimes divided, and each may be separated into
two quite distinct parts provided individually with groups of muscle
fibers. In such cases there will appear to be, as in some of the Hy-
menoptera, a pair of movable lobes on each side of the genital
apparatus.

A more detailed analysis of the structure of the organs of ovi-
position and copulation, as shown in the principal orders of insects,
will form the subject matter of Part II of this paper, wherein will be
presented also a larger body of evidence in support of some of the
statements that seem arbitrary in the brief discussion given above.

THE CERCI (UROPODS)

The prevalence of cerci in so many orders of insects, and the almost
universal occurrence of the organs in the more generalized groups
leave little doubt that the cercal appendages are primitive structures,
and that, in some form, they must have been characteristic features
of the early insect ancestors. The anatomy and structural variations
of the cerci are well known; functionally the appendages are in most
cases sensory organs, though they are frequently modified in form
to serve mechanical purposes; morphologically they are subjects of
diverse opinion among speculative entomologists. The essential facts
known concerning the cerci can be briefly stated.

In the Thysanura the cerci evidently belong to the eleventh ab-
dominal segment. The last typical segment of the body in such forms
as Nesomachilis (fig. 7A) is the tenth (X), which is a complete
annulus. From within the posterior margin of this segment there
project the three terminal filaments, of which the lateral pair are
the cerci (Cer). If the group of filaments is pulled out of the tenth
segment, it is seen that the three of them arise from a common basal
ring (B, XJ), which has all the aspects of a reduced segment, in this
case the eleventh, normally concealed within the tenth. The eleventh
annulus presents a wide dorsal region (C, XJT) prolonged into the

No. 6 INSECT ABDOMEN—SNODGRASS 93

median caudal filament (cf), a narrow ventral region (D, X/Stn)
bearing a pair of broad posterior flaps (Papt), and two prominent
lateral lobes supporting the cerci (Cer). There can be little doubt,
therefore, that the caudal filament and the cerci here belong to the
eleventh abdominal segment. The lobes of the eleventh sternum are
evidently the paraprocts (D, Papt). Projecting from beneath the base
of the caudal filament (D, cf) is a small median lobe (sa), possibiy
a remnant of the true telson, represented by the lamina supra-analis of
the twelfth segment, better developed in odonate larvae (fig. 12 A, sa).

The terminal parts of the abdomen are less simple in some other
thysanurans than they are in Nesomachilis. In Thermobia (fig. 7 E,
F), for example, the eleventh segment is largely obliterated except
for a distinct tergal plate, or epiproct (E, Eppt), which is connected

Fic. 40.—The cerci and associated parts.

A, posterior segments of Periplaneta orientalis, dorsal view, showing union of
tenth and eleventh abdominal terga. B, ventral view of terminal segment of
same. C, cerci of Heterojapyx gallardi, and muscles of tenth segment that move
them.

laterally by a pair of small sclerites with the bases of the cerci (Cer).
The paraprocts of Thermobia are large sclerotic plates (F, Papt)
supporting the cerci. They would thus appear to correspond with
the lateral lobes of the eleventh segment and the median paraproct
plates of Nesomachilis (D). A sternal region of the eleventh segment
distinct from the paraprocts is not evident in Thermobia.

In the Dicellura (Campodeidae and Japygidae) the abdominal
segments beyond the tenth are obliterated, and the cerci are sup-
ported directly by the end of the tenth segment (fig. 40 C, Cer).

The cerci of the Pterygota most commonly appear to belong to the
tenth abdominal segment, since they arise at the posterior edge of
the latter, usually from membranous areas inclosed by the adjacent
angles of the tenth tergum, the epiproct, and the paraprocts (fig. 12 A,
B, Cer). In Periplaneta the cerci of the adult insect arise from be-

94 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

tween the lateral angles of the last tergal plate and the upper
angles of the paraprocts (fig. 40 A, B), but the terminal plate
of the dorsum is here clearly a composite sclerite formed of the united
tenth tergum (A, X) and the epiproct (XJ). Frequently the cerci
are more closely associated with the paraprocts than with the tergal
plates. In any case, however, the intermediate position of the cerci
in adult Pterygota gives no positive evidence of the segmental rela-
tions of these appendages in this group of insects.

On the other hand, the ontogenetic evidence of the nature of the
pterygote cerci seems to be quite definite, for it is stated by Ayers
(1884), Cholodkowsky (1891), Wheeler (1893), and Heymons
(1896a) that the cerci in the embryos of Orthoptera are formed
directly from the appendages of the eleventh abdominal segment (figs.
5 A, 9A, B, Cer). Heymons claims that the eleventh segment itself
disappears from the adult abdomen, and that the cerci thus come
to have an apparent intersegmental position between the tenth and
the twelfth segments. As already shown, however, it appears more
probable that the eleventh segment is usually represented in the
adult by the epiproct and the paraprocts, and that it is the twelfth
segment which is lacking, or reduced to a circumanal fold (fig. 12 A,
Prpt).

The association of the cerci with the upper basal angles of the
paraprocts, or their actual connection with these plates in some cases,
as in Thermobia (fig. 7 F), has given rise to the idea that the cerci
and the paraprocts have a genetic relation to each other. Thus,
Crampton (1920, 1921) contends that the paraproct is the base of a
segmental appendage of which the cercus is the distal part. Accord-
ing to Heymons (1896), on the other hand, the embryonic cercus
represents the entire appendage of the eleventh segment, including the
basis, which in the pregenital segments unites with the primary seg-
mental sternum to become a lateral part of the definitive sternal plate.
In the adult insect, Heymons says, the cercal base usually disappears
as an evident lobe, though a rudiment of it is retained in young
nymphs of Gryllus and Decticus as a small basal ring supporting the
free part of the organ (fig. 8 B).

The musculature of the cerci, so far as it is known, is always dorsal,
there being no muscles from the sternal region of the abdomen or from
the paraprocts in any way associated with the cerci. The origins of the
muscles present, however, give no clew to the segmental relations of
the cerci, since the muscles arise either on both the tenth and eleventh
terga, or on the tenth tergum alone. In her study of the abdominal
muscles of Orthoptera, Ford (1923) finds that each cercus is typically

\
|
j
}
f
t
|
(
\

no. 6 INSECT ABDOMEN—SNODGRASS 95

provided with four muscles. Three of these muscles, distinguished
as an abductor, a depressor, and an elevator muscle, take their origins
on the tenth abdominal tergum; the fourth arises on the supra-anal
plate or epiproct. The muscle from the epiproct, Ford says, is absent
in Gryllotalpa, but it is present in Gryllus, Neoconocephalus, Ceutho-
philus, and Melanoplus. In Gryllus, because of the union of the tenth
tergum and the epiproct, this muscle, however, has shifted forward to
the area of the tenth tergum. The writer has found only three mus-
cles in Dissosteira inserted directly on the base of the cercus, two
arising on the tenth tergum, and one on the epiproct.

The origin of the anterior muscles of the cerci on the tenth segment
might be construed as evidence in favor of the view that the cercal
appendages belong to the tenth segment; or, on the other hand, it
might be taken as favoring Heymons’ claim that the true eleventh
segment has been obliterated. However, it is not necessary to assume
that the muscles associated with the cerci are primarily muscles of
these appendages. The great bundles of fibers that operate the
pincer-like cerci of Japyx almost fill the large tenth abdominal seg-
ment (fig. 40 C, mcl), but they appear to be the longitudinal dorsal
muscles normal to this segment, which secondarily function as cercal
muscles by reason of their posterior attachments at the bases of the
cerci. Ford (1923), observing that most of the cercal muscles in
Orthoptera arise from the tenth tergum, asserts that these muscles are
““intersegmental muscles between the tenth and eleventh segments,”
while the muscles from the epiproct, she says, represent “ the inter-
segmental muscles between the eleventh and twelfth terga.” (Her
reference of the posterior muscles to the twelfth tergum is in accord
with her acceptance of Heymons’ claim that the eleventh segment
has disappeared in the adult.)

Whatever may be the nature of the dorsal muscles of the cerci,
the fact is significant that the organs have no ventral musculature—
in this respect cerci differ from styli and gonapophyses. The absence
of muscles from the paraprocts to the cerci, moreover, weakens the
comparison between the paraprocts and the stylus-bearing plates of
the preceding abdominal segments, since the stylus muscles always
take their origin in these plates. The termination of the ventral mus-
culature of the abdomen in the paraprocts, on the other hand, makes
it almost certain that the paraprocts are terminal lobes of the eleventh
sternum.

Cerci are usually absent in holometabolous insects, but cercus-
like appendages occur on the eleventh abdominal segment in females
of Panorpa (fig. 8H, Cer?), and on the terminal segment of adult

96 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Tenthredinidae, which is numerically the tenth abdominal segment.
‘The panorpid appendages may be true cerci. Appendages occurring
on an apparent tenth segment might be suspected of being cerci if
there is evidence that this segment is composed of the tenth and
eleventh somites, a condition which frequently occurs in orthopteroid
insects, where there is no doubt that the terminal appendages are
the cerci. In the Tenthredinidae, however, there is reason to believe,
as will be shown later, that the terminal appendages of the adult are
not the cerci, but are appendicular organs of the tenth segment cor-
responding with the socii of adult Lepidoptera, and that they are de-
rived from the postpedes of the tenth somite of the larva.

THE TERMINAL APPENDAGES OF ENDOPTERYGOTE LARVAE

Appendicular organs representative of abdominal limbs are present
on the last abdominal segment in some or most of the larvae of
Neuroptera, Trichoptera, Lepidoptera, and chalastogastrous Hymen-
optera. These larval appendages of the terminal segment have a
lateral or latero-ventral position, and are movable by muscles in some
cases attached on their bases, but more generally inserted within
their distal parts. The appendages most resemble jointed limbs in
the Trichoptera. In Neuroptera, Lepidoptera, and Hymenoptera they
commonly have the structure typical of the abdominal legs of the
caterpillars. Since the terminal segment in these larvae is evidently
the true tenth somite of the abdomen, or the tenth and the eleventh
somites combined, there is little doubt that the terminal appendages
are the pygopods. The Endopterygota differ thus from the more gen-
eralized Exopterygota in that some of them retain the tenth segment
appendages in postembryonic stages.

In addition to the true appendicular organs, there may be in endop-
terygote larvae also processes developed from the dorsum of the last
segment, and lobes of various forms associated with the anal opening,
or protruded from within the rectum. Processes resembling cerci
occur in some coleopterous larvae, but their morphology is uncertain.

It seems probable that the pygopods of endopterygote larvae are,
in certain orders, caried over to the adult stage as processes which
sometimes occur on the proctiger, or terminal segment of the imago.
These processes have various forms in the Trichoptera and Lepidop-
tera, and are termed the soci by students of the latter; in chalasto-
gastrous Hymenoptera they resemble cerci, and are frequently called
“cerci.”” Busck and Heinrich (1922) have observed that in the micro-
lepidopteron Ethmia machelhosiella the anal prolegs, with their

no. 6 INSECT ABDOMEN—SNODGRASS 97

crochets, are retained in the pupa, and Bottimer (1926, fig. 3 A)
shows a similar retention of the anal legs in the pupa of Chaetocampa
crotonella. It is unfortunate that socii are absent in these species,
for we might expect to find the socii developed within the anal pro-
legs of the pupae. More positive evidence of the identity of the termi-
nal larval appendages with the terminal appendages of the adult is pre-
sented by Middleton (1921) in a study of the chalastogastrous Hy-
menoptera. Middleton claims that the anal prolegs, or postpedes, of the
larva of Pteronidea ribesti are transformed during metamorphosis into
anal lobes of the pupa, and that within these lobes are developed the
so-called ‘‘ cerci” of the adult sawfly. These appendages of the adult
insect, he points out, are not borne by the tergum of the anal segment,
but arise from a lateral membranous area of the venter of this seg-
ment, and thus morphologically have the same position as have the
postpedes of the larva. The anal segment in both the larval and the
adult sawfly is numerically the tenth abdominal segment, and its ap-
pendages are therefore not the true cerci in either case.

Trichoptera—The abdomen of trichopterous larvae lacks appen-
dages except on the terminal segment. In some forms, as in Platyphy-
lax designatus, a fringe of slender setae along each side of the abdomen
from the second to the end of the eighth segment (fig. 41 A, B, a)
evidently marks the dorso-pleural line separating the dorsum of the
abdomen from the region of the limb bases (LB), since, if the line of
this fringe were carried into the thorax, it would run dorsad of the
thoracic subcoxae (A, Scv;). On the ninth segment there is nothing
to mark the dorso-pleural boundary; but on the terminal segment
the base of the appendage (B, Pf) has a lateral position corresponding
with the limb base areas of the segments preceding the ninth. The
terminal segment of Platyphylax is a hemispherical lobe with a long,
median anal cleft on the ventral part of its distal surface (An). It is
evidently the tenth somite, or pygidial segment.

The pygopods of trichopterous larvae differ considerably in different
families and genera. Their principal variations have been described by
Ulmer (1903) and by Krafka (1924). According to Ulmer there
are two principal types of these appendages. Those of one type are
short ; those of the other, characteristic of Hydropsychidae and Rhya-
cophilinae, are long and leg-like. In both types the limb terminates in
a hook-like claw. The structure of the two forms of appendages
is here illustrated from Platyphylax designatus (fig. 41 B, Pp) and
an unidentified species of Hydropsyche (F, Pp).

In Platyphylax designatus each larval pygopod together with its
supporting structure (fig. 41 B, Pp) consists of a large basal plate (C,

98 SMITHSONIAN MISCELLANEOUS COLLECTIONS ~— VOL. 85_

b), a smaller intermediate plate (c), and a free apical claw (d) hav-
ing its point turned downward and outward. The apical claw has a
movement of adduction on the middle plate, and the latter is movable
on its hinge with the basal plate. The muscles of this appendage may
be studied by cutting the terminal body segment into lateral halves.
It is first seen that the base of the appendage is crossed internally by
several slender transverse muscles attached on the segmental walls.
The muscles inserted on the appendage consist of three flat groups of

Fic. 41.—Structural details of trichopterous larvae.

A, metathorax and anterior part of abdomen of Platyphylax designatus.
B, posterior abdominal segments of same. C, appendage of tenth segment of
same. D, mesal view of right appendage, showing muscles. E, mesal view of
terminal appendage of Hydropsyche sp., showing muscles. F, posterior end of
abdomen of same, with intestinal processes (ip) extruded from anus.

a, line of dorso-pleural groove; An, anus; b, proximal sclerite of appendage;
Cx, coxa; d, claw of appendage; e, f, intersegmental folds between eighth and
ninth, and ninth and tenth abdominal segments ; 7, intestinal processes; LB, area
of abdominal limb base; Pp, pygopod; Scx, subcoxa.

fibers (fig. 41 D), all arising on the anterior margin of the basal
plate (b). Two of these muscles (za, rb) converge downward and
posteriorly to a common point of insertion on the inner margin of
the base of the apical claw (d), and thus evidently function as ad-
ductors of the latter. The third muscle is a broad sheet of fibers
(2) lying external to the others, and inserted on the dorsal margin of
the intermediate plate (c). Associated with the appendage is a pair of
slender, vertical fibers (5) crossing the inner face of the basal plate
and attached above and below it to the membranous walls of the

No. 6 INSECT ABDOMEN—SNODGRASS 99

terminal body segment. The ventral muscles of the ninth body seg-
ment (C, 4) are attached posteriorly on the intersegmental fold (f)
just before the lower part of the basal plate of the appendage, but
no muscles are inserted directly on the latter.

The terminal appendages of the larva of Hydropsyche (fig. 41,
F, Pp) represent quite a different type of appendicular organ. Each
pygopod here is a large, two-segmented, leglike structure projecting
from the ventro-lateral part of the terminal body segment. The
proximal segment of the appendage (E, b) is elongate and cylindrical ;
the distal segment (d) is a strong, decurved claw with a high, nar-
row base articulated by its dorsal end with the dorsal extremity of

_the basal segment. Ventrally, just before the base of the claw, there

is a small triangular sclerite (c) in the lower, membranous wall of
the proximal segment.

The musculature of the terminal appendage of Hydropsyche (fig.
41 E) differs in several respects from that of Platyphylax (D). As
in the latter, however, there are no muscles inserted on the base of
the organ, but there are an adductor muscle (E, 1) inserted on the
base of the apical claw (d), and two flexor muscles (2, 3) inserted
on the small ventral sclerite (c). The adductor (1) arises, not in
the basal segment of the appendage as in Platyphylax (D, 1a, 1b),
but in the proximal part of the last body segment. One of the flexor
muscles (E, 2) is a broad fan of fibers arising on the dorsal wall of the
basal segment, and thus suggests an identity with muscle 2 of Platy-
phylax (D) arising on the basal plate of the appendage. The other
flexor muscle is a long bundle of fibers (E, 3) arising on the interseg-
mental fold (e) between the eighth and ninth abdominal segments
along with the fibers of the ventral longitudinal muscles of the ninth
segment (4), but separating from the latter posteriorly to enter the
appendage. This muscle apparently has no representative in Platy-
bhylax (D).

The relation between the two types of appendages described above
is obscure. The only suggestion that can be made is that the basal
plate of the Platyphylax appendage (fig. 41 D, b) corresponds with
the basal segment of the Hydropsyche appendage (FE, b) and that the
intermediate plate of the former (c) is represented by the small ven-
tral sclerite (c) of the latter. The apical claw (d) is evidently the
same in both. Still more difficult is it to find possible homologies
between either of these two types of trichopterous appendages and
the terminal appendages of the sialid larvae, Chauliodes and Corydalus
(figs. 35 D, E, 36 B, E), or the abdominal legs of lepidopterous larvae
(figs. 37, 42 C). The basal plate or basal segment (0) in the tri-

7

100 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

chopterous appendage, however, may be the limb basis of a more
typical abdominal appendage, while the apical claw (d) and the asso-
ciated middle plate (c) may possibly represent the stylus. Judging
from the structure there is little probability that the trichopterous
claw is homologous with the claws of Corydalus or Chauliodes, or
with the crochets of lepidopterous larvae. The lack of similarity
in the larval abdominal appendages of Trichoptera and Lepidoptera
is somewhat surprising, considering the many other structural like-
nesses between these two orders.

Many trichopterous larvae are provided with a group of slender
processes protractile through the anus (fig. 41 F, ip), which arise
from the intestinal wall. In the species of Hydropsyche figured there
are from four to six of these processes. The structures are hollow,
thin-walled tubules containing no tracheae, and are therefore usually
termed “blood gills”; but the idea of their respiratory nature is
based on their structure and on the fact that they can be entirely
exposed by protraction through the anus, for no one apparently
has made any physiological experiments on their function. According
to Branch (1922) the organs arise as diverticula of the intestinal
wall produced posteriorly from the six folds of the pre-rectal part of
the proctodeum, and each is provided with a three-branched muscle
taking its origin on the intersegmental membrane between segments
VIII and IX of the abdomen, and extending through the lumen to
the tip of the processes. When retracted the processes lie in the rec-
tum with only their extremities exposed in the anal aperture.. Pro-
traction evidently is accomplished by internal pressure resulting from
a contraction of the abdominal walls.

Neuroptera——tThe pygidial appendages in the larvae of the stalid
genera Chauliodes and Corydalus, as already noted, are long, thick
structures projecting posteriorly and ventrally from the terminal

segment of the abdomen (fig. 35 D, E, Pp, fig. 36 B, E). Each organ ~

consists of a large basis (fig. 36 E, LB), supporting laterally a flex-
ible, tapering process representing the stylus (Sty), and bearing dis-
tally a short, cylindrical lobe (V's) armed with two strong claws (d).
The appendage has a striking resemblance to the typical abdominal
leg of a caterpillar, except that the latter has no representative of the
stylus. The distal lobe, which clearly is serially homologous with
the gill-bearing tubercles of the preceding appendages in Corydalus
(fig. 36 B, C, D, Vs), is remarkably like the planta of the caterpillar’s
leg (fig. 37 D, Vs). It is retracted by a strong muscle (fig. 36 F, rvs)
arising on the dorsal wall of the terminal body segment, and inserted
on the distal wall of the lobe at the posterior or upper end of the long

no. 6 INSECT ABDOMEN—SNODGRASS IOI

bases of the claws. The claws themselves are in every way suggestive
of the crochets of the caterpillar’s foot.

Notwithstanding the general structural resemblance between the
terminal appendages of Chauliodes or Corydalus and the abdominal
legs of caterpillars, it can scarcely be supposed that the foot struc-
ture in either case has been derived immediately from that of the
other, since the two-clawed condition would be a highly specialized
one in the Lepidoptera. All that may be claimed is that the funda-
mental structure of the larval abdominal limbs is the same in both
the Neuroptera and the Lepidoptera. The neuropterous appendage
is the more primitive in that it retains the stylus, which has the form of
a segmented appendage in Sialis (fig. 35 A, C). We may assume that
the gill-bearing tubercles of the Corydalus larva have been formed
secondarily from the foot lobes, or retractile vesicles, as a better adap-
tation to aquatic life in this genus, and that the vesicles have been
lost from all the appendages in Sialis, and from all but the terminal
appendages in Chauliodes.

Lepidoptera.—The so-called anal legs, or postpedes, of caterpillars
are so similar to the legs of the preceding abdominal segments as
scarcely to need a separate description. The musculature of the two
sets of appendages, however, differs in some respects. The plantar
lobe of each anal leg is retracted by a large dorsal muscle (fig. 42 A,
B, C, rvsd) and a small ventral muscle (rvsv), both arising from the
intersegmental fold (f) before the ninth abdominal segment. The
lateral muscles of the leg are reduced to a few fibers (D, b) lying
external to the large dorsal retractor of the planta. Between the bases
of the legs there is a sheet of transverse ventral muscles (B, tv),
which appear to belong to the wall of the last body segment rather than
to the appendages.

The great development of the dorsal retractor muscles of the anal
legs, the reduction of the lateral muscles, and the presence of the
ventral retractors of the plantae are all features correlated with the
function of the postpedes in the caterpillar, which usually have a
stronger independent forward movement than do the legs of the pre-
ceding segments.

The large terminal segment of lepidopterous larvae appears to be
a compound segment composed of the tenth abdominal somite, with
its appendages, the pygopods, and the reduced eleventh segment, bear-
ing the anus, but lacking cerci. Figures of the embryo of Pieris rapac
given by Eastham (1930) show clearly a well-developed tenth abdom-
inal segment bearing the last pair of appendages, and beyond it a
large terminal lobe, containing the anus, which is evidently the

102 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

eleventh somite. In the caterpillar (fig. 11) the eleventh somite is
apparently represented in the terminal segment by the four postpedal
lobes surrounding the anus (C, D), which may be supposed to be the
epiproct (Eppt), the paraprocts (Papt), and a hypoproct (Hypt).
There is, however, no evidence of a dual composition of the terminal
segment of the caterpillar furnished by the musculature of this seg-
ment. Both the dorsal and the ventral internal longitudinal muscles

Fic. 42—Musculature of the terminal segment and pygopods of a noctuid
caterpillar.

A, dorsal muscles and leg muscles seen from below in terminal segment cut
open along mid-ventral line and spread out. B, ventral musculature of same
segment, seen from above. C, posterior end of abdomen, showing origins of leg
muscles. D, basal rim of left leg (L) and group of small muscles (b) lying
beneath dorsal retractors of planta (rvsd).

An, anus; b, group of small antero-lateral lez muscles; de, external dorsal
muscles; di, internal dorsal muscles; Eppt, epiproct; f, intersegmental fold
between ninth and tenth segments; Hypt, hypoproct; L, basal rim of leg; J,
lateral muscles; Papt, paraproct; rvsd, dorsal retractor of planta; rvsv, ventral
retractor of planta; tv, transverse ventral muscles; vi, internal ventral muscles;
Vs, planta.

(fig. 42 A, B, di, vi) extend continuously from the anterior inter-
segmental fold (f) to the epiproct and the paraprocts. Two large
sets of internal dorsal fibers (A, de) arise in the notches between
the epiproct and paraprocts and are inserted on the dorsal plate of the
segment. The eleventh segment, therefore, if represented here at all,
is reduced to the circumanal lobes; and the fibers of the longitudinal
muscles of the tenth and eleventh segments have become continuous.

No. 6 INSECT ABDOMEN—SNODGRASS 103

Continuity of muscle fibers is of frequent occurrence wherever the
intersegmental connections are lost, or where segmental boundaries
are obliterated.

Chalastogastrous Hymenoptera.—The larvae of the sawflies and
horntails are provided with terminal appendicular organs of several
varieties, all borne by the last abdominal segment, some arising from
the dorsum, others from the venter.

On the dorsum there is in some forms, as in Pteronidea ribesii
(fig. 43 A), a pair of small, immovable lateral processes (a) arising
from the end of the last segment above the anal opening (An). These
processes have been regarded as rudimentary cerci (Crampton, 1919),
but they are more evidently mere cuticular outgrowths, as claimed
by Middleton (1921), analogous with the urogomphi of coleopterous
larvae (fig. 44 C, ug), since they have none of the characteristics of

Fic. 43.—Terminal appendicular structures of larvae of chalastogastrous
Hymenoptera.

A, Pteronidea ribesti (Tenthredinidae). B, Cimbex americana (Cimbicidae).
C, Cephaleis sp. (Pamphiliidae).

a, paired processes of dorsum of terminal segment; An, anus; b, median proc-
ess of dorsum of terminal segment; Pp, postpedes, or pygopods; rvs, retractor
muscle of plantar lobe; ls, planta.

true cerci, and, according to Middleton, take no part in the formation
of the lateral, cercus-like appendages of adult Tenthredinidae. In
certain other chalastogastrous larvae a median process, or postcornu
(fig. 43 C, b), is borne on the end of the tenth abdominal segment.
This process varies characteristically in shape and size in different
families, as tabulated by Middleton (1921). It occurs in larvae that
bore into stems or that live in silk-spun tents or within the hollow of
curled leaves.

The ventral appendages of the pygidial segment of chalastogastrous
larvae likewise vary in form in different groups. Those of species
that live in the open closely resemble the anal legs of lepidopterous
larvae. In Pteronidea ribesii, for example (fig. 43 A), a ventral pro-
jection of the tenth segment beneath the anus, bearing two small
terminal lobes (ls), is clearly the homologue of a pair of appendages

104 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

such as occur on the more anterior segments of the abdomen. The
free lobes (V's) are evidently comparable with the plantae of the ab-
dominal legs of caterpillars (fig. 37 D, Vs), though the retractor mus-
cles of the lobes in the sawfly larva take their origins from a point
on the side of the segment just anterior to the cleft of the anal open-
ing (fig. 43 A, rvs). In other forms, as in Cimbex americana (B),
the ventral appendages (Pp) consist apparently of the plantar lobes
only, which arise directly from the flattened venter of the pygidial
segment. Again, as in the Cephidae and Pamphiliidae, typical “ pro-
legs ” are replaced on the tenth segment by slender jointed appendages
(C, Pp); but these “ arthrostyli,” as they have been called (Cramp-
ton, 1919), are evidently alternative forms of the anal “ prolegs,” or
postpedes, since they arise at approximately the same points as do
the typical postpedes in other forms (A, Pp), and do not occur con-
jointly with the latter. They are not, however, provided with muscles,

so far as the writer could discover; but each is penetrated by a large

nerve, and bears sense organs on the distal segment having the form
of minute disks, in addition to setae on the proximal and middle
segments. The jointed form of the terminal appendage is, therefore,
evidently an adaptation to a sensory function instead of a locomotory
one.

Crampton (1919) suggests that the jointed appendages, or “ arthro-
styli,’ of chalastogastrous larvae do not represent the “ prolegs ”
directly, but that each has the relation to the latter of a stylus, that
is, it is an appendicular part of the true appendage. Middleton (1921),
on the other hand, thinks that the jointed appendages are direct repre-
sentatives of the unjointed postpedes, because the two organs have
identical positions on the tenth segment, and do not occur together.
He would attribute the difference in form to the different habits of the
larvae, since those species having typical, fleshy postpedes feed in the
open on leaves and grasp the edges of the latter with the terminal
appendages, while those having slender, jointed postpedes, bore into
the stems of plants, or live in the protection of web nests or curled
leaves. To the writer it appears most probable that the two forms
of appendages are identical organs, and that the jointed variety is a
secondary modification of the typical postpedes for a sensory function,
the jointing being a mere subdivision of the appendage and not a true
segmentation.

The observation made by Middleton (1921) that the postpedes
of the larva become the cercus-like appendages of the adult sawfly has
been discussed in the introductory part of this section, and need be
given no further attention here, except to point out its importance,

y

;
.

a

No. 6 INSECT ABDOMEN—SNODGRASS 105

if true, in establishing an identity between the terminal appendages
of larval insects and the appendicular processes of the tenth segment
in certain adult holometabolous insects.

Coleoptera.—In many families of the Coleoptera the abdomen of
the larva is provided with a pair of appendicular processes arising
from the dorsum of the ninth segment. These structures have been
variously termed styli, cerci, pseudocerci, corniculi, and urogomphi.

The abdomen of the larva of Dytiscus ends with a transverse pos-
terior surface of the narrow eighth segment (fig. 44 A), in the upper
part of which the last pair of spiracles open through a median, vertical
slit (VIIISp). From the membrane below the spiracular area of the
eighth segment there arise laterally two slender, tapering processes
(ug?) fringed with long hairs. Between the bases of these organs

Vill NIISp an

r An
/ el ect? \G i

Fic. 44.—Terminal appendages of coleopterous larvae.

A, posterior end of abdomen of Dytiscus circumcinctus larva. B, end of ab-
domen of a silphid larva, Thanatophilus sp., with exserted anal lobes (from
Kemner, 1918). C, end of abdomen of carabid larva, Oodes helopioides, with
exserted anal lobes (from Kemner, 1018).

An, anus; anl, anal lobes; ug, urogomph.

is a small, median plate (JX7?), which appears to be a remnant of
the ninth abdominal tergum. Ventral to it is the anal opening (An).
According to Speyer (1922, Korschelt, 1924), each of the terminal
appendages of the larva of Dytiscus marginalis is provided with three
muscles, two inserted dorsally on its base, and one ventrally, all of
which arise on the tergum of the eighth segment.

The morphology of the larval appendages of Dytiscus is difficult
to determine. The apparent position of the organs on the rudimentary
ninth segment makes it doubtful that they are true cerci, and the dorsal
origins of their muscles is not in accord with the musculature of
styli. It is claimed by Blunck (1918, Korschelt, 1924) that in all
Coleoptera the first two primitive somites of the abdomen are united
in the first definitive segment, and that, therefore, the segment bear-
ing the terminal appendages is really the tenth. In this case we might

106 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

regard the terminal appendages of the Dytiscus larva as the cerci.
However, the evidence presented by Blunck of the fusion of the first
and second segments of the abdomen is not generally accepted by
students of Coleoptera, and is not convincing, while the fact that the
segment preceding the appendicular processes bears the eighth pair of
abdominal spiracles would ordinarily be taken as conclusive evidence
that this latter segment is the true eighth somite of the abdomen,
especially since these respiratory apertures, as described by Blunck,
have the structure typical of lateral abdominal spiracles.

For the present, therefore, the nature of the terminal appendages of
the Dytiscus larva must be left in doubt, but if the organs do not
belong to the series of lateral, stylus-like appendages, fully repre-
sented in Coptotomus and Gyrinidae, it is possible that they are struc-
tures homologous with those dorsal processes often developed on the
ninth abdominal segment of other coleopterous larvae, which appear
to be mere cuticular outgrowths, though they may become movable
at their bases. The muscles inserted on the processes in Dytiscus are
evidently not specific muscles of the appendages, but groups of seg-
mental or intersegmental fibers that, by reason of their attachments
at the bases of the processes, serve to move the latter.

The abdomen of most other coleopterous larvae consists of 10
distinct segments (fig. 44 B, C). The ninth segment is usually well
developed and frequently supports a pair of processes (B, C, ug)
arising from its dorsum. These processes are termed urogomphi by
Boving, since from their segmental position it is clear that they can
not be cerci. In some species the urogomphi are fixed outgrowths
of the posterior end of the ninth tergum (C) ; in others they arise
from the membrane behind the ninth tergal plate (B), and are then
flexible at their bases. They vary much in size and shape from
short, spine-like points to long, thick processes or multiarticulate fila-
ments, and they are sometimes distinctly jointed. The urogomphi
appear to be equivalent structures in all cases, and are probably but
cuticular outgrowths of the ninth abdominal tergum, which in
certain species become movable by a membranization of the tergal
wall at their bases. The mobile appendages of the Dytiscus larva,
therefore, are perhaps urogomphi of the rudimentary ninth segment,
secondarily movable by some of the intersegmental muscles normally
extending between the eighth and ninth segments.

The tenth abdominal segment in many coleopterous larvae is a
short, often tubular structure (fig. 44B, C, X), at the extremity
of which is a small retractile and eversible membranous pad sur-
rounding the anus, known as the “ pygopodium ” (C, XJ), which may

No. 6 INSECT ABDOMEN—SNODGRASS 107

be produced into two or more soft lobes (anl). It is claimed by Kem-
ner (1918), from embryological evidence, that the so-called pygo-
podium is the rudimentary eleventh abdominal somite. The eversible
pygopodial lobes, however, in no way represent appendicular organs,
Kemner asserts, nor are they evaginations of the wall of the rectum,
as they have been supposed to be; they are merely productions of the
cercumanal area on which the longitudinal muscles from the tenth
to the eleventh segment are attached. If the organs in question, there-
fore, are not of an appendicular nature, the term “ pygopodia ”’ should
not be given to them, since it is convenient to apply this name spe-
cifically to the true appendages of the tenth, or pygidial, segment.
The presence of a distinct though rudimentary eleventh segment
in larvae of Coleoptera is of interest because of the general suppres-
sion of this segment in holometabolous insects.

TERMINAL LOBES OF THE PARAPROCTS

In a few of the lower Pterygota an appendicular lobe is borne by
each of the paraprocts. These processes have been termed “ para-
processi”’ by Crampton (1920). Examples of paraproct lobes occur
in the Odonata and in the tridactylid Orthoptera.

The paraproct processes of Odonata occur in adult Anisoptera in
the form of small, seta-bearing lobes projecting posteriorly from the
ends of the paraprocts. Corresponding lobes are not present in the
larvae of this group of dragonflies, in which the paraprocts, together
with the elongate epiproct, form the valves that close the anal open-
ing (fig. 12A, B). In the larvae of Zygoptera, however, paraproct
lobes are highly developed as the large, flat, tracheated plates that form
the lateral caudal gills (C, paptl). The median gill (cf) is a similar
lobe of the epiproct (Eppt), and is evidently comparable with the
median caudal filament of Thysanura (fig. 7 A, B, C, cf).

In the Orthoptera paraproct lobes are well developed in the Tri-
dactylidae, where they have the form of long processes resembling
the cerci (fig. 45 A, B, paptl). In some species of Ripipteryx those of
the male are incurved at the ends and are said to be used as claspers
during copulation. In Ellipes (fig. 45 A, B) each process is borne
on a membranous area at the end of the short paraproct (Papf).
Crampton (1918, 1921) has given considerable significance to the
“ paraprocessi ”’ of the Tridactylidae, which he regards as homologous
with the styli of the preceding segments, and as representatives of
the exopodites of crustacean appendages (fig. 45 C, D, expd). Most
other writers, including Walker (1919), regard the processes as sec-

7

108 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

ondary lobes of the paraprocts. It is highly unlikely that the struc-
tures in question are styli, since, as already shown, the paraprocts are
not limb bases. Crampton’s view regarding the relation of the para-
proct process to crustacean exopodites will be discussed in the fol-
lowing Section of this paper.

Milly "IX xX. Ger

Fic. 45.—Terminal appendages of a tridactylid insect and an isopod crustacean.

A, posterior part of abdomen of Ellipes minuta (tridactylid orthopteron).
B, the same, dorsal view. C, end of abdomen of Porcellio (isopod crustacean),
ventral view. D, the same, dorsal view.

An, anus; Cer, cercus; Endpd, endopodite; Eppt, epiproct; Expd, exopodite;
la, lamina sub-analis; Papt, paraproct; paptl, lobe of paraproct; Prtpd, protop-
odite; sa, lamina supra-analis; Tel, telsqgn (fused with tergum of twentieth
segment).

MORPHOLOGY OF THE ABDOMINAL APPENDAGES

There is no need at present to offer proof of the serial identity
of the appendicular organs of the insect abdomen with the segmental
appendages of the thoracic and cephalic regions of the body. Embry-
ologists have amply demonstrated the continuity of appendage rudi-
ments on the entire series of primitive somites, at least 20 in all, leav-
ing only a prostomial lobe and a periproctial region devoid of true
limb structures. To determine the homologies of the parts of the
abdominal appendages with the parts of the better developed append-

No. 6 INSECT ABDOMEN—SNODGRASS 109

ages on the anterior regions of the body, or with those of theoretically
more generalized appendages, is quite another matter.

Theoretical Structure of Arthropod Appendages.—At the outset of
an attempt to study the morphology of the rudimentary abdominal
appendages of insects a difficulty is encountered arising from the
lack of uniform opinion as to the structure of a generalized but
fully segmented arthropod limb. Particularly is it necessary in a
study of rudimentary appendages to know the structure of the basal
part of a primitive appendage. In the thorax of modern insects
the functional base of a leg is the coxa, and yet, it seems almost
certain that at an earlier stage the true basis of the limb must have
included the subcoxal region now forming the so-called pleuron and in
some cases a lateral part of the definitive sternal plate of the support-
ing body segment. Then, the further question arises as to whether the
subcoxa was once a free segment of the appendage, or whether it has
been evolved secondarily by a differentiation of the primitive limb
basis into subcoxal and coxal parts. Differences of opinion on such
questions as these have led immediately to different interpretations of
the basal parts of the abdominal appendages, and consequently to
different views concerning the nature of the distal parts.

A simple condition of the limb base occurs throughout the Arach-
nida, which is well shown in any one of the legs of a phalangid
(fig. 46 A). It is to be seen here that the leg is supported on a large
basal segment (LB) that occupies the lateral wall of a segmental area
of the body, and that it turns slightly forward and rearward on a
dorsoventral axis (a-b) extending from the tergum above to the
sternum below. To this large basal segment the telopodite is articu-
lated by a dicondylic hinge on a horizontal axis (f-g). The first seg-
ment of the telopodite is a trochanter (z77r).

In the majority of the Crustacea the proximal region of the leg has
the same structure as in the Arachnida, there being a single basal
segment, the coxopodite (fig. 46 B, C, LB), implanted directly in the
ventro-lateral wall of the supporting body segment, and often articu-
lated dorsally (a) with the tergal plate (B, 7). The basis, however,
is not prolonged ventrally as in the phalangid (A), and is inclined to
be cylindrical (C). The telopodite articulates with the basis by its
proximal segment (the first trochanter, or basipodite, B, C, rT7r)
on a horizontal, dicondylic hinge (f-g) having the same type of struc-
ture as that in the arachnid (A).

If, now, we look at a typical thoracic leg of an insect (fig. 46D),
it is seen that the proximal joint in the appendage corresponding struc-
turally and functionally with the joint between the basis and the telop-

Ilo SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

odite in the phalangid or crustacean leg (A, B, C, f-g) is that be-
tween the coxa and the trochanter (D, f-g). The coxa (Cx), how-
ever, is separated from the tergum (7), and often from the sternum
(Stn), by a sclerotization, known as the “ pleuron ” (Scx), occupying
the position in the lateral wall of the supporting body segment that is
occupied by the limb basis in the phalangid and crustacean (A, B, C,
LB). In other words, the basal region of the appendage in the case
of the insect leg (D, LB) is composed of two parts (Scx, Cx) cor-

/
1Tr+2Tr

Fic. 46.—Basal parts of arthropod legs.

A, third right leg of a phalangid, Liobunum, showing limb basis (coxopodite)
elongate ventrally (LB), turning antero-posteriorly on dorsoventral axis (a-b),
with telopodite movable dorsoventrally on horizontal axis (f-g).

B, last right ambulatory leg of a crustacean, Anaspides tasmaniae, with small
basis, or coxopodite (LB), on which the telopodite turns dorsoventrally on hori-
zontal axis (f-g) of first trochanter, or basipodite (177).

C, leg of an isopod crustacean, Porcellio, with elongate basis, or coxopodite
(E 2 } Ae which the telopodite is articulated by a horizontal hinge (f-g) as in
A and B.

D, mesothorax and middle leg of a young acridid nymph, Melanoplus, showing
region of limb basis (LB), between the horizontal baso-telopodite hinge (f-g)
and the tergum (7), subdivided into subcoxa (Sc) and coxa (Cr).

a, dorsal articulation of limb basis with body; a-b, axis of articulation of
limb basis with body; c, dorsal subcoxo-coxal articulation; Cx, coxa; Cxrpd,
coxopodite (limb basis) ; f-g, axis of baso-telopodite hinge; Fm, femur (merop-
odite) ; h-i, axis of hinge between first and second trochanters; 7-k, axis of
trochantero-femoral hinge; LB, limb basis (coxopodite); Scr, subcoxa; Stn,
sternum; 7, tergum; 17,, first trochanter (basipodite) ; 27r, second trochanter
(ischiopodite).

responding with the single plate or segment (coxopodite) forming the
basis of the arachnid and crustacean appendages illustrated (A, B, C).
A condition similar to that in the insect thorax is found, however,
in the thoracic region of the decapod crustaceans, where a large
pleural plate occurs on each side of the body intervening between
the coxae and the tergum. This plate is clearly a composite structure,
the segmental areas of which are evidently derived from the sub-
coxal parts of the leg bases, because they support the gills of the

no. 6 INSECT ABDOMEN—SNODGRASS III

branchial chamber. Subcoxal sclerites occupying the lateral walls
of the body segments are again found in the Chilopoda (fig. 27 A,
Scx), and in the last leg-bearing segment they are here united with
the coxa to form a single basal plate (LB) of the appendage, occupy-
ing the pleural region of the segment.

From comparative studies such as those just cited, the writer has
come to regard the pleural subcoxa, or sclerotic area of the lateral
body wall supporting the free part of an appendage, as the proximal
part of the primary limb basis secondarily separated, wherever it
occurs, from the distal part of the basis, which becomes the movable
coxa, or functional base of the limb. On the other hand, many
students of Arthropoda regard the subcoxa as a primitive limb seg-
ment, which has become suppressed or fused with the coxa wherever
traces of it are not to be found in modern forms. The entire absence
of a subcoxa in all Arachnida and in the majority of Crustacea, how-
ever, is against this view; and the lack of uniformity in the subcoxo-
coxal musculature, when a subcoxa is present, suggests that the sub-
coxo-coxal joint is a recent division of the limb basis that has occurred
principally in the Chilopoda and the Hexapoda. If the subcoxa is a
secondary formation, then it must be assumed that the subcoxo-coxal
muscles likewise are secondary, and that, as the subcoxa became
differentiated from the coxa, most of the primitive basal muscles of
the appendage were transferred to the coxa.

The subcoxa in its more primitive condition is best seen in the
Chilopoda and in the thoracic segments of apterygote Hexapoda. It
here consists of a circular fold or slight elevation of the body wall
supporting the leg, containing one or more small sclerites, particularly
in the region above the coxa. The large “pleura” of the thoracic
region of pterygote insects, or the pleural plates of the branchial cham-
bers of decapod crustaceans, undoubtedly represent highly specialized
developments of the subcoxae, adapting the latter to uses quite inde-
pendent of any function connected with the legs. The subcoxal plates
in the thorax of holometabolous insect larvae, however, are relatively
small and are closely associated with the coxae (figs. 3 B, C, 41 A,
Scx;). The region of the subcoxa surrounds the base of the coxa,
but its ventral arc is reduced to a fold, which generally in the thorax
of adult insects unites with the sternum. The sclerotized area of
the subcoxa may be broken up into several small sclerites; in the
thorax of pterygote insects there is typically a large supracoxal plate
known as the ‘‘ pleuron.”

In a former paper the writer (1928) has given reasons for believing
that the body of a gnathal appendage represents the basis of a leg. It

I1I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

is similarly divided secondarily into two parts (cardo and stipes),
though it is not clear that the two parts exactly correspond with the
subcoxa and coxa of a thoracic leg. The palpi, however, are thus
seen to be the telopodites of the head appendages. If the interpretation
concerning the primary segmentation of a gnathal or a thoracic
appendage into basis and telopodite is now carried to the appendages
of the abdomen, the basal plates or basal lobes of the latter become
the true limb bases, and the telopodites should be freely movable ap-
pendicular processes of the bases.

In general, then, it appears that the arthropod limb is divided by a
joint near its base into a proximal part, the primary limb basis, and
into a distal part, or telopodite. The baso-telopodite joint is the coxo-
trochanteral joint of a fully segmented limb, which is the joint be-
tween the coxopodite and the basipodite in terms usually employed
by carcinologists. The use of the term “basipodite” by Borner
(1904, 1921) to designate the subcoxo-coxal base of the limb creates
a duplication in nomenclature that is likely to be confusing. The
movement of the telopodite on the basis is typically in a vertical
plane, produced by levator and depressor muscles arising in the basis
(hee 25205) O78).

On the assumption that the basal mechanism of all the limbs is
fundamentally the same in all groups of arthropods, we can imagine
a simple primitive condition in the arthropodan ancestors in which
the entire series of appendages had a uniform line of flexure near
the body, along which the distal parts of the limbs, or telopodites,
were movable in a vertical plane on their bases. The bases, on the
other hand, turned forward and rearward on the axes of their attach-
ments on the body. Wherever the basis is differentiated into a coxa
and a subcoxa, the primitive basal movement of the appendage on the
body is lost, but is replaced by a vertical axis of promotion and remo-
tion between the subcoxa and the coxa, as the latter becomes sec-
ondarily the functional base of the limb. Finally, if the limb becomes
rudimentary and loses its basal musculature, the basis might become
transformed to a simple immovable lobe or plate of the wall of
the supporting body segment, with the telopodite reduced to an ap-
pendicular process movable by muscles arising in the basis.

Other Theories on the Morphology of the Abdominal Appen-
dages.—The principal problem encountered in a study of the abdom-
inal appendages of insects is that of determining the homologies of the
parts of the appendages with those of a generalized limb. The basis,
the stylus, the gonapophysis, the eversible vesicle, each raise ques-
tions as to its nature and derivation.

’

no. 6 INSECT ABDOMEN—SNODGRASS 113

Discussions on the morphology of the abdominal appendages of
insects, and speculations on their possible homologies with the limbs

of Crustacea have continued for half a century without leading to

definite conclusions. They began at. least with Wood-Mason (1879),
who, in a paper on the origin of insects, interpreted the stylus-bear-
ing plates of the abdomen of Machilis as the protopodites of primi-
tively biramous appendages, of which the endopodites are represented
by the gonapophyses, and the exopodites by the styli. The eversible
vesicles, Wood-Mason suggested, may be homologues of nephridia,
those of the eighth and ninth segments being converted into the
genital ducts.

Diversities of opinion soon followed the publication of a more
widely read paper by Haase (1889) on the morphology of the seg-
mental appendages, containing not only a clear exposition of the
appendicular nature of the stylus-bearing plates in the abdomen of
Thysanura, but also a demonstration of the triple origin of the de-
finitive abdominal sterna of insects in general from the union of the
rudimentary embryonic limbs with the median sternal area in each
segment. The styli, Haase claimed, are not the leg rudiments, but
secondary structures of the nature of hairs, which have been con-
verted into locomotory organs from sensory organs. The eversible
vesicles he believed function as blood gills, but are to be traced back
in all cases to coxal glands.

Wheeler (1893) and earlier students of the embryology of insects
regarded the gonapophyses of the genital segments as the direct
representatives of the appendages of these segments. Considering
the late development of the gonapophyses, however, and their in-
variable median position on the body of the insect, Heymons (1896a)
contended that the genital processes are secondary integumental out-
growths having no relation to the appendages, and that the latter
are preserved in the styli and cerci. Heymons’ heterodox opinion
brought a severe criticism from Verhoeff (1896), who defended the
limb nature of the gonapophyses as an established fact, and main-
tained that the identity of the abdominal styli with the thoracic styli
of Machilis could not be disputed, and that therefore both are merely
appendicular processes of the legs. In reply to Verhoeff, Heymons
(1896b) emphasized his former statements in evidence of his view
concerning the nature of the appendicular parts of the abdomen,
pointing out that, during embryonic development of the Orthoptera,
the abdominal appendages disappear, and the gonapophyses are later
formed quite independent of the limbs. He argued that if the
gonapophyses are the limbs, intermediate stages should be found some-

y

114 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

where between gonapophyses and legs. The styli, Heymons reasserted,
are direct derivatives of the abdominal appendages, and occur in
primitive forms such as Campodea and Japyx in which gonapophyses
are lacking. From the abdominal limb rudiments of the embryo, Hey-
mons showed, are produced not only the typical styli, but also the
cerci, and the lateral gills of the larvae of Ephemerida and Svalis.

Since Heymons’ views are based on embryological studies they
deserve more attention than purely theoretical considerations. In his
studies of the development of the appendages of Periplaneta, Ectobia,
and Mantis, Heymons (1896a) observes that each limb rudiment
of the ninth segment is early marked by a circular constriction, which
divides the appendage into a broader proximal part and a slenderer
distal part. The proximal part flattens out and finally is incorporated
in the definitive sternal plate of the segment, while the distal part
becomes longer and slenderer and develops directly into the stylus.
The gonapophyses, on the other hand, Heymons claims, are secondary
outgrowths of the sterna in the Orthoptera. Concerning them he
says: “In Gryllus there can be no doubt of the nature of the
gonapophyses. Abdominal extremities are present in the embryo on
the eighth and ninth segments, but they later degenerate, and in old
embryos as well as in young larvae leave not the slightest rudiments.
It is only later, in older larvae, that the gonapophyses appear, and
they are therefore undoubtedly to be regarded as secondary integu-
mental outgrowths.” Thus, according to Heymons, the stylus and
not the gonapophysis is the representative of the telopodite in an
abdominal appendage.

It must be conceded that the facts of embryonic development do not
necessarily recapitulate phylogenetic evolution, since we can never
be sure that the early stage of an organ reproduces the primitive form
of that organ, and this must be particularly true of a rudimentary
structure. Thus, if the telopodite of a limb bearing a basal exite
process has long been lost, the limb rudiment in the embryo might be
supposed to consist of the limb basis and the accessory process, and
to lack the telopodite element entirely. Hence, while Heymons’
evidence of the nature of the styli is highly suggestive that the styli
are the rudiments of the telopodites, it does not demonstrate the
point. On the other hand, Heymons’ line of reasoning concerning
the gonapophyses makes it seem almost certain that the genital proc-
esses are not the telopodites of the gonopods, but the facts of develop-
ment can scarcely be taken as evidence that the gonapophyses do
not belong to the genital appendages. It is amply proven in the Thy-
sanura that the gonapophyses are processes of the gonopods, and in

/

| No. 6 INSECT ABDOMEN—SNODGRASS It5

ee

the Orthoptera they are undoubtedly outgrowths of the parts of the
definitive sterna derived from the bases of these appendages.
Verhoeff (1903), retaining his former views on the homologies

between the abdominal and thoracic appendages, but going more into

detail, proposed, on theoretical grounds, that the stylus-bearing plates
of the insect abdomen represent the coxae, which in the abdomen
he distinguished as “ coxites,” or in the genital segments as “ gono-
coxites.” According to Verhoeff’s theory, the telopodites are lost
from all the abdominal appendages except those of the eighth and
ninth segments, where they become the gonapophyses; the styli are
secondary lateral outgrowths of the coxae, equivalent to the coxal
spurs of the thoracic legs of Machilis; and the eversible vesicles are
median coxal structures comparable with the coxal glands of Dip-
lopoda.

Borner, though at first taking Heymons’ view of the nature of the
gonapophyses, later (1904) agrees with Verhoeff that the genital
processes represent the telopodites of the abdominal appendages, pre-
served only on the gonopods. Regarding the supporting plates, how-
ever, Borner differs from Verhoeff in that he identifies them as
“basipodites,” meaning by this term that each plate is the equivalent
of the coxa and subcoxa of a thoracic leg. (The same idea concerning
the nature of the basal plates is followed in the present paper, but
Borner’s term “ basipodite” is replaced with “limb basis” to avoid
confusion with the more usual application of the other word to the
first trochanter. )

Silvestri (1903, 1905) regards the basal segment of an arthropod
limb as being in all cases a subcoxa (including the so-called coxop-
odite of Crustacea), and he would divide the appendage into a basis
(subcoxa) and a telopodite at the subcoxo-coxal joint. Applying this
interpretation of the basal structure of the limb to the abdominal
appendages of insects, Silvestri (1905) identifies the stylus-bearing
plates of the Thysanura with the subcoxae. The styli he regards as
the rudiments of the telopodites, with their bases representing the
coxae. Silvestri, therefore, admits no homology between the abdominal
styli and those of the thorax in Machilis; the leg styli he claims are
secondary outgrowths of the coxal integument. Verhoeff (1903) had
figured a coxal muscle attached to the leg stylus of Machilis, but
this supposed muscle Silvestri shows does not exist—an observation
in accord with statements by earlier as well as by subsequent writers,
and one easily verified.

The most interesting feature in Silvestri’s interpretation of the
morphology of the abdominal limbs is his proposal that the genital

8

‘

116 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

apophyses are serially homologous with the eversible sacs of the pre-
genital segments. Silvestri argues as follows: On the first abdominal
segment of Projapyx and Anajapyx there are borne on each sub-
coxa (stylus-bearing plate) a typical stylus and, mesad of it, a
cylindrical or conical process. In Machilis and Nicoletia each sub-
coxa of the first segment has a retractile vesicle, and in Campodea
only a cylindrical process. The following six segments of Machilidae
bear on the subcoxae both styli and vesicles, but on the next two, the
genital segments, each subcoxa has a stylus and, in some genera, mesad
to it a genital process. Thus Silvestri contends that the gonapophysis
is evidently an eversible sac permanently everted. Muscles he ob-
serves are attached to each appendage, though he does not point out
that those of the gonapophysis are inserted on the base of the process,
while those of the vesicles traverse the organ to be inserted in its
extremity.

The abdominal appendages of insects have not lacked attention
from students of arthropod phylogeny, because their several parts
make up a composite limb pattern that may be supposed to conform
with the biramous structure of crustacean appendages, and thus
indicate either that insects are closely related to the Crustacea, or that
the primitive arthropod limb was a biramous structure. Writers who
espouse the idea of a crustacean ancestry for insects, following
Wood-Mason (1879), interpret the stylus as the exopodite, and find
the homologue of the endopodite in the gonapophyses of the genital
appendages. The theory must assume that the endopodites have been
suppressed on the pregenital segments, since the eversible vesicles
are eliminated as possible telopodite homologues by the fact that
they sometimes occur in duplicate.

The theoretical possibilities of aligning the appendages of insects
with those of Crustacea have been exhaustively searched by Cramp-
ton. In a study of the terminal appendages of the tridactylid orthop-
teron, Ellipes, Crampton (1921) adduces evidence that he takes
to be conclusive of the biramous nature of insect appendages. The
dorsal pair of terminal appendicular processes in the Tridactylidae
are undoubtedly the cerci (fig. 45 A, B, Cer) ; the ventral pair (paptl)
are the lobes of the paraprocts (‘‘ paraprocessi”). After removing
the end of the abdomen and spreading the parts out from below until
they lie in one plane, Crampton makes a comparison of the tridactylid
appendages in this position with the uropods of an isopod crustacean
in the normal position (fig. 45 C, D), and arrives at the conclusion
that the cerci of the former correspond with the endopodites of the
latter, and that the paraproct lobes of Ellipes represent the exopodites

no. 6 INSECT ABDOMEN—SNODGRASS 117

of the isopod. The paraprocts themselves Crampton regards as the
protopodites (i. e., combined coxopodites and basipodites) of the
terminal appendages. The interpretation of the paraproct lobes as
exopodites is consistent with other evidence that the styli are exop-
odites, for the paraproct lobes fall in line with the styli, or would
do so if styli were present in the tridactylids, and the abdominal styli
are serially continuous with the thoracic styli of Machilis, which
appear to have an exopodite status, and therefore suggest that the
entire series of styliform organs are exopodites.

There are several weak places in the above line of reasoning. In
the first place, the writer fails entirely to get Crampton’s view from
the comparison between the tridactylid and the isopod, since, with
the terminal parts of both in the normal condition (fig: 45 B, D),
the cercus of the insect (Cer) surely has the position of the exopodite
(Expd) of the crustacean uropod, while the paraproct lobe (papil)
corresponds in position with the small endopodite (Endpd) borne
by the basal plate of the uropod (C). In the second place, a more
careful examination of details shows that the cercus (B, Cer) has
no anatomical relation with the paraproct (Papt), being situated
dorsad of the latter in a position corresponding with that of the base
of the crustacean uropod (D). Furthermore, as has already been
shown, there is no evidence whatever to support the idea that the

-paraprocts of insects are parts of the appendages. Their musculature

indicates that they are mere lateroventral, subanal lobes of the eleventh
abdominal sternum. The cerci have no muscles arising in the para-
procts. Finally, the embryological evidence concerning the nature
of the cerci appears to show definitely that the cerci are the entire
appendages of the eleventh segment, and that their bases, if present
at all, are retained in a basal ring of each organ. Hence, until some
radically new information comes to light concerning the cerci, there
is no question of exopodite or endopodite connected with them. Our
present information is to the effect simply that the cerci are the appen-
dages of the eleventh abdominal segment.

The lobes of the paraprocts, whether the ‘“ paraprocessi”’ of the
tridactylids, the small lobate ends of the paraprocts of the Anisoptera,
or the lateral gill plates of the Zygoptera, have no validated claim
to an appendicular origin. They must, then, for the present be re-
garded as secondary outgrowths of the subanal lobes of the sternum
of the eleventh abdominal segment, comparable to the various median
outgrowths of the supra-anal plate of the same segment.

The most nearly convincing evidence of the biramous nature of
insect appendages is, admittedly, the presence of styliform processes

‘ ’

118 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

on the mesothoracic and metathoracic legs of Machilis closely resem-
bling the styli of the abdomen. Without this coincidence, or if
Machilis and its coxal spurs had not survived to modern times, it is
doubtful if entomologists would ever have thought of regarding the
abdominal styli or the cerci as other than direct rudiments of legs.

Conclusions —The abdominal appendages of insects are rudimen-
tary limbs. Each consists of a basis and usually one or more distal
appendicular parts, including a stylus, an eversible or retractile vesicle,
and a gonapophysis. From the facts known of the comparative struc-
ture of the abdominal appendages, and from theoretical considera-
tions we may draw the following tentative conclusions relative to the
homologies of the parts of the appendages, but it must be admitted
that the evidence at hand is not sufficiently definite to establish any
particular view concerning them.

The limb bases of the abdominal appendages are the lobes or plates
of the walls of the body segments that support the appendicular proc-
esses. They are usually well developed in larval insects, but in most
adults they are partially or wholly united or blended with the sternum,
or in the male genital segment fused also with the tergum; in the
eleventh segment they are reduced to small basal rings of the cerci,
or entirely obliterated. Generally there is no distinction between coxal
and subcoxal regions in the limb bases of the abdomen. In position
on the body the abdominal limb bases usually fall in line with the
subcoxae or pleural areas of the thorax. But since there is no
apparent reason for the development of large subcoxal plates on
the abdominal segments, such as those of the pterygote thorax,
it is not to be supposed that the limb bases of the abdomen repre-
sent the subcoxae alone. It is more probable that their principal parts
are derived from the flattened coxae, or that the structures in most
cases may represent primitive limb bases undifferentiated into coxae
and subcoxae. In the caterpillars and sawfly larvae, however, the
principal segment of each abdominal leg appears to be the coxa, which
is quite distinct from a subcoxal lobe of the wall of the body segment
to which it is attached. The abdominal limb bases are seldom movable
on the body, since they commonly lack muscles corresponding with
the basal muscles of the gnathal and thoracic appendages arising on the
body wall. Exceptions to this rule are found in the male genital seg-
ment, where the bases of the gonopods are occasionally provided with
muscles and are independently movable ; but in such cases it is to be
suspected, at least, that the muscles are secondary developments.

The styli, or other homologous appendicular processes of the ab-
domen, are of wide occurrence in insects, and serve a variety of

¥

no. 6 INSECT ABDOMEN—SNODGRASS 11g

functions, to which they are adapted by structural modifications. The
styliform type is not necessarily the primitive form of the organs.
The abdominal styli may be serially homologous with the thoracic
coxal styli of Machilis, or the latter may be merely large spurs re-
sembling the abdominal styli, from which they differ in lacking mus-
cles. The abdominal styli are individually movable on the limb bases
by muscles arising in the latter. If they are not the rudimentary main
shafts of the abdominal limbs, they are exite lobes of the coxae analo-
gous with the epipodites of crustacean appendages. They are not
comparable with the exopodites of Crustacea, because the exopodite
branch of a typical biramous limb is an exite of the first trochanter,
or basipodite.

There are many reasons for regarding the abdominal styli or their
derivatives in insects as the rudimentary telopodites of the abdominal
appendages. The styli seldom lose their muscles, except when they
are immovably united with the bases; in some insects they take an
active part in locomotion; they may be jointed in a manner sug-
gesting at least a true segmentation, and in the larvae of Sialis they
have intrinsic muscles in their basal segments. The styli of the gono-
pods in male pterygote insects, especially in the holometabolous orders,
are commonly modified to serve as grasping or clasping organs during
copulation. The styli are the most generally persistent of the distal
parts of the abdominal appendages. If it were not for their likeness
in apterygote and orthopteroid insects to the coxal spurs of Machilis,
it seems doubtful if the abdominal styli would ever have been regarded
as anything else than the rudimentary telopodites of the abdominal
appendages, represented in a similar form by the cerci on the eleventh
segment.

The vesicles of the abdominal appendages of Apterygota, the gill
tubercles of the larva of Corydalus, and the plantar lobes of the larval
abdominal legs of Lepidoptera and chalastogastrous Hymenoptera are
all organs of a similar and unique type of structure. They are essen-
tially exserted or invaginated lobes of the coxal areas of the limb bases
lying mesad of the bases of the styli, and are retractile by muscies
inserted within their distal parts. In the case of the Apterygota the
muscles arise in the limb bases; in the others they arise from the
lateral walls of the body. We might, with Verhoeff, regard these
sacs as derivatives of coxal glands, since integumentary glands some-
times take the form of eversible and retractile pouches. The coxal
vesicles, however, serve a variety of purposes, and they are more
simply explained as endite lobes of the coxae, which in some cases
have become normally invaginated. They may thus be likened to the

¥

I20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

endite lobes of the gnathal appendages, and, as the latter, they some-
times occur in duplicate on each appendage.

The gonapophyses are hollow, median outgrowths of the bases of
the gonopods, developed in both male and female insects during
postembryonic stages. They are movable by muscles arising in the
limb bases, or in areas of the genital segments derived from the latter,
except where they are operated by the segmental muscles of the body.
Theoretically the gonapophyses may be supposed to be either the
telopodites (endopodites) of the gonopods, or endite lobes of the
bases of these appendages. There are several objections to the first |
view. The gonapophyses, for example, are never truly segmented, and
never have a form suggestive of a leg structure; they occur only
on the appendages of the genital segments, unless the cylindrical :
processes of the first abdominal segment in certain Dicellura are |
homologous structures; and finally, they serve only in a particular |
capacity in connection with oviposition and copulation, except in |
insects where they are secondarily adapted to form a stinging organ.
These facts, together with the invariable median position of the
gonapophyses on the gonopod bases, suggest that the gonapophyses
are basal endites of the gonopods, movable by muscles arising in the
basis, as such endites usually are, and specially adapted to the repro-
ductive functions. The genital apophyses might then be regarded as_ |
serially homologous with the eversible or retractile vesicles of the
pregenital appendages and certain larval pygopods, in which case
Silvestri’s idea that they represent “ permanently everted eversible
vesicles” is better stated in the reverse, namely, that the eversible
vesicles are inverted gonapophyses. However, the vesicles do not
occur at points on the gonopod bases corresponding with the origins _
of the gonapophyses, and the musculature of the two sets of organs
is characteristically different. It is perhaps possible that the gona-
pophyses are subcoxal endites, and the vesicles coxal endites.

If none of the appendicular processes of the abdominal limps can _
be satisfactorily identified with the telopodite of a primitive appendage,
we must conclude that the abdominal telopodites have been lost from
all but the terminal segment, where they form the cerci, and that
the various persisting appendicular structures are accessory processes
of the limb bases. Otherwise, we must choose between the gona-
pophyses and the styli as possible representatives of the telopodites.
Of the two, the styli certainly present better credentials, considering
their occurrence on many segments of the abdomen as contrasted
with the segmental localization of the gonapophyses, their structural
and functional versatility by comparison with the limitations of the

a

a a i ny

no. 6 INSECT ABDOMEN—SNODGRASS 121

genital apophyses in form and use, and their leglike relations to the
limb bases as opposed to the median, proximal origins of the gona-
pophyses. All these points qualify the abdominal styli for true limb
rudiments, and give the genital processes the status of basendites
specially developed for reproductive purposes. The pretensions of the
abdominal styli to telopodite origin are opposed only by their similar-
ity to the thoracic styli of Machilis; but there is nothing to show
that these leg structures are not mere coxal spurs resembling in form
but fundamentally unlike the musculated appendages of the abdomen,
the styliform shape of which is but one of their many structural
adaptations.

A discussion of the phylogeny of insects, or particularly of the
possible origin of insects from any other group of existing arthropods
is beyond the intended scope of the present paper. A recent work by
Tillyard (1930) on the evolution of the Insecta, though somewhat
partisan in favor of myriapodan descent, gives many reasons for
believing that insects are not directly related to the Crustacea. And
yet, the weight of evidence, whether put forth by claimants of a
myriapodan or a crustacean ancestry for insects, seems to depend
largely on minimizing or disqualifying the evidence on the other side.
However, if we were to give equal weight to arguments on both sides
of the question, the insects would be cut off from all ancestral ties,
and thereby deprived of a respectable pedigree—unless they are able
to take care of themselves through all the unknown ages of time
before they are first known to us as fully-winged hexapods in the
Carboniferous deposits. To the writer it appears that all the principal
arthropod groups must represent independent lines of descent from
some remote ancestral forms embodying the potentialities of a spider,
a crab, a centipede, or an insect. It has recently been emphasized
by Clark (1930) that the chronic inability of the evolutionary theory
in its usual form to explain the lack of intermediates between the
major groups of animals constitutes a real weakness of the theory,
which calls for a new concept of the method by which distinct types
of organisms have been produced. The condition to which Clark
refers is well exemplified within the Arthropoda, where connective
forms between the classes are unknown. Moreover, it is impossible
to construct imaginary arthropods that will fill the blanks, as, for
example, the three-cornered gap between the crustaceans, the myria-
pods, and the insects. Considering that embryos develop before
our eyes by ways that are still inscrutable, it takes a strong faith in
established ideas to believe that organic evolution has proceeded
entirely by the means we have furnished for its guidance.

122

ABBREVIATIONS USED ON

a-a, dorso-pleural groove.

a-b, axis of articulation of limb base
on body.

ab, abductor muscle.

Ac, antecosta.

acs, antecostal suture.

ad, adductor muscle.

Aed, aedeagus.

An, anus,

Ap, apodeme.

Apd, segmental appendage.

Apt, sternal apotome.

Bnd, basendite.

Brn, branchia (gill).

brncls, branchial muscles.

Bspd, basipodite (first trochanter ).

c-d, axis of subcoxo-coxal joint,
Cer, cercus (uropod).

cf, caudal filament.

Col, collophore.

Con, nerve connective.

ct, coxo-trochanteral joint.

Cx, coxa.
Cxpd, coxopodite.

D, dorsum.

d, dorsal muscles.

del, lateral external dorsal muscles.

dem, median external dorsal muscles.

dil, lateral internal dorsal muscles.

dim, median internal dorsal muscles.

dl, dilator muscle.

dm, median dorsal muscles.

DMcl, longitudinal dorsal muscles,

DTra, dorsal longitudinal tracheal
trunk.

Endpd, endopodite.
Eppt, epiproct.
Expd, exopodite.

f-g, axis of baso-telopodite joint.
Fm, femur (meropodite).

ft, femoro-tibial joint.

Fur, furcula,

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VoL. 85

THE FIGURES

gmcls, muscles of gonapophysis.
Gng, ganglion.

Gon, gonapophysis.

Gp, gonopod.

GSeg, genital segment.

Hpn, hypandrium.
Ht, heart.
Flypt, hypoproct,

/, tergal promotor muscle of appen-
dage.

ip, intestinal process.

/sg, intersegmental fold.

/st, intersternite.

J, tergal remotor muscle of appendage.
k, interfurcal sternal ridge.

L, leg.

!, lateral body muscles,

Ja, lamina subanalis.

LB, limb basis.

le, external lateral muscles.

li, internal lateral muscles.

lStn, last pregenital sternum.

IT, last pregenital tergum.

LTra, lateral “longitudinal
trunk,

tracheal

Mb, intersegmental membrane (con-
junctiva).

mb, membrane.

Mcl, longitudinal muscles.

mcls, muscles.

mn, manubrium.

NIG, neural groove.

O, levator muscle of first trochanter.
Ovd, oviduct.

p, paratergal muscle.
Papt, paraproct.

paptl, lobe of paraproct.
patg, paratergite.

Pc, precosta.

NO. 6

Pen, penis.

Ph, phragma.

Pl, pleuron.

pl, pleurite.

Pm, paramere (male gonapophysis of

ninth segment).

meta-postnotum (precosta of

first abdominal segment).

pnb, pons basalis.

Pp, pygopod.

Prpt, periproct.

Prtpd, protopodite (united coxopodite
and basipodite).

Ptar, praetarsus (dactylopodite).

PNs,

O, depressor muscle of first trochanter,

rvs, retractor muscles of vesicle of leg
base, or of its derivatives.

rvsd, dorsal retractor of planta.

rusv, ventral retractor of planta.

S, definitive sternum (usually includ-
ing limb bases).

SAsz, metasternal apophysis.

sa, lamina supra-analis.

Sex, subcoxa.

Sp, spiracle,

s-p, sterno-pleural muscle.

smcls, muscles of stylus, or of its de-
rivatives.

Stn, primary segmental sternum.

INSECT ABDOMEN—SNODGRASS 123

Sty, stylus of leg base, or its deriva-
tives.

T, tergum.

Tar, tarsus (propodite).

Tb, tibia (carpopodite).

td, dorsal transverse muscles,

Tel, telson.

Tlpd, telopodite.

t-p, tergo-pleural muscle.

Tr, trochanter.

1Tr, first trochanter (basipodite).
2Ty, second trochanter (ischiopodite).
Tra, trachea.

t-s, tergo-sternal muscle,

tv, ventral transverse muscles.

ug, urogomph.

I’, venter.

v, ventral muscles.

vel, lateral external ventral muscles.

vem, median external ventral muscles.

vil, lateral internal ventral muscles.

zim, median internal ventral muscles.

V1, valvula.

Vif, valvifer.

V Mel, ventral longitudinal muscles.

V NC, ventral nerve cord.

I’s, vesicle of limb base, or its deriva-
tives.

1-3, thoracic segments.
I-XIT, abdominal segments.

REFERENCES

Ayers, H, (1884). On the development of Oecanthus niveus and its parasite
Teleas. Mem. Boston Soc. Nat. Hist. 3: 225-281, pls. 18-25.

Backuorr, P. (1910). Die Entwicklung des Copulationsapparates von Agrion.
Zeitschr. wiss. Zool. 95: 647-706, 29 text figs., pl. 21.

Baurr, A. (1910). Die Muskulatur von Dytiscus marginalis. Zeitschr. wiss.
Zool. 95: 594-646, 19 figs. Also in Korschelt (1924), Chapter 12.

Becker, E. (1926). Zur phylogenetischen Entwicklung des Skeletts und der
Muskulatur der Ateloceraten (Tracheaten). I. Das Tergalskelett und die
Dorsalmuskulatur von Chilopoden. Revue Zool. Russ, 6:2 3-67, 4 pls. (In

Russian and German. )

BerKELEY, ALFREDA A. (1928). The musculature of Pandalus danae Stimson.
Trans. R, Canadian Institute 16: Pt. 2: 181-231, pls. 1-8.

Bertese, A. (1906). Sopra una anomalia negli organi sessuali esterni femminei
di Locusta viridissima L. Redia 3: 305-314, 7 figs.

124 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

(1909). Gli insetti, Vol. 1. Milan.

(1910). Monografia dei Myrientomata. Redia 6: 1-182, pls. 1-17.

Betts, ANNIE D. (1923). Practical Bee Anatomy. 76 pp., 12 pls. The Apis
Club.

Biunck, H. (1918). Die Entwicklung des Dytiscus marginallis L. vom Ei bis
zur Imago. 2 Teil. Die Metamorphose (der Habitus der Larve). Zeitschr.
wiss. Zool. 147: 1-120, 57 figs.

Borner, C. (1904). Zur Klarung der Beingliederung der Ateloceraten. Zool.
Ans. 27: 226-243, 5 figs.

(1909). Die Tracheenkiemen der Ephemeriden. Zool. Anz, 33: 806-

823, 4 figs. i

(1921). Die Gliedmassen der Arthropoden. In Lang’s Handbuch der
wirbellosen Tiere 4: 649-604, 57 figs. Jena,

Bortimer, L. J. (1926). Notes on some Lepidoptera from eastern Texas. Journ.
Agr. Research 33: 797-810, 3 figs.

Bovine, A. G. (1914). On the abdominal structure of certain beetle larvae of
the campodeiform type. A study of the relation between the structure of the
integument and the muscles. Proc. Ent. Soc. Washington 16: 55-61, 8 figs.

BrancH, Hazer E. (1922). A contribution to the knowledge of the internal
anatomy of Trichoptera. Ann. Ent. Soc, America 15: 256-275, pls. 14-18.

Breen, R. S. (1903). The changes which occur in the muscles of a beetle.
Thymalus marginicollis Chevr., during metamorphosis. Bull. Mus. Comp.
Zool. 40: 317-382, pls. 1-7.

Brocuer, F. (1922). Etude experimentale sur le foncionnement du vaisseau dor-
sal et sur la circulation du sang chez les insectes. 5. La Periplaneta orien-
talis. Ann. Soc. Entom. France 91: 156-164, 1 fig.

Busck, A., and Hernricu, C. (1922). Life history of Ethmia machelhosiella
Busck. (Lep.). Proc. Ent. Soc. Washington 24: 1-9, pls. 1-3.

Catvert, P. P. (1911). Studies on Costa Rican Odonata. I. The larva of Cora.
Ent. News 22: 49-64, pl. 3.

(1915). Studies on Costa Rican Odonata. VII. The waterfall dwellers:
The internal organs of Thaumatoneura larva and the respiration and rectal
tracheation of Zygopterous larvae in general. Ent, News 26: 385-395, 435-
447, pls. 15-17.

CAMPBELL, F. L. (1929). The detection and estimation of insect chitin; and the
irrelation of “ chitinization” to hardness and pigmentation of the cuticula of
the American cockroach, Periplaneta americana L. Ann. Ent. Soc. America
22: 401-426.

Cartet, G. (1890). Sur les muscles de l’abdomen de I’abeille. Ann. Sci. Nat.
Zool., 7 Ser., 9:2 1-17, pl. 1.

CuHotopKowsky, N. (1891). Die Embryonalentwicklung von Phyllodromia
(Blatta) germanica. Mém. Acad. Sci. St. Petersbourg, 7 sér., 38, No. 5,
120 pp., 6 pls.

Crark, A. H. (1930). The New Evolution. Baltimore.

CRAIGHEAD, F. C. (1916). The determination of the abdominal and thoracic
areas of the cerambycid larvae as based on a study of the muscles. Proc.
Ent. Soc, Washington 18: 129-142, pls. 7-9.

Crampton, G. C. (1918). A phylogenetic study of the terminal abdominal struc-
tures and genitalia of male Apterygota, ephemerids, Odonata, Plecoptera,
Neuroptera, Orthoptera, and their allies. Bull. Brooklyn Ent. Soc. 13:
49-68, pls. 2-7.

/
|

no. 6 INSECT ABDOMEN—SNODGRASS 125

(1919). The genitalia and terminal abdominal structures of males, and
the terminal structures of the larvae of “ chalastogastrous ”” Hymenoptera.
Proc. Ent. Soc. Washington 21: 129-151, pls. 9-12.

(1920). Remarks on the basic plan of the terminal abdominal struc-

tures of the males of winged insects. Canad. Ent. 52: 178-183, pl. 6.

(1921). A comparison of the terminal abdominal structures of insects
and Crustacea. Ent. News 32: 257-264, pl. 5.

Drees, L. (1928). Das larvale Muskelsystem und die Entwicklung der imagi-
nalen Flugmuskulatur von Psychoda alternata Say. Zeitschr. Morph.
Okol. Tiere 11: 182-228, 14 text figs.

Du Porte, E. M. (1920). The muscular system of Gryllus assimilis Fabr. Ann,
Ent. Soc. America 13: 16-52, pls. 1-7.

DtrKen, B. (1907). Die Tracheenkiemenmuskulatur der Ephemeriden unter
Berticksichtigung der Morphologie des Insektenfliigels. Zeitschr. wiss. Zool.
87: 435-550, 29 text figs., pls. 24-26.

EastHam, L. E. S. (19030). The embryology of Pieris rapae.—Organogeny.
Phil. Trans. Roy. Soc. London, ser. B, 219: 1-50, pls. 1-9.

Ferris, G. F., and CHEMBERLIN, J. C. (1928). On the use of the word “ chitin-
ized.” Ent. News 39: 212-215.

FLorENCE, LAurA (1921). The hog louse, Haematopinus suis Linné: its biology,
anatomy, and histology. Cornell Univ. Agric. Exp. Sta., Memoir 51: 641-
725, pls. 58-66,

Forses, W. T. M. (1914). A structural study of the caterpillars: III, the
somatic muscles. Ann. Ent. Soc. America 7: 109-124, pls. 17-25.

(1916). On certain caterpillar homologies. Journ. New York Ent, Soc.
24: 137-142, I fig.

Forp, Norma (1923). A comparative study of the abdominal musculature of
orthopteroid insects. Trans. Roy. Canadian Institute 14: 207-310, pls. 7-23.

Fracker, S. B. (1915). The classification of lepidopterous larvae. /llinois Biol.
Monographs 2, No. 1, 169 pp., 10 pls.

GrassI, B. (1887). Anatomia comparata dei Tisanuri e considerazioni generali
sull’ organizzazione degli insetti. Atti de’ Lincei, Mem. Cl. Sci. Fis., Ser. 4,
4: 543-606, 5 pls.

Haass, E. (1889). Die Abdominalanhange der Insekten mit Berticksichtigung
der Myriopoden. Morphol. Jahrb. 15: 331-435, pls. 14, 15.

Hansen, H. J. (1930). Studies on Arthropoda. III. On the comparative mor-
phology of the appendages in the Arthropoda, 376 pp., 16 pls. Copenhagen.

Hewitt, C. G. (1914). The house fly, Musca domestica Linn., 382 pp., 104 figs.
Cambridge Zoological Series.

Heymons, R. (1892). Die Entwicklung der weiblichen Geschlechtsorgane von
Phyllodromia (Blatta) germanica L. Zeitschr. wiss. Zool. 53: 434-536,
pls. 18-20.

(1895). Die Segmentirung des Insektenk6rpers. Abhandl. Akad. Wiss.

Berlin 1895, 39 pp., 1 pl.

(1896). Uber die Lebensweise und Entwicklung von Ephemera vul-

gata L. Sitzsungs-Ber. Gesell. Naturf. Fr. Berlin 1896: 82-06.

(1896a). Zur Morphologie der Abdominalanhange bei den Insekten.

Morph. Jahrb. 24: 178-204, pl. 1.

(1896b). Uber die abdominalen K6rperanhange der Insekten. Biol.

Centralb. 16: 854-864.

126 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

(1897). Entwicklungsgeschichtliche Untersuchungen an Lepisma sac-

charina. Zeitschr. wiss. Zool. 62: 583-631, 2 pls.

- (1899). Der morphologische Bau des Insektenabdomens. Zool. Cen-

tralb. 63 537-556.

(1901). Die Entwicklungsgeschichte der Scolopender. Zoologica. Orig-
Abhandl. Gesamtg. Zool. 33, 244 pp., 8 pls.

——— (1904). Die Hinterleibsanhange der Libellen und ihrer Larven. Ann. K.
Naturhist. Hofmuseums, Wien 19: 21-58, pl. I.

Hopkins, A. D. (1909). The genus Dendroctonus. U. S. D. A., Bur. Ent.,
Tech. Ser., No. 17, Part 1, 164 pp., 8 pls.

Imms, A. D. (1908). On the larval and pupal stages of Anopheles maculipennis, |
Meigen. Parasitology 1: 103-133, pls. 9, 10.

JeziorskI, Lucie (1918). Der Thorax yon Dixippus morosus (Carausius),

Zeitschr. wiss. Zool. 117: 727-815, pls. 17-19.

Kemner, N. A. (1918). Vergleichende Studien iiber das Analsegment und das |
Pygopodium einiger Koleopterenlarven, 104 pp., 44 figs. Uppsala. |

KorscHELt, E. (1924). Bearbeitung einheimischer Tiere. Der Gelbrand Dytis-
cus marginalis L., Vols. 1 and 2. Leipzig. 3

KraFka, J., Jr. (1924). Morphology of the prolegs of trichopterous larvae.
Ann, Ent. Soc. Amer. 17: 70-103, pls. 16, 17.

KUNCKEL D’ Hercurats, J. (1875). Recherches sur l’organisation et le dével-
oppement des Volucelles, 208 pp., 26 pls. Paris.

KunIkE, G. (1626). Chitin und Chitinseide. Kunstseide, 8, 182, 183.

LaNnkKEsTER, E. R. (1909). A treatise on zoology, pt. 7, fasc. 3 (Calman, W. T.).
London.

Luspock, J. (1859). On the arrangement of the cutaneous muscles of the larva
of Pygaera bucephala. Trans. Linn, Soc. London 22: 173-191, pls. 34, 35.

(1873). Monograph of the collembola and thysanura, 276 pp., 78 pls.
London.

Lyonet, P. (1762). Traité anatomique de la chenille qui ronge le bois de saule,
616 pp., 18 pls.

Miatz, L. C., and Hammonp, A. R. (1900). The structure and life history of
the harlequin fly (Chironomus), 196 pp., 129 figs. Oxford.

MippLeton, W. (1921). Some notes on the terminal abdominal structures of
saw-flies. Proc. Ent. Soc. Washington 23: 139-144, pl. 11.

Morison, G. D. (1927). The muscles of the adult honey-bee (Apis mellifera
L.). Quart. Journ. Micr. Sct. 71: 395-463.

Netson, J. A. (1915). The embryology of the honey-bee, 282 pp., 95 text figs..
6 pls. Princeton Univ. Press.

(1924). Morphology of the honey-bee larva. Journ. Agric. Research
28: 1167-1213, 5 text figs., 8 pls.

Paterson, NELLIE (1930). Bionomics and morphology of the early stages of
Paraphaedon tumidulus Germ. (Chrysomelidae). Proc. Zool. Soc. Lond.
1930, pt. III : 627-676.

Prett, H. (1913). Das Chitinskelett yon Eosentomon. Zoologica. Orig.-
Abhandl. Gesamtg. Zool, 25, 58 pp., 6 pls.

QuteL, G. (1915). Anatomische Untersuchungen an Collembola. Zeitschr. wiss.
Zool. 113: 113-164, pls. 6, 7.

Rrester, A, (1931). Muskulatur von Copilia Dana. Zool. Jahrb., Anat. 53: 317-
404, 50 figs.

No. 6 INSECT ABDOMEN—SNODGRASS 127

SAMTLEBEN, B. (1929). Anatomie und Histologie der Abdominal- und Thorax-
muskulatur von Stechmiickenlarven. Zeitschr. wiss. Zool. 134: 180-260,
18 figs.

Scumipt, E. (1915). Vergleichende Morphologie des 2. und 3. Abdominalseg-
ments bei mannlichen Libellen. Zool, Jahrb., Anat. 39: 87-200, 25 text
figs., pls. Q-IT.

Scumipt, W. (1915). Die Muskulatur von Astacus fluviatilis (Potamobius
astacus L.). Ein Beitrag zur Morphologie der Decapoden. Zeitschr. wiss.
Zool. 11413: 165-251, 26 figs.

Sepewick, A. (1909). A student’s text-book on Zoology, Vol. 3. London.

Sirvestri, F. (19003). Classis Diplopoda, Vol. 1, Anatome, 272 pp., 345 figs.,
4 pls. Portici.

(1905). Thysanura. Zool. Jahrb., 3, Suppl. 6: 773-806, pls. 38-44.
Snoperass, R. E. (1924). Anatomy and metamorphosis of the apple maggot,
Rhagoletis pomonella Walsh. Journ. Agric. Research 28: 1-36, 6 pls.
(1925). Anatomy and physiology of the honey-bee, 327 pp., 108 figs.

New York.

(1928). Morphology and evolution of the insect head and its appen-
dages. Smithsonian Misc. Coll., 81, No. 3, 158 pp., 57 figs.

Speyer, W. (1922). Die Muskulatur der Larve von Dytiscus marginalis L.
Ein Beitrag zur Morphologie des Insektenkorpers. Zeitschr. wiss. Zool.
119: 423-402, 21 figs., pl. 7. Also in Korschelt (1924), Chapter 13.

STEINER, L. F. (1929). Homologies of tracheal branches in the nymph of Anax
junius based on their correlation with the muscles they supply. Ann. Ent.
Soc. America 22: 297-308, 1 pl.

Straus-DUrcKHEIM, H. (1828). Considérationes générales sur l’anatomie com-
parée des animaux articules, auxquelles on a joint l’anatomie descriptive du
Melolontha vulgaris, 434 pp., 10 pls. Paris, Strasbourg, Bruxelles.

Tiecs, O. W. (1922). Researches on the insect metamorphosis. Trans. and
Proc. Roy. Soc. S. Australia 46% 319-527, pls. 15-30.

Tittyarp, R. J. (1917). The biology of Dragonflies, 306 pp., 188 figs.
Cambridge.

(1930). The evolution of the Class Insecta. Roy. Soc. Tasmania, Papers

. and Proc. 1930: 809 pp., 19 figs.

Utmer, G. (1903). Uber die Metamorphose der Trichopteren. Abh. Gebiete
Naturwiss. naturwiss. Verein Hamburg 18, 154 pp., 4 pls.

VerHoerF, K. W. (1896). Zur Morphologie der Segmentanhange bei Insecten
und Myriopoden. Zool. Anz. 19: 378-383, 385-388.

(1903). Zur vergleichenden Morphologie der Coxalorgane und Geni-
talanhange der Tracheaten. Zool. Anz. 26: 60-77, 15 figs.

Voss, F. (1905). Uber den Thorax von Gryllus domesticus. II. Die Muskula-
tur. Zeitschr. wiss. Zool. 78: 355-521, 14 figs.

Watker, E. M. (1922). The terminal structures of orthopteroid insects: a
phylogenetic study. Part II. Ann. Ent. Soc. America 15: 1-76, 11 pls.
WaALLENGREN, H. (1914). Die Atmung bei den Arthropoden. II. Die Mechanik
der Atembewegunden bei Aeschnalarven. Lunds Univ. Arrskrift, N. F.

10: 1-24, I pl.

128 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

Weser, H. (1928). Skelett, Muskulatur und Darm der schwarzen Blattlaus
Aphis fabae Scop. Zoologica. Orig.-Abhanal. Gesamtg. Zool. 28, 120 pp.,
12 text figs., 12 pls.

Wuepon, A. D. (1919). The comparative morphology and possible adaptations
of the abdomen in the Odonata. Trans. Amer. Ent. Soc. 44% 373-437,
pls. 21-20.

WueEeE Ler, W. M. (1893). A contribution to insect embryology. Journ. Morph.
8: 1-160, pls. 1-6.

Woop-Mason, J. (1879). Morphological notes bearing on the origin of insects.
Trans. Ent. Soc. London 1879: 145-167, 9 figs.

}

a a

or
a

SMITHSONIAN MISCELLANEOUS COLLECTIONS
VOLUME 85, NUMBER 7

EFFECTIVENESS IN NATURE OF THE SO-CALLED
PROTECTIVE ADAPTATIONS IN THE ANIMAL
KINGDOM, CHIEFLY AS ILLUSTRATED BY
THE FOOD HABITS OF NEARCTIC BIRDS

BY
W. L. McATEE
Bureau of Biological Survey
5.

Wis epartment of Agriculture

(PUBLICATION 3125)

GITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
MARGH 15, 1932

SMITHSONIAN MISCELLANEOUS COLLECTIONS

ViGl..85; INO:\7

EFFECTIVENESS IN NATURE OF THE SO-CALLED PROTECTIVE
ADAPTATIONS IN THE ANIMAL KINGDOM, CHIEFLY
AS ILLUSTRATED BY THE FOOD HABITS OF
NEARCTIC BIRDS

By W. L. McATEE
ERRATA

Page 56. Inthe table Identifications of Lepidoptera, the middle column is a relic
from a set of calculations of the percentages of identifications among
all insects. The appended figures are to be substituted as represent-
ing the percentages of identifications among all Lepidoptera. In
explanation of the third column in this table, it may be said that it
differs from others given further on in the work by omission of
figures for families not represented among the food identifications.

Percentage of
identifications
among all
Lepidoptera
2.6992
.0270
.0270
8060
.OO19
.2055
.O108
.0270
0108
.0108
-4700
.0324
.O412
.0378
6383
6.1015
.0216
3078
-3570
1508
8438
11.4458
25.3300
.0487

0054
GLOL
.0054
-O108
2218
0112
7248
68.5670
1.2279
3.5376

Page 86. The figure 1 before the decimal in the entry for the family Diopsidae
should be deleted.

Pages 102-105. Insert the word “aquatic ” after the word “all” in the heading
for the middle column on each of these pages, with the exception of
that at the bottom of page 105.

mePrPECTIVENESS IN NATURE OF -THE SO-CALLED
BROLteGhivi ADAPTATIONS IN THE ANIMAL
RINGDOM, CHIEFLY AS ILLUSTRATED BY THE
BOOD HABITS OF NEARCTIC BIRDS

By W. L. McATEE
BUREAU OF BIOLOGICAL SURVEY,
U. S. DEPARTMENT OF AGRICULTURE

CONTENTS chee
BMG TIGL ONY mhrae arse citer enn are a csiske ie DEE weed Shiale mn sey Ea em edvelee eS
WMBrotective adaptations ............ecececececcccccccccececuceeevuccuces 4
Banimalsneaten by, WNearctie birds. ccs. acces cee vw sce es setae oe ccn ae es 6
Datarcited andehow. Obtained... 2 ccs sees. oes tee seems sessed cones 6
IG@entinicatlons Ol vanimeals 100d es. ec Oleg. fee os eo eee cise esse cer at mance 8
Protozoa (one-celled animals)................. cece ee cece cease 9
I GRIeT deve OMONEES ) a atrsc wise eS aes aks ¥ <8 dime bate au she wae neg 10
Coelenterata (hydras, jellyfishes, sea-anemones)................. II
Platyhelminthes (flatworms, flukes).............. 0000 c cece eeees 13
Nemathelminthes (threadworms, roundworms).................. 13
aliochelminthes u(rotikers)) se s.40s «ese cece sions sea cece nee ae 13
Molluscoida (corallines, lampshells)................ 000 cece eee 13
Echinodermata (sea-cucumbers, sea urchins, starfishes)........... 14
PTV ia Cea tcta am (GVO LTIIS)) 8 repehey er eyete eye arenes sierra onsies eevee Sines Se ra ce Anesth Rhy 15
Arthropoda’ (jointed “animals)..22 2.2.22. 0.0<s..02 ste. st eeene 16
@rustaceas(crabs) shrimps, SOWwbUeS )\: g0. 42.2 ce cee bone ees i
Myriapoda (thousandlegs, centipedes)....................5- 22
SEC tame CITISECESD) pameittepiere etek niekcieds aitieys chs. dle'csurare 2% 0 o.shae. ote 2
EN Puc hate (AWA C1 eSS)ettISECES!) penton n oss sce cratic: <. osia ce aieuccelanci eBs 2
Odonata (dragonflies, damselflies)...............0000005 28
Nomeathiaa (may ilies) et wtatetaee suse a eae cess eect eee 20
Plecoptera i(stonefiies)) 20.220... cae ce ce aecdssesss 30
NSO teraen CLEtIILeS) geese wyatt ea yeneete stn daeserd,= a/steptonk 31
Orthopteroidea (embracing the following 5 groups)...... 32
WW ermaptena | Canwiesi ae aie ses ner ae © icacere esau eceeie oe BD
Cheleutoptera (walleinestichks i. 22.siideawesiuws teen cae: 33
Saltatoria (grasshoppers, locusts, crickets).............. 34
| Paleo pretramn( HOACIES )maremitts suis acts octete otis sees eaaieea 38
Dictyopteral (meantids)): sees s «sees wes sees eee ease oe 39
| (Worrodentiay GpsocidS)) en. cec feeds ce es see cease. e eu: 30
| Matlophaga (bilinear Hee) 22 cc diese rent debe encatoten ot 40
Sipnoneaptencaem heas\mawera teas ese ceciaeete seen see eta: 40
MDVSANOp Grae AiTipS) | etandhoes vnc Wie seated siete gas rw eee d 4o
Rhynchota (bugs, cicadas, leafhoppers, scale insects)..... 4!

Neuropteroidea (dobsonflies, snakeflies, scorpionflies, ant-
Lions Caddishies i) a emisscee ees asic ead aage cence de cece 49
Lepidoptera (moths, butterflies)................0..00 00 52

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, NO. 7.

2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

PAGE

Coleoptera’ (beetles): Seeion- tis 2m: steceitercls.- eee 63

Meeaptera (scorpiontlies)) 2: asics sors 21 aleve elem ellen 84

Diptera, (ilies) joceicaastpeisoed sto nie ta cee eee 84

Hymenoptera’ (ants, ‘bees, wasps)!:.----.- ce «ee 88

Arachnida (scorpions, spiders, ticks; ‘etC.)).. 2.0.2.2... 9. seeee 07

Mollusca (snails, slags, mussels, limpets).....2 00-9. ----- eee IOI
Chordata (lancelets, tunicates, vertebrates).............-+- scare 110

Pisces: (fishes )* 5.02). ae staternctteten net aos ok on ee aie te eee 112

Amphibia (salamanders, toads, frogs) .............+-sssesme 120

Reptilia (turtles, lizards, snakes).........:.2.. ..-++s09ame 123

Aves iCbitds))! (fs tadten 2nGoe wee rte ictus 6 Soe oni eee 126

Mammalia’ (mamimeals)! 2.55 2c clele es = aleiele 2 oye eves ee 131

IDF orci SREP ASAeUOM cH oddone dnadinenoccubsono aos moggDoowonedooococs 135
Indiscriminancy of predators other than birds.................se.m0e 136
More theoretical aspects of indiscriminancy by predators............. 140
Indiscriminancy of normal checks other than predators.............+. 141
Relative importance of natural’ checks.2..0c 2. s.ieeas + ee eee 141
Summary —acax ews ore Sos o cies etolsias slnmtare\ercrers seta alee ia tt te 143
Bibliographiy: << woes cis niet mocteraieve, ote olen tet sieve piers ieee) wate eva eta ea 145

INTRODUCTION

In a previous paper’ the writer set forth reasons for believing
that the results of experimental tests of the effectiveness of the so-
called protective adaptations in protecting animals from their enemies
are not trustworthy indications of what occurs under natural condi-
tions. In the present contribution he proposes to show just what
insects and other animals are actually preyed upon by wild birds
of the United States, Canada, and Alaska, giving also incidental
notes on other enemies. This evidence reflecting food habits under
natural conditions goes far to show how little the alleged protective
devices have to do with choice of food by vertebrates.

Judging from the literature of the subject since 1912, the con-
tentions of the article on the experimental study of the food habits
of animals seem to have been generally admitted, or at least regarded
as too well supported to be lightly attacked. Only one essay has been
seen by the writer, that seems in any way a reply, namely an account
of “ Experiments and Observations Bearing on the Explanation of
Form and Colouring,” * by C. F. M. Swynnerton, who refers to my
criticism of the experimental method as “ rather over-vigorous.” The
vigor of the criticism is admitted but in view of the absurdity of the
arguments against which it was directed, it can hardly be considered

‘The experimental method of testing the efficiency of warning and cryptic
coloration in protecting animals from their enemies. Proc. Acad. Nat. Sci.
Philadelphia, June 1912, pp. 281-364 (Sept. 6, 1912).

? Journ. Linn. Soc., Zool., vol. 33, pp. 203-385, London, June 30, 1919.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 3

excessive. Undeniably selectionists have been absurd in their dis-
quisitions on adaptations ; for instance “ eye-spots ” on a butterfly’s
wings are to direct the attack of enemies to a nonvital spot, while
“eye-spots”” on a caterpillar are “terrifying” and prevent even
a touch where merely a touch would be fatal; in numerous species
of birds the male is colored red and black or orange and black,
characteristics that selectionists say have been developed by sexual
selection as an attraction to the opposite sex, yet the females of
these birds are supposed to be repelled by the same colors in possible
insect prey ; red insects are said to be warningly, red fruits invitingly
colored, and so on. A popular foible of similar type is that of sports-
men who hold up to admiration the marvelous protective coloration
of game birds, and in the next breath complain of severe depredations
on these birds by “‘ vermin.”

But this is digressing and the writer is glad to acknowledge that
if all of the experimenters had been as critical of their methods and
conclusions as Mr. Swynnerton, the tone of his former paper would
have been quite different. For instance Mr. Swynnerton carried
on more experiments than any of the authors reviewed in the previous
communication, before he, according to his own confession, learned
how to experiment. This in itself confirms the writer’s charges that
the experiments he reviewed were both inadequate and misinterpreted.
It may further be stated that the principal conclusions Mr. Swynner-
ton draws from his experiments and observations would have been
agreed to in advance by anyone experienced in the study of bird food.
Thus he concludes that birds show preferences among the food items
_ available to them, and that predatory animals of various groups show
more or less agreement in preferences. From his general experience
with birds he decides also that ‘‘ Unless through sheer impossible
hardness, size, etc., there is practically no such thing as ‘ inedibility,’”’’
and he appreciates that a group of insects, limited in numbers as are
butterflies, will not be taken by insectivorous birds out of proportion
to their abundance as compared to all insects available.

These things did not require experimental test for they are cor-
roborated in every thorough report on the natural feeding habits
of birds. What can not be admitted, however, is that preferences
of birds learned by feeding them upon some certain group of insects
to an extent far greater than the birds ever prey upon them in nature,

‘

reflect normal feeding habits, nor that there is evidence of intensive
enough feeding by discriminating enemies upon any group of insects

*A brief preliminary statement of a few of the results of five years’ special
testing of the theories of mimicry. Proc. Ent. Soc. Lond., 1915, pp. xxxii-xliii.

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to meet the requirements of the selection theories. We further can
not admit what the experimenters imply, namely, that the analyses
of the stomach contents of birds fail to reveal the approximate num-
bers present of certain insects (such as butterflies) which they believe
are eaten to a considerable extent. This point will be discussed later,

So much for what has happened between the previous paper and
the present, which as stated, will be devoted chiefly to an exposition
of the animal food of nearctic birds, with special reference to the
so-called protective adaptations.

PROTECTIVE ADAPTATIONS

The characteristics of animals that are usually classed as protective
adaptations include resemblance to generalities or details of the en-

vironment, whether through color or other modification of the animal

itself or utilization by it of materials from the environment for
concealment, the possession of protective bristles, spines, hard in-
teguments, stings, poisonous bites, and the like, and nauseous or
irritating odors or tastes. There are animals with actually poisonous
properties among many of the phyla including species with poison
glands and special organs for using the poison in offense or defense,
among Coelenterata, Echinodermata, Arachnida, Insecta, and Pisces;
others with poison glands connected with the mouth organs among
worms, spiders, other arachnids, mites, myriapods, chilopods, insects,
fishes, and reptiles; animals with unarmed poison glands among
coelenterates, echinoderms, myriapods, insects, mollusks, amphibians ;
and others poisonous in a variety of ways so that practically all phyla
are represented. The colors of the animals possessing dangerous
qualities in many cases are said to be warning in nature, and the
colors of animals which resemble them but lack the disagreeable
qualities are termed mimetic. The subject of protective adaptations
has very largely become one of coloration especially as associated with
the qualities of animals from the supposed point of view of possible
predators.

A statement of the various classes of color adaptations is here
quoted from Prof. E. B. Poulton, the leading advocate of the view
that these adaptations are really protective and that they have been
developed by natural selection.

Protective and Aggressive Resemblance—By far the most widespread use
of colour is to assist an animal in escaping from its enemies or in capturing its
prey; the former is Protective, the latter Aggressive. It is probable that these
were the first uses to which non-significant colours were put. The resemblances

are of various kinds; the commonest cases are those of simple concealment.
The animal passes undetected by resembling some common object which is of

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 5

no interest to its enemies or prey respectively, or by harmonising with the
general effect of its surroundings; the former is Special, the latter General
Resemblance, and both may be Protective or Aggressive. Among the most
interesting Special Aggressive Resemblances are the cases of Alluring Colour-
ing, in which the animal, or some part of it, resembles an object which is
attractive to its prey.’

Protective and Aggressive Mimicry.—Mimicry is in reality a very important
section of Special Resemblance. The animal gains advantage by a superficial
resemblance to some other, and generally very different, species which is well
known and dreaded because of some unpleasant quality, such as a sting or an
offensive taste or smell, &c., or it may even be protected from the animal it
resembles: this is Protective Mimicry. When, however, the animal resembles
another so as to be able to injure the latter or some other form which accom-
panies it or is not afraid of it, the Mimicry is Aggressive?

Warning Colours——When an animal possesses an unpleasant attribute, it is
often to its advantage to advertise the fact as publicly as possible. In this way
it escapes a great deal of experimental “tasting.” The conspicuous patterns
and strongly contrasted colours which serve as the signal of danger or inedibility
are known as Warning Colours. In other cases such colours or markings enable
individuals of the same species easily to follow those in front to a place of safety,
or assist them in keeping together when safety depends upon numbers. It is these
Warning Colours which are nearly always the objects of Protective Mimicry.’

Following is a copy of Poulton’s table ‘ classifying color adaptations :

I. Apatetic colours.— Il. Sematic Ill. Epigamic

Colours resembling some part of the | colouwrs—Warning colours——Colours
environment or the appearance of an- | and signalling col- displayed in court-
other species. ours. ' ship.

A. Cryptic col- | B. Pseudo- |
ours.—Protective | sematic colours.— |
and Aggressive False warning and
Resemblances. signalling colours.

1. Procryptic 1. Pseudapose- 1. Aposematic |
colours——Protec- | matic colours.— colours—Warning |
tive Resem- | Protective Mimi- | colours. |
blances. | cry.

2. Anticryptic 2. Pseudepise- 2. Episematic
colours —Ag- | matic colours.— colours.—Recog-
gressive Resem- Ageressive Mimi- | nition Markings.
blances. ‘ery and Alluring

Colouration. | |

Having presented the foregoing outline of protective color and
other adaptations, references to them in succeeding pages will be made
without further explanation of the terms involved.

* The colours of animals, pages 19-20, 1890.
* Idem, p. 20. *Tdem, p. 21. *Tdem, p. 338.

¥

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

ANIMALS EATEN BY NEARCTIC BIRDS

Data CITED AND How OBTAINED !

The main body of data used herein consists of the records of
animals identified in the contents of the stomachs of about 80,000
nearctic birds examined in the United States Biological Survey since
1885. These stomachs represent a wide range of species of all of
the families of birds occurring in the region; the birds were collected
at all seasons and in practically all parts of nearctic America. While
not evenly distributed in any of these senses, the collection is very
satisfactory for the purpose in hand, and yields a mass of precise
information on bird food that far surpasses anything of the kind
available elsewhere.

A word about the methods of investigating bird food may be
desirable. The gizzards of birds, together with the gullets or crops
when they contain food, are received chiefly from persons collecting
birds for some other scientific purpose, although in some cases they
are especially obtained to throw light on the relations of birds to some
crop, or useful or injurious animal. They are preserved usually with
formalin in the field and in alcohol after receipt at the laboratory.
Contents of a stomach being examined are removed either wet or dry
as best fits the particular case and transferred to watch glasses or
small white blotters for sorting and identification of the material
under compound binocular dissecting microscopes. A great deal of
the analysis is done at a magnification of 8 diameters but special
study of difficult subjects is continued when necessary under higher
powers.

At this point it may be well to comment on the popular misconcep-
tion that anything found in a bird’s stomach necessarily is ground
up and in all but unrecognizable state. As a matter of fact the reverse
is true. Most birds swallow their food whole ; consequently in any col-
lection of birds a certain proportion will have swallowed some food
items just before death. These things often are in perfect condition ;
they may be, and sometimes are, used for cabinet specimens. The
nearly or quite whole objects usually furnish clues to the fragmentary
material, and in the great majority of cases it is possible to sort out
completely all components of the food. It is the exception when the
finely ground food remains defy separation and identification. De-
terminations are carried as far as practicable; each member of the
staff of analysts is a specialist in some line and they cooperate freely ;
specimens defying their combined efforts, if in fair or better condi-
tion, are submitted to advanced investigators elsewhere. The records

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 7
quoted in the following pages include more or less of the handiwork
of practically every prominent American systematic zoologist of the
period. Nevertheless everything is not identified, far from it ; expert
assistance has not been available in some cases when needed, too far
digested residues sometimes occur, and rarely we find also well-pre-
served but puzzling objects that indefinitely defy classification.

These, however, are but minor flaws in the system; the ground-
work of our faith in the results of stomach analysis is the law of
averages. Given good distribution geographically and seasonally,
which necessarily follows from miscellaneous collecting carried on
for so many years, the reliability of results varies directly with the
number of stomachs. The collection (80,000) here reported upon
is believed sufficient to furnish fairly dependable data, although addi-
tions are made almost daily to the list of animals identified from bird
stomachs.

The total number of identifications of animals from these stomachs,
counting those of whatever degree, once for each time identified
irrespective of the number of individual specimens concerned, is
237,399.

It was impracticable to compute the total number of individual ani-
mals concerned for the reason that these were not counted in all
cases. Moreover this figure would not have been especially useful
in the absence of estimates for comparison of the actual animal popu-
lation of significant areas. In casting about for a standard which
would afford some idea of the frequency of occurrence of animals of
various groups, the estimated number of species therein proved to be
the only one available for the whole range of the animal kingdom.
That the number of species in taxonomic groups bears a general re-
lation to the number of individuals can not be questioned. It is easy
to point out exceptions, but remember we can only deal with this
problem in an approximate way, and it goes without saying that on
the average a group more numerous in individuals will have devel-
oped more species than one less numerous. The correspondence is
not exact, but it is sufficient to give a fair working idea of the position
of the various groups in the scale of frequency of occurrence, the

* The tabulation necessary to yield this figure was an immense one (covering
nearly a thousand typewritten pages) and has been found, it is not surprising,
to contain some errors. These are so small, however, that rectification of them
would not cause changes of more than a fraction of one per cent in any part
of the results, except in the table for Coleoptera, pp. 65-67. Hence they do not
invalidate the figures at all for the purpose here used of showing in a general
way the tendencies exhibited by our birds in their choice of animal food.

v

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

very standard we wish for comparison with frequency of identification
in the stomachs of collected wild birds. The general correlation of
these factors to be noted in the tables presented in subsequent pages
increases confidence in the value of the method.’

IDENTIFICATIONS OF ANIMAL Foop

To illustrate the way in which data was assembled, that for the
phyla may be given in rather more detail than is planned for the
balance of the report. The figures for number of species in the various
phyla are compiled from various estimates * of this nature; the facts
that these are not strictly up-to-date nor anything like exact are of
no consequence in a field where only approximations may be hoped for.

The subjoined table shows the estimates used for the number of
described species in each phylum and the percentage that figure bears
to the total number of animals known.

Phyla of Animals and the Number and Percentage of Species in Each

Percentage

of species
Estimated in this
number of phylum

species among the

Phylum known whole number

ProtozOals manna asia OEE ome arose 8,000 1.4272
PORUEET areeva shy vee ease cree rece aero tebe: 2,500 .4400
Coelenterata: 225 sce ake oto aero 4,500 8028
Platyhelminthes werner ose aes 5,000 .8920
INemathelminthesy saeeciaenineseeee rec 1,500 .2076
Erochel minthes sy ere lee caress ake 500 0892
Moliitiscoidameereesce teeter eon: 1,700 3032
Echinodermata seni. de cies acer 4,000 -7136
Aninitilatay 632 Asem een Sct cee Ghinee 4,000 7136
ATT OPOddimn ean clten cone or nee 418,250 74.6188
Molluscapce ence satya ec eer 61,000 10.8828
Chordata? sa. .ures sehen en ae eee eee 49,505 8.8427
AROtallsingersetn eos ee oe Cite acer 560,515 99.9995

“Here may be mentioned the law demonstrated by Olaf Arrhenius (Journ.
Ecol., vol. 9, no. 1, p. 99, Sept., 1921) that among plants ‘ The number of species
increases continuously as the area increases.” Since as a rule the number of
individuals also increases with the area, the parallelism between the number
of individuals and that of species is further confirmed.

* Pratt, H. S., On the number of known species of animals. Science, vol. 35,
pp. 467-468, March 22, 1912.

Henshaw, H. W., Number of species of living vertebrates. Science, vol. 36,
pp. 317-318, Sept. 6, 1912.

Handlirsch, A., Die fossilen Insekten, pp. 1182-1188, 1908.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 9

Classifying the 237,399 identifications of the animal food of
nearctic birds and calculating the percentage of the determinations, by
phyla, we reach the results shown in the next table, the percentage
of species in each phylum among the whole number of known species
being repeated for ease in comparison.

Identifications of Animal Food by Phyla

Percentage

of species

in this

phylum
* among the

Percentage of whole
identifications number of

among those animal

Number of of all species

Phylum identifications animals known

PeVOCOZOA: ss cess cece set seee I2 .0050 1.4272

IROTITEra 26. sees. sss ene os 2 .0008 .4460
@oclenterata (2.04.42... 122 .O514 8028
Nemathelminthes ......... 24 .OIOL 2676
Molluscoida ~ ...6...ssse0% 134 0564 3032
Echinodermata ........... 125 .0526 -7136
Aminulata ..sccsiecsescses 1,131 4764 7136
mrthropoda .........+..--. 210,752 88.7751 74.6188
IMMOINISCA, .0 5c. oe cece eee 1770 4.9583 10.8828
| Ghordata ce. hedics ceca von 13,326 Biobes 8.8427

Without going into details, it is apparent that the percentage of
| identifications preserves very well a relative ratio to that of the num-
_ ber of species and presumably, therefore, to the abundance of indi-
viduals in the phyla. What variations there are seem obviously due
_ to differences in the availability to birds of the differing types of

animals.

Taking up the phyla in order, we begin with the

PROTOZOA (ONE-CELLED ANIMALS)

Protective adaptations —Judging from what is asserted about other
phyla, phosphorescence and the possession of bright colors in some
groups and of silicious or calcareous, often tuberculate or spinose,
tests or shells or of exoskeletons formed of foreign bodies in others,
are characters that would be deemed of protective significance in
Protozoa.

Bird enemies.—Protozoa are too small to engage the attention ot
birds, those found in stomachs being Foraminifera strained from
water or mud, or picked up as gravel by ducks. It is probable also
that protozoa are consumed, along with the stems and leaves of
aquatic plants upon which they often are abundant, by wild ducks

®

1D

IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

which feed upon such vegetation. Stomach analysis, however, has
not been directed toward the recognition of such minute material.

Number of identifications, 12; percentage of identifications among
those of all animals, .o050 ; percentage of species in this phylum among
the whole number of animal species known, 1.4272.

Other enemies.—Protozoa are the prey of others of their kind—
of bacteria, of rotifers, of flatworms, of amphipods and other small
crustacea, and of mollusks; they are eaten also by the young of
numerous species of fishes, by the adults of specialized forms (men-
haden, gizzard shad), and by the larvae of batrachians.

Discussion.—Protozoa, because of their minute size and general
inaccessibility to birds, would not be expected to enter largely into
the food of this class. The forms eaten by birds are among the best
“protected” protozoa, but the possession of shells can hardly be
considered as an adaptation for protection from enemies in the case
of animals so small as to be easily devoured by almost any carnivorous
animals encountering them and which exist in such enormous numbers
that vast areas of sea bottom are covered with remains of their
shells. In this case as in many others, numbers are so large and re-
production so great that the inroads of all enemies are fully dis-
counted. Losses to predatory enemies are only a fraction of the total
death rate.

PORIFERA (SPONGES)

Many sponges are pervaded by calcareous or silicious spicules
which may render them more or less undesirable as food for pre-
datory animals. Some are brightly colored and some phosphorescent.
‘Sponges do not appear to be edible by Fishes or even the higher
Crustaceans or Molluscs. Countless lower animal forms, however,
burrow in their substance, if not for food, at least for shelter, and
the interior of a sponge is frequently found to be teeming with
small Crustaceans, Annelids, Molluscs and other Invertebrates.” *

Sponges have been identified from only 2 stomachs of nearctic
birds (Canada goose and lesser scaup) and from their low degree
of accessibility to birds, not many cases of feeding upon them would
be expected.

1Entries under this head for the various groups treated are intended as
suggestive rather than as exhaustive. A list of papers from which much of
this information has been gleaned forms the special bibliography on pp. 145-201.
Notes on the food of reptiles, amphibians, and mammals are mostly from
analyses of stomach contents in the Biological Survey.

? Parker, T. J., and Haswell, W. A., A text-book of zoology, vol. I, p. 126,
IQI0.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE Ve

Number of identifications, 2; percentage of identifications among
those of all animals, .oo08; percentage of species in this phylum
among the whole number of animal species known, .8028.

Discussion.—Information at hand seems to indicate that sponges
are used very little as food by other animals; sea-urchins, marine
worms, amphipods, and mollusks, however, are recorded as predators.
Fresh water sponges are eaten to some extent by fishes. Whether
this is credited to their “ protective adaptations” is of little moment
as the fact remains that sponges do not multiply excessively nor
overrun the earth as forms that lack enemies are in theory supposed
to do.

Sponges have bright and varied colors and one case of mimicry has
been pointed out.’ If it be true as apparent from observations thus
far, that they have few or no enemies, natural selection can not be
advanced as an explanation of their color phenomena. If sponges
without enemies have adaptations of the same character as other
groups with numerous enemies, it would seem evident that selection
by predatory animals has no necessary connection with the adaptations.

COELENTERATA (HYDRAS, JELLYFISHES, SEA-ANEMONES)

Protective adaptations-—Some coelenterates have a chitinous cuti-
cle, others have a calcareous skeleton, and many of them have nema-
tocysts or stinging cells. Numbers of them are brilliantly colored or
phosphorescent but it must be noted also that many are transparent
or nearly so, showing that possession of protective devices (as the
nematocysts) is not always accompanied by the development of
“warning colors.”

Bird enenties.—The Coelenterata most often found in bird stomachs
are the Hydrozoa (such as Abictinaria, Sertularclla, and Thuiaria).
They have been identified 113 times from the stomachs of 13 species
of ducks, 2 of gulls, and one each of murre, murrelet, and shearwater.
Sea-anemones (Anthopleura, Aulactinia) have been identified four
times from stomachs of a scoter, an eider, an oyster-catcher, and a
gull, Aleyonaria from two ducks, and coral from one.

Number of identifications, 122; percentage of identifications among
those of all animals, .o514 ; percentage of species in this phylum among
the whole number of animal species known, .8028.

Other enemies—Hydroids are eaten by marine worms, by sea-
urchins and sea-anemones, and also by fishes, as the cod, haddock,

*McIntosh, W. C., The coloration of marine animals. Ann. Mag. Nat. Hist.
7th ser., vol. 7, p. 223, Mar., 1901.

'
12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

sand launce, lump sucker, cunner, scup, filefish, and flatfishes ; cteno-
phores are eaten by the spiny dogfish, flatfishes, whiting, and cod;
sea-anemones are eaten by cod,’ haddock, tilefish, flatfishes, the sun-
fish Mola, spiny dogfish, whiting, and by the so-called jellyfishes and
by whales. Holothurians and some fishes (Scarus) feed on corals.

McIntosh notes that the brightly colored jellyfishes “ have precisely
the same habits as the uncoloured and transparent,” which raises
doubt as to the validity of the selectionist interpretation of the facts.
The brightly hued and the translucent forms are equally palatable
to whales and other animals using jellyfishes as food. He adds with
regard to sea-anemones that “the view that the gaudy colors ....
act as a warning is not borne out by the eagerness with which the
cod swallows the brightest, such as Stomphia, while the smaller flat-
fishes fill their stomachs with Edwardsiae.’” (Ann. Mag. Nat. Hist.
7th ser., vol. 7, pp. 2242225, 1901.)

Discussion.—Coelenterates are another group of animals but
slightly available to birds and seem to be taken in full proportion to
the degree of availability. The nematocysts seem a futile defense
against animals of the groups here mentioned as coelenterate enemies,
and must be also in the case of the myriads of crustacea (possible
enemies) all of which have a chitinous exterior and which more-
over manipulate their food in the chelae before chewing it, a process
that would result in the harmless discharge of the stinging cells.
It is alleged that hermit crabs have a commensal relation with certain
hydroids which grow upon the shells they inhabit and that they are
protected from their enemies by the presence of the inedible stinging
hydroids.” This is not the case where the bird enemies are concerned,
as the sea ducks which are the principal bird enemies of hydroids,
often swallow the hermit crab, shell, hydroids and all. Many of the
examples identified from bird stomachs came from precisely this
source. With respect to the practical aspect of the case, it would
appear that in its shell retreat and its own strong claws the hermit
crab has much more efficient defenses than the nettlelike hydroids.
It seems more likely that the latter merely grow on mollusk shells as
a convenient substratum. From the habit some hermit crabs have of
frequently changing their abode, the advantage held by a “com-
mensal”’ hydroid may be lost at any moment.

"McIntosh notes that sea-anemones are a valued bait for cod.
* Parker and Haswell, Zoology, vol. 1, p. 144, 1910.

Ss

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 13

PLATYHELMINTHES (FLATWORMS, FLUKES )

The majority of organisms of this phylum are parasitic and there-
fore not available to predatory animals. Perhaps some of the fresh-
water planaria and the marine nemerteans have been found but not
identified in the stomachs of shore birds, but so far we have no
positive determination of a worm of this phylum as bird food. Forbes
reports a small catfish (Noturus) feeding on fresh-water planaria.
McIntosh says that marine planarians of both plainly and brightly
colored forms are eaten by sea-anemones and fishes. Fresh-water
planarians also are eaten by fishes. Stiles intimates that carp destroy
large numbers of the liver fluke (Fasciola hepatica) in the cercaria
stage.

NEMATHELMINTHES (THREADWORMS, ROUNDWORMS)

Again a vast number of worms of this phylum are parasites, abun-
dantly so, in fact, of birds themselves. In order to reckon as food
only those so taken, all nematodes other than Gordiidae have been kept
out of the computations. The records for Gordiidae number 24, the
percentage of these identifications among those of all animals is
oIoI, and the percentage of known species’ of Nemathelminthes
among all animals according to the estimates used in the present
paper, .2676. The nematodes have a tough cuticle but no special
defenses ; nevertheless they certainly are not eaten out of proportion
to their numbers, but considering availability to birds, they may
possibly be eaten somewhat in ratio to the frequency with which they
are encountered. They are eaten also by flatworms and by various

fishes.

TROCHELMINTHES (ROTIFERS)

None of these have yet been identified as food of nearctic birds,
though possibly rotifers taken in with foliage of aquatic plants may
have been overlooked. Rotifers are eaten by the young of a number
of fishes.

MOLLUSCOIDA (CORALLINES, LAMPSHELLS)

Protective adaptations —Except for the shells of the brachiopods,
and cuticular walls of some bryozoa, no special protective features
have been developed by the Molluscoida.

Bird enemies.—Only three brachiopods have as yet been identified
from the stomachs of nearctic birds—not a matter for surprise in

* An enormous number of Nematodes await description.

14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

view of the small number and marine habitat of the species of these
animals. The other Molluscoida that have been found in bird stomachs
include Cyclostomata (having calcareous zooecia), Cheilostomata
(with calcareous or chitinous zooecia), and Phylactolaemata. Large
numbers of the statoblasts of the latter group, the fresh-water bryozoa,
have been disclosed in the stomachs of wild ducks.

Number of identifications of Molluscoida, 134; percentage of
identifications among those of all animals, .0564; percentage of species
in this phylum among the whole number of animal species known,
.3032.

Discussion.—Considering their low availability to birds, animals of
this phylum probably are taken in due proportion. Fresh-water bryo-
zoans have been recorded from stomachs of many species of fishes,
and the marine forms from a smaller number. Marine bryozoans
are preyed upon by worms, amphipods, decapods, and starfishes.

ECHINODERMATA (SEA-CUCUMBERS, SEA-URCHINS, STARFISHES )

Protective adaptattons—All of the echinoderms have a calcareous
exoskeleton and in many the surface is beset with tubercles or spines.
The starfishes and sea-urchins are armed also with pedicellariae or
grasping organs, which in some cases in the latter group are said to
be poisonous. Some sea-cucumbers have the “ Cuvierian organs”
which throw out long viscid filaments. Starfishes, especially the
brittlestars and many crinoids, have the supposedly protective faculty
of snapping off their arms or portions thereof. The colors of echino-
derms are often conspicuous and in certain cases have been termed
warning.

Bird enemies.—Starfishes have been identified 28 times in the
stomachs of nearctic birds here reported upon; sea-urchins (Strongyl-
ocentrus, Echinarachnius) 92 times; and sea-cucumbers, 3 times.
The birds (19 species) eating them were chiefly ducks collected im
northern seas.

Number of identifications, 125 ; percentage of identifications among
those of all animals, .0526; percentage of species in this phylum among
the whole number of animal species known, .7136.

Other enemics.—Starfishes and sea-urchins prey upon one another,
and are very commonly eaten by cod, haddock and other species of
Gadus, by argentines, dragonets, rocklings, wolffishes, rays, sharks,
tautog, scup, smelt, flatfishes, and others. Sea-cucumbers are less
commonly taken by the same predators. Blue foxes on the Pribilof
Islands feed on sea-urchins in winter. Sea-urchins and starfishes are
consumed also by crabs, sea-anemones, and marine worms.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 15

McIntosh comments interestingly on the enemies of echinoderms:
“The colours of Echinoderms are often most conspicuously bright,”
but Hippasterias, which is brilliant orange-red, is eaten by gulls,
cod, catfishes and by other starfishes. “ The sand-stars (e. g. Ophiura
lacertosa) are often tinted of a hue resembling their surroundings,
yet they and the more brightly tinted forms are common in the
stomachs of fishes and are eagerly devoured by gulls when stranded
on the beach.” “ The brown and purple hues of sea-cucumbers may
in some way stubserve protection... . yet both they and the trans-
parent forms are found in the stomachs of fishes.” (Ann. Mag. Nat.
Hist. 7th ser., vol. 7, pp. 225-226, 1901.)

Discussion——Echinoderms have a number of protective devices
but also it would appear, numerous and effective enemies. Birds prey
upon this group to fully as large an extent as could be expected, con-
sidering the slight degree to which they come in contact with echino-

derms.
| It should be noted that while practically all starfishes and sea-
urchins have similar protective adaptations, some are very gaudily,
others modestly colored; in one case or the other, the natural selec-
| tion theory as to the connection between special defenses and color-
ing is untenable. The sea-urchins with calcareous tests, abundant
spines, and pedicellariae seem unusually well defended, but that this
does not mean freedom from enemies is shown by the great fecundity
| of sea-urchins, individuals of some species, e. g., Echinus esculentus,
yielding 20,000,000 eggs per season.

ANNULATA (WORMS)

Protective adaptations —The chaetopods including the most com-
mon marine worms and the majority of earthworms have chitinous
setae on all segments of the body. The earthworms are habitual bur-
rowers, and some of both fresh- and salt-water annelids live in tubes.
A few in each group are phosphorescent, and many of the marine
worms are highly colored. A. R. Wallace says" “Among the crea-
tures which probably have warning colors as a sign of inedibility are
_. . . those curious annelids the Nereis and the Aphrodite or sea-
mouse.”

It should be noted however that many of the brightly colored forms
live in burrows or tubes, thus taking care not to advertise their
“inedibility.” _Leeches sometimes have strongly contrasting color, as
for example greenish with red and black spots.

‘Darwinism, p. 266, 1896.
2

¥

16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Bird enemies.—The identifications of Annulata from nearctic birds
are: 690 for the Polychaeta, chiefly Nereidae, which have been found
in the stomachs of more than 70 species of birds and in numbers up
to 500 in a single stomach; 428 for the Oligochaeta or earthworms
from 44 species of birds; and 11 of Hirudinea or leeches from tro
species. The robin (Planesticus migratorius) feeds habitually and
voraciously upon earthworms and the woodcock (Philohela minor)
makes about half of its diet of these annelids.

Number of identifications, 1131; percentage of identifications
among those of all animals, .4764 ; percentage of species in this phylum
among the whole number of animal species known, .7136.

Other enemies.—Studies that have been made of the food of fishes
indicate that a very large number of marine fishes prey upon the
Nereidae and. other annelids. They are eaten also by other worms,
starfishes, sea-urchins, sea-anemones, gastropods, and crabs. Fresh-
water oligochaetes form a steady contribution to the diet of the fishes
of their environment. Earthworms are eaten by predacious beetles,
by most batrachians, by some turtles, snakes and by various mam-
mals including shrews, skunks, and the armadillo, but especially by
the moles (Parascalops brewert, 26 per cent of the food; Scalopus
aquaticus, 31 per cent; Scapanus townsendi, 40 per cent; Condylura
cristata, 50 per cent.) Leeches are eaten by a variety of mammals,
birds, reptiles, amphibians, fishes, crustaceans, snails, and insects.

Discussion.—Both of the annelids (Nereis and Aphrodite) Wallace
mentions as being warningly colored are eaten by birds and fishes,
Nereis frequently and in large numbers. Considering the aquatic
habits of most of the annelids it would appear that they are taken
by birds as often as could be expected. It is evident furthermore
that they have numerous other predatory foes and that they probably
contribute their full quota of food toward the dietary requirements
of the animal kingdom.

ARTHROPODA (JOINTED ANIMALS)

As recorded previously in the table of phyla, the Arthropoda, in-
cluding the exceedingly numerous class of insects, furnish, as would
be expected, a very large preponderance of the animals eaten by birds.

Number of identifications, 210,752; percentage of identifications
among those of all animals, 88.7551; percentage of species in this
phylum among the whole number of animal species known, 74.6188.
The disproportion of the percentage of capture to that of frequency
reflects the relatively greater availability to birds of arthropods over
the other phyla.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 17

Since arthropods are by far the most important phylum of animals
as bird food and since it is with reference to the included class of
insects that the theories of protective adaptations have been most
highly elaborated, it is desirable to subdivide the phylum for the pur-
poses of the present discussion.

The tabulation below shows the number of identifications and their
percentages for the four classes of Arthropoda that are available for
food to nearctic birds.

Arthropoda
Percentage
of species
Percentage of in this class
identifications among the
among those whole number
Number of of all of arthropod
Class identifications arthropods species known
(nustacea: 2s... .2 os skeeacies 6,086 2.8877 3.8254
IMvGIA POG ceecce. cece. 2,862 1.3530 4781
ISG Ctae aaron is at e-wieaiiare we hiee 190,919 90.5801 91.8589
PNG ACINITC aw sed els soins swe aoe 10,885 5.1648 3.8254

It is evident that the percentage of identifications for each class
corresponds very well with the frequency of such animals as indicated
by the number of species. The validity of such comparisons used
throughout this paper thus receives further corroboration.

CruSsTACEA (CRABS, SHRIMPS, SOWBUGS)

Protective adaptations ——The exoskeleton of crustacea is either
chitinous or much thickened or calcified. Several groups have bivalved
carapaces in which the animal is nearly or entirely enclosed. The
barnacles have a more complicated covering of calcareous plates, some-
times thick and hard. Most of the decapods have strong grasping fore-
legs, and have furthermore the faculty of snapping these off when
properly stimulated. Some crustacea have burrowing habits (Hip-
pidae) and others (hermit crabs) use the shells of univalves for
shelter, while the terrestrial sowbugs roll themselves up into a ball
when disturbed.

As to color, some of the smaller aquatic forms are translucent or
transparent; the ostracods are said to assimilate with the general

color of their environment, while some copepods and decapods are

brilliantly colored. As to form, Mortensen says: “‘ The typical cases
of adaptation to life among algae are especially found among the
Caprellids ; they might be said to represent the Phasmids and Geo-
metrid larvae, among marine animals ” (p. 77). “ Zdothea marina and

18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

related species . . . . bear a striking resemblance to the plants (es-
pecially Zostera leaves) on which they are found” (p. 78).’

“Many Crustacea,’ according to Parker and Haswell, “ present
instances of protective and aggressive characters, i. e., modifications
in form, colour, etc., which serve to conceal them from their enemies
or from their prey. Probably the most striking example is that of
certain crabs (Paramithrax) which deliberately plant Sea-weeds,
Sponges, Alcyonarians, Zoophytes, etc., all over the carapace, and
are thus perfectly concealed except when in motion.” *

Poulton expresses the same view in the following language: “ Cer-
tain palatable animals make use of the Special Defence and Warning
Colours of other forms. Thus, the common English hermit-crab,
Pagurus bernhardus, commonly carries on its borrowed shell the con-
spicuous stinging sea-anemone Sagartia parasitica; while another
English species, Pagurus prideauxit, inhabits a shell which is invari-
ably clothed by the flattened Adamsia palliata. Two crabs (Polydectus
cupulifer and Melia tessellata), from Mauritius, described by Mobius,
invariably held a sea-anemone in each claw. Two other groups of
animals, sponges and Ascidians, in addition to sea-anemones, are
avoided by enemies of the Crustacea, and these are also employed
by the latter. Thus the British hermit-crab Pagurus cuanensis is
found in shells which are covered with a (generally) brightly-coloured
sponge (Suberites domuncula). Mobius also describes a Mauritian
hermit-crab (Ascidiophilus caphyraeformis) which lives in a case
formed by an Ascidian.” ®

Bird enenues—Most of the leading subdivisions of the Crustacea
contribute to the food of birds, apparently about in proportion to their
accessibility. The tabulation of numbers of species of Crustacea used
for comparison with those of percentages of identifications was made
from “A list of the Crustacea of New Jersey including the adjacent
region or that of the Middle Atlantic States,’ * the only check list
available for any considerable area of our region.

The Anostraca (fairy shrimps) are locally eaten more extensively
than indicated above and the fact is an illustration of the principle

* Mortensen, T. H., Observations on protective adaptations and habits, mainly
in marine animals. Vidensk. Meddel. fra Dansk. naturh. For. Kjob., bd. 69, 1920.

* Zoology, vol. 1, p. 601, 1910.

* Poulton, E. B., Essays on evolution, pp. 356-357, 1908.

Thomas Scott records a copepod (Acidicola rosea) which lives within the
branchial sac of an ascidian as having been eaten by a sole Pleuronectes micro-
cephalus. (20th Ann. Rep. Fishery Bd. Scotland, p. 525, (1901) 1902.)

“Fowler, H. W., Ann. Rep. N. J. State Mus., pp. 463-5908 (1911), 1912.

NO. 7 PROTECTIVE ADAPTATIONS

McATEE 19

that in natural history as elsewhere sweeping statements based on
partial or negative evidence are dangerous. No exception may be
noted for one which claims that “enemies play no part in keeping
down the numbers of Artemia (brine shrimps), or of Ephydra
(alkali flies) in the larval stage.” * Dr. Alexander Wetmore, of the
Smithsonian Institution, who has had considerable experience about
Great Salt Lake to which locality the quoted assertion relates, has
pointed out * that Artemia and Ephydra are by no means free from
enemies. Shovellers, lesser scaups, golden-eyes, green-winged teal,
Wilson’s and northern phalaropes, avocets, and black-necked stilts all
feed extensively upon both of these animals. But for the fact that

Identifications of Crustacea

Percentage
of species
Percentage of in this group
identifications among those
among those of all Crustacea
Number of of all of the Middle
Group identifications Crustacea Atlantic States
RIMIGOHtIAEE — 6s: sess ids oso 573 9.4150 ee
PNTIOStEACA-speeyeiae sav se eee ee 28 .4601 1.3667
ladOCera: Wesrs oH ey ss sieie 90 1.4788 4.1002
PU STLACOU A. “org Qe site van ts See os 207 3.4012 3.8724
COPEPOdS -o.0.5 60d von sane ceess 13 2136 18.2232
@irripedia: .sse..0..2se4 00 4o1 6.5888 5.9225
MSOPOMA Ss aces cecacmencs ses 385 6.3259 14.3508
Amphipoda ..6..i26....0800. 986 16.2010 9.3394
OTMACEA is. oie 20 ean ess be dun t as ee 3.1891
ptomatopoda: Fss.i0<cssecs0s een seeee Q112
CMIZOPOddy s5..06 eee. se cia ne 48 .7887 1.1389
Wecapodare sce saeeeeee ee 3355 55.1260 37.5853

stomach analyses have not been made of other birds collected at the
same place, it would undoubtedly be possible to add the names of
a number of species to this list. Doctor Wetmore states that “ the
toll taken by birds from the brine shrimp and alkali fly larvae and
pupae during the course of a season constitutes a mass of individuals

almost beyond comprehension . . . . The immense numbers of these
creatures . . . . must be attributed to the large number of offspring

produced rather than to an absence of enemies.” The number of
records for the minute Cladocera is fully up to expectations. Among
other items of this group the egg capsules or ephippia of Daphnia
have been found in numerous stomachs of grebes and wild ducks, and
in number up to 250 in a single stomach.

*Vorhies, Charles T., Notes on the fauna of Great Salt Lake. Amer. Nat.,

vol. 51, pp. 494-499, Aug., 1917.
” Amer. Nat., vol. 51, pp. 753-755, Dec., 1917.

20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The ostracods have been identified chiefly from the stomachs of
15 species of wild ducks, and no fewer than 1,200 have in two in-
stances been taken from a single stomach. Most of the barnacles
were found in the stomachs of 22 species of ducks, gulls, and shore-
birds from northern waters. Amphipods were eaten by more than 80
species of birds, largely shorebirds, ducks, and other waterfowl ; nearly
70 species of amphipods were identified and the number of individuals
taken by single birds ran up as high as 2,500. The isopods were con-
sumed by more than 75 species of birds, the land-forms or pillbugs
alone by about half that many; the greatest number of individuals
of terrestrial sowbugs found in a single stomach was 60, of aquatic
forms, 256. In the Decapoda it may be of interest to note that 392
of the identifications were of shrimps, hundreds of individuals being
present in some stomachs ; 1,592 of Astacidae (crawfishes), the great-
est number taken by one bird being 49; and 794 of crabs of various
kinds. Among groups of crabs represented, there were the following
numbers of captures, the figures in parentheses in each case denoting
the largest number of individual crabs found in a single stomach:
Sand crabs or sandbugs, Hippidae, 61 (14) ; stone crabs, Lithodidae,
go (16) ; hermit crabs, Paguridae, 35 (40) ; mud crabs, Pilumnidae,
186 (36) ; swimming crabs, Portunidae, 39 (16) ; edible crabs, Can-
cridae, 41 (18) ; shore crabs, Grapsidae, 180 (26) ; and fiddler crabs,
Ocypodidae, 272 (19).

Number of identifications 6,086; percentage of identifications
among those of all arthopods 2.8877; percentage of species in this
group among the whole number of arthropod species known, 3.8254.

Other enemies.—Crabs of various genera (including hermits) are
a staple item of food for many fishes, such as the dogfish, rays, eels,
sea bass, squeteague, scup, tautog, swellfish, toadfish, tilefish, hake, cod,
haddock, sculpins, and flounders. Crawfishes are relished by fresh-
water fishes and are eaten also by snakes, turtles, and various mam-
mals such as the muskrat, raccoon, skunks, mink, and otter. Such large
and powerful forms as lobsters are eaten by sea bass, rockfish, tautog,
sharks, dogfish, rays, and skates. Amphipods are captured by the
plant Utricularia, by insects, hydras, sea-anemones, and starfishes.
Practically all adult fresh-water fishes eat amphipods and isopods,
and when young prey upon Cladocera, Copepoda and Ostracoda.
Starfishes and bony fishes such as Coregonus, Salvelinus, Alosa, her-
ring, sticklebacks, and roaches continue feeding on these small forms
when adult. Marine fishes take similar crustacea available to them,
particularly the abundant shrimps and Mysidacea. Whales and seals
consume enormous quantities of isopods and Euphasiacea. Caprellids

NO. 7 PROTECTIVE ADAPTATIONS—McATEE Zi

are eaten by fishes as well as by birds. Cladocera and Copepoda are
eaten freely by larval salamanders and to a lesser extent by recently
transformed frogs. They and all other small fresh-water crustaceans
fall a prey to Hydra and aquatic insects. Small marine forms are
engulfed even by protozoans. Barnacles are eaten by the tautog, and
by sea-anemones and sea-urchins. More than 80 kinds of crustacea
have been identified from stomachs of haddock taken in waters about
Scotland (Thomas Scott). Crustacea have parasites from among
their own ranks, and from among the worms.

Discussion—Most of the small crustacea are translucent or trans-
parent but this does not save them from their foes. Practically all
aquatic animals “ get their start’ by feeding on these crustacea, the
list including a great variety of insects, fishes, and batrachians. Many
of them continue feeding upon crustacea when adult, and so com-
mon is this habit that in many cases small crustacea are the animal
basis of the food for the entire fauna of certain waters. This is true
of the Artemia of Great Salt Lake, previously discussed, and con-
spicuously so of the Mysidacea, Amphipoda, Isopoda, Euphasiacea
and Macrura of northern oceans, where everything from other
crustacea, through fishes and birds up to whales preys incessantly upon
them. There is no question of special protection here but solely of
numbers and fecundity.

The protection that crustacea might be supposed to derive from their
more or less indurated exoskeleton is discounted by the fact that in
most cases there are plenty of enemies large enough to swallow them
whole. Of what avail for instance is the bivalved shell of the almost
microscopic Ostracoda? The same principle applies all along the line
up to and including the crabs, for most crab-eaters swallow their
prey entire ; however there are some crabs that grow so large they are
possibly almost free from enemies when adult.

The claws of the large decapods naturally are of little avail against
enemies so voracious as to swallow the crustaceans whole and there
is no evidence known to the writer that the self-mutilation practiced
by decapods results in enemies swallowing the claw and letting its
owner escape.

Birds find the Hippidae or sandbugs, despite their burrowing habits,
and hermit crabs, adopted shell and all, are freely eaten by birds and
fishes. In numerous cases the hermit crabs found in bird stomachs
were those with hydroids and bryozoa growing on their carapaces
or shelters. Why should it even be supposed that such combinations
of animals are protective when the enemies of one of them are in
most cases enemies of all? For instance the diving ducks and fishes

7

22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

which relish crabs, including hermits, also eat mollusks, bryozoa, and
hydroids. What difference is it to them that a mollusk shell con-
tains a hermit crab rather than its original occupant, or that hydroids
are growing on it when these animals are browsed from rocks, etc.,
elsewhere? McIntosh has brought up this same point with regard to
species of Hyas which become covered with a growth of algae and
invertebrates, yet covered with parasites as they are, abound in
stomachs of the cod.’ They are eaten by other fishes and by birds
also. Conclusions of a similar nature undoubtedly must be drawn in
the case of those crabs associating sea-anemones and ascidians with
themselves. Both of these classes of animals have their enemies
which probably would engulf crab and all in cases where the animals
were together.

The caprellids noted by Mortensen as resembling algae and by
Parker and Haswell as so closely assimilated in form and color to
Hydrozoa and Polyzoa as to be difficult of detection nevertheless are
detected and eaten by some birds and by numerous fishes, and the
protectively formed and colored isopods of the genus /dothea are
represented by 51 records for 6 species in stomachs of 18 kinds of
birds.

The fiddler crabs (Uca), so abundant and conspicuous on the mud
flats of the southeastern coast of the United States, have one claw
enormously developed, thus having the principal characteristics of the
so-called protected species, a special mode of defense, and living ex-
posed and conspicuously in large numbers. They are freely eaten by
birds however and for this single genus of a few species, we have
271 records from 24 species of birds.

Myriapopa (THOUSANDLEGS, CENTIPEDS )

Protective adaptations.—The centipeds and millipeds exhibit differ-
ences that would warrant their being treated as separate classes;
customarily, however, they are considered together. The following re-
marks on their protective adaptations are quoted from F. G, Sinclair.’

The means of defence possessed by these animals ... . differ very much
in the different species of Myriapods. In the Centipedes the animals are
provided with a powerful weapon in the great poison claws which lie just
beneath the mouth, and which are provided with large poison glands, which
supply a fluid which runs through a canal in the hard substance of the claw
and passes into the wound made by the latter. The effect of this fluid is

Ann. Mag. Nat. Hist. 7th ser., vol. 7, p. 229, Mar., 1901.
* Cambridge Nat. Hist., vol. 5, pp. 36-37, 1910.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 23

instantaneous on the small animals which form the food of the Centipedes. I
have myself watched Lithobius in this country creep up to a blue-bottle fly
and seize it between the poison claws. One powerful nip and the blue-bottle
was dead, as if struck by lightning. I have also seen them kill worms and also
other Lithobius in the same way. When another Lithobius is wounded by the
poison claws it seems to be paralyzed behind the wound. The Millepedes, on
the other hand, have no such offensive and defensive weapon. They rely for
protection on the fluid secreted by the stigmata repugnatoria (or glandulae
odoriferae) mentioned before. This fluid has been shown to contain prussic
acid, and has a very unpleasant odour. Most of the Millepedes are provided
with these glands; but in the cave Myriapods mentioned before, the animals
have not to contend against so many adversaries, and these glands almost
disappear. Other Myriapods defend themselves by means of the long and stiff
bristles with which they are provided, e. g., the little Polyxvenus.

Bird enemics—Centipeds have been identified 236 times from the
stomachs of 65 species of nearctic birds, and millipeds 2,598 times
from 98 species. The latter were identified more than 50 times in the
case of each of 12 species of birds. The highest number of millipeds
found in a single stomach—that of a starling—was 40. More than a
tenth of the starling’s annual food in the United States consists of
millipeds.

Number of identifications, 2,862; percentage of identifications
among those of all Arthropoda, 1.3580; percentage of species in this
class among the whole number of arthropod species known, .4781.

Other enemies.—A. H. Wirkland in his report on the ‘‘ Usefulness
of the American toad”’’ states that Io per cent of the food of 149
individuals examined consisted of millipeds and that as many as 77
were found in a single stomach. Myriapods are eaten also by frogs,
salamanders, lizards, snakes and turtles. Among mammals the com-
mon mole (Scalopus), armadillo (Tatu) and civetcat (Bassariscus )
(and the mongoose as introduced into Trinidad) are known to feed
on centipeds, and Brewer’s mole (Parascalops) and the armadillo on
millipeds; shrews prey upon both groups. Centipeds are eaten by
predacious beetles, frequently prey upon each other, and it appears
that often the male is consumed by the female following pairing.
Millipeds are the chief food also of certain Lampyrid larvae, are
eaten by ground beetles, and are parasitized by phorid flies.

Discussion.—Vhere is a more or less prevalent belief that myriapods
are “ specially protected” animals. This idea is reflected in an article
on “The hothouse milliped as a new genus,” in which the author,

‘Farmers’ Bull. 196, U. S. Dep. Agr., 16 pp., 1904.

24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

O. F. Cook, says:* “ Prussic acid and other corrosive secretions

. .render . . . . the millipeds distasteful to birds and other animals
that might prey upon them.” This statement implies that millipeds
have no natural enemies, an Utopian condition that probably no
animal enjoys. In fact the evidence here adduced shows that milli-
peds and centipeds as well, have numerous effective bird enemies,
which together with special enemies in other groups, no doubt prey
upon them about in proportion to their availability. From the com-
paratively small numbers of myriapods and their secretive habits, it
could not be expected that they form a very high percentage of the
food of carnivorous animals. This reasonable expectation certainly
is fully satisfied by the showing here made of the activities of their
natural enemies.

InsEctTA (INSECTS)

From tabulations appearing earlier in this article, it will have been
noted that arthropods contribute more than 88 per cent of all the
records of the animal food of nearctic birds and insects more than
go per cent of the arthropods. To repeat the figures for the latter
group, insects furnish 190,919 identifications, which is 90.5891 per
cent of those of all arthropods. The percentage of species of the class
Insecta among the whole number of arthropod species known is
91.8589.

Not only are insects the most numerous class of jointed animals,
and the most important item of the animal food of birds, but they are
also the group about which most has been written in a theoretical way
as to protective adaptations (especially color) and as to the relation of
these adaptations to predatory foes. On all these accounts it is de-
sirable to discuss the insects in greater detail, certainly in most cases
by orders and in some instances by families. Tabulations have been
prepared, therefore, showing numbers of identifications by orders and
families, with their relative percentages. The first of these is a
distribution of the total number of identifications by orders.

The reader may have wondered why some of the tabulations as
to relative numbers of insects have not been based on the inventories
of some of the larger museums. However, this matter has been con-
sidered and the invalidating factor in such statistics is that such col-
lections are always more or less specialized either as a result of
policies of the museum or of the receipt of collections from special-
ists. Thus among insects, such favorite groups of the amateur as

* Proc. U. S. Nat. Mus., vol. 40, p. 625, IQII.

|
Tr |
Order pe | Beene
hysantira ......... | * 100 | * 700
@dondta ........:; : 705 16,642
MSOPteTa ss csc. ebsss 173 / 1+ 100,000
| Ephemerida ........ |
| Plecoptera ......... |
Gorrodentia ........ | ve | .
Mecoptera ......... a aan
| Trichoptera ........ |
Neuroptera ......... |
| Mallophaga ........ 472 | '1,250
Dermaptera ........ 180 1,098
Orthoptera .......... 2,556 25,988
Hemiptera ......... | * 3876 244,637
Lepidoptera ........ 30,653 275,920
MUIPUEPA, occ cess eas 10,253 210,880
Siphonaptera ....... *130 * 432
Coleoptera ......... * 32,500 * 738,000
Hymenoptera ....... | 17,638 493,757
Thysanoptera ....... 200 750
Strepsiptera ........ 159 414
|
BPOLAI! fees 4.0 serdns | 98,925 2,125,189

1 Estimated.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE

Percentage
of species
in this
group

1010
.7120
1747

6534

.1262
1818
2.5915
3.9147
30.9595
10.3555
fists
32.8250
17.8143
.2020
1605

25

Lepidoptera and Coleoptera are always copiously represented, while
such orders as Thysanura and Thysanoptera in most cases are
obviously neglected. However, one museum tabulation with percent-
age designations added, is herewith appended as a further demon-
stration (notwithstanding the defects just pointed out and the ap-
proximate nature of some of its figures) that multiplicity of species is
more or less closely correlated with abundance of individuals. This
table is adapted from one presented by Dr. J. M. Aldrich in the
Smithsonian Report for 1g19 (1921), p. 373.

Sununary of U. S. National Museum Collection, June 1919

Percentage
of specimens
in this
group

.0329
7821
4.7000

.0587
.0510
1.2214
11.4979
12.9082
9.9113
.0203
34.6860
23.2005
.0352
0194

SMITHSONIAN MISCELLANEOUS

COLLECTIONS — VOL. 85

Identifications of Insects

Number of

Order 1 identifications
APC A aera crn cxetorere trees 5
Odonata (further unidenti-

{LCC Been iare Cicer ee 2,082
LVS Optenas sence reece 245
Anisopteramereiecie noe 707
YAW (OWA) Soa cacnoccost 3,03
Agnathalarpciceriereincicie ai 484
Plecopteras bination tant 80
ESoptetar: tyes asi aie 129
Dermaptera eee ae oe 18
Cheleutoptera ............ 26
Diphtheropterawaesetoe ces 5,005
Orthoptera (Sens. str.)... 6,280
Raleoptera wee: ere ce 117
Dicty,optenape amram cter 58
Saltatoria (further uniden-

LIMEUM eects cee errors 6,450
Orthopteroidea (further

tinidentifled)\ meses 3590
All Orthopteroidea ...... 109,003
Corrodentiaiessericiceeaee 17,
Moallophagarere 2 sacs ciel 6
SIPHONAPtehAagecdectome sh I
Fleteropterae sacra bose 11,530
Elomopteram: qaee ree creme 5,215
Hemiptera (further uni-

dented) hit, sade cneineni 5,650
Allg ICHMCHOLC ae ects 22,395
Neuroptera (Sens. lat.)... 119
Megalopteray sh aecmeece es 167
Rhaphidioideai vs. csc -m92 «ae 54
Neuroptera (Sens. str.)... 108
Phryganoideay aeatraseccee 866
All Neuropteroidea ...... 1,314
Wepidopteta meres sect 18,487
Coleopteramaacas coer 85,322
iMecapteramemeenercenecte. 5
Da pteramtetca te svar metres 10,836
Fiymienopterarja-tees. «cette 27,02
Unidentified) i 52 sc0.0.4 00 2,676

Percentage of
identifications
among those
of all
insects

.0026

1.0905
.1283
-3703

1.5891
-2535
.O419
.0077
0094
.0136

2.9829

3.2893
.0613
.0304

3.3784

.1880
9.9534
.0089
0031
.0005
6.0392
2.7315

2.9594
11.7300
.0623
.0875
.0283
.0506
-4536
6882
9.0831
44.6899
.0026
5.6757
14.1551
1.4016

Percentage
of species
in this group
among the
whole number
of insect
species known

1691

2003
.3383
5986
1041
.0780
.OOII
.1301
.6507
.7809
.8589
3123
.2082

2.9410
.0780
-3383
.0130

4.9457

3.6442

8.5899
.0156
.O104
.3383
.3044
7187

15.6180
40.2032
.0260

11.4432
17.1798

‘The arrangement of the orders of insects in this tabulation is a compromise
among several systems.

There are no accounts given of Platyptera (embiids), Zoraptera, Notoptera
(grylloblattids), Siphunculata (body lice), Apocoleoptera (beaver beetles),
and Suctoria (fleas) because we have no records of bird enemies.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 27

Total number of identifications, 190,919; percentage of identifi-
cations among those of all arthropods, 90.5891 ; percentage of species
in this class among the whole number of arthropod species known,

91.8580.
APTERA (WINGLESS INSECTS)

Protective adaptations —The springtails and their allies appear to
have few adaptations such as are commonly called protective, their
defense being agility in some cases, and secretive habits in others.
Some species have coxal glands supposed to be repugnatorial.

Bird enemies.—While only five records of Thysanura are included
in the tabulations here reported upon others have been made since
and it seems probable that birds which feed about small pools, on
the quiet surface of which Collembola sometimes abound, or on the
edges of snowfields, will be found to pay due attention to thysanurans.

Total number of identifications, 5; percentage of identifications
among those of all insects, .0026; percentage of species in this
group among the whole number of insect species known, .1691.

Other enemies.—In reports of the Pennsylvania Department of
Agriculture and others treating the same groups of animals, seven
species of salamanders, four of frogs, and one toad are recorded as
feeding on Thysanura. Hamilton, reporting on 400 stomach con-
tents of young toads, says: “ Collembola comprised 6.2 per cent of
the diet. The springtails sometimes occurred in large numbers in
the stomachs examined, and together with thrips appeared to be
an important food of all small anurans” (Copeia, 1930, p. 45).
Forbes reports them being eaten by a Coccinellid beetle’ and a
small fish* (Labidesthes sicculus), Needham, by the brook trout,’
and Pearce by two species of fishes, a killifish and the mudminnow.*
They are known to be preyed upon also by aquatic hemiptera, and
are cannibalistic.

Discussion —Thysanura are chiefly minute insects, many of which
spend their whole lives in well-concealed places. The forms which
live more or less exposed appear to have enemies among animals
interested in such small morsels of food. However information on
the subject thus far is inadequate and no doubt will be increased by
more intensive investigation of potential predators.

* Bull. Ill. State Lab. Nat. Hist., vol. 1, no. 6, p. 52, May, 1883.
7 Op. cit, vol. 2, p. 525, 1888.

* Bull. 68, N. Y. State Mus., p. 205, 1903.

“Bull. U. S. Bur. Fisheries, vol. 35, p. 285 (1915-16), 1918.

28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

ODONATA (DRAGONFLIES, DAMSELFLIES )

Protective adaptations.—Dragonflies are fairly large, powerful,
predacious insects with remarkable ability for flight. They are
held in fear by illiterate people, a feeling possibly inspired by the
large mobile head consisting chiefly of eyes, and the strikingly con-
trasted color-pattern of many of them. A dark ground color with
vivid spots of green or yellow, answering to the description of warning
color, is common among dragonflies; some also have brilliant red,
blue, and metallic colors.

On the other hand, members of the order known as damselflies in
general are weak on the wing and of slighter and more delicate
structure. Some of them also are brightly colored but many are
dull. The immature stages of both dragonflies and damselflies’ are
aquatic, and predacious, and invariably inconspicuously colored.

Bird enemies.—It might perhaps be expected that damselflies would
be more frequently captured by birds than dragonflies, but this does
not seem to be the case, the determinations for these groups so far
standing at 245 damselflies and 707 dragonflies. However, 2,082
identifications do not indicate which suborder is concerned. About
200 species of birds are known to eat Odonata, and nymphs as well as
adults are freely taken. No fewer than 100-125 nymphs have been
taken from the gullet and gizzard of individual ducks, yellow-legs, and
magpies. Regarding the adults, Needham says: “It is doubtful
whether anything that flies is able to capture in flight one of the
swiftest dragon flies.” " However, we have records of birds eating
Epiaeschna heros, one of the largest and swiftest of the dragonflies of
the United States, and Anax junius, another of the giant species, is
commonly eaten by the pigeon hawk. No fewer than 28 individuals
of Anax were found in a single stomach of this falcon, and adult
dragonflies, mostly Anax junius, were found in 120 out of 181
stomachs of the species. In a lot of dragonfly wings, picked up under
the home of a colony of purple martins at West Chester, Pa., were
represented about 63 individual dragonflies, largely Epiaeschna heros,
but including also, Anax junius, Libellula pulchella, and Anax
longipes.

Number of identifications, 3,034; percentage of identifications
among those of all insects, 1.5891 ; percentage of species in this group
among the whole number of insect species known, .5986.

Other enemies.—Odonata are notoriously cannibalistic both in the
nymphal and adult stages. Diving beetles, water scorpions and other

“In Ward and Whipple, Fresh-water biology, p. 890, 1918.

*

Ty

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 29

aquatic hemiptera, salamanders, frogs, turtles, and many kinds of
fishes prey upon the nymphs. Ants, spiders, robber flies, chipmunks,
snakes, frogs, toads, and fishes feed to some extent also on adult
dragonflies, obtaining most of them no doubt when teneral. They are
parasitized by nematodes, mites, and flies.’

Discussion.—Odonata, both immature and adult, are freely preyed
upon by a variety of enemies and no special defense can be assumed
except the great expertness in flight of some of the dragonflies. This
we have seen does not foil various birds nor of course predators from
their own ranks. All in all it would seem that Odonata are preyed
upon fully in proportion to their abundance.

AGNATHA (MAYFLIES )

Protective adaptations-—The nymphs that live in water are plainly
colored; some cling closely to various objects in their environment,
while others swim in a rapid darting manner. The adults also are
usually inconspicuously colored.

Bird enemies.—Our tabulation shows mayflies to have been identi-
fied from the stomachs of 108 species of nearctic birds. A nighthawk
has been known to contain 400 adults at one time or many thousands
of eggs, the remains of the digestion of adults. As many as 250
nymphs have been found in a godwit’s stomach. Mayflies periodically
are exceedingly abundant and then are preyed upon by practically all
kinds of insectivorous birds. An interesting account of the behavior
of birds in the presence of a swarm of ephemerids is given by Dr. S. D.
Judd in his “ Birds of a Maryland Farm” ;* on this occasion 40
species of birds were observed eating mayflies. This list adds nine
to the species of birds known from stomach examination to feed
upon mayflies.

Number of identifications, 484; percentage of identifications among
those of all insects, .2535; percentage of species in this group among
the whole number of insect species known, .1041.

Other enemies.—Mayfly nymphs are eaten by the nymphs of stone-
flies and dragonflies, by water bugs, most fresh-water fishes, and to
some extent by salamanders and turtles; the adults are preyed upon
by fishes and adult dragonflies, spiders, toads, and bats.

Discussion—Mayflies are good food for predacious animals and are
eaten freely, so much so as to cause David Sharp to remark:* ‘‘ That

*For full discussion of dragonfly enemies, see Bull. U. S. Bur. Fisheries,
vol. 36, pp. 209-211 and 222-232 (1917-18), 1920.

* Bull. 17, U. S. Biol. Survey, pp. 22-24, 1902.

*Cambridge Nat. Hist., vol. 5, p. 442, 1910.

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, 85

insects so fragile, so highly organized, with a host of powerful
enemies, but themselves destitute of means of attack or defense,
should contrive to exist at all, is remarkable.” Doctor Sharp here
falls into the same error the selectionists often do, namely, of taking
the struggle for existence too seriously. While mayflies are a favorite
food of many predators, the evidence does not indicate that they are
eaten out of proportion to their numbers. They are also very fecund,
practically the whole body content of a female mayfly consisting of
the two egg masses. The annual occurrence of swarms covering the
foliage along streams (dating back as far as such things were re-
corded) is proof enough that enemies do not permanently reduce the
numbers of mayflies, and furthermore that the so-called defenses or
protective adaptations, of which mayflies are so nearly destitute, are
not essential to the maintenance of species in large, even overwhelm-
ing numbers.

PLECOPTERA (STONEFLIES )

Protective adaptations—tThe stoneflies are mostly plainly colored
but some are rather bright yellow; they are poor fliers but some of
them are said to emit a liquid from the basal articulations of the legs,
a performance usually classed as protective. The nymphs are aquatic
in habit, good swimmers, and obscure in color.

Bird enemies.—Stoneflies have been identified in the stomachs of
AI species of nearctic birds, usually in no very large numbers. The
total number of identifications is 80; the percentage of identifications
among those of all insects, .0419; and the percentage of species in
this order among those of all insect species known, .0780.

Other enemies ——Dragonfly nymphs prey upon those of stoneflies,
and a few fishes, salamanders, frogs, and turtles feed upon these
insects, either in the immature or mature condition according to
availability. Needham says: “‘ Hudson has demonstrated the im-
portance of stoneflies as fish food in the mountain streams of New
Zealand”? (Fresh-water Biology, 1918, p. 884), and Muttkowski
reports that 90 per cent of the food of trout in Yellowstone National
Park consists of them.

Discussion——The Plecoptera are a small group of insects of re-
stricted habitat, one we should therefore not expect to find preyed
upon extensively. They are eaten by various enemies, however, more
or less in proportion to their abundance, and the evidence does not
seem to indicate that special defenses of any kind enter into the
equation.

|
|
|
|
|
|
|
|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 31

ISOPTERA (TERMITES )

Protective adaptations —Termites pass most of their lives concealed
in galleries in wood or underground or in well-built nests. They have
strong jaws, a caste of soldiers especially well-armed in this respect,
and they emit a corrosive secretion. The color is usually yellowish
to brownish, but some species have the body reddish and the wings
dark, nearly black, thus having a coloration approaching that termed
warning.

Bird enemies.—Stomach examination has revealed termites in the
dietary of 38 species of nearctic birds. The occasions when termites
are available to most birds are infrequent, but when they come, the
insects usually are in great abundance. Accordingly large numbers
are eaten, and single stomachs have yielded as many as 215 termites
in the case of a nighthawk, 400 in that of a pileated woodpecker, and
| 1,100 in that of a flicker. The writer has twice observed numbers of
English sparrows gobbling up termites upon emergence and Hagen
_ has recorded’ a case in which 15 species of birds were in attendance
ona swarm of white ants, the robins among them so gorging them-
selves that their bills stood open.
| Number of identifications, 129; percentage of identifications among

those of all insects, .0677; percentage of species in this group among
| the whole number of insect species known, .ogIT.

Other enemics—Termites are as much sought after by some other
| animals as they are by birds and even are eaten by man. It has been
said that in the Tropics “ The flight of the winged termites is a great

event in the animal year.”’* In India cockroaches, frogs, lizards, rats,
bats, jackals, mongooses, jungle cats, and dogs have been observed *
preying upon them. In the United States, besides wild birds and
domestic fowls, salamanders, frogs, toads, lizards, spiders, centipeds,
crickets, robberflies, ants, and beetle larvae prey upon termites. The
insects have parasites also among the fungi, protozoa, nematodes,
and mites."

Discussion—The enemies of termites are comparatively well-
known, not wholly because they are numerous or active, but also
because termites are ‘‘ economic ” insects and have therefore been the
subject of considerable study from many points of view. Although

* Hagen, H., Proc. Boston Soc. Nat. Hist., vol. 20, p. 118, (1878-1880) 1881.

*Longstaff, G. B., in Shelford, R., Naturalist in Borneo, p. 37, 1916.

*Rothney, G. A. J., Proc. Ent. Soc. London, 1918, pp. Ixiv-lxvi.

*For an account of these miscellaneous enemies, see Snyder, T. E., U. S. Nat.
Mus. Bull. 108, pp. 116-118, 1920.

3

32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

they have protective adaptations of various kinds, termites are eaten
freely by numerous animals. Birds prey upon them eagerly when
occasion offers, but on the whole not out of proportion to the abun-
dance of the insects.

ORTHOPTEROIDEA
Percentage
of species
Percentage of in this group
identifications among the
among those whole number
Number of of all of nearctic
Group identifications Orthopteriodea Orthopteroidea 1 |
Winidentitiediee a seeeaeoe 6,800 35.8310 wae
Dermapteram ease eee 18 0047 1.8404
Paleopterayeacsaceeeoe ace 117 .6157 4.41904
Dicky Optenaauasmecteemietiac 58 3052 2.1136
@heleuitoptera We sa-n esos 26 .1368 1.4531
Diphtheroptera ss... 04 5,605 29.9688 69.3525
@xnthopterayyees sere 6,280 33.0472 20.7307

DERMAPTERA (EARWIGS) 4

All earwigs have pincer-like appendages at the end of the abdomen,
the function of which is little understood. One suggestion is that they
are for defense, but in what way they might serve for this purpose
is not clear. Many earwigs have glands producing a fetid secretion.
These insects in general are inconspicuous but a few have brilliant
colors. Earwigs are seldom met with in the United States and the
record of their bird enemies is short—18 identifications in the
stomachs of 15 species of birds. Percentage of identifications among
those of all insects, .o094; percentage of species in this group among
the whole number of insect species known, .1301. That this result is
merely a reflection of the infrequency of earwigs is indicated by the |
fact that in Great Britain where these insects are much more common,
the records of birds eating them are proportionately higher. Thus
Robert Newstead, treating of a mere fraction of our number of |
stomach examinations gives records for seven species of birds and |
notes that 23 earwigs were found in the stomach of a green wood-
pecker and 4o in that of a whimbrel. F. V. Theobald and William
McGowan in their report * on ‘' The Food of the Rook, Starling and
Chaffinch,” note that each of these birds prey upon earwigs, and

‘Computed from Scudder, S. H., Catalogue of the described Orthoptera of
the United States and Canada, Proc. Davenport (Iowa) Acad. Nat. Sci., vol. 8,
IOl pp., 3 pls., 1990. |

* Suppl. Journ. Board Agr. [London], vol. 15, no. 9, Dec., 1908. |

* Suppl. Journ. Board Agr. [London], vol. 15, no. 15, May, 1916.

A A SL OE gS est

nO. 7 PROTECTIVE ADAPTATIONS—McATEE 33

W. E. Collinge records* five species of birds as feeding on these
insects. Among other enemies of earwigs are batrachians, of which
6 species of salamanders, and 16 of frogs have been recorded in
the United States as feeding on Dermaptera. The earwigs are neither
an extensive nor an abundant group of insects and we should not
expect to find them preyed upon by insectivorous animals to any
marked degree.

CHELEUTOPTERA (WALKINGSTICKS )

Protective adaptations —As their vernacular names, stick and leaf
insects imply, these insects bear resemblances to objects in the vege-
table kingdom that have caused them to be considered as having
reached the very acme of protective adaptation. “Some,” says David
Sharp, “look like sticks, or stems of grass; some have a moss-like
appearance, while others resemble pieces of lichen-covered bark. The
members of the tribe Phyllides are leaf-like. A certain number
are covered with strong spines, like thorns. Some, if not all, of the
Phasmidae,” he adds, “ have the habit of ejecting a stinking fluid that
is said to be very acrid” (264). The eggs of walkingsticks are

peculiar in shape and sculpturing and many of them resemble seeds.

Bird enemics.—Records of walkingsticks in the identifications of
bird food here discussed total 26, and pertain to 18 species of birds.
The crow blackbird heads the list with seven captures.

Percentage of identifications among those of all insects, .0136;
percentage of species in this group among the whole number of insect
species known, .6507.

Other enemics.—Predacious hemiptera, mantids, lizards, and sper-
mophiles may be mentioned among the enemies of stick-insects, and
ichneumon flies are said to parasitize both adults and eggs.

Discussion.—The apparent discrepancy in the indices of frequency
of occurrence of stick-insects in bird food and in nature is to be
explained by the relatively poor representation of this group in the
United States, we having but 11 species. If we grant that the form,
color, and sluggishness of these insects has a protective value in
relation to predators, we must admit that these qualities facilitate also
the destruction of the walkingsticks by grazing animals, which engulf
indiscriminately huge mouthfuls of browse together with any insects
thereon that are not agile enough to beat an instantaneous retreat.
In the same way if the resemblance of the eggs to seeds is to be

*The food of some British wild birds, 1913.

* Cambridge Nat. Hist., vol. 5, p. 260, 1910.

""

34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

regarded as significant, it would appear to put these objects in
jeopardy, as the proportion of the food of birds and other animals
made up of seeds is immensely greater than that composed of insect
eggs. Judged as a protective adaptation, therefore, this case would
seem to fit the old adage of “out of the frying-pan into the fire.”
Two authorities who have paid special attention to the subject, how-
ever, conclude that the resemblance to seeds of these eggs has no
bionomic importance.’

SALTATORIA (GRASSHOPPERS, LOCUSTS, CRICKETS )

Owing to the facts that identifications in hundreds of cases were
not carried as far as they might have been, and that it is impracticable
to tabulate them by families, we put all the leaping orthoptera together,
rather than consider separately the two orders into which these forms
are usually grouped.

For convenience part of the tabulation of identifications is here
repeated in revised form.

Identifications of Saltatoria

Percentage of

identifications

among those
Number of of all

Group identifications Saltatoria

Diphtheroptera

(Grasshoppers, locusts) <.gct js cieels ease ore 5,005 30.3185
Orthoptera (Sens. str.)

(Green grasshoppers, katydids, crickets). 6,280 33.4328
Saltatoria

(Burther unidentified) ) os...sc.6e64- ee 6,450 34.3378
Orthopteroidea

(Further unidentified,no doubt Saltatoria) 359 T.QII2
AllSaltatoriarse crear er eo tree cen orris 18,784

Protective adaptations —A. R. Wallace says:* ‘‘ The whole order
of Orthoptera, grasshoppers, locusts, crickets, etc., are protected by
their colours harmonising with that of the vegetation or the soil on
which they live, and in no other group have we such striking examples
of special resemblance.”

With special reference to American insects, A. P. Morse makes the
following statement:* The coloration

is, with few exceptions, highly sympathetic in character, harmonizing with or
resembling very closely, often to a marvelous degree, the background of the

* Sharp, D., Willey Zool. Results, Cambridge, 1808, p. 75-04.
Severin, H. H. P., Ann. Ent. Soc. Amer., vol. 3, pp. 83-92, 1910.
* Natural selection and Tropical nature, p. 46, 1891.

“Proc. Boston Soc. Nat. Hist., vol. 35, no. 6, p. 244, 1920.

INO. 7 PROTECTIVE ADAPTATIONS—McATEE 35

insect’s environment. Earth tints, rock and sand textures, the infinitely varied
browns, greens, and grays of living and dead vegetation, yellow, orange, rose,
and silvery white are all represented in spots and streaks, the effect being to
merge the insect indistinguishably into its background while at rest, thus shield-
ing it in a very high degree from the observation of its foes. These colors
are of great protective value at the present time, natural selection continually
acting to preserve and perfect them, but though highly protective in character,
they are without doubt primarily due to physiological processes and influences
as yet imperfectly understood.

This type of coloration is admirably illustrated among New England species
by the Seaside Locust and Sand Locust which live on sandy backgrounds, the
Snapping and the Ledge-loving Locusts on rock habitats, the Coral-winged
and the Clear-winged Locusts in fields; and in the plant-perching species the
Pine-tree Locust with its background of lichened pine bark, the Red-legged
and the Two-striped Locusts among the yellowish green of herbage, and other
species of Melanoplus—M. mancus, M. fasciatus, etc..—whose darker tints
resemble those of fallen leaves from the Vaccinium thickets amid which they
live.

One who has not watched these creatures out of doors can appreciate to
but a slight degree the effectiveness of sympathetic coloring as a means of
concealment. Let him but try to pick out from its background immobile grass-
green Cone-head, leaf-brown Shield-backed Grasshopper, or any of the Locusts
just mentioned, and he will realize as never before the importance to the
defenceless insect of Mother Nature’s protective mantle of invisibility.

The wing-covers of certain katydids and allied forms are very
leaflike, the resemblance being carried so far in certain cases, it 1s
said, that the spots like those due to fungi and the tracks of leaf-
mining insects are closely imitated.

The leaping powers of the Saltatoria, remarkably developed in some
forms, must be classed as defensive; most of these insects have
powerful mandibles also, a few of them indeed being markedly
carnivorous.

Locusts of the subfamily Oedipodinae, especially, have another
adaptation some consider protective. For instance E. B. Poulton
says:* “ The brightly coloured hind wings of many moths (Catocala,
Tryphaena, etc.) and grasshoppers (Oedipoda, etc.) which flash out
conspicuously when the insect becomes active, and disappear equally
suddenly when it alights, probably serve, as Lord Walshingham has
suggested [Proc. Ent. Soc. Lond. 1890, pp. l-liii], to confuse a
pursuing enemy.” It may be noted that Morse considers these colors
as recognition markings.

Finally, among protective adaptations, certain Orthoptera are said
to mimic other insects, as for instance Membracidae, Phasmidae, ants,

‘Essays on evolution, p. 303, 1908.

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

and beetles, though we have none of these forms in the United States ;
the mole crickets and a few other forms have special fetid secretions,
and the brown drop that so many orthoptera exude from the mouth
when captured is said to be a protective device.

Bird enemies.—Nearly a tenth of all the identifications of insects
in bird stomachs are of leaping orthoptera. To name the birds that
eat grasshoppers is to name all birds not strict vegetarians. When
these insects are abundant, birds of all sizes turn their attention to
the Orthoptera and for the time being make them a staple food. Asa
constant article of diet also, they are important to many birds. The
number of identifications of Saltatoria from stomach contents was
50 or more in the case of over 20 species of birds, more than 100 in
22 additional species, more than 200 in Io other species, in excess of
T,000 in two cases, namely, of the common crow, and the meadowlark,
and more than 1,500 for the starling and crow blackbird. Expressed
in proportions of the annual subsistence of certain birds most fond
of the insects, we find according to Biological Survey records that
Saltatoria compose 21.29 per cent of the food of the western bluebird
(based on the examination of 217 stomachs), 22.01 per cent for the
eastern bluebird (855) ; grasshopper sparrow (170), 23 per cent; the
eastern and western meadowlarks combined (1,514), 26.08 per cent;
the Arkansas kingbird (109), 27.76 per cent; Franklin’s gull (93),
43.43 per cent; and the scissor-tailed flycatcher (129), 46.07 per cent.

These are illustrations of the relations of birds to leaping orthoptera
under normal conditions. When species of these insects become exces-
sively abundant as they frequently do, the gathering of the bird clans
to feed upon them is proverbial. No instance is more celebrated than
that studied by Prof. Samuel Aughey during an invasion of the Rocky
Mountain locust in Nebraska. He found locusts in the stomachs of
no fewer than 172 species of birds varying in size from the tiny
hummingbirds up to the largest hawks, and including such usually
exclusively vegetarian birds as the passenger pigeon and mourning
dove. Professor Aughey was eye-witness also to 33 additional species
of birds preying upon the locusts.’

For a modern illustration of the same phenomenon, we may cite a
brief investigation made by the Biological Survey during a grasshopper
outbreak in South Dakota in 1920. Out of the 26 species of birds
collected, representatives of 24 had been eating the hoppers; of 19
species every bird collected had taken grasshoppers, and for the

4 Notes on the nature of the food of the birds of Nebraska. First Ann. Rep.
U. S. Ent. Comm. (1877), App. I], pp. [13]-[62], 1878.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 37

species eating them, the insects composed from 40 per cent to 90 per
cent of the total food.

In another study of the effect of birds upon a severe irruption of
grasshoppers in California, H. C. Bryant estimated that birds were
destroying daily more than 120,000 grasshoppers per square mile in
the infested area.’

In countries, notably Africa, where migrations of large numbers
of locusts are of regular occurrence, various species of birds have more
or less specialized in following these flights and feeding on the mi-
grants, so much so, in fact, as to earn for themselves the name of
locust birds.”

Number of identifications of Saltatoria, 18,784; percentage of
identifications among those of all insects, 9.6506; percentage of
species in this group aniong all insect species known, 1.6308.

Other enemies —All stages of the Saltatoria are much sought for
by various animals. The larvae of Cantharid beetles and of bee flies
(Bombyliidae) subsist upon the eggs, as do also certain mites, and
the egg masses of some species are dug up and devoured by various
mammals, as moles, mice, spermophiles, and skunks. The nymphs
and adults fall a prey to vertebrates of nearly all sizes and descrip-
tions, ranging from bears, through coyotes, foxes, badgers, skunks,
civetcats, weasels, wood rats, squirrels, spermophiles, moles, shrews,
and mice, to lizards, tortoises, snakes, salamanders, frogs, and toads.
If any seek to escape their land enemies by jumping into the water,
they are snapped up by fishes. The adults are destroyed in large
numbers by parasitic diptera and hymenoptera. Most of the predatory
invertebrates are fond of grasshoppers, this being particularly true of
dragonflies, tiger beetles, ground beetles, robber flies, digger wasps,
and spiders. In the case of the latter, S. W. Bilsing found grass-
hoppers in 20 per cent of the webs of Epeira trifolium, in 35 per cent
of those of Argiope riparia, 44 per cent of those of A. trifasciata, and
in 53 per cent of those of Agalena naevia. Grasshoppers are parasit-
ized by nematodes and protozoa and are subject to bacterial and fungal
diseases, which last are said sometimes to destroy them “ in myriads.”

Discussion—tThe Saltatoria or leaping Orthopteroidea are promi-
nent in the insect world more through average large size and the

*Univ. California Publ. Zool., vol. 11, no. 1, pp. 16-17, Nov. 1, 1912.

* See Badenoch, L. N., True tales of the insects, pp. 127-128, 1899; La Baume,
W., Beihefte z. Tropenpflanzer, vol. 11, no. 2, pp. 65-128, Apr., 1910; Agr. Journ.
Cape of Good Hope, vol. 18, pp. 820-833, 1901; vol. 19, pp. 99-106, 165-171,
248-262, 1901 ; vol. 28, pp. 364-366, 1906; Trans. South African Phil. Soc., vol. 1,
p. 103, 1880.

38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

abundance of individuals rather than through abundance of species.
Almost everywhere in the United States that herbage is plentiful,
grasshoppers in the late summer rattle away from the approaching
pedestrian in such numbers as to form a veritable rolling barrage of
insect projectiles. No insects are more conspicuous in action, yet on
close examination the observer finds that the individual hopper is dull
and obscure in color. The point is worthy of attention because it proves
that the formula that abundant and conspicuous insects tend to be
warningly colored and inedible has numerous exceptions. None of
our grasshoppers of the northeastern United States are warningly
colored, unless the Oedipodinae with brightly and contrastingly colored
hind-wings, and in many instances a loudly rattling flight, may be so
considered. Whatever their status in adaptation theories such genera
as Arphia, Dissosteira, and Hippiscus seem to supply their full quota
to the food of birds and other predatory enemies. On the other hand
some of the “sympathetically” colored species mentioned in the
remarks on adaptations quoted from Morse are the very bread of
avian diet. Grasshoppers of the genus Melanoplus for instance were
identified 543 times among the records here considered, and were
found in the stomachs of more than 85 species of birds. These and
other Acridids are taken not only frequently but often in quantity, for
instance, the remains of no fewer than 123 specimens were found at
one time in the stomach of a common crow and 340 in that of a
Franklin’s gull. Judging from the records, the green grasshoppers or
Locustidae and the crickets also bear their appropriate burden of
predatory attack.

The imposing total of 17,641 identifications of Saltatoria, more than
a tenth of all insect determinations, shows what an important staple
for the birds these creatures are, and how poorly their prevailing
elaborately cryptic coloration succeeds in foiling their enemies. They
are preyed upon voraciously not only by birds but by a host of other
animals, but the effect of the attacks of predators, parasites, and
diseases together in no way suggests that the Saltatoria are a dis-
appearing race. Despite persecution, these insects abound and the
reasons are high fecundity and the great surplus of food available to
them ; these are substantial realities and outweigh immeasurably those
airy intangibilities classed as protective adaptations.

PALEOPTERA (ROACHES )

Protective adaptations——The comparatively few native species of
roaches in the United States are secretive and nocturnal in habit but
appear to have no other special protective adaptations. The introduced

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 39

species live chiefly in structures of man hence have little relation to
the indigenous fauna.

Bird enemies—Thirty-six species of birds share the 117 identifica-
tions of roaches in the food of nearctic birds. The number of records
was 10 or more in the case of four species of birds, and the number
of specimens eaten was as high as 10 in two instances but usually
was less.

Percentage of identifications among those of all insects, .0613;
percentage of species in this group among the whole number of insect
species known, .3123.

Other enemies.—Roaches seem to be more or less regularly eaten
by toads, frogs, the armadillo, spiders, rats, scorpions, and wasps.
They have specific parasites among the Evaniidae.

Discussion.—Owing to the poverty of the roach fauna of the United
States, research here is not likely to throw much light on relations of
these insects and their adaptations to predators. Tropical species are
said to resemble various other organisms, including isopods, myrio-
pods, longicorn, and coccinellid beetles, and hemiptera of the family

| Miridae. But since all of these models themselves are freely eaten
| by predators, the significance of the resemblances is hardly that usually
| attributed. In the United States natural enemies would seem to be
| proportional to the scanty population of native roaches.

}

|

|

DICTYOPTERA (MANTIDS)

Like the roaches, the mantids of the United States are few in
number and do not exhibit the unusual modifications displayed by
some of the tropical representatives of the group. The principal
defenses of our species must be their comparatively large size among
insects and their highly predatory nature. However, these character-
istics are of little avail against still larger predators and we find these
insects taken by birds in numbers probably bearing no distant relation
to the frequency of mantids in the country. Number of identifica-
tions, 58; percentage of identifications among those of all insects,
.0304; percentage of species in this group among all insect species,
.2082 (for the world, of course). The number of species of birds
~ concerned in the records here cited is 21. Mantids are eaten also

by lizards.

CORRODENTIA (PSOCIDS )
The Psocidae, which include the booklice seen in houses, are

delicate and minute insects. Many of the out-of-door species are
winged and the wings bear color patterns which may assimilate the

7

40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

insects more or less to the bark surface upon which many of them
dwell. These insects mostly below the size of food objects ordinarily
taken by birds were identified 17 times in the stomachs of nine species
of birds. In one case, that of a chimney swift the stomach contained
hundreds of specimens, gleaned no doubt from a swarm on the wing.

Percentage of identifications among those of all insects, .0089 ; per-
centage of species in this group among the whole number of insect
species, .0780.

MALLOPHAGA (BITING LICE)

The only opportunity birds have to get these usually minute insects
is to capture those parasitic on their own bodies, or in the case of
raptorial birds to engulf some with their prey. Apparently either of
these occurrences is rare; six records for as many species of birds
being all included in the present tabulation. Percentage of identifica-
tions among those of all insects, .0031; percentage of species in this
group among all insect species, .3383.

SIPHONAPTERA (FLEAS)

Only a single instance of a flea being eaten by a bird has thus far
come to light ; the opportunities for getting these small agile insects
must be very few since our native birds are parasitized by fleas to only
a very slight extent. That fleas are in no way distasteful (as food)
to some of their hosts is evident to anyone who has observed dogs,
monkeys, and other animals in their persistent and often successful
search for these pests.

Percentage of identifications among those of all insects, .0005 ; per-
centage of species in this group among all insect species, .0130.

THYSANOPTERA (THRIPS)

Protective adaptations—Some are contrastingly black and white
colored and the immature stages of many are red. It is doubtful
however if these colors have any warning significance. ‘The small size
and secretive habits of these insects doubtless are the most effective
factors in restricting predation upon them.

Bird enemies.—No identifications of thrips appear in the analyses
of the stomach contents of nearctic birds here reported upon.
Wetmore reports a thrips from the stomach of a hummingbird
(Anthracothorax viridis) from Porto Rico. (Bull. 326. U. S. Dep.
Agr. p:°73) 1916:)

Other enemies.—Thrips are eaten by small predacious hemiptera,
especially Anthocoridae, and egg parasites are known. Hamilton

i Ny ‘ ss
a ys

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 4!

(Copeia, 1930, p. 45) says of 400 young toads examined, “ Thrips
formed 10.1 per cent of the food, but were found in all but a few
stomachs. These small insects appear to be a staple article of diet
for young Bufo.”

Discussion——Thrips are too small for most birds to notice, but
considering our ignorance of the subject, the notes on enemies given
indicate that they have foes, the character and number of which,
probably as in other cases, are regulated largely by the factor of
availability.

RHYNCHOTA (BUGS, CICADAS, LEAFHOPPERS, SCALE INSECTS)

For the reason that the term Hemiptera in a broad sense was used
for about one-fourth of all the identifications of Rhynchota, it is not
practicable entirely to separate Heteroptera and Homoptera. However
the identifications of these groups are distributed as far as possible to
families in the tables presented. In using these tables, it should be
kept in mind that could the incomplete determinations have been dis-
tributed, the figures would average about a fourth higher throughout.

Protective adaptations—The popular expression ‘a nasty bug’
undoubtedly has reference, in most instances, to insects of this order,
many of which produce scents disagreeable to human senses. Theorists
have assumed these must also be repulsive to animal predators, a
doctrine briefly stated in the following quotation from E. B. Poulton:
“The Heteroptera (Hemiptera) are obviously, as a whole, a specially
protected group, commonly defended by taste or smell from large
numbers of insect-eating animals.” *

A great series of Heteroptera are more or less aquatic in habit and
thus are screened from the attacks of purely terrestrial enemies.
Some are very active, as the Saldidae and many Miridae; some are
said to be “ mimics,” as for example immature Nabidae resembling
ants and certain Reduviidae resembling wasps.

Mimicry, so-called, is exemplified among the Homoptera, also, as
some Fulgoridae are considered to resemble Lepidoptera in appear-
ance. The Membracidae with a host of bizarre forms, are thought to
present cases of mimicry to ants, and of resemblance to thorns and
seed pods of plants. One author further remarks: “ Evidently the
strong pronotal processes, which are often sharp and hard enough to
pierce the skin if the insect is seized suddenly, are unpalatable and
irritating.” “ Quoting Poulton again (op. cit., p. 4): ‘Allusion must

‘In Buckton, G. B., A monograph of the Membracidae, separate, p. 3, 1903.
*Funkhouser, W. D., Mem. 11, Cornell Univ. Agr. Exp. Sta., p. 417, June,
IQI7.

42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. Ser

be made to the special and curious defence by a waxy secretion which
is common in the Homoptera. The method may be compared to the
defensive silken walls of the cocoon in other insects, while the long
trailing filaments of wax borne by certain species of Homoptera may
play the same part as the ‘ tails’ on the hind wings of many Lepidop-
tera, or the ‘tussocks’ of hair on some of their larvae—all these
probably acting as directive structures which divert the attention of
an enemy from the vital parts.”

Many plant lice have the waxy filaments alluded to by Poulton,
while most of them exude special secretions from the cornicles, sup-
posed to be protective. Leafhoppers of various groups have been
thought to resemble color or structural details of plants they frequent,
and as for scale insects, their small size, waxy secretions and great
resemblance to the bark upon which they rest, have given them high
rank among the theoretically protected insects. Indeed they have been
thought well-nigh immune to attack and one author has intimated that
birds never eat scale insects. (Smith, J. B., Proc. State Hort. Soc.
N.. J: vol. 20; p: 90; 1904.)

Bird enemies.—Below are tabulations of the identifications of
Hemiptera in the stomach contents of nearctic birds followed by sup-
plemental comment. Comparative percentages are not given for the
plant lice, scale insects, and mealybugs as these have not been cata-
logued with the same degree of thoroughness as the other groups.

Total number of identifications of Hemiptera, 22,395; percentage
of identifications among those of all insects, 11.7301 ; percentage of
species in this order among those of all insect species known, 8.5899.

The Corixidae, although they spend practically all of their existence
in water and usually on the bottom, do not thereby secure immunity
from bird enemies.

Like other hemiptera, however, they are supposed to be specially
protected, one author saying:

As to the function of the stink-apparatus in the adult Corixa, there is no
need to look beyond defence. The insect frequently leaves the water, and it is
then exposed to all the dangers met with by the land Heteroptera. Also there
is no reason to doubt that the odoriferous secretion is equally efficacious against
enemies in water.’

Results indicate that this efficacy is nothing remarkable; indeed it
is a fallacy to suppose that so abundant and accessible a group does
not pay due toll to predators. The number of species of birds that

‘Brindley, Maud D. Haviland, On the repugnatorial glands of Corixa,
Trans. Ent. Soc. London, vol. 77, p. 13, 1929.

PROTECTIVE ADAPTATIONS—McATEE 43

Identifications of Rhynchota

|

Percentage
of species
Percentage of in this group
identifications among the
among those whole number
Number of of all of nearctic
Group identifications Rhynchota Rhynchota 1
Unidentified .............. 5,050 27.5325
Heteroptera (further uni-
dentified)) to..c 00.00.0608 389 1.8956 Sete
Scutelleridae ............. 187 -QII2 8673
MOGIGACE eis sock e cate wcs 232 1.1305 1.5077
Rentatomidae ........+0:. 5,582 27.2011 5.5037
Coreidae: 26-566. ssc caren c 305 1.9248 4.1361
PNTadidae® Gi hocaasdetu selene 15 .0731 2.0014
Neididae .2.........05-55. 7 0828 2668
VY SACIGAG “oc sc 5k. se esses 524 2.5334 2.8353
Pyrrhocoridae ............ 18 .0877 7338
Mingitidae .............:. 66 3216 2.3349
Enicocephalidae .......... ee ee 0667
Phymatidae .............. 19 1.0926 .4003
IREGUVIIdAG .oecce.. es. sss 633 3.4846 3.7692
BICDLIMAe 2.6.6. nsec nan ee 2 .0097 1334
| Mesoveliidae ............. 16 0780 0334
INADTGAG acces ai hoa 4 ne oes 163 7943 7005
| Gimicidae’ eae ces ats one 2 .0097 1334
| Amnthocoridae ............ 3 .0146 1.1007
Termatophylidae ......... a eee .0334
IMMTIGAE. isco oes. 0d ae oa. 518 2.5242 14.8101
Isometopidae ............. 1334
Dipsocoridae ............. .I001
Schizopteridae ............ or wie 0334
Hydrometridae ........... 22 .1072 0067
PREETIGdE 629% ocees ce ga ds. 228 I.11T0 6671
Veliidae ..............000. 32 -1559 6004
Saldidaes so.a. 6+ cece ciee se 74 3606 1.0674
Notonectidae ............. 327 1.5935 6004
Naucoridae .............. 306 1.4911 .4336
INGDICAE: chs-0. 0:0 0s o ecioe secs 40 .1949 2668
Belostomidae ............. 326 1.5886 .6671
| Gelastocoridae ............ 3 £0146 2001
| Ochteridae ............... a ae 1001
| Worixidae. 225... .02..--:- 4391 6.7783 2.0347
| Homoptera (further uni-
Gentine )...asescess can 107 5214 we
Oi rahdae..aiaisais ficewiete way 556 2.7004 2.7685
| Gercopidae ..scs..s.-0..0 6 102 .4970 8673
| Membracidae ............. 960 4.6781 6.2042
| rcadellidae’ 3... ... 22.6521 1,435 6.9927 25.0837
| Fulgoridae ............... 590 2875 IIl.9414.
| PS VMNIGAC 23. a)3.e 00,644 cece ma's-s 122 5945 4.7365
|
|
;

P

California Publ. Ent., vol. 2, 902 pp., 1917.

*Computed from Van Duzee, E. P., Catalogue of the Hemiptera of America
north of Mexico, excepting the Aphididae, Coccidae and Aleurodidae. Univ.

44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

feed upon them shown by the present tabulation is 85 and the number
of specimens taken at a meal ran over 200 in several cases, and up to.
1,300 in one instance (eared grebe). The Belostomidae or giant water
bugs, including the largest of North American Heteroptera, have
strong, grasping forelegs and a stout beak which readily pierces the
skin of man, making an aching, evidently envenomed wound. Not-
withstanding these characteristics they do not escape the birds. Fifty-
three species are on our list of captors, ducks, herons, and the like
preponderating ; in two cases, both herons, as many as IO specimens
were found in a stomach at one time. The Notonectidae again are
sharply biting and exceedingly active under-water bugs ; but the larger
types are eaten by no fewer than 44 species of birds, sometimes in
considerable number (30-57), while the little crawling and obscure
Plea were identified in 60 stomachs of 12 species of birds, in numbers
up to 40 in a single instance. The Gerridae, so very active on the
water surface, fall a prey to at least 49 kinds of birds, sometimes
being taken in considerable numbers (20-40). The Miridae or plant
bugs are agile and rapid in their movements and of great variety in
form and color, but corresponding with their abundance and wide
distribution, we find them preyed upon by 108 species of birds. The
Anthocoridae and Cimicidae, both odoriferous families, seem poorly
represented in our tables, but from their habits we should hardly
expect the latter to be found at all, while most Anthocoridae also live
largely hidden lives. We have found Nabidae in the stomachs of
52 species of birds, Emesidae in 10, and Reduviidae in 115; these
highly predatory forms therefore seem to have bird enemies about in
proportion to their abundance.

There are only a few species of Pyrrhocoridae in the United States
and none of them are abundant ; hence the 18 captures by nine species
of birds are perhaps not below proportion ; while the relations of birds
to the Lygaeidae shows again that an abundant and diversified group
is sure to be frequently taken by a large variety of birds. In this
family may be specially mentioned Myodocha serripes, a bug with an
extraordinarily long neck, for what purpose is unknown; at any rate
it is one of the most bizarre of the group in our area, but it is eaten
by more than 20 species of birds and no fewer than 27 specimens have
been found in a single stomach (purple martin). The large red and
black Lygaeus species were taken by 14 species of birds, and the
superabundant chinch bug (Blissus leucopterus), frequently observed
in prodigious numbers, by 29. Three species of birds, the bobwhite,
meadowlark, and brown thrasher, had records of a hundred or more
chinch bugs at a meal. These facts contrast strongly with the state-

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 45

ment that ‘‘ Very few birds prey upon it because of its repulsive smell
and taste. It is questionable whether any of them are fond of it.”
(Garman, H., Bull. 74, Ky. Agr. Exp. Sta., p. 56, May, 1898.) In the
series of Heteroptera composing the Coreidae and the groups aggre-
gated as the Pentatomidae or Pentatomoidea, we have the typically
stinking bugs. Practically all of them have powerfully scented secre-
tions usually of a character obnoxious to man, but it is not evident
that they are equally so to birds. Some of our Coreids (Thasus) are
too large for most of our birds to prey upon, but those of the next
rank in size are more or less freely taken, as Acanthocephala by 12
species, in numbers as high as 14-22 by Franklin’s gull; and the
various species of Leptoglossus by 16, 10-15 individuals at a meal by
the same gull; Alydinae, nearly as large and equally smelly, are preyed
upon by 21 kinds of birds. All Pentatomids are eaten so freely that
it is difficult to pick the most representative examples. However, to
begin, let us consider Podisus, a predacious, but nevertheless highly
scented genus; it has been found in the stomachs of 29 species of
birds, the most remarkable record being for a bird not included in
these tabulations, namely a black duck collected in Maine, which had
in its gullet alone 525 specimens of Podisus sereiventris.

One of the largest and most highly scented stink bugs of our fauna
(Acrosternum hilaris) was found in the stomachs of 37 species of
birds, in number up to 26 in one instance (purple martin), while for
our typical and most abundant genus ( Fuschistus) 62 avian predators
are known. The number of specimens found in a stomach exceeded
10 in a number of cases, and in one, that of a Franklin’s gull,
reached 175. The little Thyreocoridae, polished black with touches
of yellow on the costa, were found in the stomachs of 65 kinds of
birds, and the Scutelleridae in 60.

FE. A. D’Abreu in his report on “ Some insect prey of birds in the
(Rep. Proc. Third Int. Meeting, Pusa,

Central Provinces ” [of India
1919, Vol. iti, p. 866, 1920) says “ Pentatomids seem a favorite diet
with birds.” He gives notes also on bird enemies of 16 other families
of Rhynchota.

The only report on the food habits of birds in the American Tropics,
namely, the “ Birds of Porto Rico” (Bull. 326, U. S. Dep. Agr.,
1916), by Alexander Wetmore, in the accounts of the species through-
out shows Hemiptera to be taken in due proportion.

The Cicadidae are chiefly large insects, a factor which to some
extent must limit the number of their bird enemies ; however the list
here drawn upon shows 87 species and there are four records for one
of our smallest birds, the house wren. Some of the larger birds

40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

devour considerable numbers of the smaller cicadas, for example,
30 Okanagana rimosa were found in a nighthawk’s stomach and from
19 to 41 Tibicina septendecim in each of several crow stomachs. It is
of interest in this connection that adult as well as immature domestic
fowls have been killed by crop-binding due to eating too many cicadas.
(Weekly News Letter, U. S. Dep. Agr., vol. 6, no. 46, p. 14, June 18,
1919.) Wild birds, however, not only take large numbers of cicadas,
but feed on them steadily day after day when the chance comes. The
English sparrow and the crow blackbird are notable examples of this
and it has been concluded by entomologists that broods of the
periodical cicada issuing in parks and other places, where exposed to
concentrated attacks of these species, are doomed to extinction.
(Smith, J. B., Economic entomology, pp. 142-143, [1896]; Mar-
latt, C. L.,. Bull. 90, U.S. Dep: Ags. py 10, 1804.)

Our records do not show whether any immature Cercopidae
(spittle insects) are eaten by birds, but the adults are taken by 41
species. One chimney swift had eaten about 100 cercopids of the
genus Clastoptera. Despite the numerous defenses they are said to
have, Membracidae were eaten by no fewer than 136 species of birds
represented in the present tabulation and in numbers up to 26 indi-
viduals in a single stomach. They have been found in 15 or more
stomachs of each of the following species: Least, great-crested and
ash-throated flycatchers, wood pewee, meadowlark, Brewer’s black-
bird, Bullock’s oriole, English sparrow, cliff swallow, red-eyed, soli-
tary, and warbling vireos, bush-tit, and ruby-crowned kinglet. The
tree hoppers identified belong to 21 different genera, indicating that no
partiality is shown. Membracids with the most prominent horns and
spines of any in our fauna, such as those of the genera Campylenchia,
Platycotis, Thelia, Ceresa, and Platycentrus, are taken with the rest.
During stomach examinations 175 kinds of nearctic birds have yielded
leafhoppers (Jassidae sens. lat.) and 10 or more stomachs of no
fewer than 35 species have disclosed them. In a number of cases
from 20 to 50 leafhoppers were found in single stomachs and in one
case (barn swallow) a thousand.

The fulgorid fauna of the United States is scanty and our records
of birds feeding on these insects correspond. Beyond the fact that
they are well distributed through the various groups of the family
and pertain to 18 species of birds, there is little of special interest
concerning them. Some lesser yellow-legs had eaten from 50 to
400 each. The Psyllidae were found in the stomachs of 46 kinds
of birds and the Aphididae in 86. Cases are known in which the

*

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 47

former have been devoured very extensively by birds, an entire
orchard having been cleared of the pear psylla by nuthatches. (Zool.
Bull. Pennsylvania Dep. Agr., vol. 3, p. 79, July, 1907.) Plant lice
were found in large numbers in the stomachs of some birds, up to
200 or more in each of five species of the finch family, 300 or more
in three of them (pine siskin and two goldfinches), about 650 in the
stomach of a nighthawk and 1,600 in that of a wood duck. On a
200-acre farm in North Carolina, birds were found to be destroying
more than a million grain aphids daily. (McAtee, W. L., Yearbook,
U. S. Dep. Agr. (1912), pp. 397-404, 1913.) Aleurodidae have not
as yet been identified from stomachs of nearctic birds ; possibly some
may have been confused with scale-insects. The latter, notwith-
standing deprecatory statements that have been made relative to birds
as predators upon them, have been found in the stomachs of 88 species
of nearctic birds. No fewer than 100 Eulecanium cerasifex were
found in the stomach of a rose-breasted grosbeak, 300 Margarodes
in one of a scaled quail, 304 Saissetia oleae in that of a black-headed
grosbeak and 200, 700, and 800 of the same scale, respectively, in
three stomachs of the pine siskin.

Other enemies.—Salamanders, toads, and frogs are recorded in the
| Pennsylvania reports as feeding upon both Heteroptera and Homop-
' tera, as are also the common swift lizard (Sceloporus undulatus) and
_ the copperhead (Agkistrodon contortrix) and the hog-nosed snake
(Heterodon platirhinos). The same source credits five species of
| turtles with eating Heteroptera and one with devouring Homoptera.

Munz found that all the common frogs feed on Hemiptera about as

freely as upon any other insects, and Garman found bugs in 6 out of
20 stomachs of the common toad. Winton reports the Texas horned
|
|
|
.

lizard (Phrynosoma cornutum) as eating stink bugs.

Aquatic hemiptera, particularly Corixidae, are eaten by most fresh-
water fishes, while scattering representatives of the terrestrial families
are taken now and then as opportunities occur. Forbes records from
fish stomachs representatives of 14 families of Heteroptera and three
of Homoptera. Among mammals, the common mole is known to take
leafhoppers, chinch bugs, and other species; shrews do not entirely
neglect Hemiptera; the nine-banded armadillo devours Cydnidae and
Pentatomidae.

Insect enemies of Hemiptera include both nymphal and adult
dragonflies, the former getting considerable numbers of Corixidae and
the latter representatives of various families. Robber flies feed freely
upon Hemiptera, ground beetles and ladybirds devour them, and the

4

ai

48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Nyssonidae, Mimesidae, and Crabronidae, among predacious hymen-
optera, prey more or less selectively upon Homoptera; in the eastern
States a large Sphegid wasp is a special foe of cicadas. Other enemies
of cicadas include dragonflies, wasps, predatory beetles and bugs,
mantids, spiders, mites, hymenopterous and dipterous parasites,
fishes, snakes, turtles, squirrels, badgers, armadillos, skunks, moles,
and fungi. Spiders consume many Hemiptera of a wide variety and
are credited with being among the most important natural enemies of
leafhoppers. The latter insects are heavily parasitized by the Dry-
inidae, and by at least five other families of Hymenoptera, by Pipun-
culidae, and Strepsiptera; and are preyed upon by larvae of Chryso-
pidae, and by Coccinellidae, Reduviidae, and certain other insects.
The Pyrrhocoridae said to be specially protected are preyed upon
by spiders, pseudoscorpions, thrips (the eggs), tachinid flies, reduviid
bugs, and lizards. The Coreidae have special parasitic foes among the
Tachinidae; while the order of Rhynchota in general is subject to
hymenopterous parasites, the abundance of plant lice and scale insects
in particular depending to a large degree in many cases upon the
relative numbers of these destructive foes. Lycaenid caterpillars feed
upon aphids, coccids, jassids, and membracids. A page would scarcely
suffice to list the numerous enemies of plant lice which include, besides
parasites, coccinellid, lampyrid, syrphid, hemerobiid, and chrysopid
larvae, in addition to adult ladybird beetles, assassin bugs, and other
insects, mites, and spiders. Fungi are known to destroy, at times,
large numbers of hemiptera, among which may be mentioned plant
lice, scale insects, mealybugs, and the chinch bug.
Discussion.—Despite their malodorous secretions and other “ pro-
tective devices ’’ there can be no doubt that Rhynchota are taken fully
in proportion to their abundance by nearctic birds, and the evidence
is that their other enemies are numerous and effective. If we consider
the most pronouncedly repugnant species found in the United States,
such as the harlequin bug (Murgantia histrionica) and the squash bug
(Anasa tristis),’ we find that severe infestations of the former have
been kept in check by English sparrows (Sherman, F., Bull. North
Carolina Dep. Agr., vol. 32, no. 7; p. 21, July, 1911) and that the
squash bug has a number of deadly enemies. It has been shown that
the volatilized secretions of squash bugs if confined in a glass con-
tainer are capable of killing toads (Weed and Conradi, Bull. 89, New

‘It is worth noting that of these two common and exceedingly malodorous
bugs, one is warningly, one obscurely colored. Where is the correlation that
theories as to warning colors demand ?

————

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 49

Hampshire Agr. Exp. Sta., pp. 21-23, Feb., 1902), and the conclusion
was drawn that “ toads do not ordinarily devour many of these pests.”
Perhaps they do not “devour many ” of them, nor, with the whole
insect world available for them to prey upon, should they be expected
to specialize upon squash bugs, but they do eat them, as found by
Kirkland (Bull. 46, Hatch Exp. Sta., p. 26, 1897) and also by
Biological Survey investigators. Bird enemies also are not lacking,
present records showing six species of birds known to feed upon
Anasa tristis and four upon other species of the genus. The harlequin
bug is sometimes heavily parasitized also, while the squash bug has
both tachinid and hymenopterous parasites and is subject to a bacterial
disease.

Disregarding the “ protective adaptations”’ and reasoning alone
from the prevalence of hemiptera, there would be no presumption
that these insects would constitute a tenth of the food of any species
of birds, yet they actually do contribute 10 per cent or more of the
subsistence of the following 12 species in the United States: Nuttall
woodpecker (number of stomachs examined 53), percentage of
Rhynchota in the food, 14.76 per cent ; Scissor-tailed flycatcher (129),
10.17 per cent; eastern phoebe (370), 10.38 per cent; black phoebe
(344), 10.56 per cent; crested flycatcher (265), 14.26 per cent; least
mycatcher (177), 11.12 per cent; Bullock’s oriole (162), 10 per cent;
sharp-tailed sparrow (51), 12 per cent ; spotted towhee (139), 14 per
cent; purple martin (205), 14.58 per cent; barn swallow (467),
15.1 per cent ; and rough-winged swallow (136), 14.9 per cent ; more
than 20 per cent of the food of two birds, namely the black-headed
grosbeak (225), 21 per cent, and cliff swallow (375), 26.32 per cent,
and more than 30 per cent of the total diet of the violet-green swallow
(110), 35.96 per cent.

If the glandular secretions of hemiptera had the repugnatorial, not
to say dangerous, qualities attributed to them, there would be no such
wholesale preying upon them as is shown in the foregoing data.
Descending to milder forms of “ protection” as afforded by pointed
protuberances and secretions of wax, we find that the “ hardihood ”
of birds (from the selectionist point of view), or in other words their
tendency not to be bound by human criteria, is so great that such
devices simply do not count.

NEUROPTEROIDEA (DOBSONFLIES, SNAKEFLIES, SCORPIONFLIES, ANT-LIONS,
CADDISFLIES )
In the period during which the records of bird food here discussed
were obtained, the conception of the group of insects broadly termed
Neuroptera has gradually evolved from that of a catch-all for net-

50 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

veined insects to a restricted group of very few families. Hence
many of the determinations (in want of re-examination) cannot be
definitely correlated with modern classification. With little doubt,
therefore, there has been confusion of records between Neuropter-
oidea (in the present sense) and some of the orders elsewhere discussed
as Agnatha, Mecaptera, and especially Plecoptera. Hence the tabula-
tion figures given are only approximations, and a better conception
of the relations of birds to the insects would be obtained by lumping
all of the so-called net-veined insects together.

Protective adaptations—The Neuropteroidea have not been given
so much attention by adaptationists as some other groups of insects,
but certain supposedly protective features have been pointed out or are
suggested by analogy with other described cases. Dobsonflies are
large, the larvae and females have powerful biting-jaws, while in
some cases the males have enormously developed mandibles of less
sturdy construction, and the coloration of the wings of some presents
strong contrasts. The latter characteristic is possessed by the Myrme-
leonidae also, while their larvae, the ant-lions, have large, strong jaws
and an antlike odor. Some Ascalaphidae are said to resemble dragon-
flies both in appearance and habits; Chrysopidae have vile smells
earning them the name of stink flies; and Mantispidae not only have
predatory forelegs but are said to be protected by their resemblance
to Hymenoptera. (Poulton, E. B., Trans. Ent. Soc. London, 1902,
p- 536.) “ The well-known cases of Caddice-worms (Trichoptera)
are partly for concealment and partly for defence.” (Poulton, Colours
of Animals, p. 77, 1890.) Some of them resemble snail-shells.

Bird enemies.—While Neuroptera (sens. lat.) have been identified
from the stomachs of 56 species of birds there is little object in dis-
cussing further this heterogenous assemblage. Sialidae (dobsonflies
etc.), despite their average large size and biting powers, were taken
by 38 kinds of birds; 58 specimens were found in the stomach of a
Bonaparte’s gull and from 55 to 93 larvae in three stomachs of lesser
scaups and 192 in one of a canvas-back. Snakeflies, of bizarre ap-
pearance, and of limited distribution in the United States were
identified in the food of 22 species of birds; and Mantispidae, “ pro-
tected by their resemblance to Hymenoptera,” and also by considerable
rarity in our fauna, were found in the stomach of 11 species. Stink
flies (Chrysopidae) were eaten by 18 kinds of birds, and Myrme-
leonidae by 20. Ascalaphus was identified but once, quite in keeping
with its extreme rarity, and Hemerobiidae 11 times. The figures for
identifications are low for scarce or locally distributed groups, but

4
|
|
|
4
|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE SI

when we come to one of common and general occurrence, the corre-
sponding rise in frequency of capture by birds is apparent at once.
The caddisflies, more numerous in species and individuals than all our
other Neuropteroidea together, appropriately contribute nearly two-
thirds of the total number of records for the group. The number of
species of birds feeding upon them is 113, and of these 45 or more had
taken the “ specially protected’ larvae. The number of records of
caddisflies determined was 10 or more for 23 species of birds, and
more than 20, 30, and 40 in the case of four, three, and three species
respectively. The number of specimens taken by single birds exceeded
30 of larvae in a number of instances and ran as high as 207 (in a
scaup duck), and of adults reached such figures as 280 and 400 in
the case of the nighthawk.

Identifications of Neuropteroidea

Percentage
of species

Percentage of in this group
identifications among the
among those whole number
Number of of all of nearctic
Group identifications Neuropteroidea Neuropteroidea +
Neuroptera (sens. lat.).... 119 9.0504 sha
Megaloptera ....5...6.00% 167 12.7094 3.5648
Rhapidioidea ............. 54 4.1096 1.6886
Neuroptera (sens. str.).... 108 8.2192 32.2700
IPhryeanoidea ......+s0+5% 866 65.9061 62.4774

All Neuropteroidea ....... 1,314

Other enemies —Forbes reports that neuropteroid larvae compose
about 10 per cent of the food of the sucker and catfish families in
Illinois ; he found caddis larvae in the stomachs of 17 species of fishes.
According to various authors, these larvae are an important element
in the food of most kinds of trout. Salamanders, frogs, larvae of
stoneflies, and parasitic hymenoptera also are enumerated among the
enemies of caddis larvae. Forbes found larvae of Sialidae in seven
species of fishes, and these are known to be eaten also by frogs and
turtles. Chrysopidae have been seen to be eaten by frogs, salamanders,
and ants, and they have numerous hymenopterous parasites sometimes
destroying inmates of about half of the cocoons. (McGregor, E. A.,
Can. Ent., vol. 46, pp. 306-308, 1914.) Frogs are recorded also as
capturing Mecaptera, as are also lizards and larvae of ant-lions.
Robber flies and dragonflies apparently devour any Neuroptera chance
throws their way.

*Computed from Banks, Nathan, Catalogue of neuropteroid insects (except
Odonata) of the United States, 53 pp., 1907.

52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Discussion.—lIt is obvious from the available data on enemies of

Neuropteroidea that the small or rare groups have few, the large and
abundant families many foes, the result that would be predicted with
‘protective adaptations” discounted. The group most numerous in
species and individuals, namely the caddisflies, has the most enemies,
and their larvae, said to be well defended from enemies, form one of
the staple elements of the food of fresh-water fishes all over the globe,
as well as a favorite prey of aquatic birds.

LEPIDOPTERA (MOTHS, BUTTERFLIES )

Protective adaptations—In the space that can be devoted in this
paper to protective adaptations of Lepidoptera it is impossible to do
more than call attention to general aspects of theoretical considera-
tions, since what has been written on the subject would fill many
volumes. This flood of literature is due principally to the fact that
Lepidoptera have been regarded as the chief examples of the phe-
nomena of warning colors and of mimicry, subjects that have been
expounded and discussed at great length.

Warning coloration, it need hardly be stated, designates the con-
spicuous, often highly contrasted, patterns, which it is held may be
assumed with relative impunity by tough, distasteful, or dangerous
species. Batesian mimicry is the more or less pronounced resemblance
to these species by others supposedly less qualified to cope with the
struggle for existence, while Mullerian mimicry is mutual approach
in appearance by species all of which belong to “ specially protected ”
groups. As remarked in my 1912 paper, these theories were chiefly
built up at a time when there was almost complete ignorance of the
actual feeding habits of predacious animals, and attempts to secure
evidence on the subject by experiment were in most cases characterized
by a singular lack of appreciation of the vital factors involved and of
realities in nature.

The following statement by Alfred Russell Wallace gives the gist
of the principal nearctic instances of mimicry among Lepidoptera:
“In North America, the large and handsome Danais archippus with
rich reddish-brown wings is very common, and it is closely imitated
by Limenitis misippus, a butterfly . . . . which has acquired a color
quite distinct from that of the great bulk of its allies. In the same
country there is a more interesting case. The beautiful dark bronzy-
green butterfly, Papilio philenor, is inedible both in larva and perfect
insect, and it is mimicked by the equally dark Limenitis ursula. There
is also in the Southern and Western States a dark female form of the

NO. 7 PROTECTIVE ADAPTATIONS—-McATEE 53

yellow Papilio turnus, which in all probability obtains protection from
its general resemblance to P. philenor.” (Darwinism, p. 248, 1896.)

Mimicry of another order of insects, the Hymenoptera, is shown
by many of the clear-winged moths (Syntomidae and Sesiidae) as
adults; and of black-and-yellow ringed larvae, it is said they gain
great advantages from resemblance to the justly respected appearance
of hornets and wasps.

The majority of adults of Lepidoptera, especially the moths, exhibit
in greater or less perfection what is called cryptic coloration, that is
resemblance to details of the environment, exemplified by the species
that are inconspicuous on bark, old leaves and the like. This style of
protective adaptation also is attributed to many larvae and pupae. On
this topic Poulton says: “ There is no better instance of special pro-
tective resemblance than that afforded by the larvae of Geometrae,
‘stick caterpillars’ or loopers as they are often called. These cater-
pillars are extremely common and between two and three hundred
species are found in this country [Great Britain]; but the great
majority are rarely seen because of their perfect resemblance to the
twigs of the plants upon which they feed.” (Poulton, E. B., The
colours of animals, p. 26, 1890.)

This idea is pushed to an extreme by another author as shown by
the following quotation relating to the caterpillar of “a geometrid
moth. In the larval state the insect bears a very close resemblance
toa twig. Its habit of clinging to a real twig with its posterior ‘legs’
and allowing the body to swing out, adds to the illusion. The head
of the caterpillar resembles a leaf bud, while in color the entire
creature is an exact counterpart of a rough apple twig, the plant upon
which it naturally feeds. Thus complete immunity is secured from the
attacks of birds and all enemies which depend chiefly upon sight.” *
(Howes, Paul Griswold, Insect behavior, pp. 164-165, 1919.)

Adaptations of caterpillars supposed to repel enemies, which have
received the most attention from writers on the subject, include:
armatures of hairs or spines, repugnant odors, warning colors, and
terrifying attitudes, in addition to various special resemblances.
Among the latter, Howes considers especially remarkable those that
“rely for their protection upon their mimicry of the excreta of birds.
I have been completely fooled by these larvae on more than one
occasion. They frequently rest in the center of a green leaf and while
conspicuous, never suggest a living insect to the uninitiated. In color,
the upper and lower portions of the body are dark chocolate brown,

*See pages 56 and 85 for the facts as to immunity of loopers.

7

54 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

banded through the center with pure white, which suggests the lime
so often seen in the excreta of birds. The entire creature is highly
glossed, which gives a fresh and moist appearance to the object, which
makes no attempt to conceal itself, depending entirely upon its
strangely camouflaged body for protection.” (Insect behavior, p. 165,
TOLOS) 9

Poulton summarizes the purpose of caterpillar adaptations as fol-
lows: “In the remarkable. abundance and variety of methods by
which concealment is effected in Lepidopterous larvae, we probably
see a result of their peculiarly defenceless condition. .... Hence
larvae are so colored as to avoid detection or to warn of some un-
pleasant attribute, the object in both cases being the same—to leave
the larva untouched, a touch being practically fatal.’ (The colours of
animals, p. 51, 1890.)

On the concealment of lepidopterous pupae, the same author says:
‘Protective Resemblance, either Special or General, is seen in nearly
all exposed pupae, but most chrysalides are buried in the earth or
protected by cocoons. The cocoons are often sufficient defense,
because the silk is very unpleasant in the mouth; but such protection
only applies in the warmer weather when there is an abundance of
insect food. In the winter, insectivorous animals are pinched by
hunger, and would devour the pupa in spite of the cocoon. We there-
fore find that all cocoons which contain pupae during the winter are |
well concealed, either spun between leaves which fall off and become
brown, or hidden under bark or moss, or constructed on the surface
of bark with a color and texture which renders them extremely diffi-
cult to detect.” (Ops cit; pp. S1-52-)

Pausing only long enough to note the incorrectness of the statement
‘all cocoons which contain pupae during the winter are well con- |
cealed ” (witness those of Saturniidae, not to speak of the cases of
many Tineidae), we may pass to Howes’ more imaginative account.

We find, for instance, the chrysalis of a butterfly, a species of Vanessa. It |
hangs by a tiny silk-fastened stem under a protecting fence rail. Within the
shell of the chrysalis, there is nothing but a mass of disintegrating tissues, a
thick fluid, studded with globules of fat. It is neither caterpillar nor butterfly.
It cannot thrash about from side to side or make a demonstration, there are no_
spines to pierce a would-be enemy, no wings by which the creature might take
flight. It is as helpless now as so much custard, for the insect is in the process
of change from one form to another.

*This comment ignores the fact that a great many birds habitually devour |
the excreta of their young, even returning to it when accidentally dropped, and
this nestling excreta is exactly of the luscious appearance described by Howes.

a

NO: 7 PROTECTIVE ADAPTATIONS—McATEE 55

Such is the actual condition of the pupal butterfly, but let us examine its
outer covering. It is a frightful-looking object, armored, and covered with
sharp spikes between which beady false eyes peer out. It is absolutely harmless
but appears otherwise. To birds it is doubtless a thing to beware of, yet one
tiny puncture of its brittle covering would reveal a delicious feast within.

Many insects are thus protected, ones that could not compete in any form
of battle. They are given immunity from attack because they could not ward
it off themselves. In the case of the transforming pupa, some such form of
protection becomes a necessity. A butterfly in the making is as helpless as the
egg from which it sprung, so Nature resorts to camouflage to terrorize the
destroyers of her children. (Insect behavior, p. 168, 1910.)

Aside from the fact that Vanessa pupae do not enjoy immunity
(see p. 62), we may well inquire whether birds are not Nature’s
children just as much as the butterflies, and just as fully entitled to
be her beneficiaries ?

Bird enemies.—Identification to species especially has lagged more
in the case of lepidopterous items of food, than in those of any of the
other larger orders of insects, due chiefly to poor condition of the
remains of adults, and to lack of knowledge of larvae. Unidentified
Lepidoptera exceed 2.85 times those in some degree identified, and in
considering the relation of the percentages of identifications to those
of the number of species of various groups, the former figures should
be multiplied by 2.85.

In view of the very unsatisfactory distribution of identifications of
_ Lepidoptera to families (over 70 per cent of the whole number being
| merely as Lepidoptera), it would be of little avail to discuss the

relative importance of family groups as bird food. Rather it will be
better to treat the subject along lines of general interest already
developed, as the preference between larval and adult Lepidoptera,
the extent to which hairy caterpillars are eaten, and the relation of
birds to butterflies, the chief illustrations of mimicry theories.

The question as to which is eaten most extensively, adult or larval
Lepidoptera, is easily answered in favor of the latter. As the table
shows 68 per cent of all records of Lepidoptera are for larvae, further
unidentified ; moreover, it is certain that the great bulk of specimens
identified to families also were larvae. Thus the Noctuidae deter-

-mined were chiefly cutworms, the Geometridae were mostly loopers,
the Tineidae principally case-bearers, and so on. Caterpillars, not
further identified, were found from 50 to 100 times in the stomachs
of 23 species of birds; from 100 to 200 times in 21 species ; from 200
to 300 times in eight species (downy woodpecker, blue jay, red-winged
and Brewer’s blackbirds, warbling vireo, black-capped chickadee,
hermit thrush, and bluebird) ; from 300 to 400 times in two species

:

SMITHSONIAN MISCELLANEOUS COLLECTIONS

Identifications of Lepidoptera

Gelechtidde 445-6 elena
drortricidacwencts scien
Grambidaeweeeee neers
Pyralidae: Fess ccm ere
Sesiidaet tans hum. c oven
(Cossidaemeprer cee een ete

Psychiddensstraaus. ete ees
Geometridae ee ne
Bombycidae en) snes nee
[Easicampidae mse 2 hate:
[iparidae f.hcoraen eat
Notodontidae 4 2i2¢25..20.
INoctuidae tne ae iae ens
INGarIStidacumaiee recite
AT ctiidaeweree eee eee ene
Ceratocampidae- 5-45-42 -
Satunntidaeweee see eee eer
Sphingidae haere erates
Moths (further unidenti-

Hed) L aie wmech seen ors
ALLE VOLS oe eer eee
Elespeniidaem amen
Teyana ci daeaeemreertiee tree
Nymphalidae) .:2...-....+-
Pieridaes sce sce ae sien Oe
Bapilionidae seer mere
Butterflies (further un-

Ad entitled) eee reer
AU DULLCRTUVE St rtaiacioieis eee’
Lepidopterous eggs ......
Lepidopterous larvae .....
Lepidopterous cocoons and

chysalidestys ca -peeiies
Lepidopterous adults .....

Number of
identifications

499

149

38

oO
AN NH MN ND

174

118
1,128

68

29
156

2,116
4,083

Percentage of
identifications
among all
Lepidoptera

2008
0021
0021
.0626
.0071
.O160
.0008
.0021
.0008
.0008
.0306
" 0025
.0731
.0029
.0496
4742
.OO17
0286
.0277
.O122
.0056

.8896
1.9085
.0038
0004
.0248
.0004
.0008

0172
.0474
4233
5.3201

0954
2749

Percentage
of species
in this family
among
described
nearctic
species of
Lepidoptera 1

6.3575
2.9296
4.4548
7.2183
10.0573
1.5252
3I70
4983
-1963
12.2318
JOISI
3624
.2265
1.2533
32.1049
.2265
1.7668
1812
4681
1.2533

90.1541
2.0447
-0453
2.5218
9664
Bit

9.8459

*Computed from Dyar, H. G., A list of North American Lepidoptera, etc.,
U.S. Nat. Mus. Bull. 52, 723 pp., 1902.

VOL. 85

|
t
(
{
|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 57

(red-eyed vireo and robin) ; and more than 400 times in the following
five species: crow (438), starling (727), meadowlark (474), crow
blackbird (600), and English sparrow (466). One hundred or more
caterpillars further unidentified were found in single stomach con-
tents of each of the following birds: sparrow hawk, downy wood-
pecker, hairy woodpecker, black-billed cuckoo, yellow-billed cuckoo,
crow, starling, crow blackbird, hermit thrush, wood thrush, and robin.
A very characteristic phase of the destruction of caterpillars by birds
is their use as a special food for the young ; numerous species of birds
make a practice of feeding the young a very much higher proportion
of caterpillars than is taken by the adults.

It has often been asserted that hairs and spines are very effective
in protecting certain caterpillars from birds." Bastin for instance,
states that “stinging hairs defend their possessors from almost all
birds except the cuckoos.” (Insects, their life-histories and habits,
p. 168, 1913.) These claims ignore the fact that birds are véry well
equipped with relatively insensitive bills and feet for removing spines
and hairs from larvae if they choose. Some birds do this, others
actually dissect caterpillars, eating parts they want from the inside,
piecemeal. Hairy and spiny armature is no bar to birds with such
feeding habits, and, furthermore, do not seem to be of any great
service in relation to numerous birds which swallow entire larvae thus
defended. A characteristic statement about hairy caterpillars is:
“Tent caterpillars have few enemies..... Our two species of
Cuckoos make it a regular business to feed upon these worms which
no other birds will eat.” (Lugger, Otto, Fourth Ann. Rep. Ent. Minn.
(1898), p. 142, 1899.)

Seventeen of the species of birds included in the tabulations on
which this paper is based had eaten tent caterpillars or the eggs from
which they hatch; numbers of larvae taken at a meal ran up as high
as 200 in case of the black-billed cuckoo, and of eggs as high as 1,047
in that of a blue jay. Compiling records from the reports of ento-
mologists and others who have found birds feeding upon tent cater-
pillars, we get a list of 43 species of bird predators upon the so-called
“Orchard” species (Malacosoma americana) and 32 upon the
“Forest” species (IM. disstria). Caterpillars even more offensively

*In the case of this as in other similar claims, we may, well ask why such
theoretically effective defenses have not been developed by a larger proportion or
in fact by all larvae? The most cursory consideration of the subject shows
that hairiness of caterpillars is in the main a phyletic character. A few related
families include the great bulk of the hairy larvae.

*

58 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

hairy than these are those of the gipsy and the brown-tailed moths ;
the hairs of the latter species especially cause a troublesome and
painful rash upon the skin of man. Nevertheless 46 kinds of birds
are known to eat caterpillars of the former species and 31 those of the
latter. (For a valuable article on bird enemies of these and other hairy
caterpillars, see Forbush, E. H., Bull. 20, n. s., U. S. Div. Enum
pp- 85-93, 1899. ) |
The larvae of the tussock moth (Orgyia leucostigma) are supposed
to be especially distasteful to birds, but Forbush records (Mass. Crop
Rep., July, 1900, p. 36) nine species of birds as feeding upon them.
The writer*has observed English sparrows and robins eating them; |
in the spring of 1921 in Washington, D. C., the larvae were quite
common and robins were feeding freely upon them, carrying them
to their young, I believe, as it was a common sight to see the birds |
flying with the white tufts showing at the tips of their bills. |
Again records of birds feeding on fall webworms (Hyphantria

|

\

fextor) are relatively scanty, only six species being named, yet careful
observation in the field has proved one of them to be a very effective
foe. Dr. C. Gordon Hewitt informs us that of the various factors
operating in the reduction of this insect in Nova Scotia in 1916, the
red-eyed vireo was most important and “it was estimated that about
40 per cent of the larvae had been destroyed in the webs by this bird.”
(Rep. Dominion Ent. 1917, p. 8.) A later report shows an average |
destruction of 68 per cent. |

It would not be necessary to refer to the preying of birds upon
smooth caterpillars, a thing universally done, except for theoretical dis-
quisitions as to the “protected nature” of certain groups. The
Geometridae, loopers or measuring worms, are said to be protected by
resemblance to twigs, etc., a statement made without giving due weight
to the fact that such a defense depends upon immobility whereas these
caterpillars must be in motion the greater part of the time while
searching for and devouring food. Forty-four species of birds are
recorded as feeding upon Geometridae in our tabulation and numbers
of specimens as high as 20 were taken at one meal by the starling and
go by the robin.

Larvae of the Sphingidae are said to be protected by their “ horns ”
and by “ terrifying attitudes,” but 44 species of birds covered by the
present investigation do not seem to agree with the theorists on these
points. Ten species are known to prey upon Dielephila lineata alone,
and in field observation the crow has been known to clear tomato
patches of the hornworm (Phlegethontius sexta).

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 59

Many printed pages have been devoted to discussion of the question:
“To birds eat butterflies? ’”’ but the natural answer: “ Certainly, but
probably not out of proportion to their abundance,” seems not to have
occurred to the disputants.

At this point it will be well to say a word about the alleged difficulty
of identifying adult Lepidoptera, especially butterflies in the stomach
contents of birds. But for this, some argue, the number of records
of butterflies eaten would be much larger. The assumption is made
that the scales are necessary to identification, and since they are so
easily rubbed off, determination will usually be impossible. This ob-
jection serves mainly to exhibit the ignorance of its proposers relative
to the analysis of the contents of bird stomachs. In the first place
when adult Lepidoptera have been eaten at all recently, that fact is
evident to the practiced eye, even unaided, on first glance at the
stomach contents. A characteristic fuzzy, felted appearance, due to
the distribution of the hundreds of scales throughout the mass, tells
the tale at once. Even after digestion is far advanced the scales do
not disappear because they are so numerous and stick to everything,
and they are evident under magnifications used in the analysis of
practically every stomach contents. Moreover were all scales absent,
it would be possible to unroll the wing membrane, if swallowed, and
examine the venation; the antennae also would usually be present ;
and the form of the head, thorax, and body, which are characteristic,
could be made out.

In addition we would remind the reader that all things found in
birds’ stomachs are not ground to a powder. Just the reverse in fact
is true; birds feed more or less constantly, and whenever shot they
will as a rule just have swallowed some article of food which, of
course, will be in good condition for study. In the long run all con-
stituents of the food will be found nearly or quite intact in the
stomachs in proportion to the frequency in which they are taken.

However it is unnecessary to discuss the matter further. One need
only consider the extent to which we have identified certain insects
far more fragile than butterflies, as mayflies (Ephemeridae) 484
records, midges (Chironomidae) 1,003 records, and crane flies (T1-
pulidae) 1,565 records, to be assured that there is no likelihood what-
ever of a butterfly being overlooked during careful stomach analysis.

Of the 113 records of birds eating Rhopalocera included in the
present tabulations, 24 refer to larvae and two to chrysalides. It is
worth noting that one of the larval records was for Anosia plexip pus,
two for Papilio species, and six for l’anessa species, supposedly the

60 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

best protected forms. Nymphalidae and Hesperiidae are most numer-
ously represented among the adults taken.

The 87 records of imago butterflies are distributed among T5 species
of birds, but all save 18 of them pertain to a single bird, the pigeon
hawk. The specimens of this hawk examined were taken on their
southward migration at a point that is in the migration path of butter-
flies also, so that opportunities for catching these insects were at the
best. (It is worth noting here that dragonflies, swallows, swifts, and
bats also using this same migration track were preyed upon by the
pigeon hawk.) In this case as in many others the abundance and
availability of prey is shown to have great influence upon the choice
of food by birds. Amid the butterflies, this hawk preyed upon them;
elsewhere we have no record of its doing so. Clearly the other birds
(14 species) in whose stomachs butterflies have been found (18
records). are only occasional predators upon them. This is only what
would be expected, for ordinarily butterflies, numerically, are no
considerable part of the insect fauna; when under extraordinary cir-
cumstances they do become over-abundant they are more frequently
devoured by birds. Thus Bryant found Brewer’s blackbird eating
large numbers, and three other species of birds smaller numbers, of
Eugonia californica during an unusual outbreak of the species. (Con-
dor, vol. 13, pp. 195-208, Nov. 1911.)

Summary of identifications of Lepidoptera: Total number 18,487;
percentage of identifications among those of all insects, 9.6831 ; per-
centage of species in this group among the whole number of insect
species, 15.6180.

Other enenues.—For the most part fishes are only casual devourers
of Lepidoptera, getting chiefly larvae which fall into the water, most
of which would perish anyway. However, gamy fishes such as trout
snap up adults that incautiously fly near the surface of the water.

Bullfrogs have been observed feeding freely on Papilio turnus
adults (Mallonee, Science, 1916, pp. 386-387) and half a dozen leopard
frogs have been noted as eating 500 Argynnis aphrodite in a week
(Shiras, Nat. Geogr. Mag., 1921, p. 174). Kirkland found cutworms,
tent and other caterpillars to compose 28 per cent of the total food of
149 toad stomachs examined by him, and Munz found lepidopterous
larvae in stomachs of four species of frogs. In 209 leopard frogs,
Drake found one imago, one chrysalid, and 121 larvae of Lepidoptera.
Surface reports remains of Lepidoptera in stomachs of eight species
of salamanders, one toad, and nine frogs. In the Tropics lizards are
said to be important enemies of adults of this order and our lizards

- a 2 amen ieaeredeiedeenetie
$$ c rere

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 61

are known to eat both larvae and imagos. Surface records Lepidop-
tera from the stomachs of five species each of snakes and turtles.

Cutworms are commonly taken and other caterpillars and chrysa-
lides are devoured to a smaller extent by moles. A number of small
mammals, such as opossums, spermophiles, ground squirrels, tree
squirrels, prairiedogs, grasshopper mice, skunks, raccoons, shrews,
armadillos, the mongoose, and Nasua feed more or less regularly on
caterpillars, and take an occasional pupa or imago. Bird has observed
that field mice and skunks are effective enemies of the gall-making
larvae of Papaipema. (Can. Ent., vol. 41, no. 2, pp. 67-68, Feb., 1909. )
Haskin has reported squirrels devouring large numbers of Aelitaea
chalcedon adults (Ent. News, vol. 27, no. 8, p. 370, Oct., 1916), and
Attwater found wings of several hundred Danais archippus that had
been eaten by the Texas grasshopper mouse (Bull. Amer. Mus. Nat.
Hist., vol. 6, p. r81, 1894). Bats catch moths, and monkeys also have
been reported as eating butterflies commonly. (Trans. Ent. Soc.
London, 1912, p. iv, XVii-Xviil.)

The insect enemies of Lepidoptera also are numerous and some ot
them are exceedingly destructive. Robber flies and dragonflies are
frequently observed devouring adult Lepidoptera, and a Natal col-
lector considers Mantidae the chief enemies of butterflies. (Proc. Ent.
Soc. London, 1906, p. lii.) Spiders catch them directly or trap them
in their webs, Phymatidae lie in wait for them, and predacious beetles
sometimes capture them. However, the latter predators are more
serious foes of caterpillars, in the pursuit of which they have as
fellows numerous wasps. Ants, chrysopid larvae, and other insects
and mites feed upon the eggs; and parasites often destroy large pro-
portions of the eggs laid. Parasites of lepidopterous larvae also are
legion, including numerous species of Hymenoptera and Diptera, and
they take a large toll from every generation of the insects. Exceed-
ingly high percentages of parasitism have frequently been observed,
reaching locally in a few cases even to 75 and 100 per cent. It has
been found in one case at least that no fewer than 63 species of
hymenopterous parasites attack a single species of moth. (Cambridge
Nat. Hist., vol. 5, p. 521, 1910.)

Bacterial diseases frequently kill large numbers of caterpillars and
sometimes locally extirpate certain species.

Discussion.—It is one thing to record a proved fact, but quite
another to assert that a certain thing does not occur in nature. Our
stock of verified data stands as an imperishable record and addition
to it, not subtraction, is the rule. Let none be tempted therefore to add

62 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to the vast body of speculation that has proceeded from slight premises
by data in a preceding paragraph apparently indicating that birds do
not eat many chrysalides of butterflies. For in that case we must refer
him to Bryant’s statement that about 15 per cent of the pupae of
Eugonia californica at a time when they were very abundant showed
evidences of attack by birds (Condor, vol. 13, p. 200, Nov., 1g1r),
and to Chittenden’s that “in one case it was found that during the
winter the number of pupae of the cabbage butterflies was reduced
more than go per cent by birds feeding upon them.” ( Farmers’
Bull. 766;,°Ur S: Dep. Ars p: ©; TOr6s)

In this paper we cannot possibly discuss all of the data relating to
predators upon insects and other animals, but the evidence we present
in our tabulations surely goes far to prove that no groups are neglected
by predators (except as availability or sheer size dictates) and that
the various groups are preyed upon more or less in proportion to their
numbers. As applied to Lepidoptera this rule is apparent in the
greater number of records for such large families as the Noctuidae,
Tineidae, and Tortricidae for instance as contrasted to such smaller
ones as the Sphingidae, Arctiidae, and Bombycidae or of the more
numerous Nymphalidae to the less numerous Papilionidae. Due to
the high proportion of unidentified Lepidoptera, our tables are not as
complete and informing as could be desired, but where there are ap-
parent exceptions to the rule of proportional loss to predators, data
from other sources usually indicates unreliability of the apparently
negative evidence. For instance the records of Geometridae in our
table seem too low for this rather important family which is un-
doubtedly numerous in individuals. But that this is due solely to the
make-up of our material is proved at once by reference to the litera-
ture; no fewer than 73 species of nearctic birds have been observed
feeding on cankerworms (Paleacrita and Alsophila) for instance.
Wellhouse, who reports finding cankerworms in 98 of 100 stomachs
of birds (36 species) collected near Lawrence, Kans., in 34 of which
they composed the total food, says: “ Probably no insect is a favorite
food of more species of birds than the cankerworm larva.” (Bull.
Univ. Kansas, vol. 18, no. 1, p. 301, Oct. 1917.) In a study of birds
in relation to cankerworms in Illinois, Forbes found these larvae to
compose 45 per cent of the food of a collection of 55 birds (15 species)
and that one species, the cedarbird, was destroying them at the rate
of at least 90,000 per month. (Forbes, S. A., Trans. Illinois State
Hort. Soc. (1881), pp. 123-130, 1882.)

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 63

The snow-white linden moth (Ennomos subsignarius) has a typical
twiglike caterpillar, but several entomologists have testified that it was
practically exterminated in cities by the English sparrow. (See
Merrick, G. W., Bull. 286, Cornell Univ. Agr. Exp. Sta., p. 62,
Nov., 1910.)

The only other important family of moths which our tabulations
might indicate to be neglected by birds is the Pyralidae. With little
doubt this condition is due either to the larvae not being recognized
er to our stomach material not being fully representative. Certainly
birds are known to be enemies of our pyralid larvae, a little search
revealing records of avian predators upon Lowvostege similalis, L.
sticticalis, Pilocrocis tripunctata, Pinipestis simmermanm, Diatraea

-saccharalis, Acrobasis nebulella, and Pyrausta nubilalis. Five species

are known to feed on the last-named, the corn rootborer, while of
Loxostege sticticalis, the beet webworm, it is recorded that: “‘ Insect-
eating birds devour the worms in large quantities. Where the worms
were abundant [in Colorado] ... . blackbirds were attracted in
flocks of thousands and in several instances . . . . the worms were all
cleaned out of fields by them in the course of two or three days.”’ (Gil-
lette, C. P., Bull. 98, Colorado Agr. Exp. Sta., p. 10, Mar., 1905.)

These instances emphasize the universal scope of the predatory
activities of birds; in general the enemies of economic species of
insects are better known, and fully discounting the fact that they are
most studied, this is only another way of saying that the most abun-
dant species have the most numerous enemies.

COLEOPTERA (BEETLES )

Protective adapiations —While more pages have been written about
warning colors, mimicry and the like in Lepidoptera, which insects
furnished the inspiration for this line of speculation, the important
and extensive order of Coleoptera has been far from neglected and
perhaps the most positive statements of all have been made regarding
the “protected”? status of some of its members. In conclusions
derived from G. A. K. Marshall’s data on ‘“ The Bionomics of South
African Insects” (Trans. Ent. Soc. London, 1902, pp. 393-584),
Prof. E. B. Poulton in discussing the chief specially defended
Coleoptera mentions: “ The groups about which there seems to be no
doubt at all—conspicuous, constantly refused by insect-eaters, and
hable to be mimicked by other Coleoptera are the following: FEroty-
lidae, Coccinellidae, Malacodermidae, including the Lycinae, Lam-
pyrinae and Telephorinae, Melyridae, Cantharidae, Chrysomelidae,

5

64 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

Endomychidae, and Pyrochroidae.”” The Cleridae are cited as a family
that while undoubtedly distasteful, in forming color associations take
the colors and patterns of other insects “rather than impress the
’ The following four
‘at any rate partially distasteful ” : Scarabae-
idae, Cetoniidae, Tenebrionidae, and Lagriidae. The longicorns are
thought to include a few distasteful species in addition to many that

stamp of their own likeness on the assemblage.’
families are said to be ‘

mimic aculeate Hymenoptera and other specially defended insects.
Cicindelidae are said by Wallace to be protected by cryptic coloration,
the refuge of the weak, while Poulton and Shelford have recorded
them as models mimicked by species less prepared for the struggle
for existence—a tribute to the strong.

“The Carabidae are a powerful specially defended group,” writes
Poulton (op. cit., pp. 513-514) “and it is of advantage to be recognized
as belonging to the group, even though it is no doubt of still greater
advantage to be mistaken, as may happen at a distance, or on a super-
ficial view, or during rapid movement, for the still more formidable
Mutillidae and ants”’..... “ Dr. A. R. Wallace has always thought
that the extreme hardness of the mimicked Curculionidae and An-
thribidae is the character which protects them.” (Poulton, op. cit.,
PP. 522-523.)

Comment of this kind could be cited indefinitely, for something or
other has been claimed to be “ special protection ”’ for practically every
group of beetles. It is undesirable and unnecessary to cite this matter
in detail, but some attention should be given to the subject of repug-
natorial secretions which has figured considerably in accounts of
protective adaptations of beetles. For convenience, a summary of the
occurrence of such secretions is quoted from a recent article on the
topic:

“It has been well understood that the presence of defensive or
repugnatorial scent glands in certain insects exists in direct adaptation
to the needs and habits of their owners and in close response to their
environment ; also that such glands are of very frequent occurrence
and with much variation as to position, form, and function; and that
their presence is of value to the insect for repellent, defensive and
warning purposes. .... Biologically speaking, the principle involved
in such cases, though often modified, is practically identical with that
of the mephitic, sulphuretted, oil-like fluid ejected by the skunks. Thus
far anal glands are known to be present in the following families of
Coleoptera: Cicindelidae, Carabidae, Dytiscidae, Gyrinidae, Staphy-
linidae, Silphidae, and Tenebrionidae. The blood itself serves as a

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 65

repellent fluid in the Meloidae, and in the Coccinellidae and Lam-
pyridae, and it issues from a pore at the end of the femur as a yellowish
fluid. The cantharidin in the blood of some species of Meloidae,
commonly known as ‘ Spanish Fly,’ forms an especially caustic pro-
tection against birds, predacious insects and reptiles.” (Wade, J. S.,
Notes on defensive scent glands of certain Coleoptera, Psyche, vol. 28,
nos. 5-6, p. 146, Oct.-Dec., 1921.)

Bird enemies.—It is worth pointing out that about 15 per cent of
all the determinations of beetles were not carried to the family, and
consequently that the percentages for the various families should be,
on the average, about a seventh larger than shown in the tabulation.

Identifications of Coleoptera

Percentage
of species
in this family
Percentage of aeomaued
identifications nearctic
Number of among all species of
Family identifications ! Coleoptera Coleoptera 2
Gicindelidae ......-....... 649 -7606 6146
Carabidae Josie cs nds eeas 15,887 18.6200 11.6730
| EATMPMIZOIdAG 23.5 0e0es208 jes ae Aono ens
| Omophronidae ........... 16 (Olay) 0808 (15)
| Ip Idae.* <akeswecweseas’ 363 4254 .2210
ID tISCIMAC. ey cic cess 4 cowie es 1,729 2.0264 1.79054
MV PAMIGAC een oaig waieie Sa eas 64 .0750 .0808
Hydrophilidae ............ 2,418 2.8340 1.0244
Platypsyllidae .......00.. eee Shi .0053 (1)
Brathinidae’ .....0%. 6. << ae tnd .O16I (3)
IBeptinidae o5...4 samesuecaes ee ree O16 (3)
Silphidae ...........2000: 409 4794 .7386
lambidae ..2.:...-...0.- wir ah: 0323 (6)
Scydmaenidae ............ vee a .9381 (174)
Orthoperidae ............. 4 .0047 .3073 (57)
Stapnylinidae .../....e2.60% 1,605 1.8811 14.8163
Pselaphidae .............. 3 .0035 1.9140
Clavigeridae ...........0. aoe ae 0161 (3)
IM OAC acc ecw haan tana ds ae on 4690 (87)
Sphaeritidae ............. nae ae 0053 (1)
Colydiidae ............... 5 .0059 .4520
Murmidiidae ............. ee Ka 0269 (5)
Monoedidae .............. ee oie 0053 (1)

* There is an omission of 737 records of Carabidae and 574 of Chrysomelidae,
enough to make more than 1.5 per cent of the total of beetle records.

*Computed from Leng, C. W., Catalogue of the Coleoptera of America
north of Mexico, 470 pp., 1920.

*The number of nearctic species in the family.

66

SMITHSONIAN MISCELLANEOUS COLLECTIONS

Identifications of Coleoptera—Continued

Family
athridudae hence ee oe
Mycetaeidae
See
Bhalacridacmns-eeeicree cent
Coccinellidae’ =. 22... eee ees
Allecnlidaeteeor.t os eee
Menebrionidaew mse eeaeceo
IDSKepaRGRYS sono pone boceoer
Mionomidaea- eee eee
Melandryidae.-.)ae cn eee
Ptinidae
Anobiidae }
Bostruchidaceesasaceeee cee
ley ctidacmene ec eirraee
Sphindidaemeass see seers
Gisidae, sence oe a aaeienee

Lucanidae
ee
Cerambycidae {eases aoe
Chrrysomelidae) 22..-...02-
Mylabridaey. scone cmisen ces
Scaphididae 2 see aac e ae
listeridae’ sana seenem eer ite
Lycidae

Lam.pyridae

Bheieodiinan ere
Telephoridae
Melyridaeleensscce seers
Cleridae i
Corynetidae |
@rthniidaewee ence cree
levinexylidaemeatcse eee ee
Telegeusidae .............
Micromalthidae ..........
Cupedidae’ sac. ss cece.
Cephaloidae. =. ss... ee ae
@edemeridae ~22ssss- 02.6
Mordellidaeweccae een
Rhiphiphoridae ...........
IMeloidaemsnermetiosane once
Wevialitidae secmeise ecco
Pythidaewarerscts nie cere
Pyrochroidaec,cm.c tk ser ¢

Number of
identifications

16

879

38

ONS:

34

279
10

Percentage of
identifications
among all
Coleoptera

.0187
.OO12

.0316
1.7053
.0398
2.5749
.0094

0457
£0508

.0281

.0047
15.5317
1418

1.8577
6.6407
.O551
0094
1.2459

1.0302

0445

0410

.OOI2
.0023
.O105
0398
.0023
.3270
.O1I7
.0070
-0047

VOL. 85

Percentage
of species
in this family
among
described
nearctic
species of
Coleoptera

.5007
.2048

.6308
1.9517
6685
6.1411
0916
.0323 (6)
4367

1.4550

3288
.0862 (16)
-0323 (6)
-4582
5.3701

1724

6.0548
5-2515
5014

2.0704
1.5095

1.7307
1.1806

.0269 (5)
.0107 (2)
0053 (1)
.0053 (1)
0215
0431
2641
7656
-1347
1.2239
O61
.0916
-0593

NO. 7 PROTECTIVE ADAPTATIONS—McATEE

Identifications of Coleoptera—Continued

Family

Pedilidae |
Anthicidae
Euglenidae |
Cerophytidae
Ceb.ionidae
Plastoceridae
Rhipiceridae
Bilateridae .4......secea0 303
IMielasidae: sic ..cnise sa dec 0s
Throscidae ...............
Buprestidae ..............
Psephenidae
PDEVOPIdaG ees fake oan
Helmidae
Heteroceridae ............
(GeOmysSIdae: cn. aaiewevete si oer
Dascillidae )
Riedie ere
Chelonariidae ............
Dermestidae .............
Biyiethidaes —scialeleiie esses.
PROV SSOCICAG 6 ag a5 <i cewe cs
MSiOmiddae..<aisaeces daecucs
Nitidulidae ...............
Rhizophagidae ............
Monotomidae .............
GiGi idaes ens sa- 2 eee 2
HeGOLY IGA: civ..me,2 <6 « sores
Derodontidae ............
Cryptophagidae ...........
Mycetophagidae ..........
Brentidae ...............-
Platystomidae ...:.......
Belidae 2... cc ccc ccc ee cess
@urcilionidae s.... 8.0...
Platypodidae .............
Scolytidae ...............
Water beetles (further uni-
dentified) ..............
Rhynchophora (further
unidentified) ...........
Beetles (further unidenti-
HEC) We tive fake arco See es yarts
Beetle larvae (further uni-
dentified) ..............

Number of

identifications

NI

143
17

190
B12

Percentage of
identifications
among all
Coleoptera

0716

.0082

5.2612
.0047
.0070
7759

.0387

.1676

67

Percentage
of species
in this family
among
described
nearctic
species of
Coleoptera

T53L1

1939

3.1056
-3073
-1347

2.0434

.3072

-0593
.0107 (2)

3288

.0053 (1)
.6955
.5220
.0215 (4)
.3450
LL,
-0754 (14)
.1941
.4582
3828
.0269 (5)
.7278
1725
.0323
3342
.0053
9.9153
0215
2.0488

68 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The tiger beetles “are avoided on account of their ferocity ”
(Bastin, Insects, their life-histories and habits, p. 151, 1913), and have
been referred to as “ dreaded insects ” (Poulton, Colours of animals,
p. 252, 1890), but what creatures capable of feeling dread so regard
these beetles is unexplained ; certainly the facts indicate they are not
birds. The 649 records included in the present tabulation are dis-
tributed among 99 species of birds. Eight species have 10 or more
records each, two others, the eastern meadowlark and eastern kingbird,
over 20, the crow more than 60, and the crow blackbird 94. No fewer
than 25 larvae of tiger beetles were found in a single stomach of an
eastern bluebird, and 156 adults in that of a sparrow hawk and 164 in
that of a long-billed curlew. If tiger beetles ever evade attacks by
birds it is by celerity of motion rather than by any special defenses.

With respect to Carabidae or ground beetles, Forbes in his report
on the food of thrushes may have given some comfort to protective
adaptation theorists when he said: ‘‘ We note, however, a remarkable
deficiency of the highly colored genera—such as Galerita, Brachynus,
Lebia, Platynus, Chlaenius, etc., which are either absent, or found but
rarely in these birds’ (thrushes, bluebird) food. Evidently these more
showy beetles are protected by some more effective means than ob-
scurity of color.” (Forbes, S. A., Bull. Illinois State Lab. Nat. Hist.,
vol..t; no. 6, p. 57, May, 1883:))

However, this statement is but another instance of the danger of
generalizing from insufficient data. In the study of the food of birds
and other animals we are always adding to the list of species eaten
and to the number of times they are taken ; the movement is never in
the contrary direction. We are constantly finding enemies of forms
previously held to be more or less exempt, and usually to an extent

which more than compensates for previous lack of knowledge on the .

subject.

In the present instance such progress in knowledge since Forbes’
study is indicated by 535 records of the capture of Chlaenius by
41 species of birds, 254 for Platynus by 55 species, 44 records for
Galerita by 13 species, 39 for Lebia by 21 species, and eight for
Brachynus by seven species ; figures more or less in harmony with the
relative abundance in individuals of these groups. In this connection
it may be well to note also F. H. Chittenden’s remark that Lebia
grandis “is protected by its warning color from rapacious birds.”
(Farmers’ Bull. 1020, U. S. Dep. Agr., p. 16, 1919.) Six of the 39
Lebia records here cited are for grandis, and the writer submits that
six records for this single species out of a total of 85,322 for all

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 69

beetles (18,548 nearctic species) fully satisfies expectations based on
the relative availability of the species to birds.

Species of Agonoderus, much more common than Lebia but just as
contrastingly colored, contribute 188 records to our tabulations and
were eaten by no fewer than 57 species of birds. From ro to as many
as 50 specimens had in several instances been taken at a meal. There
are 57 records for Casnonia, a small genus of “long-necked” dis-
tinctly “ warningly-colored ” beetles in stomachs of 14 species of birds.

Even black, alone, the predominant color among Carabidae has been
held to have a warning value, but Amara, Anisodactylus, Harpalus,
and Pterostichus, chiefly typically black species, are eaten by the
hundreds. There are 445 records for the powerful Pasimachus (80
individuals in one crow stomach), and 497 for the species of Calosoma
which are not only large, but some of which have contrasting blue
margins, others fiery spots, and all powerful, ill-scented excretions.
In fact, it is everywhere evident that the special defenses alleged for
the Carabidae are more in the nature of pleasing fictions for theorists
to speculate upon than practical reliances for the beetles concerned.
Eloquent is the fact that between a sixth and a fifth of all determina-
tions of beetles in the stomachs of nearctic birds are of Carabidae.

The Haliplidae, all of which have “ warning colors,’ and the
Dytiscidae and Gyrinidae, said to be protected by anal glands, all seem
to be preyed upon in proportion to their abundance. The Silphidae
quite generally have nauseous excretions and include numerous species
with distinct warning colors, but it is the latter forms such as Necro-
phorus with 102 records and Silpha with 213 that most evidently are
eaten in due proportion. The apparent falling of records of this
family below the index of frequency must be attributed to the smaller
and rarer species with more concealed habits being overlooked, rather
than to the larger familiar ones enjoying immunity on account of
alleged special defenses which they possess in the highest degree.

That the Staphylinidae is the family most numerous in species, and
probably therefore of individuals, among all Coleoptera is a fact not
realized by the average collector. It has been brought out only by the
accumulated research of generations of coleopterists, and its lack of
obviousness must be attributed to the secretive habits of so many of
these small or even minute beetles. Most of them spend their lives
chiefly under cover of various kinds, for example, in fungi, in leaf-
mold, under bark, in old logs, and in ant nests, and it must be on this
account that the records of birds capturing them are not very much
more numerous, rather than that they are disliked. In fact the 1,605

- determinations for them proves they are not disliked, and these

70 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

records are shared by more than 160 species of birds. Fifteen of these
kinds of birds had more than to records each, nine others more
than 20, six others from 30 to 60 records, one other, the chimney
swift, 76, the crow 190, and another, the starling, more than 200.
In several instances the number of specimens found in a stomach was
as many as from 20 to 50 and larger numbers were 85 for the bald-
pate and 150 for the dowitcher. Such data certainly do not indicate
distaste for Staphylinidae, hence the failure of the total number of
captures to come up to theoretical expectations must be due to some
other factor, presumably the small size and concealed habit of living
characteristic of so large a proportion of the beetles of this family.

The same causes also serve to explain why a number of the minor
families of beetles have not yet been identified in bird stomachs ; the
Platypsyllidae, and Leptinidae are parasitic upon mammals, the Scyd-
maenidae, and Clavigeridae mostly live in ant nests, the Ptiliidae are
minute, while the others most of which have five or fewer species in
our region owe their degree of immunity to their very rarity.

Passing now to one of the larger groups of beetles about whose
protected status “there seems to be no doubt,” namely the malaco-
derms, variously regarded as forming from one to four families, we
find that they are devoured in no mincing way by nearctic birds.
While various authors refer to these beetles (generally known as
Lampyridae in the United States) in terms varying from distasteful
to inedible or immune, our records show 879 determinations of them
from bird stomachs. All of the groups were preyed upon, the Lycinae
and Phengodinae least, however, because they are scantily represented
in our fauna. The adult lampyrids identified were eaten by no fewer
than 108 species of birds and the larvae by 25. Larvae in number up
to 50 were found in a bluebird’s stomach, and in three instances as
many as 100 were taken from a single stomach of the robin. Our
most common lampyrids are Chauliognathus and Telephorus. The
former genus was identified 179 times in the stomachs of 34 species
of birds. Three of these had from 30 to 38 records each and the
number of individual beetles eaten ran as high as 30 in a single in-
stance. Telephorus (Cantharis) were determined 274 times in the
stomachs of 35 species of birds; the number of imagines in a stomach
ran as high as 16 and of larvae, 100. If the Lampyridae fail in any
degree to attain proportional representation among the food items
taken by nearctic birds it is due to the nocturnal habits of a large
number of the species. The diurnal species seem to be captured as
frequently as would be expected.

|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE Fu

The Melyridae (Malachiidae) are poorly represented compared to
the Lampyridae, yet upon inspection of the records it does not seem
that they are really avoided. Six genera and at least 10 species of
these beetles were identified ; 21 species of birds had eaten them, and
for one of these birds, Say’s phoebe, there were eight records of
feeding on Collops. Identifications of the Cleridae again include
numerous (21) species distributed among an equal number of species
of birds. One of these birds, the red-eyed vireo, had eight of the
records. In our experience Cleridae occur chiefly scattered and in
small numbers, a type of distribution with which the records of birds
preying upon them seem to harmonize.

Of the Histeridae, Donisthorpe says: “All the species of this
family are protected by their oval shape and hardness. They also
‘feign death.’ ” . the “species which are spotted with red, are
probably protected by their resemblance to Coccinellidae.” (Trans.
Ent. Soc. London, 1901, p. 354.) The prevailing color in this family,
1. e. black, has also been said to have a warning significance. Our
records show 1,063 identifications of Histeridae representing 116
species of birds; they are very freely eaten by some of these birds,
the number of records per species exceeding 20 in the case of at least
12 kinds, and the number of specimens eaten at a meal running up
to as high as 200 as a maximum.

The family of blister beetles ( Mylabridae, Cantharidae, or Meloidae
as it is variously known) is especially noted for the presence in the
bodies of its members of a vesicant poison, cantharidin, of which as
small a quantity as one grain has proved a fatal dose for a human
being. Bastin says of them “the blood contains cantharidin, an
extremely caustic substance, which is an almost perfect protection
against birds, reptiles, and predacious insects.” (Insects, their life-
histories and habits, p. 167, 1913.) While these beetles are supposed
to enjoy a very high degree of protection from natural enemies, 47
species of birds included in the tabulations here discussed had fed
upon them. Seven of the species had 10 or more records apiece of
preying upon blister beetles, the eastern kingbird having no fewer
than 77. In some cases from 12 to 16 specimens of cantharids were
found in single stomachs and a maximum of 31 in the case of a
magpie; more than 30 species in all of these beetles were identified.

Pyrochroidae are said to be another specially defended group of the
first order, but in view of the fact that there are only 11 nearctic
species of the family and they usually rare, we believe that the four
records of our birds capturing them are as many as could be expected.
One of the birds eating Pyrochroidae, namely a hairy woodpecker,

Jz SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

must have had unusual luck in order to obtain the 12 specimens it
contained.

Donisthorpe says “ The Elaters ‘ feign death’ and their ability to
‘skip’... . is no doubt of great use to them. Mr. Holland points
out that many of them possess a colour and shape suggesting the ap-
pearance of bits of dry brown stick.” (Trans. Ent. Soc. London, 1901,
p. 360.) Over four thousand (4,489) records of these beetles being
eaten by nearctic birds show that the protective devices mentioned are
of no particular account. There would appear to be no doubt whatever
that birds feed upon Elateridae whenever available to them.

The larvae of Buprestidae live in wood, and the adults have hard
chitin and metallic or other brilliant coloration, but since there are
more than 650 records of their occurrence in the stomach contents of
nearctic birds, it is certain that concealment of the larvae rather than
color protection is their main defense. Heteroceridae or mud beetles
certainly seem well concealed to the human eye but the records indi-
cate they are taken fully in proportion to their abundance. Dermes-
tidae, said to be protected because they are carrion-feeders, are taken
freely considering their availability in nature. Byrrhidae are thought
to be excellent examples of cryptically defended insects. “ The legs
and antennae are packed close to the body, fitting into cavities for
their reception and the beetles then represent rabbits’ dung, or little
lumps of earth; they in no way suggest the appearance of living
beetles.” (Donisthorpe, Trans. Ent. Soc. London, 1901, p. 357.)
However 312 records for them show American birds are not especially
deceived by the alleged protective devices.

It is unnecessary to comment on every family, but coming to the
Erotylidae we have a group which though small in numbers is said
to be one of the most highly protected groups. However, in the United
States, insects of this family in general do not have the bold habits
supposed to be associated with warning colors; in fact most of them
feed concealed in fleshy fungi. Correspondingly most of the determi-
nations of beetles of this family are for the species which live exposed
as Languria, for which there are 10 records, probably all that should
be expected for a single small genus. Similarly the Endomychidae
are protected by feeding inside of fungi or on fungi growing on the
under side of logs rather than by their “ warning colors.” It should
puzzle selectionists to explain why these and other brightly colored,
supposedly distasteful insects have such retiring habits that their
“ warning coloration ” is seldom displayed.

Contrasting these elusive beetles with another brightly colored but
decidedly not secretive group they are supposed to mimic, the Coc-

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 73

cinellidae, it is easy to see what factor makes for greater depredations
by birds ; it is none other than the frequently mentioned “ availability.”
Endomychids and Erotylids are red and black or yellow and black
beetles, less abundant and much more retiring in habits; while coc-
cinellids with the same colors are more common and live unconcealed.
The former are relatively seldom captured, the latter are freely eaten.
No better example of the influence of availability in guiding choice of
food by birds could be desired. This despite universal acclaim of
coccinellids as specially defended insects. “All the lady-birds are very
gaily colored” says Donisthorpe. “They boldly walk about with-
out any attempt at concealment, as do also their larvae. Both their
larvae and pupae are very brightly spotted. The distastefulness of the
perfect insects was proved* by Jenner Weir, and has since been con-
firmed by both Poulton and Wallace.” (Trans. Ent. Soc. London,
1901, p. 354.)

Packard states that “ The Coccinellidae are . . . . protected by a
yellow mucilaginous disagreeable fluid oozing out of the sides of the
thorax,” (Journ. N. Y. Ent. Soc., vol. 3, p. 116, 1895), and Wallace
says: “ The Coccinellidae or lady-birds are another uneatable group.”
(Darwinism, p. 234, 1896.) Let us see. The total number of records
of coccinellids in the food of nearctic birds is 1,455 and these are
shared by 127 species. Twenty-seven kinds of birds had to or more
records each, nine of which ran over 50, and three over 100. Not only
is the effect of availability noted in birds eating more coccinellids than
other similar but less abundant and conspicuous beetles, but its
influence is evident in at least two other ways, namely that leaf-feeding
birds, as warblers and vireos, get the most ladybird beetles, and that
in California where coccinellids are notably more abundant than they
are in the eastern States, a larger number of birds feed upon them and
they get a great many more of the beetles. The largest numbers of
coccinellids found in individual stomachs were 12 and 18 taken by
English sparrows, 13 by the summer warbler, 14 by the warbling vireo.
and 15 by the valley quail.

We now come to the consideration of three families (the Scara-
baeidae and Cetoniidae being reckoned as one) which Poulton says are
“at any rate partially distasteful.” Regarding one of these fami-
lies, the Lagriidae, which has only 17 species in the nearctic fauna, it

* The “ proof” was experimental, of course; for the value of this proof see
my I9I2 paper. Also note that Meisner’s results on the poisonous effect of
Coccinellid juices (Ent. Bl. Nurnberg, vol. 5, no. 9, pp. 180-182, Sept. 20, 1909)
are controverted by a repetition of his experiments by Heikertinger (Wien. Ent.
Zeit., vol. 38, Heft 4-8, pp. 109-113, June 15, 1921).

74 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

may be said that almost any small number of captures by our birds
would satisfy expectations. There are eight records distributed among
six species of birds, about all that probabilities demand. As to the
Tenebrionidae so many species of which have secretions nauseous to
man, the nearly 2,200 records are eloquent of the fact that these
beetles are not disliked by birds. If they do enjoy any degree of
immunity from bird attacks, it is probably on account of their char-

acteristic nocturnal or otherwise seclusive habits. The number of —

species of birds known to prey upon Tenebrionids is in excess of 175;
nine birds had over 20 records each, five others more than 40, one
additional over 50, and two others more than 100. The number of
specimens taken at a meal ran up to 44 in two cases and to 46 in
another and 53 in still another. The number of species of Teneb-
rionidae identified was over 100, including 12 of Eleodes, the largest
and most potently odoriferous of the family.

Of these a recent article says: “ It was interesting to note that the
quantity of the secretion voided varies noticeably with the different
species under observation, both under field and under laboratory con-
ditions, and some of the species, notably Eleodes tricostata Say,
undoubtedly have the habit of erecting the abdomen in a threatening
manner when approached, though no secretion may be voided. Such
species undoubtedly find protection through imitation of the threaten-
ing movements of their more formidable associates. Two of the
characteristics of the Eleodes are their slowness of movement, and
their habit of coming out of their hiding places about sunset for
feeding purposes, and their presence is readily noticed on the bare
sandy plains by birds, skunks, and other enemies, hence their protec-
tive secretion, or, in the absence of this, their threatening maneuvers
are no doubt of highest value to them.” (Wade, J. S., Notes on
defensive scent glands of certain Coleoptera, Psyche, vol. 28, nos. 5-6,
p. 148, Oct-Dec, 1921.)

In this connection it may be said that our tabulations show 51
records of birds feeding on Eleodes tricostata and tog for the other
species. Other large Tenebrionids as Asida and Nyctobates are well
represented in the table of determinations as are also the metallic
torms such as Helops, Meracantha, and [:pitragus. Blapstinus with
286 records for 11 species is the favorite genus, and the reason is
what ?—simply that it is the most widely distributed and the most
numerous in individuals.

With respect to the other “‘ partially distasteful” family of beetles,
the Scarabaeidae (sens. lat.), the more than 13,000 records in our
tabulations speak for themselves. The selectionist protectionists have

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 75

simply made a very bad guess. Consider for instance the Cetoniini,
the best ‘“‘ protected’ tribe, some of which are said to mimic bees in
flight. Our most numerously represented genus, Euphoria, has 445
records, of which 148 are for the most beelike species of all, E. inda.
Cotinis, very poorly represented in our fauna, has 156 records, and
Cremastochilus, noted for their association with ants, 77.

Species of Onthophagus “live in and about dung and are of a
colour which conceals them well in those surroundings.” (Donis-
thorpe, Trans. Ent. Soc. London, 1901, p. 358.) However, they were
preyed upon 642 times by the birds included in the present tabulations.
The species of Aphodius also are dung-feeders and said to be protected
The number of determinations of this genus is approximately 3,565 ;
in numerous cases 100 of these beetles were found in single stomachs
and in one instance no fewer than goo. A warningly colored species,
A. fimetarius (with the thorax black and elytra red), was identified
in 913 stomachs. Consider the entirely different case of a beetle, the
rose chafer (Macrodactylus), known to be actually poisonous (see
Science, n. s., vol. 43, pp. 138-139, Jan. 28, 1916) besides having
protective (cryptic) coloration and long spiny legs: although there
are but two species in the country, we have 52 records in our tabula-
tions representing 15 species of birds. The larger numbers of speci-
mens taken were: nine by a crow, 12 by a road-runner, and from
15 to 40 in five instances by the eastern kingbird. These records show
that the most potent protective adaptations possible do not necessarily
protect. The highly significant fact about the case is that predators
do not seem to recognize the dangerous qualities of the rose chafer ;
every generation of young chicks and pheasants will pay a heavy
death toll if permitted to stuff themselves with these beetles. Even
trout kill themselves in the same way. But what advantage is all this
to the beetle? Those that cause the death of some predators, them-
selves lose their lives, that is, all of those actually proved “ fit” in
this respect are eliminated ; the only effective poisonous action is upon
young (among birds)—adults can and do eat them freely. No con-
siderable body of predators has ever been killed, and ‘“ warning color ”
has not been acquired (the rose chafer is a uniform and inconspicuous
brownish-yellow). Theories as to protective adaptations seem to suffer
from every angle of this case. (For fuller discussion of the subject
see Lamson, Geo. H., Journ. Econ. Ent., vol. 8, no. 6, pp. 547-548,
Dec., 1915; Bates, J. M., Science, n. s., vol. 43, pp. 209-210, Feb. 11,
1916; and McAtee, W. L., The Auk, vol. 33, no. 2, pp. 205-206,
Apr., 1916.)

¥

76 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Some curiosity may be felt as to the relations of birds to the large |
Scarabaeidae and Lucanidae with thoracic horns and especially strong
or greatly enlarged mandibles. In nearctic America we have few |
species in any of these groups; another limiting factor important in
relation to bird predators is the large size of these beetles. Never- |
theless all of the genera are represented in the food of birds, Passalus |
by 32 records, Platycercus by 19, Lucanus by 25, Ceruchus by three, |
Dorcus by six, Sinodendron by seven, and lucanids further unidentified
by 28 determinations ; our largest scarabaeid, Dynastes is represented |
by five identifications, Strategus by 27, Xyloryctes by seven, Copris |
by 62, and Phanaeus by 252. The latter genus, besides being |
“horned,” has brilliant metallic colors. |

The long-horned beetles or Cerambycidae include many species |
with showy colors, but selectionists as a rule have not attributed dis- |
tasteful qualities to the group; rather they have considered these |
beetles mimics of various more strongly “ protected” insects such as |
wasps and other Hymenoptera and weevils. Numerous longicorns |
have cryptic coloration also, but their chief defense must be residence
of the long larval stages in wood where they can be reached only by
a small proportion of insect predators. The imago state, only in which
the colors theorized about are displayed, is of relatively short duration.
Bearing these facts in mind we believe the records show that longi-
corns are fed upon to such an extent as to indicate that in proportion
to availability they contribute their due share to the subsistence of
birds.

The total number of determinations in the present tabulations is
1,585, shared by 162 species of birds. Twenty-one kinds of birds have
from 10 to 19 records each; six additional species from 20 to 29; six
others from 30 to 39; one other 42; still another 53; and two as many
as 169 and 173 respectively. The woodpeckers, on account of their
peculiar qualifications for obtaining the larvae, naturally are the chief
enemies of Cerambycidae. Several of the species prey upon these —
beetles to the extent of from ro to 50 per cent of their total food. The
number of adult beetles taken at a meal by these or other birds ex-
ceeded 30 in a number of cases and in four ran as high as 83, 100,
102, and 168. The last named figure is for one of our most wasplike
species, Xylotrechus colonus, in the stomach of a raven. There are
10 identifications of Nylotrechus; of the other wasp-colored long-
horns, we have the following numbers of determinations: Cyllene, 10
(28 specimens of C. robiniae in the stomach of a magpie) ; Calloides
nobilis 1, Neoclytus 11, Clytanthus 12, Clytes 5, Strangalia 6, Typo-
cerus 16, and Leptura 39. It is noticeable that the numbers appended

bad

ee — eeeeeeeeeeeeeeeEOEOEOEeeEeEeEeEeEEeEEeEEeeEeEEE—EeEeEeEeEeEeeEEEE_E_E_E_______E EO _—_—_—_V—_—_—_

eo = - —_ A A A A LLL LLL LLL LLL LLL RL LLL LD |
pa ee eee —

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 77

correspond very closely to the relative abundance in individuals of
these genera. Of the ant-suggesting genera, we have for Euderces
five records and for Cyrtophorus one; and for the ichneumonid-
mimic Neolorchus three identifications.

The figures for the distinctly warningly colored genera are
Acmaeops seven, Desmocerus two, Gaurotes two, Tetraopes nine, and
Oberea five. Such a catalogue shows that all the forms, whatever their
alleged “‘ protection” are eaten more or less, and there is no other
explanation of the comparative extents to which they are taken so
reasonable and satisfying as that it probably depends almost entirely
on their relative abundance and availability to birds.

The Chrysomelidae or leaf beetles are classed by Professor Poulton
as undoubtedly specially protected, and Donisthorpe writing of them
under another name says: “‘ The Phytophaga are considered to be all
more or less distasteful, and no doubt justly so. Many species have
been proved to be so, and the group is mimicked by various orders of
beetles throughout the world.” (Trans. Ent. Soc. London, Igor,
p. 367.) Selectionists should have been somewhat restrained in their
theorizing by the very name Phytophaga, for the leaf beetles and their
allies being groups that subsist directly upon vegetation, must ac-
cording to inevitable law in the organic world form the base of a
column of predacious life more or less exclusively dependent upon
them. Like the grazing mammals, all plant-feeding insects, no doubt,
have their lions, wolves, and eagles, their hyaenas, jackals, and
vultures.

No reason appears from the records of bird food here discussed to
warrant doubt that the leaf beetles do in fact contribute their full
quota toward the subsistence of predatory animals. The total number
of identifications of Chrysomelidae is 5,666, and these are shared by
well over 200 species of birds, so it is certain that practically all of our
birds feed more or less upon these beetles. More than 230 species of
Chrysomelidae are represented in the determinations, this in turn
indicating that all tribes of the family are preyed upon. The Cassidini,
on account of their bright colors and specialized larvae, receive fre-
quent mention as a specially protected group but our scant representa-
tion of this tribe seems to bear its share of bird predation; Cassida
3 records, Physonota 2, Coptocycla 48, and Chelymorpha 48. Again
correspondence of the number of identifications with observed fre-
quency of the insects is quite-evident.

Resemblance to caterpillar droppings always is spoken of by selec-
tionists as a prime defense, and one tribe of our leaf beetles, the
Chlamydini, exhibits this to a high degree. When feigning death, as

78 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

they do when disturbed, they “resemble the excrement of certain |
caterpillars so closely as to render their eee dificult . . . . and |
it is said that birds will not pick them up.” (Blatchley, Coleoptera
of Indiana, p. 1,114, 1910.) Two genera represent this tribe in our |
fauna and of these, Chlamys has been found 56 times in bird stomachs |
and Exema 17 times; 10 of the latter beetles were contained ina |
single stomach of a Bewick’s wren. The number of determinations |
|
!
|
|
|
|
|
|
|
|
|

cited, in view of the few species we have of this tribe, fully satis
the probabilities.

The genus Diabrotica, chiefly yellow and black species, has received
special attention from the standpoint of protective adaptations. ‘“]
believe,’ says C. J. Gahan, “that the species of Diabrotica are pro-
tected, and that the species of Lema derive advantage by mimicking
them.” (Trans. Ent. Soc. London, 1891, p. 369.) The tabulations
of bird food here discussed show 41 records of Diabrotica vittata
distributed among 17 species of birds; 107 of D. 12-punctata for 42
species of birds (18 specimens being found in a stomach of a cliff
swallow) ; and 194 records of D. soror for 22 kinds of birds (a
black-headed grosbeak had eaten 21 of these beetles). There are also
34 other records for scattering and unidentified species of the genus.
Thus there is no evidence of special protection for Diabrotica; as for
Lema the species are much less numerous in individuals, and that is
the real reason they are captured less frequently by birds; we have |
22 identifications shared by 14 species of birds.

One other Chrysomelid, the Colorado potato beetle (Leptinotarsa
z0-lineata) , has had its protective adaptations pointed out on numerous
occasions, and like the rose chafer, among the Scarabaeidae, seems to
be actually poisonous. (See Riley, Seventh Missouri Rep., 1875,
pp. 6-7.) However, our records show that 23 species of birds devour
the insect and r1 others are added by the literature of the subject.
One hundred and eighteen identifications of this pest are included in
our tabulations; the larger number of specimens found in single
stomachs are 8 in that of a starling, 10 in a sharp-tailed grouse,
12 in a black-headed grosbeak, and 14 in a rose-breasted grosbeak.
Birds such as the bob-white, crow, and rose-breasted grosbeak are
recorded as having cleared fields of these pests.

Before leaving the Phytophaga or Chrysomelidae it may be well to
cite certain records of large numbers of individuals being taken at a
meal by birds, since they show not only-that there is no restriction of
bird attack to certain tribes of the family but also that there is no
restriction of the more important avian predators to certain groups
of birds. Some of the larger records are: 36 specimens of Micro-

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 79

rhopala vittata taken by a starling ; 47 Donacia by a red-winged black-
bird; 50 of Systena sp. by a Baird’s sparrow; 50 of Disonycha
caroliniana by a horned lark; 58 of Myochrous denticollis by a house
wren; 212 Donacia subtilis by a Franklin’s gull; 250 Colaspis
brunnea by a nighthawk ; and about 300 FE pitrix cucumeris by each of
five individual tree swallows.

Bruchidae spend so much of their lives within seeds that they are
little exposed to attack by birds; an advantage which probably is
compensated for by their being devoured with the seeds by some birds
and other seed-eating animals. However, this is a subject that has
scarcely been investigated. Our 47 records represent nine or more
species of bruchids and were distributed among 29 species of birds.
Expectations based on availability of free bruchids probably are
satisfied.

The great series or suborder of beetles known as the weevils or
Rhynchophora, for the most part, are said to be cryptically colored,
resembling seeds, buds, bark, bits of earth, bird droppings, etc.
Wallace adds: ‘One of the characters by which some beetles are
protected is excessive hardness of the elytra and integuments. Several
genera of weevils (Curculionidae) are thus saved from attack and
these are often mimicked by species of softer and more eatable
groups.” (Darwinism, p. 260, 1896.) However, it should be pointed
out at once that those who predicate hardness as a defense against
predators do so without due reflection upon the digestive powers of
animals.

Recall the fragmentation and gulping down of bones by dogs; the
swallowing of snails, shells and all, by squirrels; while reptiles, am-
phibians, birds of prey, and predatory mammals either swallow their
vertebrate prey whole or in large pieces, the bones included ; waterfowl
and shorebirds habitually take shellfish entire, including such hard-
shelled forms as clams and oysters; gallinaceous birds are provided
with gizzards which grind up the hardest seeds; and finches and
numerous other birds are just as effectively equipped if on a smaller
scale. Not only do birds with gizzards grind up their food materials,
but the grit and pebbles they swallow are in most cases gradually
ground down and pass out through the intestines in the form of fine
sand. Most predators, in fact, have either a powerful mechanical or
a resistless chemical digestion that as a rule is fully competent to dis-
pose of anything entrusted to it. With such digestive powers at the
service of predators, it is extremely unlikely that hardness in the
degree possessed by weevils is any bar to their being eaten ; moreover
being jointed, weevils are thoroughly susceptible to chemical digestion.

6

8o SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

In illustration of the factor of hardness as related to bird food the
three genera Lixus, Thecesternus, and Sphenophorus, representing
as many families of weevils, may be discussed. Lixus is so hard that
the entomologist usually finds drilling a necessary preliminary to
pinning ; moreover the species are full of vitality, sometimes living
through 24 to 36 hours in the cyanide bottle. Records of this genus
in our tabulations total 102, distributed among 27 species of birds.
No fewer than 18 specimens were found in a single stomach of a
nighthawk. Thecesternus, a weevil with unusually thick and hard
integument, was identified 151 times in the stomachs of 22 species of
birds. Twelve specimens were taken by a meadowlark at a meal and
15 by arobin. The billbugs of the genus Sphenophorus not only are
hard, but like Lixus have much ability to resist the fumes of cyanide
and prolonged submersion in water. However 1,397 determinations
were made representing 34 species of these weevils. They were found
in the stomachs of no fewer than 11o species of birds. Some of the
larger numbers taken from single stomachs were: 10 in the cases of
the upland plover, clapper rail, and yellow-headed blackbird; 11 in a
robin ; 12 in an avocet ; 17 in a crow blackbird ; 20 in a killdeer ; 33 in
a crow; and 34 in a magpie.

Hardness thus appears to be of no consequence as a defense. Brief
attention may be paid to a few other of the so-called protective devices
of weevils. One of the obscurely colored genera, with the habit of
dropping to the ground and feigning death, is Rhinoncus ; such weevils
are said to resemble seeds, but what good this would do, since most
birds eat seeds, theorists have left unexplained. Rhinoncus has been
identified 73 times from the stomachs of 30 species of birds, of one
of which, the olive-backed thrush, an individual had eaten 20 of these
weevils. Rhodobaenus, our only conspicuous red and black weevil,
was identified 14 times in-the stomachs of Io species of birds, and
Tyloderma, black weevils with whitish or yellowish markings, 133
times in 48 species of birds. Fifteen specimens of Tyloderma were
taken from the stomach of a meadowlark.

To mention the relations of birds to a few representative genera of
weevils, we may record that the rare Otidocephalus were identified
six times in the stomachs of an equal number of species of birds ; that
the minute Apion were taken 91 times by 36 species; the nut weevils
(Balaninus) 380 times by 85; the cotton-boll weevil (Anthonomus
grandis) 348 times by 43 species of birds (23 other species are re-
corded as enemies in the literature); the alfalfa weevil (Hypera
murinus) 2,222 times by 50; the clover root weevils (Sitona) 1,611

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 81

times by 94; the engraver beetles (ps [Tomicus|) 120 times by 24;
and the Anthribidae 29 times by 21 species.

A few of the larger numbers of weevils found in single stomachs
also may be cited; thus 109 Dorytomus mucidus were found in one
stomach of a downy woodpecker; 153 Calandra oryzae in a barn
swallow ; 167 Barypithes pellucidalis in a starling ; 282 Hyperodes sp.
in an eared grebe; and 281 larval and adult alfalfa weevils in a
Brewer’s blackbird, 300 in a killdeer, and 317 in a valley quail.

The nearly 20,000 identifications of weevils in birds’ stomachs attest
to the frequency of their capture, and records such as those just cited
to the relish with which they are eaten.

Though to all weevils are attributed various protective adaptations,
weevils of all sorts are preyed upon; the secret of the whole relation-
ship between prey and predator in this as in other cases is distribution
of the attack. All available food supplies are sought by predators and
the amount of attention they receive is in direct proportion to their
availability.

Total number of identifications, of Coleoptera 85,322; percentage
of identifications among those of all insects, 44.6899; percentage of
species in this order among those of all insect species known, 46.2032.

Other enenues.—lIt is difficult to summarize what is known regard-
ing the predatory foes of so extensive an order as the Coleoptera.
Fresh-water fishes prey systematically upon both larvae and adults
of the aquatic beetles but secure other forms only incidentally. How-
ever it appears that falling into the water or otherwise becoming
available as prey for fishes is a more or less frequent happening to
terrestrial beetles, since most of the families are represented in the
food of these animals. (See especially Forbes, papers, bibliography
p. 188.)

Kirkland reports Coleoptera as making up the following percentages
of the food of 149 common toads: ground beetles and their allies, 8 per
cent ; May beetles and allies, 6 per cent ; wireworms and allies, 5 per
cent ; weevils, 5 per cent; potato beetles and allies, 1 per cent ; carrion
beetles, 1 per cent ; and miscellaneous beetles, 1 per cent. Drake found
Coleoptera to constitute 33 per cent of the whole number of animals
consumed by 209 leopard frogs and 54 per cent of the insects. The
number of specimens of various families identified was: Carabidae
176, Cicindelidae 44, Hydrophilidae 1, Staphylinidae 12, Coccinellidae
13, Erotylidae 1, Elateridae 1, Spondylidae 2, Cerambycidae 4, Chryso-
melidae 2, Tenebrionidae 1, and Rhynchophora 146. It is worth noting
that this author says of weevils: ‘‘ The habit of these insects of drop-
ping to the ground when disturbed gives the frog a chance to capture

82 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

them.” This is just the habit the selectionists have declared protective.
Klugh reports finding 35 Colorado potato beetles and five other
Coleoptera in 25 stomachs of the leopard frog. Surface’s report shows
that other frogs and toads feed extensively upon beetles, the larger
families contributing most heavily ; the salamanders also eat a great
many beetles, especially aquatic forms. Lizards, snakes, and turtles
also feed upon beetles, some of the smaller terrestrial snakes taking a
great many of them. Pack reports lizards feeding on beetles of such
“protected” groups as Coccinellidae, Erotylidae, Meloidae, and
Chrysomelidae.

Among mammals, moles and shrews prey freely upon beetles,
taking Scarabaeidae and their larvae more and wireworms and ground
beetles less often. Spermophiles, prairiedogs, chipmunks, squirrels,
grasshopper mice, and other rodents as well as raccoons, foxes, and
coyotes prey upon beetles occasionally, and such animals as bats,
skunks, and armadillos depend upon them to a much larger extent.
There is no reason to believe that the ‘“‘ protected” groups of beetles
fare any better with mammalian than with avian predators. However
citation of a few instances of the capture of such beetles may be
advisable. A series of three armadillo (Tatu novemcinctum) stomachs
from Texas contained Carabidae and Scarabaeidae in profusion, also
weevils, Histeridae, Lampyridae, Staphylinidae and Tenebrionidae
(including Eleodes). The stomach of a skunk (Mephitis occidentalis)
collected at Nelson, Calif., held 60 per cent of pupae of the Colorado
potato beetle; two shrews (Sorex vagrans amoenus) from Crater
Lake, Ore., had fed on Silphidae, one to the extent of 50 per cent, the
other to 100 per cent of the total food. A prairiedog (Cynomuys
gunnisoni) from Magdalena, N. Mex., had nothing but remains of
Calosoma triste in its stomach, and a badger (Taxidea taxus) from
Ash Meadows, Nev., had eaten no fewer than 150 Calosoma
prominens.

Passing to the enemies of beetles in the insect kingdom, it is well
known that the various predatory tribes make no exception of beetles
even though their generally hard integuments would seem to be a bar.
Mantids, chrysopids, robber flies, predacious bugs and beetles, wasps,
ants, dragonflies, and spiders all feed upon beetles, and every tabula-
tion of the species eaten by them shows “ protected ” forms liberally
represented. Beetles are subject to numerous parasites which attack
them in all stages from egg to imago, and like most insects, they at
times are decimated by fungal or bacterial intruders.

While it has been impossible in the limits of this paper to discuss
fully the enemies of beetles other than birds, a few cases may be cited

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 83

in more detail as showing how the activities of such foes supplement
the predatory activities of birds. Take for example the wood-boring
beetles, which although they are eaten by birds to an extent that
indicates that no special protective adaptation operates in their favor,
yet are shielded from most birds during the greater part of their lives
by living under bark or even within solid wood. However this habit
does not put them out of the reach of predatory and parasitic insects.
Thus Kleine records 159 hymenopterous parasites of Cerambycidae.
and 136 beetle predators and 157 hymenopterous parasites of Sco-
lytidae in Europe. (Ent. Bl. Nurnberg, vols. 4-5, 1908-1909.) The
cotton-boll weevil (Anthonomus grandis) again passes the larval and
pupal stages apparently well hidden from most enemies, yet some of
the 66 kinds of birds known to prey upon it remove the immature
stages from the cotton bolls, and in addition the weevil has 54 insect
enemies, about half of which attack the concealed stages. (Bull. 100,
iS. Bur. Ent. 1912.)

So it is with the mechanically protected species; all have in the
chains of their life histories weak links, of which hungry predators
and assiduous parasites are not slow to take advantage. For contrast,
consider the case of the Colorado potato beetle, an insect exposed
almost throughout its life history, and with all of the attributes—
color, reflex bleeding, nastiness, even poisonous qualities—of a most
highly “ protected”’ insect. Besides the 27 species of wild birds
known to feed upon this insect, ducks, chickens, guinea fowls, skunks,
snakes, frogs, toads, at least eight species of Pentatomidae, two of
Reduviidae among bugs, and eight of Coccinellidae and seven of
Carabidae among beetles, besides robber flies, wasps, spiders, pha-
langids, and mites prey upon the various stages. Despite all of its
protective adaptations, the Colorado potato beetle undoubtedly has its
full quota of foes; its rapid increase and spread over the United
States was due to enormous increase by cultivation of a favored food
plant and not to lack of enemies. Dr. F. H. Chittenden remarks:
“Few, if any, noxious insects have so many recorded natural enemies
as the Colorado potato beetle.” (Bull. 82, pt. 7, U. S. Bur. Ent., p. 85,
Feb., 1911.) In other words, the potato beetle, being an important
economic insect, has been much studied, and among other things we
have learned that it has numerous enemies. If less were known about
the species it would be hailed as a marvelous instance of protective
adaptation ; facts are a terrible handicap to theorizing.

Discussion—In general we have seen that whatever the beetle,
something in the way of protective adaptation has been claimed for it,
yet practically all are eaten. On the other hand we have also seen that

84 SMITHSONIAN MISCELLANEOUS COLLECTIONS ~— VOL. 85

the large families of Coleoptera, those abundant in individuals, are
most freely eaten by birds, while the small families with few species
escape with small losses. It is the old story over again of food sup-
plies (beetles in the present consideration) being drawn upon in pro-
portion to their abundance and availability.

MECAPTERA (SCORPIONFLIES )

Protective adaptations —The scorpionflies are predacious ; those of
the genus Panorpa commonly have yellow bodies and black markings
in the wings ; and the males have enlarged genitalia carried aloft some-
what like the stings of scorpions. The species of Bittacus resemble
crane flies.

Bird enemies.—We have only five records of scorpionflies being
eaten by nearctic birds, these being distributed among four species.

Number of identifications, 5; percentage of identifications among
those of all insects, .0026; percentage of species in this group among
the whole number of insect species, .0260.

Other enemies.—There seem to be no records of such.

Discussion.—Poverty of data is the chief characteristic of the record
for scorpionflies. These insects are not an obtrusive part of the insect
fauna and have been little studied. The question of the efficiency of
their protective adaptations in relation to predators can hardly be
intelligently discussed at present.

DIPTERA (FLIES)

Protective adaptations——Not much has been written about the pro-
tective adaptations of diptera, the suggestion most often made being
that a considerable number of them “ closely resemble wasps, and bees,
and no doubt derive much benefit from the wholesome dread which
those insects excite.” (Wallace, Natural selection, p. 69, 1891.) The
families that have the most numerous species supposed to resemble
Hymenoptera are the Stratiomyidae, Bombyliidae, Asilidae, Co-
nopidae, and Syrphidae. Many flies have metallic colors, which are
alleged to be warning; such insects are common among the Stratio-
myidae, Dolichopodidae, Tachinidae, and Muscidae. A large number
of Diptera pass the greater part of their lives in the larval stage and
many of these larvae are more or less protected from birds by their
habitat, as the Cecidomyidae in galls, the Mycetophilidae and others in
fungi, the Culicidae, Chironomidae, many Tipulidae, the Simuliidae,
Stratiomyidae, Tabanidae, and Ephydridae in mud or water; and
various others in excrement and other decaying organic matter. Of
course this sort of protection is of no avail in the case respectively of

~

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 85

birds which eat galls or fungi or which obtain their food chiefly in or
about water, or which feed directly upon or by tearing up carrion and
the like.

Bird enemies.—Until comparatively recently it was very difficult to
get identifications of flies found in bird stomachs, and even more so
of their larvae, hence nearly half of the identifications of Diptera
were not carried further than to the order.

The large proportion of the unidentified to the total number of
records of Diptera has one advantage, namely that it is distributed
probably among nearly as great a number of species of birds as are
the records for all flies. Since that number exceeds 250, we may be
sure that there is no group of birds that habitually avoids Diptera.
Among these records are numerous instances of from 50 to 500
specimens of flies or their larvae being taken at a single meal, good
evidence that the flies concerned were not at all distasteful. These
data are sufficient commentary also on the state of determinations of
Diptera, scores of specimens being present and not being named even
to the family. The material will be re-examined in the future to
obtain more satisfactory results, but there has been no time for that
in connection with the present paper, which is wholly a by-product.

A satisfactory discussion of the relations of birds to Diptera is
hardly possible therefore, and the best that can be done is to present
the fragmentary data available and to make allowances for deficiencies.

Identifications of Diptera

Percentage
of species
Percentage of in this family
identifications among the
among those whole number
Number of of all of nearctic
Family identifications Diptera Diptera 1
Tipulidae ................ 1,505 14.4426 5.7309
Did ACS cass eseiaysier oo ere etna he 3 LO27y, 0922
Psychodidae ............. 5 0461 .3920
Chironomidae ............ 1,003 9.2562 3.0903
Gilicidae | Seve .e sc ce sneie ne 112 1.0336 1.9257
Mycetophilidae ........... 53 4891 2.9058
Cecidomyidae ............ 15 1384 1.6835
Bibionidae ............... 140 1.2920 8648
STII AG: cis-sparsisiesaa'sieis hare 8 .0738 3173
Blepharoceridae .......... I .0092 1384
ny pada: Yates cise sete oe ere I .0092 .0807
Orphnephilidae ........... 08 Ra .OITS
Stratiomyidae ............ Vee 6.7553 3.4362
Wabanidae- :.....<-.cce~0es 336 3.1008 3.5285
Acanthomeridae .......... Lee aie S153

“Computed from Aldrich, J. M., A catalogue of North American Diptera,

etc., Smithsonian Misc. Coll., vol. 46, pp. 1-680, 1905.

86

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Identifications of Diptera—Continued

Number of

Family identifications
Meptidacmeracesen = sore 42
Nemestrinidae "22.2... Js: eae
Cyrtidaernavesaset chee I
Bombyliidae) 2asseesssee- 8
Wherevidae wary: sss cence 5
Scenopinidael wanes se. soe I
My daidaey ess eene nee
A proceridaemer aaa eee nae vine
Asiltdaet tration eae 170
Dolichopodidae 232 soe 86
Hmpididaes joann eee 48
onchopteridaess.2.0 en ae
Bhoridae: t5ccque ntsc )
latypezidae (ye sa. ones
Pipunculidae” Wem ose ee - sie
Synphidaemancecceeeeaee 259
Conopidael ae neemagce nek ae
Oestridaey tases eee 4
Machinidae jcc eecs ass 54
Dextidaen perenne oe es
Sarcopnagidaey (sen ay-ue. 102
Muscidae (sens. lat.)...... 512
“Anthomyiidae: ..5...2...:. 109
Scatophagidae: Aasnace nee 79
Heteroneuridae ........... 7
EHelomyzidae sesnee oe II
Borboridae) saceere eee 2
Phycodromidae 22... 5.22 20
Sciomyzidaene seaen sarees 13
Sapromyzidael sas... o6 oo" - 24
Ontalidae 2a csacnas asta 18
Rhopalomeridae .......... one
dinypetidae mace aa.t elena 10
Mucropezidae sa.s-- 42... ae
Sepsidacmusccr acetic ene 8
Rsilidaepuncsetcsc sore ate
Diopsidacmnn ceils ere ate
E phy dridae sas acs eee 305
©scinidde meer nea 20
Drosophiliddes a... - 24-5. 2
Geomyzidae ssn ce ccc es I
Agromyzidae’ (22... e-- ose 2
Elippoboscidae 2.....-. 4. -1- I
Nycteribiidaer cms. ee eee es
Wnidentiied 44-02 ose 4,904

Percentage of
identifications
among those
of all
Diptera
387 6

.0092
.0738

0461
.0092

1.5688
.7936
-4430

.0830
2.3902
.0369
4983
9413
4.7250
1.0059
.7290
.0646
1015
.26076
.1846
.1200

2215
1661

-0923

.0738

to

8147
.1846
10184
.0092
.0184
0092

45.2566

Percentage
of species
in this family
among the
whole number
of nearctic
Diptera

1.4644
0092
-4843

5.2581
.8187
.1268
-5074
.0807

6.3536

6.2959

5.4196
-0346
.7380
2998
.3229

9.3401

1.0378
-3459

12.6841

1.9833

1.4183

3.4478
9917
-1614
.4728
.2707
.0231
.7380

1.2338

1.7181
.0346

2.3984
-7380
.3344
.2883

I.O1I5

1.6720

1.5451
8763
.1730

1.0054
5304
.0576

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 87

From the foregoing table it is evident that crane flies (Tipulidae),
midges (Chironomidae) and mosquitos (Culicidae) are adequately
represented, and it is fair to say that an important reason for this
showing is that the groups of birds eating most of these flies and their
larvae have been examined rather recently and that in consequence
closer identification of their food items has been made. This would
indicate that records for the other families will be similarly increased
by future studies. It is worth noting that most of the larvae of
Chironomidae which are so commonly eaten by birds are red (a
warning color) so much so as to be popularly called “ bloodworms.”
There are numerous instances of hundreds of these larvae being taken
at a single meal.

The more than 10,000 records cf Diptera mark these insects as a
valuable bird food; as in other cases certain birds prey to a greater
extent upon the group than others ; of these may be cited seven species
of swallows which make 13 per cent to 40 per cent of their total food
of flies and an equal number of flycatchers consuming them to the
extent of from II per cent to 44 per cent of their entire subsistence.

Total number of identifications, 10,836; percentage of identifica-
tions among those of all insects, 5.6757; percentage of species in this
order among the whole number of insect species known, 11.4432.

Other enemies—Fishes are among the most important enemies of
flies having aquatic immature stages. Pearse, writing of the food of
33 species of fishes in Wisconsin lakes, reports 20 per cent of their
food to consist of flies and their larvae, chiefly the latter. Marine
Chironomidae are eaten by shrimps and sea-anemones. A variety of
fishes, the top minnows and killifish in particular, are such efficient
enemies of mosquito larvae that they have been widely used in
mosquito-control campaigns. Diptera are eaten quite freely by frogs
and toads and to a lesser extent by lizards, snakes, and turtles. Among
mammals, shrews, moles, and bats feed regularly and extensively upon
Diptera; other mammals that get at least some Diptera are mice,
squirrels, foxes, and armadillos.

Among their own kind, 7. e., insects, about all the predacious kinds
feed freely on flies. The latter are soft-bodied insects easily pierced
by the sucking predators or chewed up by the biting kinds. Tiger
beetles, assassin bugs, mantids, ants, panorpids, dragonflies, and robber
flies and other predacious members of their own order habitually feed
upon flies. A number of families of wasps, such as the Nyssonidae,
Bembecidae, Crabronidae, and Vespidae, prey freely upon Diptera,
and spiders gain from their ranks a considerable share of their sub-

88 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

sistence. Flies appear to be subject to parasitism only to a conpara-
tively slight extent but some of them are decimated by fungal diseases,

Discussion—While flies in the adult stage appear to have a degree
of freedom from predators, it is evident that the immature stages of
many groups of them pay a heavy toll; the chief food of predacious
fly and beetle larvae that live under bark, in decaying fungi or carrion
are the fly larvae they find there; the chief food especially of the
young of a great many fresh-water fishes again are fly larvae and
pupae; and a very important element of the food of the mold- and
earth-traversing shrews and moles are the larvae of flies. Fly larvae
perish in large numbers also because of the drying up or exhaustion
of their breeding nidus. Possibly some relative good fortune for the
adults may be only compensatory, but so little is known about the
subject that discussion is not on a very firm basis. Regarding the fate
of adults it is worth while recalling the all but universal destruction at
times wrought among the ranks of its hosts by the fungus Empusa
muscae.

Evidence showing the importance of availability as regulating the
consumption of dipterous food is presented in testimony of an Alaskan
correspondent about birds feeding on mosquitos. These insects, so
much more prominent an element of the insect fauna of that territory
than they are in the United States, apparently are fed upon much
more freely by birds. This correspondent, A. H. Twitchell, a reindeer
breeder, reports all small*birds frequenting the vicinity of his camp,
as myrtle, blackpoll, and Wilson’s warblers, Gambel’s sparrow, and
Alice’s thrushes preying regularly on mosquitos and feeding them
extensively to their young.

HYMENOPTERA (ANTS, BEES, WASPS)

Protective adaptations——In selectionist writings, Hymenoptera
usually are classed as the very acme of protected insects, and pro-
tective qualifications are broadly assigned to the whole group. Poulton
says: “Ants and wasps are known to be aggressive dominant insects
avoided by the majority of insect-eating animals.” (Essays on evolu-
tion, p. 281, 1908.) Drummond, in similar vein, declares that “ well-
armed or stinging insects are always conspicuously ornamented with
warning colours. The expense of eating a wasp, for instance, is too
great to lead to a second investment in the same insect, and wasps
therefore have been rendered as showy as possible so that they may
be at once seen and as carefully avoided. The same law applies to
bees, dragonflies, and other gaudy forms; and it may be taken as a
rule that all gaily-coloured insects belong to one or other of these two

|
|
|
|
|
|
|
|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 89

classes; that is, that they are either bad eating or bad stingers.”
(Tropical Africa, p. 163.)

Mimicry of a group is supposed to be a tribute to its specially
defended character and it is said that: “ The hymenoptera including
the formidable hornets, wasps, bees and ants are more frequently
mimicked than any other order.” (Poulton, The colours of animals,
p. 245, 1890.) “‘ Stinging hymenoptera . . . . are sedulously avoided
by insectivorous creatures in general.” (Bastin, Insects, their life-
histories and habits, p. 247, 1913.) Numerous Hymenoptera which
do not possess stings are said to mimic those that do have them and
species of one non-stinging group, the sawflies, are alleged to be
protected in the larval stage by distasteful or disagreeable internal or
external secretions.

Bird enemies.—For many years difficulty was experienced in ob-
taining identifications of hymenoptera and the following table plainly
shows the effect of this situation, more than a third of the determina-
tions being to the order only.

Identifications of Hymenoptera

Percentage
of species
of this family
among the
whole number
Percentage of of species
identifications of these
among those groups in
; Number of of all New York
Family identifications Hymenoptera State 1
BOVE CAG vhs sus Ssg'PA Kaw eee 2 .0074 .4640
Pamphiliidaé ........«.<.- 5 0185 1.2529
Tenthredinidae ........... 263 .9732 17.5407
PIpMVATUdae: 55. ovis caweled 5 0185 .3248
BUIICIGAC) ules Sawer xed «eins 16 .0592 3719
MPEDIIGAe: Co se.t sic bec wed 2 .0074 2320
O©ryssidae) ce sicc ew oct see I .0037 .1392
Tenthredinoidea (further
unidentified) .......... 85 3145 a
Mipionidae % i.<ihdwescans 2 2604 9745
PRAY SUGAG <5, fap cwlsaiseoox 13 .0481 .2320
Capitoniidae ...........0.5 I .0037 nuk
BTACONICAG: 2s sacs Saceae 28 .1036 9.0488
Fvaniidae ................ 5 0185 3719
Trigonalidae ............. I .0037 .6496
Ichneumonidae ........... 1,003 4.1184 25.6614
Ichneumonoidea (further
unidentified) ........... 13 .0481

*Computed from Bradley, J. Chester, Hymenoptera, in A list of the insects of
New York, etc., Mem. 101, Cornell Univ. Agric. Exp. Sta., pp. 870-1,033, 1926,
the most comprehensive checklist of nearctic forms available.

go

SMITHSONIAN MISCELLANEOUS COLLECTIONS

Identifications of Hymenoptera—Continued

Family
Hicitidaeteerer reise cenit:
Cynipidaewnanyae eee
Pteromalidae 7... acces ee
Eupelmidaer.... ees ccmiee
Callinomidaess. ea coe:
Eanytomidaes sees cere
Berilampidaeva. casa ede e
Ghalcididaew ater aac tr
IWeucospidael semen see er
Chalcidoidea (further uni-

denititied|)imesmirineeierrats
Platyeastnidaceen-as eerie
Scelionidaes-aeee eres
Geraphronidacesa..e acne
Diapriidae’ vere cis estan
Belytidae, facemas acca ausee
Serphidae a2c; ance ue oeice
Pelecinidae nc. aese ocr
Serphoidea (further uni-

Centitied)) eee
Hormicidacma.cneea ieee.
Miyrinicidaes.ccc.cr eee
Formicoidea (further uni-

dentitred) meee eeeericee
Ghrysididaeien een eee
Bethylidaeseamecescicmaeee.
Diryinidaes eae acter
Soolhittkeseoopedeumoddacdse
IMAGINES Gaagecoocnnnoe
Mintillidae Mearsctcetse eee
Psammocharidae ..........
Mumenidae jae seyta acre te:
Wespidacue ener erate:
Vespoidea (further uniden-

titted) Me ercacsterserta cers
Sphecidacwarecieermerce cre
Bembecidae ...5......:...
Sphecoidea (further uni-

Gentitied)) erence
Flalictidaeyciwoncicscto oe
Andrenidacmescnrcseiecicent

Number of
identifications

14

=

_
SOM OOUNT IDE RO BS ON eS

144

23
134
92

Percentage of
identifications
among those

Hymenoptera

of all

.0518
.1400
0814
OTT

0444

.0185

OIII
2701
.0037

.0518
.0259
0148
.0074
.0444
.0259
0296
0333

.0666

7.7410
4.4404

34-9715

_

2257
-0333
0148
.0540
OLLI
0851
.1332
1295
5328

8215
.2812
0148

.O851
.4958
.3404

Percentage
of species
of this family
among the
whole number
of species
of these
groups in
New York
State
0464

6.7286
.9280
.1392
.6032

1.5313
.0464

3.1091
.0464

.7889
1.1137
.4640
5508
.4176
0464

1.1137
1.2529

1.0673
1392
3248
9281
.0928
9745

3-0195

1.8562
5104

3.2019

2.5980

9745
2.4504

VOL. 85, |

|
|
|
|

7 PROTECTIVE ADAPTATIONS—McATEE gl

Identifications of Hymenoptera—Continued
Percentage
of species
of this family
among the
whole number

Percentage of of species
identifications of these
among those groups in
Number of of all New York
Family identifications Hymenoptera State
IPantireidae@ .2:....0..<s. ; 2 .0074 3248
Nomadidae ............... 12 .O444 8817
Buceridae’ .. 0.2.00 00660: 10 .0370 shan
Anthophoridae ........... 4 0148 5568
ely lACId ae! fs 2 sects cee nes I .0037 3248
@olletidaes.. 2. s.c6in.- 8 0296 3248
Megachilidae ............. 18 0666 1.6241
Geratinidae 22.2.2. 05.0..-6.- 2 .0074 .0464
SELIG Ce eee ieiec cate farats odoinn ate I .0037 .0928
Gy OCOpid ae ecm atsatie = wale 2 .0074 .0464
END a Cra tact teuay ote cit Serer ose 139 5143 8353
Apoidea (further unidenti-
CCR Maret scn ees cre eet 3 B72 1.3705
Unidentified .............. 10,682 39.5266

Examination of the preceding tabulation shows again the influence
of availability upon choice of food. It is at once evident that the
groups more numerous in species and individuals are taken most often
by birds. Whether all are taken in sufficient number to satisfy expec-
tations is subject to discussion but the relativity of capture to abun-
dance is unmistakable. Superfamilies such as the Cynipoidea and
Chalcidoidea, owing to the minute size of most of their species, could
not be expected to figure largely in the diet of birds, and the same is
true for most Serphoidea. These are just the groups and the only
ones in the table except the Sphecoidea that seem obviously to be
inadequately represented. The Sphecoidea perhaps verge toward the
opposite limit of size for bird food.

Since so many Hymenoptera were determined no further than to
the order, the number of species (over 300) of birds eating these
unidentified forms may be taken as an approximation to the entire
number of bird species consuming Hymenoptera. It is enough at any
rate to indicate that Hymenoptera are eaten by birds of all groups
studied, just as the total number of records (over 27,000) of Hymen-
optera clearly shows that these insects are one of the most important
elements of bird food.

Beginning our consideration of the Hymenoptera with the sawflies,
it may be said that some of these insects are alleged to obtain pro-
tection from their resemblance to stinging members of the order. As

Q2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to the larvae, Poulton says: ‘‘ Numerous experiments have convinced
me that the latter are almost invariably distasteful.” (Essays on
evolution, p. 238, 1908.) However, the present tabulations reveal more
than 30 species of birds as predators upon sawfly larvae and no fewer
than 50 to 100 specimens of these larvae have been found in single
stomachs of the mockingbird and from Io to 25 in those of other
species. Hewitt records seven species of British birds as feeding upon
larvae of the large larch sawfly (Nematus erichsonii) (Bull. 10, Div.
Ent. Dom. Can. Dep. Agr., p. 22, 1912), and attributes to them great
destruction of the larvae. The 380 records of Tenthredinoidea in
our table are distributed among 99 species of birds and such wasp-
suggesting forms as Cimbex, and the horntails of various kinds, with
24. records, seem to be proportionally represented.

|
|
|

Most of the Ichneumonoidea are not credited with any special
defenses besides their resemblance to stinging hymenoptera, and the
more than 1,200 records of their being eaten would seem to indicate
that this means of protection is more imaginary than real. Some of
the more interesting records may be cited as indicating the extent to
which birds eat these insects: Protapanteles: 50 specimens in the
stomach of an English sparrow (one of a series of 12 containing 10 or
more each), and 120 in one stomach each of a Brewer’s blackbird and
an Aleutian sandpiper ; Ichneumonidae, further unidentified: 68 speci-
mens in a sanderling’s stomach (19 birds have from 10 to 42 records
each) ; Ichneumon sp.: 37 specimens in the stomach of a burrowing
owl; Glypta tuberculifrons: 44 individuals taken at a meal by a
yellow-throated vireo; Ophion spp.: 54 records for these large
ichneumons which can sting.

Most ants, their size considered, can bite severely ; their body fluids
contain formic acid and other pungent substances ; and many of them
also can sting. As further tribute to their prowess the reference of
Poulton may be quoted to the “ numerous mimetic resemblances to the
aggressive, abundant, and well-defended ants.”’ (Essays on evolution,
p. 252, 1908.) Badenoch says that ant-models “as a rule are exempt
from persecution.” (Romance of the insect world, p. 300, 1893).
The confidence of selectionists in the protective nature of ant mimicry
is further shown in the following statement by Donisthorpe on Nabis
lativentris: “I consider this insect to be an ant mimic in its earlier
stages, when it is usually found in the company of ants. From this
mimicry it obtains protection from outside enemies, both as much
when away from ants as when with them.” (Ent. Monthly Mag.
ard seri, vol.7, pp; 137-138, 1921.)

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 93

But why this conclusion? The more than 12,000 records of ant-
eating by the birds represented in our tabulations certainly indicate
no sort of immunity on the part of ants from the attacks of birds.
These records are shared by well over 300 species of birds which,
practically speaking, means that all birds eat ants. Ninety-three of
the species of birds represented in our tabulations have from 10 to
4g records of ant-eating each, 18 others from 50 to 99, 17 others
from 100 to 199, four additional over 300, and one additional species,
the eastern flicker, in excess of 500. All these records are among the
Formicoidea further unidentified, three-fourths of the total for all
ants. From 200 to 300 ants at a meal is a common number; the
swallows often get 800 or more; the nighthawk 1,000; and wood-
peckers 2,000 or more. In two cases stomachs of flickers yielded more
than 3,000 ants each, and in one case more than 5,000. Out of 684
stomachs of this last named species, 524 contained ants.

In this connection the extent to which ants enter into the diet of
certain birds is of considerable interest ; our five species of thrushes
of the genus Hylocichla consume ants to an average of 12.65 per cent
of their total food, while 16 species of woodpeckers, the food of
which was tabulated by Prof. F. E. L. Beal, ate ants to extents vary-
ing from 5 to 85 per cent of their entire subsistence, the average for
the 16 species being 28.49 per cent.

The stinging ants, of course, are the most highly “ protected” of
all and it is unfortunate for our discussion that the group is so
poorly represented in the United States. Myrmicidae, including
Ponerinae and Dorylinae, are more or less generally provided with
stings, which however in the most of our species are too small to
inflict damage on a human subject. Our tabulations show 1,200
records for Myrmicidae, and they are eaten in just as large numbers
as are other ants. The harvester ants of the genus Pogonomyrmex
are larger and equipped with stings which can painfully wound a
human being. We have 66 records of these ants being taken by 25
species of birds; no fewer than 200 and 400 individuals were taken
from the stomachs of two Texan nighthawks. Mitchell and Pierce
write of birds feeding freely on Pogonomyrmex and note a case of a
group of nesting jackdaws (Megaquiscalus) cleaning up a colony in
a short time. (Proc. Ent. Soc. Washington, vol. 14, no. 2, p. 72,
June, 1912.)

Among the remaining, mostly stinging, Hymenoptera are the
Chrysididae, supposed to be protected by their hardness, abililty to
roll into a ball, and by metallic colors. We have 61 records of these
being eaten, shared by 37 species of birds. The Vespoidea or wasps

‘

QO4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

as a whole have 822 records representing 140 species of birds. Some
of the larger numbers of wasps consumed at a meal were 10 Vespula
germanica by a wild turkey, 10 Polistes sp. by a yellow-billed magpie,
and the following numbers of unidentified wasps by the birds men-
tioned: purple martin 17, olive-sided flycatcher 24, and kingbird 30.

The 103 records of Sphecoidea are distributed among 43 species of
birds, none of which took any notable number of these large insects.
Bees all sting, and the 797 records of their being eaten by the birds
examined by us would seem to indicate considerable disregard for
the stings on the part of birds. The number of species of birds repre-
sented in these bee-eating records is 144. Thirty-two species of birds
took honey bees (Apis mellifera) on a total of 118 occasions and nine
species of birds ate bumble bees a total of 18 times. These numbers
of determinations seem in fair proportion to the availability of the
bees concerned. The largest numbers of bees taken at a meal were
26 Andrenidae by a rose-breasted grosbeak, 34 honey bees by a cactus
wren, and 106 of the domestic species by a road-runner.

It is of interest to note that besides the thrushes and woodpeckers
previously mentioned, two other groups of birds are very notable
consumers of Hymenoptera. Thus the seven species of swallows make
an average of 24.9 per cent of their diet of these insects, and 14
species of flycatchers average 33 per cent.

Summary.—Number of identifications of Hymenoptera 27,025;
percentage of identifications among those of all insects, 14.1551; per-
centage of species in this group among all insect species, 17.1798.

Other enemies—Hymenoptera, having so few aquatic representa-
tives, do not figure in the diet of fishes as anything but an incidental
item, consisting of specimens, a considerable proportion of them ants,
that have approached too near or have fallen upon the surface of
the water.

Passing to batrachian enemies of Hymenoptera we may note that
Kirkland found ants to compose 19 per cent of the contents of
149 toad stomachs, and that he had evidence also of their feeding ex-
tensively upon honey bees. Garman also found not only the common
toad (Bufo lentiginosus) but also the pigmy toad (Bufo quercicus)
to be very fond of ants. Toads have been observed to feed freely
upon the larger stinging insects also, such as yellowjackets and wasps.
Drake found 25 ants and 21 other Hymenoptera in 209 stomachs of
the leopard frog. Insects of this order, especially ants, are eaten by
all frogs and toads and to a considerable extent by salamanders also.
Most lizards feed freely on ants, bees, and wasps. Winton found
agricultural ants (Pogonomyrme.x) in 80 per cent of the horned-toad

NO. 7 PROTECTIVE ADAPTATIONS—MCcATEE 95

stomachs (485) examined by him, and Mitchell and Pierce record
the extermination of a colony of these ants by horned-toads. Several
species of snakes and a few turtles feed to a slight extent upon
Hymenoptera.

Among mammals, moles prey extensively upon ants, and shrews

-and bats by no means avoid them. In our country armadillos are

destructive ant eaters and in other continents various mammals
specialize upon ants. Spermophiles and other slightly insectivorous
rodents include ants and other Hymenoptera in their bill-of-fare.
Skunks are assiduous in digging out the nests of yellowjackets (lVes-
pula), the comb, its contents and active inhabitants of the nest all
being devoured. Mice, weasels, foxes, and especially badgers simi-
larly ravage the nests of bumble bees, while bears plunder not only
these insects but also honey bees and hornets. Meadow mice and
shrews have been found to be among the most effective enemies of
sawflies, extracting the larvae from the cocoons, and these and deer
mice take a heavy toll of the Hessian fly, nibbling the stem-galls and
devouring their inmates. Squirrels feed freely upon galls produced
by Hymenoptera.

The insect enemies of Hymenoptera are numerous and effective and
strangely enough many of them are within the ranks of the order.
Philanthidae use Aculeates for food, many bees, cuckoo wasps, and
the like live parasitically in the nests of other Hymenoptera; the
surprising phenomena of hyper-parasitism reveal numerous serious
enemies of Hymenoptera among their own kin; and a number of
dipterous parasites of sawflies, bees, and wasps are known. The
so-called guests in the nests of bees and ants destroy many of the
larvae of their hosts. Predacious insects such as assassin bugs, Phy-
matidae, dragonflies, and robber flies feed freely upon Hymenoptera,
the last-named foes almost appearing to have a preference for the
larger and better armed sorts of stinging Hymenoptera. Spiders of
certain species entrap and devour large numbers of Hymenoptera.
Nematode and protozoan parasites exist and some Hymenoptera have
important fungal and bacterial diseases.

Discussion.—According to selectionists, Hymenoptera are the most
highly protected insects and the so-called mimicry of examples of this
order, such as the ants, by numerous spiders, long-horned beetles and
rove beetles, plant-bugs, and other insects is regarded as strong
evidence for the truth of the claim. Let the case be presented in the
words of an advocate (Poulton, Essays on evolution, p. 260-261,
1908) : “ The means by which the resemblance to ants is brought about
are diverse, the end—the resemblance itself—is uniform. Further-

7

96 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

more the likeness is almost always detailed and remarkable, however
it is attained, while the methods made use of differ absolutely. . . . .
We are compelled to believe that there is something advantageous in
the resemblance to an ant, and that Natural Selection has been at
work. The phenomena do not merely disprove all other suggested
causes of change ; they constitute the most powerful indirect proof of
the operation of Natural Selection.”

If the above reasoning has any application so far as the attacks of
predators upon ants are concerned, we should expect some evidence
that ants are relatively free from such attacks. Let us see what is the
case. Beginning in the very homes of the ants we find, among crea-
tures habitually living in ant colonies, that numerous Staphylinid
beetles devour the brood, besides crippled and even normal ants; the
larvae of Clythrinae (Chrysomelidae) feed on the eggs; lycaenid
caterpillars and paussid beetles eat the eggs and larvae. Numerous
ectoparasitic mites and some chalcidids also attack the ants in their
domiciles, as well as entoparasites among the Strepsiptera, Phoridae,
Conopidae, Braconidae, Chalcididae, Proctrotrupidae, and Nematoda.
Ants have very important predatory enemies in their own ranks,
namely the doryline and slave-making ants. Ant-lions of the family
Myrmeleonidae, Diptera with similar habits, predacious wasps, es-
pecially the Crabronidae, assassin bugs, ground and tiger beetles, and
spiders are serious invertebrate enemies of ants. Most toads, frogs,
and lizards, the amphisbaenids, and certain snakes feed upon ants;
fishes take them when opportunity affords; practically all birds eat
ants, several groups as the song thrushes, ant-thrushes (Formi-
cariidae), and woodpeckers depending upon them for a large part
of their food; in the same way most of the insectivorous mammals
are fond of ants and several groups of this phylum are specialized
ant eaters, namely Echidnidae (spiny anteaters) among the Mono-
tremata, the banded anteater (Myrmecobius) among the Marsupialia,
and nearly the whole order of Edentata (antbears, pangolins, and
armadillos ).”

In fact it would be difficult to name a group of insects that is so
thoroughly preyed upon as the ants, and impossible to name one that
has so many specialized foes scattered through the various animal
phyla. So far as predatory attack is concerned, it would seem that
ant-mimics court rather than avoid danger. To recapitulate: if there
is any virtue in the protective adaptations of the “aggressive, abun-

*For a comprehensive account of “The predacious enemies of ants,” see
Bequaert, Bull. Amer. Mus. Nat. Hist., vol. 45, pp. 271-331, 1922.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 97

dant, and well-defended ants,” it should be apparent in some lessening
of predatory attacks upon them. However the very opposite is true
and the case affords the best sort of proof of the thesis of this paper,
namely that the number of enemies, or perhaps better stated, the total
losses to enemies, is in direct proportion to abundance of the group
concerned.

Selectionists regard bees as very highly protected insects, but taking
the honey bee as type of the group because more is known about the
species than any other, we find that bee-keepers complain bitterly of
the numerous enemies of the inmates of their hives. Wasps, velvet-
ants, robber flies, Phymatid bugs, mantids, and dragonflies are serious
insect enemies of honey bees; spiders, toads, lizards, rats, mice, and
skunks prey upon them; numerous wild birds join in the attack (32
nearctic species according to our tabulations), and domestic ducks are
said to be insatiable in devouring bees. A mite is the primary cause of
the so-called Isle-of-Wight disease among hive-bees; ants and wax
moths destroy the comb ; and there are at least two serious infectious
diseases. At times bees rob other colonies, the rifling being accom-
plished however only after great slaughter. In the case of the honey
bee, much study has been devoted to the insect and we know consider-
able about its enemies, but the ruling principle is as clear in this case
as in that of the ants, namely, that common species have numerous
enemies.

Since ants and the honey bee fairly exemplify two of the main
phases of protective adaptation in Hymenoptera, despite which these
species clearly have their full quota of enemies, we cannot doubt that
other species of the order, when they are as well known, will prove to
have predatory foes fully in proportion to their relative abundance.

In fact the 27,000 records of Hymenoptera now available are suf-
ficient indication that the order contributes its due toll to the subsis-
tence of one of the chief groups of its enemies—the birds.

ARACHNIDA (SCORPIONS, SPIDERS, Ticks, Erc.)

Protective adaptations—Most arachnids possess venom of suf-
ficient strength, and means of injecting it into other creatures, to
enable them to overcome the animals upon which they prey. Numbers
of them have chelicerae, which in a few cases are rather powerful.
The poisonous nature of many of the species has been greatly exag-
gerated especially by primitive races of man so that they are held
in extreme dread.

98 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Besides these direct means of defense, ticks and spiders exhibit in
a high state of development that class of protective adaptations known
as cryptic coloration (both defensive and aggressive). Certain groups,
however, are brilliantly colored; some spiders also have the body
integuments hardened and produced in the form of angles or spines,
and numerous spiders resemble ants. Among the forms of this class
ordinarily coming to the attention of man, spiders are by far the most
numerous, but the small often minute mites prove, when the care
necessary to their study is bestowed, to be exceedingly abundant.
However, these small forms are beneath the notice of most birds.

The following table shows the relation of the records of the various
orders to their approximate representation by species in the nearctic
region.

Identifications of Arachnida

Percentage
Percentage of of species
identifications in this order
among those among all
Number of of all nearctic
Order identifications Arachnida Arachnida 1
Wnidentitied ence or eae 26 2386 oe
Maphostta sj sacae seas 20 .1836 .0513
Microthelyphonida ....... ae ee 0513
Sconpionidaweeeeeeeeenios 18 .1652 1.1801
Pseudoscorpionida ........ 80 7343 2.4115
Redipalpida wrmse ease: 4 .0367 .3078
Solpucidateee: sete eee oe 2 .2203 6157
Phalancidam eter eres 478 4.3873 3.4376
Araneidauicjcs «rete esuciee 9,966 91.4729 66.7044
IN CAtitlaencetoe aa eee 258 2.3681 23.0886
Pycnogonidare ss see cciae rd 1010 2.1549

Bird enemies.—Birds certainly specialize upon the group of arach-
nids—spiders—that to man seems most abundant and easily available,
over OI per cent of their arachnid captures coming from this order.
We have records of more than 300 species of nearctic birds feeding
on spiders. The freedom with which they take these creatures is
illustrated by the following citations of records; of those identified
to the order alone or about 92 per cent of the total, 81 birds have
from 10 to 49 captures each; 28 birds from 50 to 99; 15 from 100

*Computed from Comstock, J. H., The spider book, etc., 721 pp., 1912, with
numbers of Araneida and Acarina approximated from the following works,
respectively: Banks, N., Catalogue of nearctic spiders, U. S. Nat. Mus. Bull.
72, 80 pp., 1910, and Banks, N., A catalogue of the Acarina, or mites, of the
United States, Proc. U. S. Nat. Mus., vol. 32, pp. 505-625, 1907.

af

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 99

to 199; two additional above 200; one more above 300; besides the
following with greater numbers of records; English sparrow, 420;
Eastern meadowlark, 425; crow blackbird, 621; starling, 631; and
crow, 722. Some of the larger counts of spiders taken at a meal
were 25 by a Say’s phoebe, 33 by a greater yellow-legs, 46 by a
wood duck, 58 by a Louisiana heron, 187 by a starling, and 300 by
a hairy woodpecker.

A rather small proportion (less than 8 per cent) of spiders found
in bird stomachs were identified, but results obtained along this
line show the determinations are distributed to cryptically colored
groups as the Epeiridae (30 records) and Thomisidae (28) ; more
brilliant forms as the Attidae (158) and to the formidable Ly-
cosidae (370), in a way that would indicate availability to be the
principal factor in choice. There are two records of Synemosina
formica, the most antlike of our spiders, a small and rather un-
common form that one would expect no more frequently regardless
Or its “
resemblance to an ant is no protection whatever against predators.

protected”’ status. As noted in the last section, however,

There are 134 records of the cocoons or egg-cases of spiders being
eaten showing that even these quiescent stages do not escape the birds.
In bulk spiders do not ordinarily form any considerable percentage
of the total food of birds, but the proportion runs as high as 6 per
cent and 8 per cent of the annual diet in the case of certain song
thrushes and petty flycatchers.

There are 34 records in our tabulations of ticks being eaten, and
224 of mites. Of interest in connection with the latter are the finding
of 100 Parasitidae in the stomach of a red-eyed vireo; 320 mites
further unidentified in the stomach of a pipit ; 535 water mites in the
gizzard of a green-winged teal, and 594 of the same group in the
stomach of a pied-billed grebe.

Such geographically restricted and relatively uncommon forms as
the scorpions, pedipalps, and solpugids, even though having only a
small number of records each, would seem, nevertheless, to be amply
represented, considering their availability. Pseudoscorpions are
present throughout our area but lead chiefly concealed lives; the 80
records are distributed among 22 species of birds.

The daddy-long-legs, or Phalangida, with 478 determinations cer-
tainly have not been slighted; 10 of the birds taking them have Io to
1g records each; two others over 20 records ; and one each additional,
the yellow-billed cuckoo, 34; and crow blackbird, 60. Large numbers

100 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

of the arachnids captured for one meal were 20 by a yellow-billed
cuckoo and 77 by an ovenbird.

Number of identifications, 10,885; percentage of identifications
among those of all arthropods, 5.1648; percentage of species in this
class among those of all arthropods, 3.8254.

Other enenues.—Spiders are frequently snapped up by fresh-water
fishes, and mites often, especially the water mites. Pycnogonids, or
sea-spiders, have occasionally been found in the stomachs of marine
fishes and are consumed also by sea-anemones. Kirkland found
spiders to compose 2 per cent of the food of 149 common toads
examined by him, and Drake found them to constitute about 27 per
cent of the entire number of animals found by him in the stomachs
of 209 leopard frogs. However, practically all frogs, toads, and sala-
manders that have been studied have been found to feed upon spiders,
often extensively, and mites, pseudoscorpions, and phalangids are not
neglected. Lizards commonly prey upon spiders, while snakes and
turtles so far have rarely been observed to do so.

Spiders appear to constitute an important element of the food of
our shrews, and a lesser, though frequently taken, item in the diet of
the moles. We have records of the wood rat and armadillo preying
upon spiders, and evidence that the badger at times is a destructive
enemy of scorpions. Monkeys and marmosets are said to be very fond
of spiders, and anteaters also are listed among their foes.

Of invertebrate enemies various wasps (Pompilidae, Sphegidae, and
Trypoxylonidae) are among the most effective destroyers of spiders,
some species preying exclusively upon them, temporarily at least, even
upon single species. The writer has found the cells of Pelopaeus
cementarius filled entirely with flower spiders, Misumena americana,
supposedly one of the most perfectly protected (cryptically colored)
species. Dragonflies prey upon spiders. Tiger beetles and ants eat
spiders and mites, ground beetles and ladybirds also figure as enemies
of mites and spiders. Water mites are preyed upon by dragonfly
nymphs and aquatic Hemiptera. There are a number of kinds of
spiders that habitually prey upon their fellows when adult, and canni-
balism among young spiders is the rule. Mantispidae and other
spiders eat the eggs and young of spiders, and there are many proc-
trotrupid and ichneumonid parasites of the eggs. Parasitic nematodes
also afflict the adults. Scorpions are notoriously cannibalistic, so much
so in fact that it is said in some cases that whenever two of them are
found together, one is eating the other.

Ss es ee

0 OO EE EEO ns

McATEE IOI

NO. 7 PROTECTIVE ADAPTATIONS

Discussion.—All spiders have venom and some of them are large
and venomous enough to be able to kill birds. The case would seem
to be crucial for the usefulness of this direct means of defense, but
we may well say, in the light of the evidence, that the defense is
entirely disregarded by birds. Not only do our records show more
than 10,000 records of spiders having been eaten by more than 300
species of birds, but the birds emphasize their disregard for the
dangerous qualities of spiders by making them in many cases the
staple food for their callow young. Such minor protective adaptations
as those of color and form necessarily fall with the greater, and there
is no evidence whatever but that birds eat spiders under any and all
conditions as freely as they choose. The nearly 1,000 records of
arachnids other than spiders seem to be distributed among the orders
in very just proportion to the extent these creatures are available to
birds. No evidence of “ special protectedness ” obtrudes itself.

MOLLUSCA (SNAILS, SLUGS, MUSSELS, LIMPETS)

Protective adaptations The great majority of mollusks are
equipped with calcareous shells into which they can entirely withdraw.
Besides this protection more than half of the species are aquatic and
hence are more or less out of reach of many birds. Many land snails
have the apertures of their shells furnished with processes or teeth
which partly bar these openings and operculi to close them. Snails
and especially slugs secrete mucus freely ; a habit thought by some to
be protective. Numerous mollusks are colored more or less in harmony
with their environments, this being especially noted of marine forms
living on seaweeds, gorgonians, etc. The nudibranch mollusks are
characteristically brightly colored and have been said to be distasteful.
Of shelled mollusks, Wallace remarks: “ The brilliant colors of the
scallops (Pecten) and some other bivalve shells are perhaps an indica-
tion of their hardness and consequent inedibility.” (Darwinism,
p. 266-267, 1896. )

Bird enemies.—The tabulation of identifications herewith presented
is the best that could be made so far as comparative records is con-
cerned; these had to be gleaned from two sources as noted, which
between them do not include all of the families, nor, because of
disparity of data, do they give even the grand divisions comparable
treatment.

102 SMITHSONIAN MISCELLANEOUS COLLECTIONS

Identifications of Mollusca
Aquatic shells

Percentage of

identifications
among those
Number of of all
Group identifications Mollusca
Wmidentitved! 225.055 saree ee 1,032 8.7673
Pelecypoda (further uni-

Gentine)! sc. 2 scene 513 4.3582
Ostreidae) Hee kee. oswieae ten 552 4.6895
Atomitdaeanse ere teu eer
Dimyidaeteene ences
Spondylrdaey Geese sees ots ifr
Pectinidaes.mne cent 193 1.6306
TSIM a Cae Met ee ee
INWACBIINGEYS Geo ducuuabecece re Sate
IMBWGINGEI® ooncacoasoducgne 674. 5.7260
Wnioniddew eee 8 .0679
NTCIdAC aeee Gee near ere 73 .6201
Niiculidaetnaseen nee ore 45 3823
Medidaewe can eee re I £0085
Solenomyidae .....:.....%. I .0085
@anditidae@ ses. erie oer aces theta
Astantidacus.aje- caer cn 20 .1699
Grassatellidacmetsa- ease cer aide
Egycinidae, snrdas sate 2 0170
Ungulinidae nae. c25 sem:

Cyrenellidaci ses asese nee

leucinidae wasnt core ce

Diplodontidacwar.cem seme or

@hamidaey seaport cee ee oor Bre
Cardidaewesseeeeeeeeee 20 .1609
Wenilidaemeecemsseacterice: 2 .O170
lsocarditdaes we siteseiees aee <a ae
Wenenidaerts se oasecon sso 131 1.1129
Corbiculidacwemereee ees ere fo)
Betricolidae ss senterecrs cis 28 2378
Donacidaeueeemmne ar eee 122 1.0364
Psammobidae -....526 4. he Sie
drelilimidaeecarasoee ces oe 324 27525
Semelidacs ache cane ae
Gnathodontidae ........... 4 .0340
Mactnidaew-n-rrasoe cans aoe 21 1784
Anatinidae wan cmniy. ceils =i

‘Compiled from Dall, W. H., A preliminary catalogue of the shell-bearing
marine mollusks and brachiopods of the southeastern coast of the United
States, with illustrations of many of the species, U. S. Nat. Mus. Bull. 37,

221 pp., 74 pls., 1880.

Percentage
_ of species
in this group
among marine
mollusks of the
Southeast
Coast 1

2516
.1887
.0629
1887
1.8868
6918
.5031
1.6353

2.0755
6289
1.9497
.1887
.4402
6918
.1887
.5031
.6289
.0629
1.4465
2516
3145
817
0620
.1887
1.9497
1258
.4402
3773
-3773
2.3270
6280
.1258
3145
8176

VOL. 85,

PROTECTIVE ADAPTATIONS—McATEE

N

Identifications of Mollusca—Continued

Aquatic shells

Percentage of
identifications
among those
Number of of all
identifications Mollusca

Group

HEVONSiGaewees cree aoc cies
Verticordiidae ............
Cuspidariidae .............
Poromyidae ..............
IPandondae ......--6.s +-- “0 wae
@orbulidae........ 0.05200 2 7716
WWitvidaeg were cares ee wae ace 10 .0849
DHIGAVIdAG oe ae tds iietbaiets 25 2124
pap hitdacweniericls + cees sien 400 3.3982
Solentdae’ star. sacciscasinee 4 .0340
Gastrochaenidae ..........

tO ladidacmenee sans. aeee sale sie oes
sNereCIdae muerte aeons aseetesie 5 .0425
Gastropoda (further uni-

Gentitied Jee. ecco ane: 3,421 29.0631
Wentalitdaeweccs. as ose I 0085
Iimacimidaes 2 42.42.6605 0m.

Gavolinidae asses aan eae

Gyimluliidaes enw. «ese ce

G@loOnidaetses, se eee

@liopsidaesseeess sone n aes

Pneumodermatidae .......

Actaeonidae ..............

PIMP iCuidae eaters nae oe rye
sornatinidacwes ee eee 77 6541
Scaphandridae ............
PNDIUStrIidde? Giese: css cee
Tat aen etesyererte tere erecta
Philinidae) foe. ae. see eee:
Gastropteridae ............
Umbraculidae ............
INDIVSUGACG Gane cess eens -
Pleurobranchidae .........
@nchidiidae 1s) eo.
Veronicellidae ........+... ee
Auriculidae .............. 158
Siphonariidae ............
(Gacimiidace eine eee so Die
Terebridae ............... I .0085
(Conidaer scsi steer Sear ae ee
Pleurotomidae ............ 43 3653
@ancellariidde 2. ....0:22: ae Sak
Olividaes: 5 sécsscscsswitecs It 0934

1.3423

103

Percentage
of species
in this group
among marine
mollusks of the
Southeast
Coast

.4402
.6289
1.1321
.5031
-3145
6918
£0629
.1887
3145
2516
7547
5031

to

.7673
.6289
1.1950

.0629

.0029

1258

.0629

.6289

1258
1.2207
.5060
1258
.6918
“3073
.0629
1258
1258
.1887
.0629
0629
.00603
.1887
.0629
.6289
.6918
8.3019
5031
5031

+

Marginellidae

Turbinellidae

Columbellidae
Pyramidellidae
Amphiperasidae
Cerithiopsidae
Trichotropidae
Seguenzidae ......s55-64<
Murritellidae S.ese weae cs

Philomycidae
Ampullariidae
Assimineidae

Aquatic shells

Number of
identifications

LS

176

SMITHSONIAN MISCELLANEOUS COLLECTIONS

Identifications of Mollusca—Continued

Percentage of
identifications
among those

of all

Mollusca

0595

.1529
0525
3.6021
2.0394
9430

.O169
1.4952

1444
03.40
1.8520
.O169
1104
0595

VOL. Sm

Percentage
of species
in this group
among marine
mollusks of the
Southeast

Coast
2.9560
3145
1258
1.0692
1.6981
1.9407
5031
1.6352
2.6415
3.3962
2516
1.0692
3-3333
1.0063
.1887
5031
3145
3773
.6918
5031
3145
1.0063
6918
1.0063
1258
.0629
.1887
.8176
.3145
5031
3145
3145
6918
2516
.1887
8176
1.5004
3145

1258
.1887

NO. 7 PROTECTIVE ADAPTATIONS—McATEE
Identifications of Mollusca—Continued
Aquatic shells
Percentage of
identifications
among those
Number of of all
Group identifications Mollusca
Wruneatellidae’ wa... 5.006.
@horistidadl 2... cc0re ee ae se
Galyptracidae .... 1. assess > 135 1.1469
Gali emer tier eer Doe
Amaltheidae .............
Xenophoridae ............ anes Sais
INicit1 C1C Gamera en eteteme ta ie areee ie ere 47 3993
Wamieliaritdae: sees. scte we es ae Ae
Ncimaecidaey ame. aclele sane ers 43 3053
Ee petidae meee so ecees cares
Setivel limidaere. tae. cre acusc = aise
Addisonudae 5.2.00 .s.s.26:
@occulinidaes... cca ese
Phasianellidae ............
eiturbimidaew.. .oe4 ven <+0s on otis ee
MTOChIdae any saesia see eee 18 1529
Delphinulidae ............
Cyclostrematidae ......... ee ayes
Nf tetitst cl aaa ee ieee onsen tee ene 374 2773
Stomatiidae: wee qantas
Telia TOT AG a. ss eres weer
Scissunellidaé: ..4.40%a0
Pleurotomariidae .........
Taissurellidae 2.2...:2..... ee ae
@hitonidae =.c 226s s2e oo. 26 2200
Cephalopoda (further uni-
Gentified)s ves cn ere te 86 7306
Wolieinidae .22...02665+24- 3 .0254
Land shells
Percentage of
identifications
in this group
among those
of all land
Mollusca
Number of (except the
Family identifications unidentified )
Cyclostomatidae ..........
diruncatellidaey ass. .saeiee wee Bere
lelicinidde: ..22...0.+..-+ 14 7.9090
Flelicidae 2..6..e00.2.0205 70 39.5479

105

Percentage
of species
in this group
among marine
mollusks of the
Southeast
Coast

.2516
.0629
.5031
.1887
.1887
.1258
1.5004
.1887
3145
.2516
.0629
.0629
3773
.1887
8176
3.9622
6289
8805
.5031
.0629
.0629
.1887
1258
2.2012
1.4465

1258

Percentage
of species
in this group
among all
nearctic land
Mollusca +

5277
1.0554
1.0554

37-7305

*Compiled from Pilsbry, H. A., and Johnson, C. W., A classified catalogue
with localities of the land shells of America north of Mexico. Reprinted from

The Nautilus, 1897-1898, Philadelphia, 35 pp., 1808.

106 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85 .

Identifications of Mollusca—Continued
Land shells

Percentage of
identifications

in this group Percentage

among those of species

of all land in this group

Mollusca among all
. Number of (except the nearctic land
Family identifications unidentified ) Mollusca

Bulimultdae ess eee ae mie 2.9023
Wiocoptidaess yee: Natt sive 4.2216
Papidae: 23.53 tees ae cranes 42 23.7287 13.1925
Nchaniti deus I .5049 1.5831
Glandinidae 25... - ses see- Ae ae 1.0554
sRestaceliltda@ems. se tase ee ee .2038
Circinantidacene eee neces Ae 1.8469 |
Lonitidacweyeencee eee 27 15.2542 17.6779
Mimacidaciee eee nee eree 2 1.1299 1.5831
ANOLOMGLKS Goaneaoceaccega6 ae es 3-6939
Pinlomycidaes ya. I 5649 1.3192
ndodontidae --4-24442400% 6 3.2808 4.2216
Succinetdae eee I4 7.9090 5.8047
Vasinulidae’ 3s. 520.4. see ; an 2638

The outstanding impression given by the foregoing table is that
notwithstanding their relatively low availability to birds, mollusks of
practically all kinds are eaten. In general it is also true that the large
groups more numerous in species and individuals contribute most
heavily to avian subsistence, while the small groups of less abundance
get off with a light toll.

Let us see what are the relations of birds to some of the protected
mollusks. In a paper “On the Adaptive Coloration of Mollusca”
(Proc. Boston Soc. Nat. Hist., vol. 14, pp. 141-145 (1871) 1872),
Edward S. Morse alludes to protective coloration of several species.
His remarks with comment deduced from our tabulations are here-
with presented. “Among the marine forms we notice the adaptive
coloration of certain species very well marked. The common Littorina
of the coast swarms on the bladder weed, the bulbous portions of
which are olive brown in color or yellowish according to age. The
shells of the Littorina found upon it, present in their varieties these
two colors and are limited to these colors ” (p. 143). Our tabulations
show 503 records of capture of Littorina, 11 species of which were
identified from the stomachs of 46 kinds of birds. These shells are
freely eaten as the following instances of large numbers taken by
single birds testify: Pacific eider 110, surf scoter 120, black duck 150,
purple sandpiper 205, and greater scaup duck 350.

i

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 107

“The few species common to the mud flats exposed by the retreat-
ing tide are colored black or dark olive.’ Examples: //yanassa
obsoleta, Nassa trivittata, Rissoa minuta (p. 143). There are 78
records of Jlyanassa obsoleta distributed among 18 species of birds.
Thirty-two of these shells were taken at a meal by a greater scaup
duck, and from 42 to 62 by each of six knots. There are 98 records
of Nassa trivittata, two of them being 275 and 285 specimens in the
stomachs of greater scaups; and there are three determinations of
Rissoa minuta.

Lacuna vincta: The colors ‘‘ quite match the Laminarian upon
which they are found ” (p. 143). This species was identified 39 times
in nine species of birds in which numbers from 32 to 75 were found,
and in one case, that of a golden-eye, no fewer than 116.

“ Margarites helicina | have found in numbers on the large Lami-
narian and on seaweed at low water mark and its color is decidedly
protective’ (p. 144). Our tabulation shows 10 records of this species

‘

of shell, distributed among five kinds of birds.

“A very evident case of protective coloring is seen in the three
species of Crepidula found on our coast. Crepidula fornicata is drab,
variously rayed and mottled with brown, and it lives attached to
stones near the roots of the large Laminarian or upon stones clothed
with algae of similar colors, or attached to the large Mytilus. Crepi-
— dula convexa, a much smaller species, lives on the roots of seaweed.

Professor Perkins records its occurrence on the black shell of
_ Ilyanassa obsoleta, This Crepidula has a very dark brown shell, ac-
cording well with the dark color of its various places of lodgement.
Crepidula plana or unguifornus lives within the apertures of the shells
of larger species of Gasteropods, as Buccinum, Natica, Busycon and
others. The shell of this Crepidula is absolutely white ” (pp. 144-145).

All of the limpets named in the foregoing quotation have been
identified from the stomachs of nearctic birds, and the total number
of records for species of Crepidula is 135. Fifty specimens of
C. glauca were found in one gizzard of a greater scaup duck and 60 in
another. The “ protection” of C. plana is very undependable since
all of the mollusks named as its hosts are swallowed whole by birds
and other predatory enemies of shellfish. With reference to a special
enemy of limpets “ it has been calculated that a single flock of oyster-
catchers, frequenting a small Scotch Loch, must consume hundreds
of thousands of limpets in the course of a single year.” (Cooke,
Cambridge Nat. Hist., vol. 3, pp. 56-57, 1895.)

As an example of the land snails thought to be defended from some
enemies by the toothed apertures of their shells, the genus Polygyra

108 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

may be mentioned. Twelve species have been identified from the birds
represented in our tabulations, the total number of determinations
being 42. Slugs were identified three times, but our findings in this
respect probably are not representative since in Great Britain it is
said that: “ Every kind of slug and snail is eaten greedily by black-
birds, thrushes, chaffinches, and in fact by many species of birds.”
(Cooke, Cambridge Nat. Hist., vol. 3, p. 58, 1895.) With regard to
highly colored shells, such as Pecten, conjectured by Wallace to be
protected, it may be said that our table shows 193 records for Pecten
and the Biological Survey has been called upon to make a special study
of damage to the scallop industry by wild fowl in the vicinity of
Marthas Vineyard, Mass. Teredos were identified from the stomachs
of four Bachman oyster-catchers and of one egret.

It should be remarked that the very large number of records of
Ostreidae in the tabulations is due also to a special investigation of
the bird enemies of Ostrea lurida. The high records for Tellinidae
(Macoma, especially) and Paphiidae (Paphia staminea) are chiefly
by-products of this same study. The large numbers of identifications
for such dominant families as the Mytilidae, Nassidae, Columbellidae,
and Litorinidae among marine shells and Amnicolidae and Lymnae-
idae among fresh-water ones, need no explanation.

Number of identifications, 11,771; percentage of identifications
among those of all animals, 4.9583; percentage of species in this
phylum among the whole number of animal species known, 10.8828.

Other enenies——Mollusks are preyed upon to an important extent
by very many marine fishes, as well as by most of the rays and sharks ;
among these being numerous forms specialized (as by possession of
the pavement-like pharyngeal dentition) to feed upon shell fish. For
such fishes as the haddock, cod, wolffish, and flounders they are a
staple food. Field found razor-clams in 4 per cent of 388 stomachs
of the smooth dogfish (Mustelus canis); and in 3.68 per cent of
516 stomachs of the summer skate (Raja erinacea). The same author
found mollusks of various kinds in 17.64 per cent of the stomachs of
306 cunners (Tautogolabris adspersus), and in 27.2 per cent of those
of 33 toadfish (Opsanus taw). Some of the marine fishes are known
to be enemies of certain specially defended mollusks ; as predators on
Crepidula, the scup, tautog, swellfish and toadfish may be mentioned ;
upon chitons, the haddock and flounders (Pleuronectes) ; and upon
Eolis and other nudibranchs, the cods, gurnards and_ flounders.
Cephalopods, especially squids, are a favorite food of many of the
highly predacious fishes as the sharks, rays, bonito, swordfish, blue-
fish, mackerels, pollock, and haddock. It would be easy to compile a

Sena er
9

NO. 7 PROTECTIVE ADAPTATIONS—McATEE LOg

very long list of squid-eaters. As to the extent to which these cephalo-
pods are taken, Field reports squid from the following percentages
of the stomachs examined by him: Summer skate 6.39 per cent,
smooth dogfish nearly 10 per cent, and goosefish 17.39 per cent.
Cuttlefishes are known to be eaten by the bonito, cod, whiting, and
gurnard, and octopods by the ling, haddock, and conger eel.

Turning to the fresh-water mollusks, we find that they are equally
beset by enemies. Pearse reports 2 per cent of the total food of
32 species of fishes in Wisconsin lakes: consists of these animals, and
from Forbes we learn that mollusks make up about one-fourth of
the food of the dogfish (Amia) and a sheepshead (Aplodinotus),
about half that of the suckers (Catostomus), rising to 60 per cent in
the case of the red-horse (Neorostoma), and a considerable propor-
tion (14 to 16 per cent) of the food of the perch (Perca flavescens),
catfishes, sunfishes, top minnows, and shiner (Abramis). Almost all
fishes eat mollusks to some extent and practically all groups of
mollusks suffer from these predatory attentions.

Taking up the relations of amphibians to mollusks, it may be noted
that Kirkland found 1 per cent of the food of 149 toads to consist of
snails and slugs, and Drake found 29 of these mollusks in 209
stomachs of the leopard frog. In general it may be said that most
frogs consume aquatic snails when in the larval state and land snails
when adult. With reference to European conditions, Cooke adds:
“Frogs and toads are very partial to land mollusca. A garden at-
tached to the Laboratory of Agricultural Chemistry at Rouen had been
abandoned for three years to weeds and slugs. The director intro-
duced 100 toads and go frogs, and in less than a month all the slugs
were destroyed.” (Cambridge Nat. Hist., vol. 3, p. 58, 1895.) Snails
are eaten by most salamanders, the kind, whether water or land,
depending on the habits of the salamanders concerned ; small mussels
even are consumed by some of the thoroughly aquatic forms.

Reptiles do not prey very extensively upon mollusks, yet snails are
frequently eaten by lizards; slugs and snails are eaten by several
species of snakes and by most turtles, the aquatic forms of the latter
group consuming some bivalves.

Among mammals we find that some of the land forms consume
mollusks to a slight extent ; shrews, rats, white-footed mice, squirrels,
and chipmunks may be mentioned as examples; a specimen of the
eastern chipmunk (Futamias striatus) taken near Fairfax, Va., had
packed in its cheek-pouches or swallowed more than 47 Pomatiopsis
lapidana. It is well known that the muskrat preys extensively upon
fresh-water mussels, and the mink and otter must be listed as foes of

1 }

Ilo SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

fresh-water mollusks. The food of the walrus consists mainly of
shellfish. Dyche reports that the California sea-lion feeds very largely
upon squids and octopods and it is known that squids form a con-
siderable proportion of the diet of sperm and other whales.

The enemies of marine mollusks include also sea-anemones,. star-
fishes, and boring univalves of the genera Purpura, Polynices, Thais,
Lunatia, Natica, Cycotypus, Fulgur, and Urosalpinx. Fresh-water
mollusks form an important element of the food of dragonfly nymphs,
and a lesser one of horse fly larvae, water beetle larvae, water bugs,
leeches, and crayfishes. JIand snails are attacked by predacious
beetles and fly larvae. Mollusks also have enemies among such para-
sitic groups as mites, nematodes, and trematodes.

Discussion—From the abundance of their enemies and from the
extent to which these predators feed upon mollusks (more than 8,000
records for birds in our tabulations), it is evident that the possession
of a shell as a means of defense has been entirely discounted so far
as predators of any size are concerned. The relations of birds to the
protectively colored forms show that some of these (Litorinidae) are
freely eaten; the brightly colored shells (Pectenidae) also are freely
taken, as well as the very hard and thick-shelled ones (Ostreidae).
Slugs, snails, limpets, teredos, chitons, and cephalopods pay their toll
also, testimony to the all-pervading search for food by birds. In fact
the evidence is that birds feed more or less indiscriminately upon all
mollusks of suitable size that are available to them. Other enemies
follow mollusks, especially the marine forms, where most birds can
not, and it would seem that the whole molluskan world is exploited
as a source of food to as large an extent as could be expected.

CHORDATA (LANCELETS, TUNICATES, VERTEBRATES)

While the Chordata with 13,326 identifications contribute only
5.6133 per cent of the total determinations of the animal food of
birds, yet the phylum comprises so many familiar animals that it
probably will be best to treat it more in detail as was done in the case
of insects.

Number of identifications, 13,326; percentage of identifications
among those of all animals, 5.6133; percentage of species in this
phylum among the whole number of animal species known, 8.8427.

A tabulation of the records of Chordata with frequency indices de-
rived from estimates for the world fauna gives the following results:

NO. PROTECTIVE ADAPTATIONS—McATEE Gt

N

Identifications of Chordata

Percentage
of species
in this class
among those

Percentage of of the whole
identifications number of
among those Chordate
Number of of all species
Class identifications Chordates known
Wirochorda. .. ..c0. sea. e=s 5 .0375 2.6228
HEATS COSiersers eretaercistavers cao © 4,92 36.9427 26.2281
Xm piibia #e .2ss0 see eso 097 7.4816 4.4386
TED tll ice eerste teeters aiskeiseae ior 6905 5.2153 12.1053
PNGVGS: a Hisseses tei) atl eis Hit als S05 3,555 20.6771 40.3510
Mammalian er emcccys s)..<.<iek 3,151 23.6454 14.1228

The urochordates listed are ascidians, in three cases being identified
as Boltenia ovifera. While the identifications of urochords is far
from proportional to the frequency of these animals, the result is
only what would be expected in view of their strictly marine and
chiefly submerged habitat in which they are exposed to the attacks of
only a very small proportion of our birds. It may be noted here that
tunicates have numerous enemies, however, among fishes which take
the pelagic and both simple and colonial fixed ascidians. It is on
record that these animals are not uncommon in the stomachs of cod
and haddock, and they have been found also in herring, flatfishes,
tilefish, cunner, scup, the great sunfish (Mola), and a number of
other fishes. They are taken also by sea-anemones and sea-urchins.

Omitting the Urochorda and figuring frequency indices from the
fairly well known numbers * of North American species in the various
classes we obtain the following table: °

Identifications of Vertebrata

Percentage
of species

Percentage of Number of in this class

a identifications nearctic among those

among those species of all

Number of of all in this nearctic

Class identifications Vertebrata class Vertebrata
Pisces ....2 4,023 36,0565 3,054 61.3253
Amphibia .. 907 7.4844 141 2.8313
Reptilia ... 605 52073 308 6.1847
PN CS Sted seks! BISGs 26.6870 Sol 16.0843
Mammalia . 3,151 23.6542 676 13.5743

In this as in other cases we clearly observe the tendency for losses
to predators to correspond to the extent and abundance of a group.
In fishes the largest class are preyed upon the most, but less than their

‘Counts derived from standard works on the various classes, as noted later in
connection with the tabulations by classes.

8

II2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

relative abundance would seem to warrant, for the reason that, as a
group, they are relatively inaccessible to birds, many of the deeper-
water forms being entirely so.

Pisces (FIsHEs)

|
|

Protective adaptations.—It has been held that the great group of
spiny-rayed fishes is protected from enemies to a greater degree than
the soft-finned families, and in general harsh scales and spines are
deemed protective. Some fishes have poison glands connected with
certain specialized spines. Some of the species with disagreeable
qualities have colors that are said to be warning, while the great |
majority of fishes exhibit varying degrees of cryptic coloration, many |
of them having more or less ability to change in color in response to
that of their environment. Such in brief are some of the more or |
less theoretical defenses of fishes; as to actual physical protection, it |
may be said that fishes are shielded from most birds by their aquatic
habits and many of them even from water birds by their living at |
considerable depths.

Bird enemies.—It is well known that whole families of the so-called
lower orders of birds are specialized to prey upon fishes, for example
the loons, terns, cormorants, anhingas, pelicans, mergansers, herons,
and kingfishers. There are special fish eaters in other groups, and
many birds not at all specialized to prey upon fishes nevertheless con-
sume them to some extent more or less habitually. Nearctic birds
which subsist almost exclusively upon fishes include: the western
grebe, Caspian, royal and Cabot terns, black skimmer, anhinga, double-
crested cormorant, brown and white pelicans, man-o’-war bird, and
osprey. Others making fishes from 50 to go per cent of their diet are:
the common loon, Holboell’s grebe, black, Mandt’s and pigeon guille-
mots, common and Brunnich’s murre, kittiwake, glaucous-winged,
herring, and ring-billed gulls, gannet, violet-green cormorant, Ameri-
can and red-breasted mergansers, bald eagle, and belted kingfisher.

Nearly 5,000 records of fishes being eaten are contained in our
tabulations of the food of nearctic birds, and of these nearly half were
identified no further than the class. The remaining determinations
grouped by families are listed herewith:

7

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 113

Identifications of Pisces

Percentage
of species

Percentage of of this group
identifications among
among those North

Number of of all American
Group identifications fishes fishes 1
Wtdentified ...0606..000 0% 2,253 45.7052 err
Branchiostomidae ......... : : .I310
Heptatremidae ........... mie ae .0327
IMiyxinidaege. ......ce0s6css trys cine 0327
Petromyzonidae .......... se ae 3274
Chlamydoselachidae ....... or se .0327
Elexanchidae ......:.6.0.«- er Ss 0982
Heterodontidae ........... Sars aes 0055
Scylliorhinidae ........... toe ae 1965
Ginglymostomidae ........ ake ave .0327
Pseudotriakidae .......... Aer ae .0327
SLEICAG: (on ger avare Miata’ ent aves .9823
Syria! aa..sc sos siete xneee — Sa .0982
PUOPNGAS facveee die desde ne na .0327
amchiaridae <.sck..deeaeales ‘hig we .0327
IBamnidae wie lace cree se «le spots <a .1637
Cetorhinidae ............. mar Sears .0327
Rhinodontidae ..c......: ee oak .0327
Squalidae’...............- nets ee .1637
Walatidaee ascent es ecer Sieg ee 0827,
Echinorhinidae ........... ae aes .0327
Squatinidae .............. oe Sate .0327
PP TISHOAC Is. daa tee ate acnaas + ore aa 0655
Reino DatiGae’ a). siete ee ee ote .2047
Rand aes a wets See.ccuncweuie rere ae .6221
WNarcobatidae <...........- or Ae .1310
Dasyatidae o....... 08000 sn dds a 5239
Myliobatidae ............. we 2 26019
Miaritrd aes es. a s00 se00 2s sae aes .0055
Chimaeridae ............. ere rts .1310
Polyodontidae ............ mE ee .0327
Acipenseridae ............ ae .1965
| Lepisosteidae ............. 5 1016 1310
| ATHLIGAC! Bias es su vie asso uses I .0203 .0327
| Siluridae ............0.0.. 100 2.0313 3.1107
Koricaritdae ac. sse sees ee te 3274
Catostomidae ............. 62 1.2504 2.1284
@yprinidaeess.. ssm.< eee 482 9.7909 7.3346
Prythrinidae. ....3<6.00s.. aes ae .0327
Characinidae ............. ae at 6221
RpOdes: id vase ceie vdeo « 2 .0406

*Computed from Jordan, D. S., and Evermann, B. E., The fishes of North
and Middle America, etc., U. S. Nat. Mus. Bull. 47, 4 vols., 1896-1900.

SMITHSONIAN MISCELLANEOUS COLLECTIONS

Identifications of Pisces—Continued

Percentage of
identifications
among those
Number of of all
Group identifications fishes

Gymnotidae =. -se ee ooueee

Symbranchidae :...:......

Denichtyidaey aa.sscgene eer cae oe
Anenillidae 252.425. cene. 16 -3250
Simenchelyidae ...........
ivophidacmernn cen see ri ee
Synaphobranchidae .......
Leptocephalidae ..........
Muraenesocidae ..........
Nettastomidae ...:.....0+.
INemichthyidde (254. e0..6-
IW ARCELOE pers otrs Bo auuoeee
Ophichthyidaewsasssse. eee I .0203
Miitraentdaer as eeimeeite rier

Saccopharyngidae .........

Eurypharyngidae .........

Blopidacey eee scene:

Aibulidaeieen see eer

Ehodontidacwessrceee eer

Ghanidaca.-s see eee

Dorosomidae

Glugeidae I Areca eee 249 5.0579
Bneraulididae a: . 4s. esse SI 1.0360
Alepocephalidae .......... aan ers
Salmonidae, 2. sss ne ee cs 198 4.0220
ehymallida cesar eer rst Se nits
Argentinidae ............. 30 .6004
Microstomidae ...........
Synodentidae jcc ce see
Ailopidaeiecrsiic seraeree eter
Benthosauridae ...........
Bathypteroidae ...........
Ipnopidacue enone eee
Rondeletiidae .............
@etomimidae’ 3.2 )..2.56..-
Myctophidaes 2.5 reer: <-
Maunolicidae, Jc. asses

to
°
HR -
One
aN

.0406

bo:

Astronesthidde ...........
Stomiatidae <5. 2s seeeesee
Malacosteidae ............
Allepisaunidae’ s.0ce sce...
Odontostomidae ..........
Paralepididae s/.4.5.42702%

VOL. 85

Percentage
of species
of this group
among
North
American
fishes

0655
.0327
.0327
.0327
0327
.0327
.0982
.2619
-4584
0655
2292
1310
.9496
.9405
.0327
.0327
£0655 |
0327
.0982
.0327

1.3097

8841
3602
1.0478
0055
-3929
1310 |
‘sr
.1310 |
0655 |
"0655

.0327

.0327

0655

.4080 |
.0655
-2047
0982
.1965
.0327
.1637
.0327
.1965 |

-

py

NO. 7 PROTECTIVE ADAPTATIONS—McATEE

Identifications of Pisces—Continued

Percentage of

identifications
among those
Number of of all

Group identifications fishes
Sternoptychidae ..........
Idiacanthidae .
falosauridae ..24...0...+.
Notacanthidae ..........
iEmpogenyidae 2.2.00 .56 2.6%
Byam it aes Acres ziwc els ee ds a8 ee a
lWimbridae .aacase.s Ss vena: 9 .1828
NE CIT AC Uet nnd sa Ne etacsniseetasns Be .4460
IPoeciliidae 2... ..2.2.0..4-- 530 10.7659
Amblyopsidae ............ ae aoe
ESO CIA aeemne cuter earsieiiny yas 8 162
Hemiramphidae ........... 4 O82
Scombresocidae ...... av a ae
TX OCOCHGAG kavisisa acs coe se I .0203
Gasterosteidae ............ ToT 2.2547
Aulorhynchidae ..........
Aulostomidae .............
astulariidae™ ..25.....220s.
Macrorhamphosidae..... ss fae
Syngnathidae ............ 6 1210
Percopsidae oi... ...6.< . —
Aphredoderidae .......... I .0203
Atherinidae .............. 38 .7719
Mugilidae ................ 18 3656
Sphyraenidae ............
Polynemidae ............. are et
Ammodytidae ............ 25 .5078
Bathyclupeidae ...........
Stephanoberycidae ........
Trachichthyidae ..........
ELV CIAAE! 2. one es: Stasi
Holocentridae ............
OLVEMINIGAC seen. yee sions
Mullidae ................. Sede
Scombridae .............. I .0203
Gempylidae ...............
Lepidopidae ..............
Trichiuridae .............
Istiophoridae .............
Xela t1daes yaa vate es ays senate
Nematistiidae ............ ae :
Carangidae ............... 7 .1422
Pomatomidae ............ 4 0812
Rachycentridae ........... I .0203

Percentage
of species
of this group
among
North
American
fishes

.0982
.0055
2202
-1905
.0327
.0327
.0055
.1965
3.8638
.1310
.6221
2292
.0055
.6221
.2292
0327
.0327
0982
.0327
1.1460
.0055
.0327
151033
.5566
.1637
.1637
.1310
0327
.0055
.0327
-3274
-4584
.0327
2619
4912
2202
.1310
.0327
0982
0327
0327
1.0074
0327
0327

FT

116 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Identifications of Pisces—Continued

Percentage
of species
Percentage of of this group
identifications among
among those North
Number of of all American
Group identifications fishes fishes
Nomeidaer saan leciyeeieesiee see id .1637
Coryphaenidae -.-.-....-. I .0203 .0055
ampridae) -... 2d: sls steve sare syle .0327
Pterachidaey soe ..c cern a sae .0327
Branndaea aan ss others siete Rahs .1310
Stemeseniidae ta. =. she. ne cial 0327 |
Centrolophidae” Gros... 5s. ste oes 0055
Stromaterdaes meee tiels me siels 1965
Icosteidacynemsee cere ee wee ne .1310
Grammicolepididae ........ ee tre 10327
Metragonunidae te. oe eee e a. oe .0327 |
Pempheridae’ (:4.2..-21--- oe ae .1310
Bilassomidacwnssriece: tenes aes Bae .0055
Centrarchidaeia sacs eens 164 23313 1.0478
KGQhIMIMGEYS) bossastendoodacs nee ees .0055
Percidacsn eee eee 171 3.4735 2.8160 |
Cheilodipteridae .......... : ae .4Q12 |
Centropomidac) -y...-4-cr ae ae 4257 |
Serrantdae a. wa. ee eerie = 5 .1016 3.0779
Ioloyolobi Vansengepoooudoc ae acy. .0327
Priacanthidae cease eee aoe a .1310
Mutianidae 4 asec eaene se ne 1.1133
Elaemtlidae neces sree 2 .0406 1.8009
Sparidae Ae oanc cee acer 5 1016 7858
Miaenidae meee cae cien nh See 0655
Gerridaceae eee eee ter 3 .0609 5500
Keyphosidae saeeees see ae Rte 3929
Sciaenidae) Go seeee ace. sees 14 2844 3.5036
@innhitidae sea-ice eee ae sas .0655
Embiotocidae =. 2..:.000 ee 7 .1422 5804
Gichlidaeyeeane eer eee er here er toe 1.8337
Pomacentridae (2... cee ie ware .9823
aloridaer ae ds sess eer ne aleve II 2234 1.6044
Scanidageae sacnensoo eee ste Pre 1.4407
LADAC asics oc hs a ae ciclo ane ee .0982
(GEROMEV Goo ogbaccsoomucs ete nee .0055
Epliippidaein -siierieee crete nrae ee .0g82
Chaetodontidae ........... a ner 6549
Lanchdacee sere sys Seis .0327
Meuthididaeyy.c)s cess ee nes Seay .29047
Mrtacanthidae: saeco aa ae .0327
Balistidacmey.arcae eee ; mere 5239
Monacanthidae ........... 3 .0609 3002

@stractidaeweawac aciierle ter ae ord: .1310

nee. PROTECTIVE ADAPTATIONS—McATEE 7

Identifications of Pisces—Continued

Percentage
of species

Percentage of of this group
identifications among
among those North
Number of of all American
Group identifications fishes fishes
Tetraodontidae ........... ee aes 6221
Canthigasteridae ......... Su ea .0055
Diodontidae .........6.0.. I 0203 3602
IMIG OGG oo nec cds kaye em i bes Sar 00655
SCOMpaenidae sa..s%0. od es aoe Save 2.6523
Anoplopomatidae ......... Lee .0055
Hexagrammidae .......... 4 .O812 3274
(Botticla G see vis eters ensiete cadver ees 188 3.8188 4.2567
Ramphocottidae .......... on eats .0327
ENOOM1 AG maeicrser eis ct asei Se eee 2 .0406 1.1788
Gy clopteridde, mera. ce sm ane Sav. 2047
{ NEiparididaes.. ase« vol. <6 sive sas 1.2443
Merrmlidae ec ectes «00 aces oa I .0203 8186
| Peristediidae ............. Meee oe .1310
Cephalacanthidae ......... ater oe .0327
\ Callionymidae ............ Sash ae 1310
GOBNGAC: .uSee2.u.dneccawe a 10 .2031 2.8487
Echeneididae ............. ee . aes .2292
i Malacanthidae ........... See ae 51637,
| Opisthognathidae ......... setae any 3602
Bathymasteridae .......... ate Aa 0982
Chiasmodontidae .......... oe ae .0055
Chaenichthyidae .......... ayes ne .0327
| Mrichodontidae®..3.....0.«: ae ae 0055
Dactyloscopidae .......... ean A327.
Uranoscopidae ............ 2 .0400 .1637
Batrachoididae ........... 17 3453 327d
Gobiesocidae ............. I .0203 S841
Blenniidae ................ 160 3250 4.5187
Cryptacanthodidae ........ Boe a .0982
Anarhichadidae ........... I 0203 1905
@erdalidac ee eee ae iene .0982
Ptilichthyidae ............ Bes ane .0327
Scytalinidae .............. one ms .0327
| OATCICAG ee acne eee ae ae 1.0805
Derepodichthyidae ........ oe siipt .0327
Ophidiidae ............... ae wos 5506
Lycodapodidae ........... oe 2 are .1310
BPierasteridae-. 0 < nie seas wees me 0982
| Brotwhdae’ cb ccusice sha idee ee tie 8841
| Bregmacerotidae .......... — eas .0055
Merlucciidae ............. ee ee .0982
| (Graditdac eee weiter 31 .6207 1.1788
| Macrounidae ....e-0eee os: ee 3 1.0805
|
|

118 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Identifications of Pisces—Continued

Percentage
of species
Percentage of of this group
identifications among
among those North
Number of of all American
Group identifications fishes fishes
INesalecidacwer ea eo eer ae wa .0327
rachypteridae s2.4.ceeee Bee ah £0982
Stylephonidaceen seer ae aC .0327
Pleuronectidae i
TER INGEST ay aes 0 60 798
Soleidae if : 2 3/2
Lophidaes ee naa oeeae ake ae .0327
Antennariidae, sas250 4.666) ae a: .4Q12 |
Ceratiidacnckh sana eceoues tae ne 3274
Ogcocephalidae ........... ae eee 3274

Total number of identifications of fishes, 4,923; percentage of iden- |
tifications among those of all vertebrates, 36.9565; percentage of
species in this class among those of all nearctic vertebrates, 61.3253.

Commenting on this table it is obvious that a wide range of fishes |
is preyed upon, and that the families known to be most abundant in |
individuals almost invariably are those most extensively consumed by
the birds. As to the bearing of this data on protective adaptations, we
see the spined catfishes well represented, more so in fact than the
equally abundant and only negatively if at all defended suckers. No
fewer than 36 small catfishes were found in the stomach of a single
belted kingfisher. The very spiny sticklebacks are eaten enough to
show that their spines are no deterrent to the attacks of birds; no
fewer than 150 of these little fishes have been taken from the stomach
of a great blue heron. Advancing to the true prickly-scaled and spiny-
finned fishes, we note that Centrarchids (sunfishes, bass, etc.) and
perches are freely taken. High counts of sunfishes in stomachs are |
12 in that of a least bittern, 14 in an anhinga, and 18 in a little green
heron. Twenty yellow perch have been eaten at a meal by the least
bittern and the great blue heron and no fewer than 25 darters by the
little green heron. The Cottidae or sculpins often have a highly
developed armature of spines about the head, but there is no evidence
that it protects them from birds. Flatfishes (Pleuronectidae) repre-
sent almost the acme of protective coloration, especially of power to
simulate the background, but they seem to be proportionally repre-
sented in our table. One double-crested cormorant had eaten 16
Symphurus plagiusa.

The unidentified fishes were distributed among approximately 165
species of birds to which a considerable number would have to be
added to give the total number of fish-consuming species. A family

a

NO. 7 PROTECTIVE ADAPTATIONS—MCcATEE 119
of fishes that almost everything “ picks on,’ such as the minnows
(Cyprinidae), was represented in the stomachs of 44 species of birds.
Numbers up to 50 of these little fishes were found in stomachs of the
belted kingfisher and the hooded merganser, and in the case of young
of the common carp a high count of 106 was made from the stomach
contents of a glossy ibis. Thirty-nine species of birds are known to
prey upon the common killifishes and their allies ; and numbers were
taken from many stomachs, the maximum being 526 from a little
blue heron.

Other enemies.—Fishes have no more destructive enemies than the
predacious element among their own kind. Among the highly preda-
tory marine forms may be mentioned the dogfishes and other sharks,
swordfish, bluefish, squeateague, conger eel, and the angler, and among
fresh-water fishes, the gars, sculpins, trout, amia, pikes, and bass.
In a study of the fisheries of Buzzard’s Bay, Mass., Field estimated
that two species of sharks destroy more than 500,000 fishes annually
in that body of water. Pearse found fishes to compose 12.3 per cent
of the food of 32 species of their class in Wisconsin lakes. Forbes
notes that the principal piscivorous fishes of Illinois, those which
obtain three-fourths or more of their total subsistence from their
fellow fin-bearers, are Lota, Stizostedion, Esox, Micropterus, Icta-
lurus, Leptops, and Lepidosteus. Six other species are listed as taking
from 25 per cent to 65 per cent of fish food.

Predators devour fishes in all stages, and there are numerous
special enemies of fish spawn; worst among these are other fishes
such as the suckers, sculpins, minnows, sticklebacks, killifishes, top
minnows, and trout.

Not many enemies of fishes are numbered among our batrachians
and reptiles, those worthy of note including only the bullfrog, Nec-
turus and Cryptobranchus of amphibians; the king, garter, and water
snakes, copperhead, rattlesnake, and cottonmouth moccasin among
snakes; and the painted terrapin, and snapping and _ soft-shelled
turtles.

Some mammals are important enemies of fishes but the number is
not large; we may mention the raccoon, mink, otter, seals, sea-lions,
porpoises, and whales.

The young of fishes especially fall a prey to a variety of insects, as
the larvae of aquatic beetles and of dragonflies, and to several kinds
of water bugs and to hydras. Insects also, and crawfishes and leeches
prey upon the eggs of fishes, and squids are said to be among the most
destructive foes of adult fishes. Parasites of fishes abound and are
recruited from the ranks of such diverse groups as bacteria, proto-

120 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

zoans, cestodes, trematodes, and crustaceans. Fishes are destroyed in
large numbers sometimes by fungoid diseases ; and enormous numbers
of them perish by being stranded in pools, overflowed by high tides or
freshets, which later dry up.

Discussion.—Forbes in his discussion of the ‘‘ Food Relations of
Fresh-water Fishes” notes that: “ The soft-finned fishes were not
very much more abundant, on the whole, in the stomachs of other
species than were those with ctenoid scales, spiny fins, and other de-
fensive structures,—an unexpected circumstance which I cannot at
present explain” (p. 479). The natural comment upon this remark
is that the fact detailed does not need to be explained, only accepted,
theoretical bias being cast aside. He goes on to say: “ Only the
catfishes seem to have acquired defensive structures equal to their
protection, the predatory apparatus of the carnivorous fishes having
elsewhere outrun in development the protective equipment of the best-
defended species” (p. 480). Examining the basis for this statement
we find that Forbes examined the stomachs of about goo adult or
nearly adult fishes, and that he found catfishes in five of these
stomachs; darters were identified only four times, whitefish only
twice, and round suckers only three times, yet all of these are groups
which equal or exceed catfishes in abundance. There is no reason
therefore for saying their defenses are unusually efficient ; from the
table on p. 113 we see that birds take catfishes in due proportion.

Some kind of protectedness is claimed for practically every kind
of fish, yet we see that all groups of them are devoured by natural
enemies, and where data is available, predation seems to be very much
in proportion to abundance. This principle is especially evident in
depredations upon fishes if carried back through the life history of
these animals; young fishes are more abundant than adults and they
are greedily devoured by many piscivorous animals; while fish eggs,
most abundant of all, are sought by a perfect swarm of predators.
The grand principle of predation proportional to population is well
supported by the known relations of fishes and their foes.

AMPHIBIA (SALAMANDERS, Toaps, FroGs)

Protective adaptations ——All amphibians have skin glands that se-
crete a slime which some have thought to function partly as a defense.
Toads in particular, frogs to a lesser extent, possess poison glands
also, and “‘ experiments have proved that toad poison injected into the
system will kill any vertebrate, the dose being proportionate to the
size of the animal.’’ (Dickerson, Mary C., The frog book, p. 17,

-

NO. 7 PROTECTIVE ADAPTATIONS—McATEE I21I

1906.) Some amphibians have warning colors, but it is noticeable that
the nearctic species having such coloration (certain Ambystoma) do
not possess especially noxious secretions, while our toads are not at
all warningly colored. The real defense of most amphibians lies in
their habits, such as aquatic life, nocturnal activity, and lying in
seclusion in burrows or under logs and rocks. Most of the species are
very fecund also.

Bird enemies.—The extent to which the various families of Am-
phibia have been identified from the stomachs of nearctic birds is
shown in the subjoined table.

Identifications of Amphibia

Percentage
of species
Percentage of of this group
identifications among
among those North
Number of of all American
Group identifications amphibians amphibians ?
Unidentified .............. 132 13.2396
Urodela (further unidenti-

TEC) oer eat cont aayreroainees esac 124 12.4372 ees
iINe@cturidae=ssa0.5. sents 4 .4012 1.4184
Typhlomoigidae .......... sists oe .0709
Amphuimidae ............ oe oe 1.4184
Cryptobranchidae ......... — ee £0709
Salamandridae ............ I 1003 227
Ambystomidae ........... 16 1.6048 14.8932
Plethodontidae ........... 8 .802 31.9140
DSIFEMICAC ssc aiele cesses - a Fett 1.4184
Anura (further unidenti-

LIEGE) 5 Maes sts teyercee tks ce sae, 3 4o 4.0120 ae
Discoglossidae ............ ae oan: .0709
Scaphiopodidae ........... wae Bie. 2.8368
Bttonmidaey sees eels ee 60 6.0180 9.9288
Hylidae ................-- 77 7.7231 14.1840
Leptodactylidae ........... mene ae 2.8368
INatidae eee ees esis 535 53.6605 12.0564
Brevyicipitidae, .... seen cr ie 2.8368

While the Ranidae are more abundant and accessible to birds than
most of the other amphibians, even so they seem considerably over-
represented in the preceding tabulation, a circumstance that is ex-
plained in part by the fact that greater numbers proportionally of
the stomachs of aquatic birds have been examined than of any other

group.

* Computed from Stejneger, L., and Barbour, T., A check list of North Ameri-
can amphibians and reptiles, pp. 5-40, 1917.

I22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85 ;

Fifty-three species of birds are recorded as preying on Ranidae, of
which the common crow is the most voracious (numbers as high as
24 and 29 individual frogs being counted in stomachs of this species),
and has the largest number of records (197). Thirty-four frogs were
found in the stomach of one little blue heron.

Among 14 species of toad eaters, the following are most important :
common crow 16 records, red-tailed hawk 10, red-shouldered hawk
9g, and broad-winged hawk 9. The most frequent consumer of
salamanders also is the common crow with 83 records.

Total number of identifications of amphibians, 997 ; percentage of
identifications among those of all vertebrates, 7.4844; percentage of
species in this class among those of all nearctic vertebrates, 2.8313.

Other enemics.—Fishes occasionally eat the eggs of toads and
frequently devour tadpoles of both toads and frogs, and the larger
predacious fresh-water fishes are fond of frogs. The bullfrog
especially preys upon other frogs and the gopher frog is a special
enemy of toads. The Anura more or less frequently are cannibalistic
upon the young of their kind, while larvae of salamanders regularly
devour their brethren. Aquatic salamanders also eat the eggs and
larvae of frogs. Snapping turtles, soft-shelled turtles, and alligators
prey upon frogs, but it is particularly among snakes that the most
deadly enemies of the Anura occur. The garter snakes and hog-nosed
snake are especially fond of toads, while snakes in general eat frogs
and also salamanders. In their account of the “* Snakes of Okefinokee
Swamp,” Wright and Bishop report that: ‘‘ With the larger snakes,
the food most generally sought is Anura or Amphibia in general.
It is par excellence the food of the aquatic snakes, and with these four
or five species it is usually some species of Rana, though Acris, Choro-
philus or Hyla may rarely appear as their prey. [qually important
are frogs in the food of the larger land snakes, five species being
addicted to them. With these the southern and oak toads (Bufo) are
easily of first importance, with the tree frogs (//yla) and the narrow-
mouthed frog (Engystoma) occupying second and third places. In
fact, these 10 snakes prefer the soft-bodied frogs and toads to any
other food of the swamp (reptilian eggs not considered).” (Proc.
Acad. Nat. Sci. Philadelphia, vol. 67, p. 147, Apr. 1915.)

Among mammals the skunk is known to be fond of toads, and
coyotes, skunks, weasels, minks, otters, wildcats, and the brown rat
feed upon frogs. No doubt most of these animals will take sala-
manders also when the opportunity occurs; the little spotted skunk
and coyote are definitely known to do so, one stomach of the latter
animal yielding 15 Ambystoma. The mongoose was found to feed

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 123
commonly on toads and frogs in three separate investigations of its
food habits in Trinidad.

Leeches prey upon both eggs and young of amphibians and there
are numerous insects which destroy tadpoles. Such are the giant water
bugs, backswimmers, water scorpions, predacious diving beetles, and
their larvae, and the nymphs of dragonflies. Finally, it should be
mentioned that myriads of amphibian eggs and young perish because
of the unwise choice by their parents of too temporary bodies of
water for their egg-laying.

Discussion—The relations of predators to amphibians throw an
interesting light on the efficiency of protective adaptations in averting
the attacks of foes. Clearly the Ranidae or frogs are more preyed
upon than any other group, certainly much more so than the toads.
The theorist on adaptations attributes this to the superior special
defenses of toads, but with no doubt whatever the difference in
amount of predation on these two groups is a direct reflection of their
relative abundance.

If toads really were specially protected, if their so-called defenses
actually saved them from a certain proportion of predatory attacks,
should they not increase continually relative to the Ranidae? The fact
that they do not is the best proof that could be asked that their
“ special defenses ” do not actually function in nature. In short there
is no reason to believe in the case of amphibians but that the attacks
of predatory enemies bear a close relation to abundance and availa-
bility of the various orders and families. Where a certain group
appears to have an advantage in escaping certain foes, to a degree, it
invariably proves that it suffers extraordinarily from attacks of other
enemies.

Repritia (TurtLes, LizArps, SNAKES)

Protective adaptations —Although turtles have the direct defenses
of their shells, jaws and claws, several of the species have also a
strong musky odor, and some exhibit warning colors. Numerous
lizards have cryptic coloration; one of our species is poisonous and
one has the faculty of changing its color considerably. Many lizards
drop their tails easily, a device said to aid them in eluding enemies.
The horned-toads besides their protective coloration have more or
less prominent spines on the back of the head. Many snakes exhibit
cryptic coloration and a number of them have offensively odorous
secretions. Certain serpents practice intimidatory actions and a con-
siderable number of our species are dangerously venomous.

124 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Bird enemies.—Following is a tabulation of the records of Reptilia
found in the stomachs of nearctic birds. The total number 695 seems
proportional to the abundance of animals of this group in the United
States.

Identifications of Reptilia

Percentage
of species
Percentage of of this group
identifications among
among those North
Number of of all American
Group identifications reptiles reptiles 1
Winidentittedit ere oe 21 3.0215 aah
Crocodylidae .. 5.06. e ee I .1438 .6404
Lacertilia (further uniden-

tied) eer anc ee 140 20.1432 Soe
Gekkontdaemeons mene ee ae Ae 9741
Kublepharidae) s+. cm. eee war Sais -6494
lewanidaeweeeateteeeeee 47 6.7624 22.0790
PMGTAUIOAS san goadacasaeoct 2 2878 2.9223
Anntellidae Maecenas Pies ie 6494
Helodermatidae ........... ee Weg 0325
Nantusiidacw eerie aa ae 1.2988
Mendae qoecc vse eeertor: 5 7194 4.8705
Scincidaes cates emceicrr 26 3.7409 4.8705
Ophidia (further unidenti-

HEM IES te Meer te Cy aone 250 35.9700 et
Bipedidaehns tasers ae ees .0325
Léeposternidae ...........- oe ie .0325
Leptotyphlopidae ......... Bich ae .6404
Boidaeenyoascaren ere Bre Mee .O741
Golubuidacie enone III 15.9707 35.0076
Elapidae ..... sect tree hecreeaee Stas ait .6404
Grotalidace sseeseee eee I .1438 5.8446
Chelonia (further unidenti-

TEC rack isrs ciseicree exaiere toate 84 12.0859 vie
Kanosternidae) =e eee oe 2 2878 2.2729
Chelydridaemeen. sade I 1438 6404
Mestudinidde=.eesdae sce: 3 4316 0.7410
Ghelonudae......0.-.40...: ae ae 2.2729
Dermochelidae ............ Ban we 6494
eri ony Chtdae. .jesew oitearsl ee I .1438 1.2988

In commenting on the foregoing table the obvious fact is recalled
that our birds can hardly assume the role of predators upon turtles
except in the case of rather small young of these animals. This
limitation considered, g1 records seems fully as many as could be

*Computed from Stejneger, L., and Barbour, T., A check list of North
American amphibians and reptiles, pp. 41-125, 1917.

ee

NO. 7 PROTECTIVE ADAPTATIONS—MCcATEE 125
expected. Forty-five species of birds participated in the lizard-eating,
including some surprisingly diminutive ones such as the canyon and
Carolina wrens and the white-eyed vireo. The road-runner, crows,
jays, butcherbirds, and the Carolina wren took lizards most frequently.
The chameleon, despite its powers of color change, was identified more
often than any other species of lizard, namely, 24 times in the
stomachs of 10 species of birds. One swallow-tailed kite had eaten
seven specimens at a meal. Horned-toads and swifts, notwithstanding
their defenses, which as it happens are diametrically opposed in
character, were “among those present ”’ in the stomachs.

In contrast to the comparatively wide distribution of the lizard
determinations, those of snakes were shared by only 26 species of
birds. Crows, hawks, and owls were the most important of these
predators; and it is worth mentioning that the little Carolina wren
again unexpectedly appears in the list. The superlatively cryptic
green snake (Cyclophis aestivus) was eaten by red-shouldered and
broad-winged hawks ; the swift racers (Bascanion) by five species of
hawks and the crow; the desperately bluffing hog-nosed snakes by the
red-tailed and Swainson’s hawks; the stinking garter and water
snakes by several kinds of birds; and the redoubtable and warningly
colored king snakes by the red-shouldered hawk. A great blue heron
had swallowed a water snake (Natrix fasciatus) slightly over 25
inches long. The only venomous snake identified in the stomachs was
Crotalus confluentus from a great horned owl but field observers
credit another of our birds, the road-runner, with occasional depreda-
tions on rattlesnakes.

Total number of identifications of reptiles, 695; percentage of
identifications among those of all vertebrates, 5.2173; percentage of
species in this class among those of all nearctic vertebrates, 6.1847.

Other enemies—While some of the larger predatory fresh-water
fishes may occasionally devour a young turtle or small snake, actual
records of the occurrence have not come to hand. The only one of
our amphibians known to be a reptile eater 1s the bullfrog, which has
been observed to eat snakes and newly-hatched turtles and alligators.
Reptiles have numerous destructive enemies among their own ranks.
Snapping turtles eat snakes; several kinds of snakes eat turtle eggs
and a few the young; a few species of lizards prey upon other lizards,
and a number of snakes devour both these animals and their eggs.
Snakes are the worst enemies of snakes, such species as the racers,
king snakes, ring-necked snakes, coral snakes, water moccasin, and
copperhead being conspicuous in this respect. The king snakes are

126 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

immune to the poison of the venomous serpents and kill them when-
ever they run across them. Among mammals, skunks, raccoons, and
bears dig up and devour the eggs of turtles; skunks, foxes, and wild-
cats eat snakes and lizards ; the badger is known to feed upon tortoises
and snakes, the coyotes on horned-toads and garter snakes, the opos-
sum on horned-toads, and ground squirrels and grasshopper mice upon
lizards.

Discussion—The reptiles are not a very numerous group in our
fauna and it would appear that they have natural enemies in due
proportion. While some of the turtles are monarchs of the waters
they inhabit when adult, yet their young must run the gauntlet of
numerous enemies which cut the number down so that there are no
indications whatever of an increase in the number of these species.
So it is apparently with all the forms that when adult seem too large
to have many enemies to fear; they are small and relatively helpless
in the earlier stages of their life, and it is then that predators do
great execution. In the class of reptiles, fratricide in almost every
direction seems to be one of the most important elements of natural
control. That such control is effectively exercised, the relatively
stationary character of the reptile population sufficiently attests.

Aves (Brrps)

Protective adaptations—Much has been written about protective
coloration in the bird world, including the nests, the eggs, the sitting
bird upon the nest, and later the nestlings, the fledglings with their
special plumages, and extending to the adults of hundreds of species,
some of which (Anatidae) have a special protective dress, the eclipse
plumage, during the season when the flight feathers are moulted. The
ringed plovers of numerous species are said to have ruptive color
patterns tending to break up the outline of the birds and render them
inconspicuous. (The phylogenetic significance of this group character
apparently is ignored. )

Crests of birds in some cases are said to be used to frighten their
enemies, as are various sudden displays of contrastingly colored
feathers elsewhere. Boldly marked birds of colors held to be warning
in other classes of animals are numerous and the unusual often in-
tense and striking coloration of the lining of the mouth of certain
nestlings is held to be warning in effect. It has even been claimed that
the color of some bird eggs advertises their low digestibility and that
they are therefore avoided by all but ravenously hungry predators.

Bird enemies.—Birds, not content with preying upon animals of
every class from protozoans to mammals, also draw upon their own

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 127

kind to the extent of a fourth (26.6 per cent numerically) of all their
vertebrate food. The following table shows the distribution to families
and two more inclusive groups of the determinations that have thus

Identifications of Aves

Percentage

| far been made of birds in the stomachs of nearctic birds.
Percentage of of species
identifications in this group
among those among
Number of of all nearctic
Group identifications birds birds 1
Birds (further unidenti-
LOC) ercpasha tie ero is us bees 301 8.4669
Egg-shell .............0.. 463 13.0327 avr
Golymbidae sas sos seean ee 6 .1688 -7490
Gaviidaeime nese ecierciess < sete Live 6242
PRICIOAE + fta earn an am pias aener4 3 .0844 2.7465
SLEECOLATNGAC aa 6 specs ee eT .4994
Waridae: Gaeecmncds asiesne 3 0844 5.2433
ReynchoOpidaes a. ser. cee eet res ee 1248
Diomedeidae .........-:.. Ge vee 6242
Procellaridae .........-.. is are 3.8700
Phaethontidae ............ ars ae 3745
Stlidae ieyesec. ate ase) cena aca tendes aol Se -7490
PNTININO 1CAC cele a saws iets eee ead Se .1248
Phalacrocoracidae: . oss. ne ses -7490
elG@CANIGAG tas stc% ce araialenee oe Bae .2407
Fregatidae ............... oe wa 1248
PNTIALICAGH wreccisseese-a susteserse stees 26 7314 7.2407
Phoenicopteridae ......... si ate .1248
Plataleidae 2... 2ae de.ees te ele .1248
rdid ae wate twice cause stare Rate joer .4004
W@icomidde Viste .cms noes ele — .2407
INT CEIMAE fos csi sie ew es ook 2 10563 1.7478
Gri da eins Soyo sie Cate ee aor ee 3745
PXGAIM GAG? ec taciscsaiaesee ee ate 1248
wall Cla Ge eres creye: aiefevetensue areieiss 19 25345 1.9974
Phalaropodidae ........... a 0844 3745
Recurvirostridae .......... 2 .0563 2497
Scolopacidae ............. 53 1.4908 5.2433
Gharadnidae 35.....-6.-- 5 .1406 1.7478
IADMGIZACACS es0cse -s eee oe ce ere 49904
Haematopodidae .......... eure nets -4994
Wacanidae . caes cc. 2s css eee ae ae 1248
Odontophoridae .......... 2 6470 .8739
Tetraonidae ............:- 26 FATA. 1.7478
yieisiairil Gale eanerasere eaters ates 265 7.4542
Meleagridae .............. ne ; ane

*Computed from Check list of North American birds, prepared by a Com-
mittee of the American Ornithologists’ Union, 3rd ed. (rev.), 430 pp., 1910.
9

128 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Identifications of Aves—Continued

Percentage

Percentage of of species

identifications in this group

among those among

Number of of all nearctic
Group identifications birds birds

Gracid tet res ate prss nochanaiters bate beta 1248
Columbidaeweacc. eee II 3094 1.6229 |
Cathartidae: .c.-.0< ceoeemiee aia 3745 \
Buteomidae, so. dav eoveees ae a 2.8713
Bal conidaewy sero aaeeiae 14 3938 1.4980 |
Pandionidae’ 2.52.25 02000: oe ae .1248 if
Alticonidden®scas2 css satrns: sé eats .1248 |
Steisidaetsc.asncseos gos 15 4219. 2.3720 | |
Psittacidaencrcstr. Acts notin tea aes .1248
Guculidies 0 ee: 6 1688 8730 |
‘Arogonidae mass. ae we .1248 |
Alcedinidae 22.044 a et 3745 ‘|
Picidaenstsat kee SI 1.4346 2.9962
Caprimulsidae (2..-22..4.. I .0281 -7400
Micropodidaei-.. 2.402. . 4. 17 .4782 .4004 |
irochilidac.. 2400.64 eee ae 2.2471
Cotingidae 2. ke" ee ae sine eats .1248
Pyraniidae. si... 10 2813 3.8700
Ataudidae. 32. 1.:b ase 12 3375 2497 |
Cotvidaeatrss. ere eee 20 5626 2.8713
Sturnidae ye a eee eee 2 .0563 .1248
heteridae 22k aber III 3.1223 2.3720 |
Fringillidae J... 006.602. .2. 992 27.9040 11.6101
anpaniddes.ete yee ae Il 3004 4904
Hirundinidae ............. 76 2.1378 1.6229
Bombyeillidae -.23..5..5 -- 34 .9564 24907
Ptilogonatidae a. -..0. hs sai nets .1248
Keanidael es. coseGcs oe I 0281 .2407
Wareonidaen -. eee 88 2.4754 1.4980 |
Goerebidae 4. 25). es. ee ust .1248
Mniotiitidae’ 22 ee. a. 2 488 13.7270 6.8662
Miotacililidaes {2 -eni eee 7 .1969 8739
Cinclidach eee eee Soe can .1248
Mimidae. ta. cc. seve oes oe 45 1.2658 1.3732
Miroclodytidaes tees 17 .4782 1.7478
@erthndde ait. Sakae. 4 .II25 1248
Sittidaenneeoa ees 15 .4219 .4004
Pakidacmeemnne ea 38 1.0689 1.8726
Chamacidacen eee I .0281 1248
Sylvildacwy i eee 25 -7032 -7490
durdidacaeeces eee eee 243 6.8353 1.8726

About one-eighth (13.02 per cent) of all the records are for bird
eggs, and the number of species feeding upon eggs is so considerable

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 129

and represents so great a variety of birds (55 species) that the late
Prof. F. E. L. Beal, taking these facts in connection with his field
observations, was constrained to express the belief that scarcely a
species of bird exists that upon good opportunity, can resist the
temptation to eat another bird’s eggs. Numbers of identifications
such as 6 for the yellow-billed cuckoo, 10 for the brown towhee,
12 for the Baltimore oriole, 10 for the California towhee, and 11 for
the bank swallow prove that egg or at least egg-shell eating is not
confined to birds of the recognized predatory groups. Probably a
number of the records are due to birds swallowing bits of their own
egg-shells. On the other hand eggs may be punctured as by the house
wren, or eaten without swallowing any of the shell, occurrences not
likely to be registered in the evidence brought to light by stomach
examination.

Of the records for predation upon the various families of birds, it
may be said that the high number for Phasianidae represents domestic
poultry almost entirely, and that of the other families, the two—
sparrows and warblers—undoubtedly most numerous in individuals
are those which bear the brunt of predatory attack. The rather high
number of determinations of Turdidae reflect the abundance of the
robin which contributed nearly 45 (43.6 per cent) of the total. The
Icteridae, next in line, are birds of great abundance, which might be
expected to rank still higher among the avian contributors to the
subsistence of their predatory relatives. However, there is no evidence
that they are at all immune to attack, as the great flocks of blackbirds
wintering in our southern latitudes are constantly harried by pre-
dacious birds in variety and force.

The sparrows, most persecuted of all, because most available, repre-
sent almost the acme of protectively colored birds; the bob-whites
(16 records), ruffed grouse (11), and their allies, also cited, as
marvels of cryptic coloration are certainly eaten freely considering
their relative numbers. It is of interest that birds of prey by no
means spare each other, and it seems that a slight advantage in size
is all that is needed to induce this strained predation; indeed there
are records of intra-specific cannibalism. The pugnacious kingbird
and other members of the family of tyrant flycatchers do not escape ;
the aerially expert swifts and swallows pay their due toll; and the
green-coated vireos, best blended with foliage of any of our birds,
are freely eaten.

Birds “ warningly colored” that are represented in the dietary of
other birds as illustrated by our tabulations include the bobolink
(19 records), Baltimore oriole 4, orchard oriole 3, lark bunting 6,

130 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85.

cardinal 3, rose-breasted grosbeak 1, black-headed grosbeak 3, scarlet
tanager 11, Blackburnian warbler 5, bay-breasted warbler 12, myrtle
warbler 16, magnolia warbler 16, Canadian warbler 13, Wilson’s
warbler 10, hooded warbler 1, and the robin 106. These birds cer-
tainly have the colors and arrangement of colors said to be warn-
ing in the case of other animals, but brought home in the instances
of these familiar and practically defenseless species, for none of
which can any degree of inedibility be assumed, and in the light
of the fact that all are eaten, some freely, some less so, in relation to
their numbers, the theory of warning coloration becomes a wraith
of the imagination so tenuous that one cannot understand why it ever
received serious consideration.

Total number of identifications of birds, 3,555; percentage of
identifications among those of all vertebrates, 26.6870 ; percentage of
species in this class among those of all nearctic vertebrates, 16.0843.

Other enemies.—Fishes are not recorded as serious enemies of
birds, but it is probable that sharks and some other highly predacious
forms take some toll of birds that rest on the surface of the ocean.
The goosefish is known to have eaten seven wild ducks at a meal and
to have attacked such large birds as geese and loons. In fresh-water,
bass have been observed to capture swallows. (Fins, feathers, and
fur, p. 8, Dec. 1921.) The bullfrog is the only one of our amphibians
known to eat birds, but records of its so doing are fairly numerous
and some of the birds taken are surprisingly large (e. g. woodcock).

Among the snakes we find very serious enemies of birds, some of
the expert climbing species especially, making birds, their eggs and
young a considerable part of their diet. Most noteworthy in this
respect are the pilot snake and black snake. Other bird-eaters are the
garter, house, hog-nosed, king, and all of the Crotaline snakes.

The larger predacious mammals are very fond of birds and must
be numbered among their worst enemies. Such are the opossum,
wild cats, foxes, coyotes, raccoon, badger, and skunks. Smaller species
as the weasels and mink are no less destructive and even the highly
vegetarian squirrels never lose an opportunity to devour the eggs and
young of birds. The red or pine squirrels are universally acknowl-
edged to be among the most destructive foes of birds. The domestic
cat, large numbers of which lead a more or less feral life, possibly is
the most deadly single enemy of birds.

Recently much evidence has been gathered showing that the larvae
of certain flesh flies (family Muscidae, sens. lat.) parasitize the nest-
lings of various birds, this activity resulting in the destruction of

EEE EE LLL eS

INO. 7 PROTECTIVE ADAPTATIONS—McATEE I31

numerous broods. Birds have other external as well as internal para-
sites also, the relation of which to mortality is not well known. An
occasional bird falls a victim to mussels or other bivalves, to cray-
fishes, and to mantids and spiders.

Discussion —“ Warningly colored ” nearctic birds, according to our
tabulations, are eaten along with the others, the common ones fre-
quently, the rarer ones to a lesser extent. Our most extensive family
and the one most numerous in individuals, occupies the logical, if
unenviable niche, as the most important contributor to the subsistence
of predatory species. This family, the finches, includes many of the
most “ protectively colored” species. Fortunately there is other direct
evidence of the way in which nearctic predators react to protective
coloration. I refer to Dr. Raymond Pearl’s paper on the “ Relative
Conspicuousness of Barred and Self-colored Fowls” (Amer. Nat.,
vol. 45, pp. 107-117, Feb., 1911). Natural enemies captured in one
year 325 individuals out of a total of 3,443, a flock which contained
both barred and solid-colored fowls. By all theories of protective
coloration, the latter are the more conspicuous and should pay a
higher toll to predatory enemies. Of the total number of birds 10.05
per cent were self-colored and of all the eliminated birds 10.77 per
cent were self-colored. Thus these monochrome birds were taken
almost exactly in proportion to their numbers in the whole flock. This
is precisely the result that would be expected by those who have
learned by study of the subject that availability is the one strongest
factor in choice of food by predators. With availability as the con-
trolling factor it follows that in the long run, and on the average,
losses to predators will be very closely in proportion to the relative
abundance of the group concerned.

MamMMatiA (MAMMALS)

Protective adaptations.—Many of the mammals are conceived to be
very perfect exemplifications of protective or cryptic coloration. A
few are credited with noxious qualities, accompanied in the case of
the skunks only, in our fauna, by warning coloration. The short dense
fur of moles and shrews is said to be a deterrent to predators and
these animals are thought to be protected by a strong musky secretion
also; shrews have even been credited with poisonous bites. However
the most potent defenses of mammals in general against birds are
their large size, and their teeth and claws.

Bird enemies.—Despite the size and direct means of defense of
many species, mammals pay a heavy toll to bird predators. In our

132 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

complete tables, the group without doubt is over-represented, owing
to the fact that stomachs of the hawks and owls have been kept
examined practically up to date. However this fact probably does not
materially affect the relative numbers of identifications for the differ-
ent families, as shown in the following table for mammals alone.

Identifications of Mammalia

(Land mammals only)

Percentage of Percentage
identifications of nearctic
among those species
Number of of all in this
Group identifications mammals group 1
Further unidentified (in

many cases carrion).... 331 10.5046
Carrion (identified to spe-

GIES) is ites, Serle eee oes 18 5712 nee
Didelphtidae” fn aca. eee a 2 0635 2058
Malpidaeden wi cen ee eee 22 .6982 1.4790
Sortcidaee eee eee ee 274 8.6957 6.5076
Phyllostomidae ........... no os 2958
Vespertilionidae .......... 19 .6030 3.5490
Molossidaesnaatceie eee aM ae 1479
Wirsiddeern a ume roe Ot: 3.2538
(Grint Saunoweansecodsnds I 0317 5.6202
iRrocyonidaers: = sania. s aa: wee ae 5916
Miistelidaew esas. cre eee I 0317 8.5782
Heldaents tose sa eee eer I 10317 f
Rodentia (further unidenti-

fied)\\ Ge aatcnen inti cr 86 2.7203 See
IMiuridae) ses assent eect 1,816 57.6326 24.1077
Geomvyidaeesc cee mates 31 .9838 7.9866
Heteromyidae ............ 16 .5078 8.5782
Aapodidaewierrieen rece I4 4443 1.7748
Enithizontidae: 2.54... « ee a 2058
Aplodontidaey 22.5. ae saves I .0317 7395
Sciunidaean nena ee ec ote 173 5.4903 14.6421
Petauristidae secs 15 .4760 -7395
@astoridae iyce. scat ene oe oe 2958
©chotonidaen... sense sone Sie seus 1.7748
ILNSOGES Ao oadodasoocoone 330 10.4729 2.9580
Dasypodidacueias alos: age vide 1479
Mayassuidae .n.a7ceceeen ee are Sree .1479
Gerviddewt yt. sate cere aye aaah 3.9933
Antilocapridae ............ ae Ate .1479
Boviddersueerni-cscrcseue te ae esr

*Compiled from Miller, Gerrit S., Jr., List of North American land mammals
in the U. S. National Museum, 1911, U. S. Nat. Mus. Bull. 79, 455 pp., 1912.

NO. 7 PROTECTIVE ADAPTATIONS—MCcATEE 133

Let us now take up some of the groups of interest in the order of
their appearance in the tabulation. From the large number (274)
of records for shrews it would appear certain that the alleged special
defenses of these animals are no protection against birds. Thirteen
species of shrews were identified in the stomachs ; 27 species of birds
are known to prey upon our common short-tailed shrew and 23 upon
unidentified species of Sorex. Shrews are by no means gregarious,
nevertheless five specimens of Sorex personatus were taken at a meal
by a great gray owl. Considering their almost exclusively under-
ground life, moles were captured fully as often as would be expected ;
the number of species of birds preying upon them is 12.

Bats, again on account of their nocturnal activity, are not greatly
exposed to the attacks of birds. Six predators upon them are recorded
in our tabulation with a total of 19 identifications. While the Mus-
telidae are provided with unusually strong musky scents, they are also
rather above the size for many birds to attack. The single determina-
tion in our table, attributed to a crow, might perhaps be more correctly
added to the records of carrion. Skunks, of this family, customarily
cited as examples of animals having noxious qualities and warning
coloration to advertise them certainly are too large for all except a
very few species of our raptors to conquer. However there are a
number of published and other records of the great horned owl
preying upon skunks.

Muridae (mice and rats) are secretive, elusive animals with what
would be called highly protective coloration, but this does not prevent
their being the staple mammal food of birds. Meadow mice, perhaps
our most ubiquitous rodents, are eaten by the largest number of
species of birds, namely 44. Twenty-six species of birds are known
to feed on the house mouse and 35 upon deer mice (Peromyscus).
We have records of five species of birds preying upon our largest
member of this family, the muskrat, and eight upon the smallest
(Reithrodontomys ).

Pocket gophers, like the moles, spend most of their lives under-
ground and this fact limits the opportunities of birds for capturing
them, yet there are 31 records for 11 species of birds; nocturnal and
burrowing habits shield also the pocket mice and kangaroo-rats.
Captures in these groups probably are in proportion to their reduced
availability. Jumping mice (14 records), a more diurnal group, seem
to be proportionately represented.

Erithizontidae (porcupines) are entirely beyond the size of prey
practicable for birds, though possibly some of them are captured

134 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

when young. Mountain-beavers also are rather large and are inacces-
sible to any but owls, of which the great horned owl contributed our
only record of their being eaten.

The large number of determinations of members of the squirrel
family, cover, it must be recalled, such diverse groups as the spermo-
philes, prairiedogs, groundhogs, tree squirrels, and chipmunks. There
are only two records of the groundhog, a very large rodent, one being
captured by a goshawk and the other by a golden eagle. The number
of identifications (15) of the chiefly nocturnal flying squirrels seems
as large as could be expected. The cryptically colored rabbits are
exceedingly common and live fully exposed to predacious birds, fac-
tors which go far toward accounting for the very large number of
records of their being eaten. The remaining families in the list all
consist of animals so large that only a few of the most formidable
birds can prey upon them, and then only upon the young. There are
observations of such occurrences, but it so happens that our records
of stomach contents do not include any of them.

Total number of identifications of mammals, 3,151 ; percentage of
identifications among those of all vertebrates, 23.6542; percentage of
species in this class among those of all nearctic vertebrates, 13.5743.

Other enemies——Fishes have few opportunities to capture mam-
mals, but trout have been known to feed upon meadow mice and
lemmings, and it is probable that other highly predacious fresh-water
fishes occasionally get small mammals that venture near or in the
water. The bullfrog is the only one of our amphibians known to eat
mammals, an occasional mouse falling to its lot. The snapping turtles
also get some mice and sometimes even capture animals as large as
rabbits. Among snakes we find many habitual predators upon mice
and other small mammals. Some results of studies of the food of
snakes by the Pennsylvania Division of Zoology may be briefly cited:
Pilot snake, mice 22 per cent of the diet, squirrels 11 per cent, weasels
4 per cent; black snake, mice 26 per cent, rabbits 4 per cent, other
mammals 7 per cent; milk snake, mice 71 per cent, other mammals
II per cent; copperhead, mice 41 per cent, shrews 4 per cent, other
mammals 8 per cent. In the case of the timber rattlesnake, mice, rats,
and rabbits composed nearly the whole diet. This is known to be true
also of most of our venomous snakes,

The worst foes of mammals, however, are their own kind, and the
diversity of their predatory habits may be indicated by brief refer-
ences to their mammal prey. Opossums take limited numbers of small
mammals, while raccoons and skunks prey more extensively upon

yy

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 135

them, especially upon mice and ground squirrels. The bob cats or
lynxes are fond of mice, ground squirrels, rabbits, and other rodents,
occasionally prey upon small domestic stock, and are known to eat
skunks and porcupines. The mountain-lion specializes upon deer, but
eats a variety of wild mammals, including foxes, skunks, coons, porcu-
pines, and bob cats. House cats take mice, rats, moles, shrews, and
rabbits. Coyotes and wolves prey upon the young of deer and
domestic stock, and upon prairie dogs, spermophiles, and other small
rodents. On the bill-of-fare of our various species of foxes are shrews,
mice, ground squirrels, pocket gophers, kangaroo-rats, and rabbits.
Badgers also take all of these mammals and in addition, prairiedogs
and mountain-beaver. The black-footed ferret is a special enemy of
the prairiedog, and relishes rabbits also. Weasels are ferocious
enemies of small mammals in general, and for their size, shrews are
fiendish predators. They commonly overpower and devour other
shrews and mice of their own or even of slightly greater bulk. The
polar bear preys especially upon seals, and the killer whales also
destroy these animals, as well as wearing down and devouring the
largest of all mammals, whales.

Discussion.—Limitations due to relative sizes allowed for, we see
the same phenomenon in the case of mammals as in those of other
elements of bird food, namely that the more available (this usually
meaning abundant) groups are preyed upon most extensively, while
those which are less abundant or whose habitat is somewhat out of
the domain of birds are not so often captured. We see that the
burrowing moles and pocket gophers escape with moderate losses, but
that the abundant mice, and the both common and relatively easier
found rabbits suffer severely. It is evident also that the mammals
outside the range of prey of birds have serious enemies, chiefly other
mammals; and it is further evident that, taking all mammal enemies
into consideration, they are most numerous in the case of so abundant
and ubiquitous a group as the mice, and proportionally less numerous
for other less abundant families.

DISCUSSION

Availability is a mighty factor in the choice of food by birds.
Within the limits imposed by special habitats, bodily modifications,
and the relative sizes of predator and prey, birds are prone to feed
upon what is abundant and easily obtained. Not only is this very
natural procedure the everyday order, but it is conspicuously exempli-

136 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

fied by the characteristic flocking * of birds to the scene of insect out-
breaks or of other occurrences of unusual abundance of food.

Constant seeking of the available leads to a wide distribution of
predatory attack because of seasonally or otherwise variable abundance
or availability of many of the food organisms, further on account of
the greater or lesser restriction of predators to specific habitats in
each of which the range of food items is different, and because of the
specialization of various predators in methods of seeking food.

That the predatory attacks of birds are amazingly distributed over
the entire animal kingdom, preceding pages bear witness. If it be
asked whether birds eat bats or moles, flyingfishes or hermit crabs,
dragonflies or mole crickets, sea-urchins or bryozoans, the ariswer is
ever in the affirmative. Given an animal group comprising only a
small number of species we find that there are only a few records of
birds preying upon it. Given one of large numbers of species we
invariably find it is an important item of bird food. If the validity
of depending upon the number of species as an index of frequency
be questioned, no matter. The tendency for feeding to be distributed
over the whole range of the available food organisms and in at least
rough proportion to the known abundance of the various groups, is
beyond dispute.

This principle, predation in proportion to population, stands out
clearly in the tabulations of the animal food of nearctic birds here
presented and discussed. Compared to it the effect of the so-called
protective adaptations on character of food is negligible. If these
adaptations controlled choice of bird food to a significant extent,
discrimination would everywhere be evident ; finding indiscriminancy,
on the contrary, we must conclude that the ruling criterion in choice
of food is availability.

INDISCRIMINANCY OF PREDATORS OTHER THAN Brrps

Nearctic birds, as a group, are little influenced by the protective
adaptations of available prey. Let us see what can be said of other
classes of predators.

Odonata.—In a general article on “‘ Predacious Insects and their
Prey,” Prof. E. B. Poulton says of a tabulation of dragonfly victims :
“ Short as it is, the list is extremely interesting, and raises the expec-
tation that dragonflies will be found to prey rather largely upon

* American instances are summarized in the following paper: McAtee, W. L.,
The rdle of vertebrates in the control of insect pests, Smithsonian Rep. 1925,
PP. 415-437, 7 pls., 1926.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 137

specially defended groups of insects.” (Trans. Ent. Soc. London
1906, p. 401, 1907.)

Agnatha, Plecoptera, et al—In a report which deals with the food
of stoneflies, mayflies, caddisflies, and Diptera in trout streams, Mutt-
kowski and Smith state that “Aquatic insects in rapid streams are
opportunists as regards food and eat whatever becomes available.” *

Orthoptera—Professor Poulton in the paper referred to says of
the prey of these insects: “ The proportion of specially protected
forms was very high.” (Op. cit., p. 408.)

Rhynchota—Quoting from Poulton again, he says of bugs, “ So
far as it is possible to judge from the ... . table it appears that
Hemiptera will prove to be extremely dangerous foes to the specially
protected groups.” (Op. cit., p. 403.)

Diptera.—Writing of the food of the larvae of aquatic midges,
A. L. Leathers says:* “ The organisms found were so similar, both
in number and variety, to those available in a given locality that there
seemed to be little or no sorting in their method of feeding.”

Professor Poulton remarks on robber flies that “A study of the
table at once shows that the Asilidae are most indiscriminate in their
attacks. The stings of the Aculeates, the distasteful qualities of
Danainae and Acraeinae, and of the odoriferous Lagria, the hard
chitinous covering of Coleoptera, the aggressive powers of Odonata,
are alike insufficient protection against these active and voracious
flies.” (Trans. Ent. Soc. Lond. 1902, p. 336.)

Parasites —“ Certain species and groups of species [of insects |
.... have, as far as we know, relatively few parasites in any
ReSiON: « . . « This is sometimes considered to be due to the posses-
sion of protective devices of certain kinds, but the explanation is not
satisfactory. Neither systems of colorations, nor nettling hairs, nor
an armour of chitinised plates, nor rapidity of movement, nor the
existence of toxic principles in the blood prevent insects from being
decimated by parasites.” *

Miscellaneous insects —‘* Many groups of predacious insects also
appear especially to attack the conspicuous, easily-captured prey pro-

vided by the groups with warning colours. This has been observed in

*Muttkowski, R. A., and Smith, G. M., The food of trout stream insects in
Yellowstone National Park, Roosevelt Wild Life Ann., vol. 2, no. 2, p. 261, Oct.,
1929.

* Bull. U. S. Bur. Fisheries, vol. 38, Doc. no. 915, p. 3, 1922.

*Thompson, W. R., On natural control, Parasitology, vol. 21, no. 3, p. 279,
Sept., 1929.

138 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85 =

the case of the predacious asilid flies, Dragonflies, Hemiptera, Man-
tidae, and Locustidae.” (Poulton, E. B., Essays on evolution, p. 318,
1908. )

Arachnida.—* Spiders are for the most part not very particular as
to the insects they catch.” (Bristowe, W. S., Proc. Zool. Soc. Lond.,
1929, p. 643.)

“Tt is quite probable . . . . that certain species of spiders, together
with Mantides and other predacious insects, will be found to be among
the chief, perhaps the chief non-parasitic enemies of aposematic
insects.” (Poulton, E; B:, Trans. Ent: Soc: Lond: 1902) ps 3278)

Pisces —‘ In general, fish are opportunists as far as their food
is concerned. They eat what animal food is available, regardless of
the origin.” *

The closeness with which the brook trout is guided by availability |
in its choice of food is indicated in the following table by Dr. P. R.
Needham based on studies near Ithaca, N. Y.:

Comparison of Available Aquatic Fish Foods in Stream Bottoms and Aquatic
Foods Consumed by Trout * |

Available aquatic | Consumed aquatic

foods oods |

Order |

Number Per cent Number Per cent |

|

Mavyitlya nymphs) asec oct | e236 30.90 | 350 30.12 |
Caddisfly larvae and pupae....... 1,335 Pieyp | 528 44.07
Stonefly, nymMps gos. seen osureeeel 921 14.67 4I 3.47
Fly jarvae, and! pupae.-secececsse | 869 13.84 | 187 15.82
Beetleslarvacecm-nce usta aac | 476 TES all 33 2.79
Crayfish and ‘shrimps... «220.0% «1. | 235 20 Aas 14 1.18
Miscellameousys-nicntce se nese ere | 125 1.99 23 1.04
MOtalstase rreneesa.giinaca tomes fates 6,277 99.98 1,182 99.99

1 Quantitative studies of the fish food supply in selected areas, Suppl. 18th Ann. Rep.
New York Conserv. Dep. 1928, p. 227, 1929.

Further testimony to the effect of availability on the food of fishes
is contained in Muttkowski’s study * of “ The Fauna of Lake Men-
dota” in which he shows that insects form about 60 per cent by bulk

*Muttkowski, R. A., The ecology of trout streams in Yellowstone National
Park, Roosevelt Wild Life Ann., vol. 2, no. 2, p. 229, Oct., 1929.

*Muttkowski, R. A., Trans. Wisconsin Acad. Sci. Arts and Letters, vol. 19,
pp. 374-482, 1918.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 139

of the macrofauna of the lake and about that proportion of the total
diet of the fishes.

Some of the above remarks tending to emphasize feeding on pro-
tected forms are special pleading because their authors felt under the
necessity of proving “ protected” insects do have enemies. The
various groups of predators thus referred to, however, prey upon
other than the specially protected insects, just as birds do, and ex-
amined in that light, the comment “ indiscriminate”? would in most
cases fit their food habits. An adaptation of Poulton’s tabulation for
robber flies illustrates the point.

Percentage
of species
in this group
among the
whole number

Number Percentage of insect
Name of prey of records of records species known
@rthopteratesss- ees ese 13 5.70 2.04
ny tehOta yaa cere eee ace ee 12 5.26 8.58
Neuropteroidea .......... 7 3.07 eit
Iepidoptera: s..c.s. cess 32 14.03 15.61
Coleoptera “i.e. .ceee ess os 40 17.54 46.20
Dipteraerccmt cme eemeetes 57 25.00 11.44
Hymenoptera ............ 67 29.34 17.17

This does not look very different from tabulations for birds, and
clearly illustrates the same propensity demonstrated for that class,
namely of preying largely upon the groups most numerous in species,
and presumably therefore in individuals—in other words upon what
is most available.

That availability does largely govern choice of food is the very
thing that creates problems in wild life economics. When man invades
the domain of wild life and in the various phases of his husbandry
makes available large supplies of new foods, they are immediately
attacked and up to a certain limit the enemies steadily increase in
variety and abundance. It is needless to cite examples of this uni-
versal phenomenon from the vegetable kingdom. All may not realize,
however, that it prevails also in the animal world. The temerity of
the pioneer establishing an orchard in a clearing in the foothills where
the crop is largely harvested by wild life, is paralleled by that of the
sheep raiser who grazes his herds in mountain meadows where they
are attacked at once by wolves, coyotes, wild cats, bears, and other
predators. Man’s taking the domestic fowl wherever he goes furnishes
material for further demonstration of the supreme influence of
availability. The chicken, a native of Asia should have no * natural

140 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85
* in other parts of the world, nevertheless in America for
instance, a large number of predators, including hawks, owls, crows,
jays, skunks, weasels, and foxes eagerly welcomed the new food.

enemies ’

More THEORETICAL ASPECTS OF INDISCRIMINANCY BY PREDATORS

The experience we have when we place inviting food supplies in
abundance before the birds indicates what must happen in nature
under similar circumstances. If we imagine a world of food available
to predators we must realize that the elements composing it will be
utilized very much in proportion to their abundance. This is only
what would be expected if there is or ever was such a thing as the
oft-mentioned “ balance of nature.” To preserve a balance, natural
checks must be in proportion to population. If they were not appar-
ently so, no balance would have been observed and the term balance
of nature would never have been invented.

Distribution of predation in proportion to population also is what
we should expect if the theory of adaptive radiation, or the occupation
of every possible ecologic niche is correct. Given the world of prey
to exploit it is inevitable that predation will extend in all possible
directions. No source of food will be left untouched if by any possi-
bility it can be drawn upon. Under so searching a campaign for food
each inevitably will be utilized in proportion to its abundance.

That this principle actually is at work is well shown by a series * of
studies by Harry B. Weiss which indicate that regardless of locality
there is a more or less fixed set of ratios between types of food habits
of insects. Thus from several widely separated areas the insect popu-
lation groups into from 45 to 55 per cent of phytophagous species,
from 15 to 27 per cent of saprophagous, from 14 to Ig per cent
harpactophagous, from 10 to 12 per cent parasitic, and from 1 to
4 per cent of species of miscellaneous feeding habits.

“Insect food habits and vegetation. Ohio Journ. Sci., vol. 24, no. 2, pp. 100-
106, Mar., 1924.

Ratios between the food habits of insects. Ent. News., vol. 35, no. 10, pp.
362-364, Dec., 1924.

Notes on the ratios of insect food habits. Proc. Biol. Soc. Wash., vol. 38,
pp. 1-4, Jan., 1925.

Insect food habit ratios on Quelpart Island. Psyche, vol. 32, no. 2, pp. 92-94,
Apr., 1925.

The similarity of insect food habit types on the Atlantic and Western Arctic
Coasts of America. Amer. Nat., vol. 60, no, 1, pp. 102-104, Jan.Feb., 1926.

INO: 7 PROTECTIVE ADAPTATIONS—McATEE I4I

Weiss’ table summarizing these interesting data is substantially
quoted as follows:

| I eee Har- | | Patten
Nai ae pee pacto- | Para- feeders,
| her | Paag phag phag- | sitic. | misc.
| ber of | ous. | ous. ous. Per | species.
| species Per | Per Dep cent | Per
cent cent | cent cent
| | | |
Quelpart Island .............. S771” se. Oe oes 19 10 | I
Western Arctic Coast of N. A.| 400} 47 27 14 10 2
SSrateror IN. Dione seen. eiisiielen | 10,500! 49 | I9 16 12 4
Dtatevor Connl........s..000%5 6,781 | 52 19 16 10 3

The similarity of the figures whether for the States of New Jersey
or Connecticut, for the Pacific Coast of Arctic America, or from far
flung Quelpart Island (Corea) shows that there is at work some
principle controlling choice of food that overrides whatever effect the
so-called protective adaptations may have.

It is almost certain that the constancy of the ratios is due to a
tendency (one might well say a compulsion) toward distribution of
predatory and parasitic attack. This distribution is one that lays every
group under tribute, that takes toll from each so long as the tax is
more easily collected there than elsewhere, but when that condition
fails turns toward more easily available supplies.

Predation is thus kept proportional to population and_ practical
indiscriminancy as to factors other than availability must result.

INDISCRIMINANCY OF NATURAL CHECKS OTHER THAN PREDATORS

It will not be questioned, we believe, that from the standpoint of
protective adaptations such checks as parasites, bacterial and fungal
diseases, heat, cold, and other climatic factors, are indiscriminate in
action.

RELATIVE IMPORTANCE OF NATURAL CHECKS

Years ago Chittenden writing of “Insects and the Weather ”'
stated: “ It also appears to me what has been observed by Mr. Marlatt
in the case of scale insects . . . . is true in general, viz., that favor-
able or unfavorable climatic conditions are of greater importance in

* Bull. 22, n. s., U. S. Div. Ent., pp. 51-64, 1900.

142 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

determining the abundance or scarcity of insects as a whole than are
other natural checks such as parasitic and other enemies, or even
fungous or bacterial diseases ’’’ (p. 63).

Recent studies have only crystallized long-held views to the effect
that the grand overwhelming factors of insect control are climatic.’
Thus Uvarov in discussing “ Weather and Climate in their Relation

2

to Insects,” * says:

Apart from the seasonal rhythm in the appearance and activities of insects,
there is a more or less strongly marked periodic fluctuation of a species from
year to year. Only relatively few insect pests are equally numerous and injurious
every year, while most of them are practically negligible, except in certain
years, when mass outbreaks occur. It would be out of place to discuss here
all the causes for these periodic fluctuations, but I would like to point out that
recent researches in this direction tend to throw some doubt on the commonly
accepted idea that the chief controlling factor is the parasites, since a number
of cases have become known in which the factors normally keeping an insect
species down are almost entirely of meteorological order. This has been ad-
mitted for the cotton boll weevil in America (Hunter and Pierce, 1912), for
the corn-borer in Europe (Thompson and Parker, 1928), for the almond sawfly
in Palestine (Bodenheimer, 1928), for the cotton seed bug in Egypt (Kirk-
patrick, 1923), for plague fleas in India (Hirst, Rogers), for vine-moths in
Europe (Stellwaag, 1925), and for some other notorious pests.

Again Bodenheimer in answering * the question “ Welche Faktoren
regulieren die Individuenzahl einer Insektenart in der Natur?” states
that parasites, predators, and scarcity of food, are rarely or only
secondarily of regulatory significance, but that climatic factors are
the real controlling influences.

Accepting the great superiority of meteorological phenomena as
regulative factors we may make some inquiry as to the relative im-
portance of other controlling agencies. Diseases sometimes are
dramatically destructive, but they rarely have a steady regulatory
influence.

Among parasitic and predacious organisms it must be presumed,
except for specific limiting factors, that their effectiveness as control
agencies will be more or less in keeping with their total numbers.
Thus we can deduce from a table such as that on page 9 that most

*This statement has general validity, for insects are nine-tenths of the
terrestrial animals above the size of nematodes, and probably a large proportion
of the smaller animals, as well as part of the tenth of larger size are subject to
similar checks.

* Uvarov, B. P., Conference of [British] Empire meteorologists, 1929, Agri-
cultural Section, pp. 17-18.

* Bodenheimer, F. S., Biol. Zentralbl., vol. 48, pp. 714-739, 1028.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 143

of the groups can play only minor roles in the whole drama of pre-
dation, and that insects must occupy the center of the stage, regardless
even of the superior individual size of the chordates.

To put the case in other language we may quote from David Sharp,’
“Tnsects form by far the larger part of the land animals of the
world; they outnumber in species all the other terrestrial animals
together, while compared with the vertebrates their numbers are
simply enormous ” (p. 83).

“Insects derive their sustenance primarily from the vegetable king-
dom. So great and rapid are the powers of assimilation of the Insect,
so prodigious its capacity for multiplication, that the mammal would
not be able to compete with it were it not that the great horde of six-
legged creatures has divided itself into two great armies, one of
which destroys the other” (p. 521).

SUMMARY

The hypotheses about protective and warning colors and mimicry
are part of the Natural Selection group of theories. These coloration
phenomena and other protective adaptations are supposed to have been
developed and perpetuated by the selective value they had in shielding
their possessors from attack by predators.

Preceding sections of this discussion call attention to the evidence
that one group of predators after another is known either to prey
habitually upon “specially protected” groups, or to be so largely
guided in choice of food by availability as practically to ignore pro-
tective adaptations.

The former is admitted to be true of dragonflies, robber flies,
mantids, predacious locustids and Hemiptera, parasitic insects, and
of spiders, while the latter is stated to be characteristic of the aquatic
immature forms of mayflies, stoneflies, caddisflies, and two-winged
flies, and of fishes. Data cited throughout the main body of the
present paper show a high degree of indiscriminancy also on the part
of amphibians and reptiles.

In fact this general indiscriminancy on the part of predators is so
evident that even ardent advocates of the selection theories have been
impressed by it and one of them, G. A. K. Marshall, in a paper on
the “‘ Bionomics of South African Insects ” says:

If the view advocated by many, that birds cannot be reckoned among the
principal enemies of butterflies in the imago state, be true, then I consider that

we may practically abandon the whole theory of mimicry as at present applied
to the Acraeinae and Danainae of South Africa at all events, for from what I

* Cambridge Nat. Hist., vol. 5, 1910.
* Trans. Ent. Soc. London, 1902, p. 356.
10

7"

144 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

have observed of these insects, J am convinced that their warning coloration
cannot have reference to either Mantises, Asilidae, or lizards, which are prac-
tically the only other enemies that can be taken into account. .... That
they [birds] have been the chief, if not the only agents in the production of
mimicry whether Batesian or Miillerian I have little doubt.

In other words selectionists practically rest their case on the re-
actions of birds to protective adaptations. The principal object of
the present paper has been to show what those reactions are so far as
nearctic birds are concerned, and there is no reason to suspect that the
results are otherwise than typical for birds of the world. |

The most outstanding feature of the records of the animal food of
nearctic birds undoubtedly is the marvellous distribution of them
through the phyla, orders, and subordinate systematic groups. Within |
size limits, animals of practically every kind accessible to birds are
preyed upon, and as we consider the records for group after group a
tendency for the number of captures to be in proportion to the abun-
dance of the animals concerned is unmistakable. Availability un-
doubtedly is the chief factor involved in the choice of food, and pre-
dation therefore tends to be in proportion to population.

Considering bird predation alone this principle leads to a high de-
gree of indiscriminancy in attack upon the whole kingdom of animal
life. The combined attack of birds plus all other predators still more
closely approaches complete indiscriminancy. In other words there is
utilization of animals of practically every kind for food approximately
in proportion to their numbers. This means that predation takes place
much the same as if there were no such thing as protective adaptations.
And this is only another way of saying that the phenomena classed by
theorists as protective adaptations have little or no effectiveness. |

Natural Selection theories assume discrimination in the choice of |
prey. The principle of proportional predation so obvious from the
data contained in this paper vitiates those theories for it denotes |
indiscrimination, the very antithesis of selection. |

Finally so far as the types of adaptations discussed in this paper are |
concerned the influence of such factors as disease and climatic factors,
the last the most important of all in reducing animal populations, is
completely indiscriminate.

The total mortality of animal groups is known normally to be in
strict proportion to their numbers, 7. e., a pair of the new generation
remains, to replace a pair of the old and it is apparent elimination of
all but that pair is very largely due to agencies indiscriminate in their
action. There would seem, therefore, to be no discriminative eliminat-
ing forces of sufficient strength to bear the very great burden put
upon them by natural selection theories.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 145

BIBLIOGRAPHY

A complete bibliography of the subject matter of this paper would fill vol-
umes and nothing like it is attempted here. Articles on food habits cited
are chiefly those from which notes supplemental to our tabulations were gleaned ;
few of foreign origin are included. The citations relating to enemies of various
groups are illustrative only, not exhaustive, and some of exotic origin are
used where helpful in indicating the wide distribution of predatory attack.

The bibliography is primarily one of predation and so far as possible entries
are distributed according to the thing eaten. When the feeding habits are
varied, entries are filed according to the group the diet of which is reported
upon. Titles classifiable under either of these criteria are arranged according
to the phyla or orders to correspond with divisions of the text. Those unclassi- |
fiable are grouped at the beginning of the bibliography as Miscellaneous.

MISCELLANEOUS

Buiecvap, H.

1915. Food and conditions of nourishment among the communities of in-
vertebrate animals found on or in the sea bottom in Danish waters.
Rep. Danish Biol. Sta. to the Board Agr., vol. 22, 1914, pp. 41-78.
Notes on food and enemy relations of many groups.

Ciark, A. H.

1925. Life in the ocean. Smithsonian Rep. 1923, pp. 309-304. Chiefly a
discussion of food relations, many notes from which are used in
this paper.

Ewine, H. E.

1929. A manual of external parasites. 225 pp., 96 figs. Includes informa-
tion on the hosts attacked by mites, ticks, Mallophaga, Anoplura,
and fleas.

Force, Epitu R.

1925. Notes on reptiles and amphibians of Okmulgee County, Oklahoma.
Copeia, no. 141, p. 26, Apr. Notes on examinations of the stomachs
of 6 species.

KyerSKOG-AGERSBORG, H. P.

1920. The utilization of echinoderms and of gasteropod mollusks. Amer.
Nat., vol. 54, pp. 414-426, Sept.-Oct. Notes on these forms as food
of fishes; short bibliography.

McATEE, W. L.

1918. Bird enemies of brine shrimps and alkali flies. The Auk, vol. 35,
no. 3, p- 372, July. Eight species of birds mentioned as feeding
extensively on both. Doctor Wetmore states: “The toll taken by
birds from the brine shrimp and alkali fly larvae and pupae during
the course of a season constitutes a mass of individuals almost
beyond comprehension.”

McIntosuH, W. C.

1899. The resources of the sea as shown in the scientific experiments to
test the effects of trawling and of the closure of certain areas off
the Scottish shores. 248 pp., 32 tables, many pls. Notes on food
of fishes, and other marine animals.

146 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Pratt, Henry S.
1923. Preliminary report on the parasitic worms of Oneida Lake, New
York. Roosevelt Wild Life Bull., vol. 2, no. I, pp. 55-71, Oct.
Parasites of fishes, birds, reptiles, frogs, and mollusks.
WEED, C. M.
1884. The food relations of birds, frogs, and toads. Ent. Lab. Mich. Agr.
Coll., pp. 20-20. Catbird 3 stomachs, robin 6, bluebird 2, crow
blackbird 2, spotted frog 8, green frog 4, and toad 7.
YonGE, C. M.
1928. Feeding mechanisms in the invertebrates. Biol. Rev., vol. 3, no. 1,
pp. 21-76, Jan. Notes on numerous food and enemy relationships.
Full bibliography.

PROTOZOA

See entries under Miscellaneous, Blegvad, McIntosh, Yonge.

PORIFERA

See entries under Miscellaneous, Blegvad, McIntosh, Yonge.

COELENTERATA

See entries under Miscellaneous, Blegvad, McIntosh, Yonge.

PLATYHELMINTHES
GAMBLE, F. W.
1910. Platyhelminthes and Mesozoa, Cambridge Nat. Hist., vol. 2, pp. 1-120.
“ Turbellaria are carnivorous. .... Land Planarians feed on earth-
worms, molluscs and wood-lice; fresh-water Planarians on Oligo-
chaet worms, water-snails, and water-beetles; marine forms devour
Polychaet worms and molluscs... .. Certain Rhabdocoelida are
mess-mates of Molluscs and Echinoderms, and a few others are
truly parasitic—a mode of life adopted by all Trematodes save
Temnocephala” (p. 4).
Stites, Cu. WARDELL.
1902. Frogs, toads, and carp (Cyprinus carpio) as eradicators of fluke
disease. 18th Ann. Rep. U. S. Bur. Animal Industry, 1901, pp.
220-222, figs. 197-203. Carp apparently destroying large numbers
of Fasciola hepatica.

NEMATHELMINTHES

VAN ZwWALUWENBURG, R. H.
1928. The interrelationships of insects and roundworms. Bull. Exp. Sta.
Hawaiian Sugar Planters’ Assoc., Ent. Ser. vol. 20, 68 pp., Jan.
Some are primary parasites of orthoptera and bumble bees.

TROCHELMINTHES
Harroc, M.
1910. Rotifera, Gastrotricha, and Kinorhyncha. Cambridge Nat. Hist.,
vol. 2, pp. 197-238. They devour algae, Infusoria, and other rotifers
Cp: 212):

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 147

MOLLUSCOIDA
OssurN, RayMmonp C.
1921. Bryozoa as food for other animals. Science, n. s., vol. 53, pp. 451-453,
May 13. Two species of ducks and 10 of fishes noted as predators
upon bryozoa.

ECHINODERMATA
CLARK, Hupert LYMAN.
1920. Echinoderms in birds’ stomachs. Science, n. s., vol. 51, pp. 594-595,
June 11. Ducks and gulls feeding on holothurians and brittlestars.
Dawson, J. W.
1867. The food of the common sea urchin. Amer. Nat., vol. 1, no. 3, pp.
124-125, May. Minute seaweeds mixed with diatoms and remains
of small sponges.

ANNULATA
BENHAM, W. B.

1910. Chaetognatha. Cambridge Nat. Hist., vol. 2, pp. 186-194. “ The food
of the Chaetognatha consists of floating diatoms, Infusoria, small
larvae, and . . . . Copepods, small Amphipods, larval fishes”; they
are also cannibalistic (p. 190).

1910. Polychaeta. Cambridge Nat. Hist., vol. 2, pp. 245-344. “ The Nereidi-
formia are mostly carnivorous, and feed on small Crustacea, Mol-
lusca, sponges, and other animals; and Polynoids are even said to
eat one another.” The Terebellids and Cryptocephala feed on minute
organisms strained from water; the deep sea forms feed on Radio-
laria and Foraminifera (p. 206).

Buiarir, W. N.

1927. Notes on Hirudo medicinalis, the medicinal leech, as a British Species.
Proc. Zool. Soc. London 1927, pp. 999-1,002. The larvae live on
frog-tadpoles and fish, and the adults on horses and cattle.

Miter, JoHN A.

1929. The leeches of Ohio. Distribution of the species together with what
is known of their occurrence, food, and habitat. Ohio State Univ.
Contr. vol. 2, Stone Labr., 38 pp. Feed on turtles, fishes, frogs,
mammals, snails, worms, and insect larvae.

Moore, J. PERcy.

1923. The control of blood-sucking leeches, with an account of the leeches
of Palisades Interstate Park. Roosevelt Wild Life Bull., vol. 2,
no. I, pp. 9-53, I pl. 17 figs., Oct. Natural enemies of leeches
include domestic and wild ducks, herons, kingfishers, crows, rats,
minks, turtles, snakes, frogs, newts, fishes, crayfishes, dragonflies,
and other predacious insects and leeches (pp. 20-30, 36).

CRUSTACEA
Bonnot, PAUL.

1930. Crayfish. California Fish and Game, vol. 16, no. 3, pp. 212-216,
figs. 65-67, July. Scavengers, will eat anything organic either alive
or dead; destroy fish spawn. Are preyed upon by many fishes,
other crayfishes, salamanders, snakes, turtles, kingfishers, raccoons,
and man.

148 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Breper, C. M., Jr.

1922. Notes on the summer food of Chilomycterus schoepfi Walbaum.
Copeia, no. 104, pp. 18-19, Mar. 20. Analysis of contents of 26
stomachs; chiefly crabs.

Cott, HucuH B.

1929. The Zoological Society’s Expedition to the Zambesi, 1927: No. 2,
Observations on the natural history of the land-crab Sesarma
meinerti, from Beira, with special reference to the theory of warn-
ing colours. Proc. Zool. Soc. London, 1929, pp. 679-6092, pl. 1,
figs. 1-4. “Crabs have numerous enemies; they are preyed upon
by many small mammals, such as jackals, civets, and mongooses,
but more especially by birds.” Pelicans, secretary-birds, herons,
ibises, storks, owls, hawks, gulls, and waders (p. 689).

Empopy, GEORGE C.

1910. A new fresh-water amphipod from Virginia, with some notes on its -

biology. Proc. U. S. Nat. Mus., vol. 38, pp. 299-305, 17 figs.,
June 18. Notes on 3 species of fish eating this Eucrangonyx, p. 305.
Forses, S. A.

1880. On the food of young fishes. Bull. Ill. State Lab. Nat. Hist., vol. 1,
no. 3, pp. 66-79, Nov. Notes on stomach examinations; entomo-
straca the most important food.

1883. The first food of the common white fish. Bull. Ill. State Lab. Nat.
Hist., vol. I, no. 6, pp. 95-109, May. Entomostraca.

HANKINSON, T. L.

1914. Young whitefish in Lake Superior. Science, n. s., vol. 40, pp. 239-240,

Aug. 14. Notes on food, chiefly entomostraca.
KENDALL, W. C.

1923. Fresh-water Crustacea as food for young fishes. Rep. U. S. Comm.
Fisheries, 1922, app. I, 32 pp., 10 figs. Copepods and ostracods in
part carnivorous; Malacostraca and isopods, and amphipods, scaven-
gers; crayfishes, carnivorous. Enemies of amphipods include fishes,
birds, insects, Hydra, and the plant Utricularia; crustaceans im-
portant food for young fishes. Bibliography.

Kuiueu, A. B.

1927. The ecology, food-relations, and culture of fresh-water Entomostraca.
Trans. Royal Can. Inst., vol. 16, pt. I, pp. 15-98, May. The chief
enemies of entomostracans are fish, dragonfly nymphs, and Hydra.
They are eaten also by Corethra, young larvae of Dytiscus, tadpoles
of Rana sylvatica, and by the entomostracans Leptodora kindtu and
Cyclops fuscus. Their chief food is planktonic Chlorophyceae. A
long bibliography.

McATEE, W. L.

1913. Some bird enemies of amphipods. The Auk, vol. 30, no. I, pp. 136-
137, Jan. Amphipods preyed upon by 30 species of birds, including
6 species of shorebirds and 14 ducks.

MYRIAPODA
Sincrarr, F. G.

1910. Myriapoda. Cambridge Nat. Hist., vol. 5, pp. 20-80. Food of milli-
peds, vegetable, of centipeds, animal, including diptera, other insects,
worms, other centipeds. Centipeds eaten by South American
Indians.

NO.. 7 PROTECTIVE ADAPTATIONS—McATEE I49

MISCELLANEOUS INSECTS
Bairp, A. B.
1923. Some notes on the natural control of the larch sawfly and larch case
bearer in New Brunswick in 1922. Proc. Acadian Ent. Soc., vol. 8,
1922, pp. 158-171. Lygaeonematus erichsonii. Birds ate about 10
per cent and tachinid flies parasitized about 15 per cent of the
larvae; a pentatomid, ants, and coccinellids were minor predators;
hymenopterous parasites of the cocoons were scarce, but shrews
consume about 40 per cent of them; natural enemies account for
about 75 per cent of each brood.
Coleophora laricella. Birds sometimes consume 75 per cent but on
the average about 25 per cent of the larvae; ants and pentatomids
take a few; hymenopterous parasites of the pupal stage were of
slight importance.
Beat, F. E. L., McAtee, W. L., and Katmsacu, E. R.
1927. Common birds of Southeastern United States in relation to agricul-
ture. Farmers’ Bull. 755, U. S. Dep. Agr. (rev.), 43 pp. 22 figs.
In the introduction are statements about the bird enemies of certain
groups of insects as 66 species vs. the cotton-boll weevil, 41 for
the cottonworm, corn ear worm 12, white grubs 57, wireworms 128,
billbugs 55, armyworm 43, cutworms 88, chinch bug 24, corn leaf
beetle 22, corn root worm 26, leafhoppers 100, clover leaf weevil
25, clover-root borers 74, cucumber beetle 19, sweet-potato flea
beetle 28, grapevine flea beetle 23, bean leaf beetle 109, rice weevil 20,
potato beetle 26, periodical cicada 33, and horse flies 40.
CLARINVALL, AM.
1928. De la disparition brusque des invasions d’ insects. Bull. Soc. Centr.
Forest. Belgique, vol. 31, pp. 266-278, 316-335, and 378-302, 9 figs.
Lack of food, climatic factors, natural enemies, and disease; the
last three causes are given special attention; parasitic Hymenoptera
and Diptera receive much comment and predators of the following
groups are discussed: mammals, birds, Neuroptera, Coleoptera,
Diptera, and Hymenoptera.
Forsusy, E. H.
1900. Birds as protectors of woodlands. Massachusetts Crop Rep., July,
1900, pp. 26-39. Contains lists of birds feeding on gipsy moth,
46 species; brown-tail moth, 29; forest tent caterpillar, 25; orchard
tent caterpillar, 32; cankerworms, 51; tussock moth, 9; may
beetles, 8; and plant lice, 34.
GrIRAULT, A. A.
1907. Hosts of insect egg-parasites in North and South America. Psyche,
vol. 14, pp. 27-39, Apr. Coleoptera, 9 species, 10 egg parasites;
Diptera, I species, I egg parasite; Hemiptera, 28 species, 43 egg
parasites; Hymenoptera, 9 species, II egg parasites; Lepidoptera,
51 species, 68 egg parasites; Neuroptera, I species, 2 egg parasites ;
Orthoptera, 26 species, 47 egg parasites.
1o1t. Hosts of insect egg-parasites in North and South America, II.
Psyche, vol. 18, no. 4, pp. 146-153, Aug. Coleoptera, 6 species,
6 parasites; Hemiptera, 10 species, 16 parasites; Hymenoptera,
2 species, 4 parasites; Lepidoptera, 15 species, 22 parasites;

150

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Neuroptera, I species, I parasite; Odonata, 2 species, 5 parasites;
Orthoptera, 4 species, 4 parasites; Platyptera, 1 species, I parasite.

McATEE, W. L.

1911. Economic ornithology in recent entomological publications. The Auk,

vol. 28, no. I, pp. 138-142, Jan. Clover root curculios (Sitones
hispidulus) are recorded as being taken by 24 species of birds.
Sorghum midge (Contarina sorghicola) apparently eaten by hum-
mingbirds. New Mexico range caterpillar (Hemileuca oliviae)—
preyed upon by robins. Crane flies (Tipulidae)—86 species of
nearctic birds are known to feed upon tipulids and their eggs.
Pentatomidae are eaten freely by a great variety of birds. Yellow
bear caterpillar (Diacrisia virginica)—two species of birds, the
black-billed cuckoo and the bob-white, are recorded as enemies.
Mosquitos are preyed upon by more than 20 species of North
American birds.

1911. Economic ornithology in recent entomological publications. The

Auk, vol. 28, no. 2, pp. 282-287, Apr. Oakpruner (Elaphidion
villosum) is recorded as preyed upon by four species of birds.
Potato beetle (Leptinotarsa decemlineata)—21 species of birds
are recorded as enemies of this pest. Hop flea beetles are eaten
by killdeers and cliff swallows. Gipsy moths are recorded as being
taken by 46 species of birds and brown-tail moths by 31. Alfalfa
leaf weevil (Phytonomus murinus)—the English sparrow and the
black-headed grosbeak are recorded as feeding on this weevil.

1911. Economic ornithology in recent entomological publications. The

Auk, vol. 28, no. 4, pp. 505-500, Oct. Muiullipeds are recorded as
being eaten by 83 species of birds, certain beetle larvae, toads,
armadillos and skunks. Cutworms are eaten on sight by prac-
tically all birds that glean their food from the ground or from low
vegetation. Flea beetles are eaten by various species of birds, in
the case of Crepidodera by 26 species. Harlequin bug (Murgantia
histrionica) is kept in check locally by the English sparrow.
Cabbageworm (Pontia rapae) eaten by English sparrow. Codling
moth—mortality during the winter as high as 90 per cent caused
chiefly by birds, the beetle larvae, Tenebrioides sp., and diseases.
Larch sawfly (Nematus erichsoni)—red-eyed vireos and cuckoos
reported as feeding upon larvae of this insect.

1912. Economic ornithology in recent entomological publications. The

Auk, vol. 29, no. 3, pp. 416-417, July. Billbug (Sphenophorus
callosus) recorded as taken by nighthawks. False wireworms
(Eleodes) preyed upon by 13 species of birds. Agricultural ant
(Pogonomyrmex barbatus molefaciens)—eight species of birds
recorded as foes. Alfalfa weevil recorded as taken by 31 species
of birds, notable mention being made of the English sparrow.

1913. Economic ornithology in recent entomological publications. The

Auk, vol. 30, no. I, pp. 128-132, Jan. Boll weevils preyed upon
by 53 species of birds. A rice weevil (Lissorhoptrus simplex)—
the only natural enemies recorded are two species of birds. Plum
curculio—seven species of birds recorded as enemies. Leafhoppers

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 151

known to be preyed upon by more than 120 kinds of birds, numerous
species taking them in abundance. Nabidae, Lygaeidae, and spiders
also mentioned as enemies.

1913. Economic ornithology in recent entomological publications. The
Auk, vol. 30, no. 4, p. 602, Oct. Eight species of birds observed
feeding on the larvae of the fruit tree leafroller (Archips argyros-
pila). May beetles and their larvae (Lachnosterna) preyed upon
by 60 species of birds, the crow and crow blackbird probably being
the most important enemies.

1914. Economic ornithology in recent entomological publications. The
Auk, vol. 31, no. 3, pp. 421-422, July. Sugar-beet wireworm
(Limonius californicus )—the California shrike an important enemy.
Crambus calignosellus—the quail and the kingbird noted feeding
on this species. Crambus laqueatellus—the wood pewee observed
taking large numbers. Rose aphid (Macrosiphum rosae)—house
finch and white-crowned sparrow feeding on these aphids. Chinch
bug—17 species of birds recorded as foes.

1915. Economic ornithology in recent entomological publications. The
Auk, vol. 32, no. 2, pp. 253-254, Apr. Wireworms (Elateridae) are
recorded as being taken by 90 species of birds. Grasshoppers—
upward of 100 species of birds are known to feed upon these insects.
Alfalfahopper (Stictocephala festina)—four species of birds re-
corded as enemies. Midges (Chironomus) are recorded as preyed
upon by six species of birds.

1915. Bird enemies of forest insects. Amer. Forestry, vol. 21, no. 6, pp. 681-
691, June. Bark beetles are preyed upon by more than 45 species
of birds. Round-headed and flat-headed wood borers—the larvae
of these insects are recorded to be eaten by all kinds of wood-
peckers. Flat-headed apple tree borers are recorded as taken by
the downy woodpecker. Carpenter ants—fully 50 species of birds
are known to eat these insects. An average of nearly 30 per cent
of the food of woodpeckers is recorded as being ants. No fewer
than 46 kinds of birds are known to feed upon the gipsy moth in
one or another of its stages. Thirty-one species of birds are re-
corded as enemies of the brown-tail moth. Orchard tent cater-
pillars are preyed upon by 43 species of birds, forest tent cater-
pillars by 32 and cankerworms by more than 50. Snow-white
linden moth—the English sparrow is recorded as an important
check on this insect. Plant lice are preyed upon by most small
birds. Scale insects are known to be taken by more than 60 species
of birds. Cicada—fishes and tortoises when opportunity presents,
frogs, toads, lizards, squirrels, and a multitude of birds prey upon
these insects.

1915. Economic ornithology in recent entomological publications. The
Auk, vol. 32, no. 4, pp. 520-521, Oct. Katydids—birds recorded as
important foes, special mention being made of chipping sparrows.
Calosoma sycophanta—crows and hairy woodpeckers recorded as
enemies of this beetle. Armyworm—more than 20 species of birds
recorded as foes.

15

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

1915. birds that feed upon pecan insects. Proc. Nat. Nut Growers’ Assoc.,

pp. 40-41, Dec. 1. Pecan leaf caterpillar—birds known to feed
upon pests of this genus (Datana) are robins, starlings, and two
species of cuckoos. Fall webworm—four species of birds noted as
enemies of this pest. Pecan weevil—64 kinds of birds are known
to feed upon these beetles and congeners. White ants—of the
27 species of birds feeding on white ants, a flicker is recorded to
have taken 1,100. Oakpruners are known to be preyed upon by
four species of birds. Cyllene—five species of birds are recorded
as enemies. Bark beetles are devoured by more than 45 kinds of
birds. Plant lice and scale insects each are taken by 60 or more
species of birds.

1916. Economic ornithology in recent entomological publications. The

Auk, vol. 33, no. 2, pp. 216-217, Apr. Twelve-spotted cucumber
beetle—27 species of birds recorded as enemies. Grasshoppers—six
species of birds feeding upon them during an outbreak in New
Mexico. Pine moth—the hairy woodpecker recorded as the most
efficient natural force in restraining the Zimmerman pine moth.

1916. Economic ornithology in recent entomological publications. The

Auk, vol. 33, no. 4, pp. 448-450, Oct. Armyworm (Cirphis uni-
puncta)—crows recorded as great destroyers of this pest; cow-
birds and grackles also reported doing good work. Meadowlarks
and robins observed eating the larvae. The armyworm has many
natural enemies, among which are insects, reptiles, birds, and
mammals. Skunks and toads undoubtedly eat thousands both of
caterpillars and pupae. Clover leafhoppers are recorded taken by
nine species of birds. Corn and cotton wireworm (Horistonotus
uhleri)—birds are the only enemies of this pest recorded. Velvet-
bean caterpillar—the “ricebird” and the mockingbird eat many
of these. California green lacewing flies (Chrysopa californica)
are recorded as taken by two species of birds, the western wood
pewee and the nighthawk.

1917. Bird enemies of a few insect pests. The Auk, vol. 34, no. 2, pp. 230-

231, Apr. Grasshoppers are eaten by practically all birds, excep-
tions being the strictly vegetarian doves and pigeons. Fall army-
worm—several common wild birds recorded as important enemies.
Cabbageworm—tbe English sparrow, chipping sparrow and house
wren recorded as enemies of this pest. Velvetbean caterpillars
are preyed upon by the red-winged blackbird, the mockingbird,
and the field sparrow.

1918. Economic ornithology in recent entomological publications. The

Auk, vol. 35, no. 2, pp. 251-253. Apr. Potato aphid (Macrosiphwmn
solanifolii)—chipping sparrows, quail, and English sparrows ob-
served feeding on this pest. Sweet-potato leaf folder (Pilocrocis
tripunctata) reported taken by the boat-tailed grackle. Cabbage
looper (Autographa brassicae)—boat-tailed grackle observed feed-
ing on adults and larvae. Pecan-leaf casebearer (Acrobasis nebu-
lella) larvae taken by three species of birds. Fall webworm
(Hyphantria textor )—red-eyed vireos recorded as destroying about
40 per cent of the larvae in Nova Scotia in 1916; other bird foes

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 153

noted are yellow-billed cuckoos and Baltimore orioles. Emperor
moth (Samia cecropia)—cocoons destroyed by woodpeckers.

1918. Economic ornithology in recent entomological publications. The
Auk, vol. 35, no. 4, pp. 493-405, Oct. Round-headed apple tree borer
(Saperda candida)—entomologists record birds as enemies; the
present note names five species as feeding upon the adults. Root-
worms—37 species of birds recorded as enemies of Diabrotica
duodecimpunctata, and 23 species as enemies of Diabrotica soror.
Green plant-bugs (Nezara spp.) identified in stomachs of 31 kinds
of birds, 100 individuals being found in the stomach of a Franklin’s
gull. Whitegrubs (Lachnosterna spp.) taken by 78 species of
birds and 2 of toads.

1919. Economic ornithology in recent entomological publications. The
Auk, vol. 36, no. 2, pp. 305-307, Apr. Woodpeckers noted as prey-
ing extensively upon larch bark beetles and borers. Grape root
borer (Memythrus polistiformis)—the crested flycatcher observed
feeding on the adults. Peach-tree borers (Sanninoidea exitiosa and
S. pictipes)—two species of birds recorded as foes. Cankerworms
preyed upon by 75 species of birds. Whitegrubs—several groups
of birds named as enemies.

1920. Economic ornithology in recent entomological publications. The
Auk, vol. 37, no. 2, pp. 322-325, Apr. False wireworms (Eleodes)—
24 species of birds recorded as enemies. Lotus borer (Pyrausta
penitalis )—red-winged blackbirds noted as foes of this pest. Round-
headed apple tree borer (Saperda candida)—‘‘ Woodpeckers de-
stroy great numbers of the borers by removing them from their
burrows. .... In some cases from 50 to 75 per cent.” Ten kinds
of birds recorded as enemies. Flat-headed apple tree borer—12
species of birds recorded as preying upon the adults of Chryso-
bothris. Striped cucumber beetle (Diabrotica vittata)—17 species
of birds noted as foes. Grainbugs (Chlorochroa spp.) recorded as
taken by eight species of birds. Whitegrubs (Phyllophaga) preyed
upon by 81 species of birds, the common crow being the most
important enemy of both adults and larvae.

1921. Economic ornithology in recent entomological publications. The
Auk, vol. 38, no. 2, pp. 302-304, Apr. Spotted apple tree borer
(Saperda cretata)—“ by far the most effective natural check to the
increase of this borer seems to be the woodpeckers.” Clover stem
borer (Languria mozardi)—hymenopterous parasites, toads, and
five kinds of birds recorded as enemies. Beet leaf beetle (Mono-ia
puncticollis)—enemies recorded are ladybird beetles, a stink bug,
parasites, toads, and birds. Cabbage flea beetles (Phyllotreta spp.)
12 kinds of birds noted as foes. Grapevine flea beetle (Altica chaly-
bea)—eight species of birds recorded as feeding upon this insect.
Clover leaf weevil (Hypera punctata) preyed upon by 42 species
of birds.

1922. Local suppression of agricultural pests by birds. Smithsonian Rep.
1920, pp. 411-438, pls. 1-3. In more than 70 cases birds apparently
exterminated one or another of 32 insect pests locally.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

1923. Economic ornithology in recent entomological publications. The

Auk, vol. 40, no. I, pp. 161-162, Jan. Pale western cutworm
(Porosagrotis orthogonia)—records the western grasshopper spar-
row, horned larks, and possibly other wild birds as enemies. Green
June beetle (Cotinis nitida)—19 kinds of birds recorded as foes;
southern corn root worm (Diabrotica 12-punctata) taken by 4c
species of birds. Potato beetle (Leptinotarsa decemlineata)—fed
upon by 25 species of birds.

1923. Economic ornithology in recent entomological publications. The

Auk, vol. 40, no. 3, pp. 557-559, July. Corn ear worm (Heliothis
obsoleta)—17 species of birds feed on this pest; more than 50
larvae were found in a single stomach of the boat-tailed grackle.
Cloverleaf weevil (Hypera punctata)—records 43 species of birds
preying upon this weevil. European corn borer (Pyrausta
nubilalis)—five species of birds recorded preying upon larvae and
three species observed catching the moths. Tussock moth (Hemero-
campa leucostigma)—-12 species of birds known to feed upon this
insect in one stage or another. Lacewing flies (Chrysopidae)—17
species of birds recorded as predators, most of them taking the
adults, but five known to eat the larvae.

1924. Economic ornithology in recent entomological publications. The

Auk, vol. 41, no. I, pp. 191-193, Jan. American silkworm (Samia
cecropia)—Dr. John Tothill concludes from his observations in
Nova Scotia that nearly three-fourths of the caterpillars are eaten
by birds (orioles, robins, etc.), and about 85 per cent of the pupae
are destroyed by woodpeckers. Apple leaf skeletonizer (Hemero-
phila pariana)—chipping sparrow observed feeding on the larvae.
Mormon cricket (Anabrus simplex )—three species of birds men-
tioned as enemies, but birds said to be important factors in the
control of these insects.

1924. Economic ornithology in recent entomological publications. The

Auk, vol. 41, no. 4, pp. 629-632, Oct. False wireworms (Eleodes
spp.)—13 kinds of birds recorded as enemies. Argus tortoise beetle
(Chelymorpha cassidea)—identified in the stomachs of 14 species
of birds, most often in those of the starling and kingbird. Codling
moth (Carpocapsa pomonella)—woodpeckers recorded as important
enemies, special mention being made of the red-bellied. Oak sap-
ling borer (Goes tesselatus)—woodpeckers noted as destroying
many larvae and pupae. Larch sawfly (Lygaconematus erich-
soni )—four species of birds recorded as feeding upon the larvae,
consuming about Io per cent of them in New Brunswick. Larch
casebearer (Coleophora laricella)—4 species of birds recorded as
enemies. Spruce budworm (Tortrix fumiferana)—several species
of birds and insect parasites noted as foes.

1926. Economic ornithology in recent entomological publications. The

Auk, vol. 43, no. 3, pp. 396-308, July. Most common birds are
enemies of the Japanese beetle (Popillia japonica). Green June
beetle (Cotinis nitida)—observations show that starlings feed on
the larvae and cardinals on the adults; in addition to these two
birds, 22 other species are known to prey upon this pest. Striped
cucumber beetle (Diabrotica vittata)—17 species of birds known

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 155

to feed on this beetle. The larvae or pupae of cattle grubs (Hypo-
derma) were found in stomachs of four species of birds; the robin
also observed feeding on the larvae. Cankerworms (Alsophila
pometaria and Paleacrita vernata)—76 species of birds listed as
predators. Cabbageworm (Pieris rapae)—‘“‘ Birds which are known
to feed upon cabbage worms are the chipping sparrow, English
sparrow, and house wren.”

1926. Relation of birds to woodlots. Roosevelt Wild Life Bull., vol. 4,
no. I, 152 pp., 22 pls., Oct. Contains a section (pp. I0I-136) on
forest insect pests and their bird enemies discussed under the
following heads: Plant lice (Aphididae), scale insects (Coccidae),
cicadas (Cicadidae), walkingsticks (Phasmidae), flat-headed wood
borers (Buprestidae), leaf chafers (Scarabaeidae), leaf beetles
(Chrysomelidae), round-headed wood borers (Cerambycidae), bark
beetles (Scolytidae), caterpillars (Lepidoptera), and sawflies, borer
wasps, and ants (Hymenoptera).

1926. The role of vertebrates in the control of insect pests. Smithsonian
Rep. 1925, pp. 415-437, 7 pls. General notes on amphibians, reptiles,
and mammals as enemies of insects. Summarizes 109 cases of con-
trol and 88 of local suppression of insects by birds.

1928. Economic ornithology in recent entomological publications. The
Auk, vol. 45, no. 4, pp. 526-528, Oct. Satin moth (Stilpnotia salicis)
—five species of birds noted as enemies in Massachusetts. Western
robins and bats reported as feeding on it in British Columbia.
Lygus elisus—26 species of birds known to feed upon plant bugs
of this genus. Cotton-stainers (Dysdercus spp.)—record of nine
kinds of birds preying upon cotton stainers, and three species
feeding upon other bugs of the same family. Fall armyworm
(Laphygma frugiperda)—lists 13 species of birds as enemies, and
notes that English sparrows have several times been observed to
eradicate local infestations. Pale western cutworm (Porosagrotis
orthogonia)—horned larks observed doing effective work against
this pest.

MurtrxKowskI, R. A., AND SMITH, G. M.

1929. The food of trout stream insects in Yellowstone National Park.
Roosevelt Wild Life Ann., vol. 2, no. 2, pp. 241-263, Oct. Stone-
flies, carnivorous forms prey chiefly on larvae and pupae of
mayflies, caddisflies, midges; mayflies are chiefly scavengers;
caddisflies: the carnivorous species are inclined to be cannibalistic,
but they take also rotifers, midge larvae and pupae, and dead
insects. Bibliography.

Patcu, E. M.

1906. White grubs and June beetles. [In circulars, finance, meteorology,
and index.] Bull. 137, Maine Exp. Sta., pp. 286-287. Enemies
noted: Skunks, moles, and ground squirrels in addition to a large
number of birds prey on the grubs. Besides toads and frogs and
possibly insectivorous snakes, a large number of birds feed on the
adult beetles. Cecropia moth, p. 294: enemies noted are chickens,
turkeys, and swine. The tent caterpillar, p. 206: natural enemies
of this caterpillar include birds and parasitic insects; it is also
susceptible to attack by bacterial and fungus diseases.

156 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

TuHompson, W. R.

1929. On the relative value of parasites and predators in the biological
control of insect pests. Bull. Ent. Research, vol. 19, pt. 4, pp. 343-
350, Mar. Mentions parasitic habits for 3 families of Coleoptera,
the Strepsiptera, 2 families of Neuroptera, 2 of Lepidoptera, 8 of
Diptera, and 19 of Hymenoptera, and predatory habits in 3 families
of Orthoptera, 9 of Neuroptera, the Odonata, some Corrodentia
and Thysanoptera, in 8 families of Hemiptera, the Dermaptera, 19
families of Coleoptera, the Mecoptera, 2 families of Lepidoptera, 15
families of Diptera, and 6 of Hymenoptera. Thinks value of preda-
tors has been underestimated.

Uvarov, B. P.

1928. Insect nutrition and metabolism. A summary of the literature. Trans.
Ent. Soc. London, pp. 255-343. Largely technical on metabolism,
but some details of food are given. There is a section on carnivo-
rous insects, pp. 269-270. Bibliography.

APTERA
MACNAMARA, CHARLES.
1924. The food of Collembola. Can. Ent., vol. 56, no. 5, pp. 99-105, May.
Feed on sap, pollen, diatoms, algae, carrion, and Collembola.

ODONATA
Burnuam, Epwarp J.
1899. Preliminary catalogue of the Anisoptera in the vicinity of Manchester,
N. H. Proc. Manchester Inst. Arts and Sci., vol. 1, pp. 32-34.
Certain birds appear to feed exclusively upon these insects while
they last. The dragonflies mentioned are Macromia illinoensis and
Tetragoneuria semiaquea.
CALVERT, PHILtp P.
1893. Catalogue of the Odonata (dragonflies) of the vicinity of Phila-
delphia, with an introduction to the study of this group of insects.
Trans. Amer. Ent. Soc., vol. 20, no. 3, pp. 205-206, July-Sept.
Notes on numerous bird enemies of dragonflies.
LAMBOoRN, Rost. H.
1890. Dragonflies vs. mosquitoes, 202 pp., 9 pls. Nymphs feed on mosquito
larvae, upon each other, upon water beetles, bugs, and small fishes.
Not worth encouraging as mosquito enemies; fish and waterfowl
also are foes of mosquitoes.

IEWCAS SVs
1908. Foe of dragonfly nymphs. Entomologist, vol. 41, p. 16. Notonecta
glauca.

Lyon, Mary B.

1915. The ecology of the dragonfly nymphs of Cascadilla Creek (Odon.).
Ent. News, vol. 26, no. 1, pp. 1-15, Jan. Notes on stomach contents
of 36 specimens, midge larvae the most prominent item of food,
but mayflies, Corixids, dytiscids, amphipods, cladocera, ostracods,
hydrachnids, and snails were eaten.

Moorr, J. PEercy.

1900. Kingbirds eating dragonflies. Ent. News, vol. 11, p. 340. Epiaeschna

heros; habitually captures them.

|
{
|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 157

NEEDHAM, JAMES G.

1808. Birds vs. dragonflies. Osprey, vol. 2, nos. 6-7, pp. 85-86, Feb.-Mar.
Review of notes by Rene Martin on European hobby and swift
as enemies; Merops persicus lines its nest with their wings; when
teneral, chipmunks, frogs, toads, snakes, ants, and birds eat them.
Various birds eat nymphs.

1918. [Food of Odonata.] In Fresh-water biology, by Ward and Whipple,
p. 890. Diptera and other dragonflies.

NEEDHAM, JAS. G., AND HEywoop, Hortense B.

1929. A handbook of the dragonflies of North America, 378 pp., illus. Food,
flies, mosquitoes, honey bees; enemies, birds, frogs, fishes, water
snakes, spiders, other dragonflies; large numbers sometimes de-
stroyed by storms.

SuaArp, Davin.

1910. [Enemies of Odonata.] Cambridge Nat. Hist., vol. 5, pp. 424-425.
Hawks, bee-eaters, other birds, fishes, snakes, newts, aquatic Cole-
optera, Hemiptera, and other Odonata.

WatkKer, E. M.

1924. The Odonata of the Thunder Bay District, Ontario. Can. Ent., vol.
56, no. 7, pp. 170-176, July; no. 8, pp. 182-189, Aug. Dragonflies
found in stomachs of sucker, whitefish, sturgeon, and golden-eye
ducks ; a dragonfly nymph observed eating an adult of same species.

Witson, CHARLES BRANCH.

1920. Dragonflies and damselflies in relation to pondfish culture, with a
list of those found near Fairport, Iowa. Bull. 36, U. S. Bur.
Fisheries, pp. 182-264, pls. 67-60, figs. 1-63, Aug. Notes on con-
tents of alimentary canals of 250 nymphal and many adult Odonata;
citation of previous studies; full bibliography. Odonate nymphs,
diving beetles, water-scorpions, other aquatic Hemiptera, Hydra,
nematodes, fungi, birds, fishes, reptiles and amphibians prey on
nymphs; Diptera and Hymenoptera parasitize the eggs; and birds,
other dragonflies, ants, spiders, robber flies, frogs, and fishes prey
upon the adults, which also have both external and internal
parasites.

AGNATHA

See various entries under Pisces; also Muttkowski and Smith under Miscel-
laneous Insects.
NEEDHAM, JAMES G.

1920. Burrowing mayflies of our larger lakes and streams. Bull. U. S.
Bur. Fisheries, vol. 36 (1917-1918), pp. 269-292, pls. 70-92. May-
flies of prime importance as food of fishes; quotations from Forbes,
Waener, and Pearse, as to their value (pp. 270-271).

PLECOPTERA

See various entries under Pisces; also Muttkowski and Smith under Miscel-
laneous Insects.

158 SMITHSONIAN MISCELLANEOUS -COLLECTIONS VOL. 85

ISOPTERA
[HacEN, H.]

1881. [Letter on birds vs. termites.] Proc. Boston Soc. Nat. Hist., vol. 20,
1878-1880, p. 118. Record of 15 species of birds following an
emigration of white ants, robins, bluebirds, and sparrows being
mentioned.

[LonestaFF, G. B.]

1918. A flight of winged termites at Barrackpore. Trans. Ent. Soc. London,
1918, pp. Ixiv-lxvi. Lizards, bullfrogs, rats, cats, dogs, jackals,
mongoose, crows, Indian mynah, bats and cockroaches observed
eating white ants.

Snyper, T. E.

1920. [Nearctic Termites.] Notes on biology and geographic distribution.
U. S. Nat. Mus. Bull. 108, pp. 87-211. Termite checks include
parasitic fungi, protozoans, nematodes, mites, and predacious ants,
robber flies, beetle larvae, crickets, spiders, centipeds, lizards, and
domestic and wild birds (pp. 116-118).

1924. New termites and hitherto unknown castes from the Canal Zone,
Panama. Journ. Agr. Research, vol. 29, no. 4, p. 182, Aug. I5.
Ants and anteaters as foes.

DERMAPTERA
BRINDLEY, H. H.

1920. Notes on certain parasites, food, and capture by birds of the common
earwig (Forficula auricularia). Proc. Cambridge Phil. Soc., vol.
19, (1916-1919) pp. 167-177. Fourteen species of British birds
known to eat earwigs; also domestic fowls.

Morcan, W. P.

1924. Notes on the function of the forceps in earwigs. Proc. Indiana
Acad. Sci., vol. 33, (1923), pp. 303-306, 7 figs. Earwigs are preda-
tory and cannibalistic, use forceps in capturing and holding prey.

Suarp, Davin.

1910. Forficulidae—earwigs. Cambridge Nat. Hist., vol. 5, pp. 202-216.

Eat larvae, snails, flowers, vegetables.

CHELEUTOPTERA
Bapenocy, L. N.
1899. [Enemies of Phasmidae.] True tales of the insects, p. 48. Birds,
lizards, mantids, bugs; eggs parasitized.
Suarp, DAvIp.
1910. Phasmidae—stick and leaf insects. Cambridge Nat. Hist., vol. 5,
pp. 260-278. Vegetarian, but sometimes cannibalistic. Enemies in-
clude birds, Hemiptera, ichneumon flies.

SALTATORIA
AUGHEY, SAMUEL.

1878. Notes on the nature of the food of the birds of Nebraska. 1st Ann.
Rep. U. S. Ent. Comm. (1877), Appendix II, pp. [13-62.] Records
migratory locusts from the stomachs of 172 species of birds and
field observations on 33 other species eating them.

|
,
\*
|
|
|
|

——

ne

——— ee eG -

INOS 7 PROTECTIVE ADAPTATIONS—McATEE 159

BapenocH, L. N.

1899. [Enemies of locusts.] True tales of the insects, pp. 127-128. Especially
“locust birds”; but bears, skunks, squirrels, mice, frogs, and lizards
are mentioned.

BRUNER, LAWRENCE.

1902. Grasshopper notes for 1901. Bull. 38, Div. Ent., pp. 39-49. Chickens,
turkeys, blackbirds, sage grouse and sharp-tail grouse mentioned
as natural enemies.

1905. Grasshopper conditions in Nebraska, Northeastern Colorado, Wyom-
ing, Montana, and Western Kansas during the summer of 1904.
Bull. 54, U. S. Bur. Ent., pp. 60-64. ‘“ Magnificent examples” of
the usefulness of gulls as grasshopper destroyers; turkeys used to
combat the insects.

Bryant, H. C.

1912. Birds in relation to a grasshopper outbreak in California. Univ.
California Publ. in Zool., vol. 11, no. 1, pp. 1-20, Nov. Los Banos;
15 species of land birds were found to eat the insects. Tame ducks
also important. The native birds were calculated to be destroying
daily 120,445 grasshoppers per square mile.

Burritu, A. C.

1920. Meadowlarks control cricket pest. California Fish and Game, vol. 6,
no. I, p. 38, Jan. Meadowlarks recorded as important enemies of
the coulee cricket.

CrippLe, NorMAN.

1920. Birds in relation to insect control. Can. Field-Nat., vol. 34, no. 8,
pp. 152-153, Nov. Crows, gulls, black terns, blackbirds, and grouse
recorded as destroying large numbers of grasshoppers.

1922. [Enemies of grasshoppers.] Can. Field-Nat., vol. 36, no. 4, pp. 66-68,
Apr. Diptera, Hymenoptera, Coleoptera, birds, mammals (skunks,
badgers, weasels, pocket mice, shrews, gophers), snakes, toads, and
frogs.

GILLETTE, C. P.

1905. The western cricket. Colorado Agr. Exp. Sta., Bull. 101, 16 pp.,
Apr. Anabrus simplex. Bears and coyotes feed upon this pest but
birds destroy them in greatest numbers; hawks, sage grouse and
blackbirds noted (p. 7).

Grasse, P.

1924. Les ennemis des Acridiens ravageurs francais. Rev. Zool. Agr. Appl.,
Bull. Soc. Zool. Agr., vol. 23, no. I, pp. 1-14, pl. 1, figs. 1-4, Jan.
Mammals, birds, reptiles, spiders, mites, wasps, beetles, flies, nema-
todes and Protozoa.

LuGGeErR, OTTOo.

1889. Notes on the Rocky Mountain locust in Otter Tail County, Minne-
sota, in 1888. 5th Bienn. Rep. Dep. Agr. Minn., Suppl. 1, pp.
305-343, 22 figs. Nematodes, mites, tachina flies, bee flies, blister
beetles, ground beetles, soldier beetles, robber flies, digger wasps,
dragonflies, birds, skunks, shrews, toads, snakes, and_ turtles
mentioned as enemies.

rT

160 SMITHSONIAN MISCELLANEOUS COLLECTIONS — VOL. 85

McAtTeEg, W. L.

1913. Economic ornithology in California. The Auk, vol. 30, no. I, pp.
132-136, Jan. H. C. Bryant records 22 species of water and shore
birds and 40 species of land birds as enemies of grasshoppers.

1917. Economic ornithology in recent entomological publications. The
Auk, vol. 34, no. 4, pp. 497-498, Oct. Grasshoppers are found on the
bill-of-fare of practically all wild birds; freely eaten also by
chickens and turkeys.

Merritt, D. E.

1916. [Enemies of grasshoppers.] Bull. 102, New Mexico Agr. Exp. Sta,
pp. 15-16, Apr. Birds; fields near breeding grounds of the black-
birds are free from grasshopper damage; poultry; skunks; mites;
parasitic flies; ground beetles; blister beetles; bee flies.

MorseE, ALBERT P.

1920. [Enemies of Orthoptera.] Proc. Boston Soc. Nat. Hist., vol. 35,
p. 271. Frogs, toads, salamanders, snakes, lizards, birds, mice,
moles, shrews, skunk, and fox.

SANDERSON, E. DwIGHT.

1906. The differential locust. Bull. 57, U. S. Bur. Ent., pp. 19-26, figs.
g-11. Melanoplus differentialis. Blackbirds and bobolinks sup-
pressing an infestation; a conopid fly parasite also mentioned.

1906. Report on miscellaneous cotton insects in Texas. Bull. 57, U. S.
Bur. Ent., p. 22. Blackbirds and bobolinks checking an outbreak
of Melanoplus differentialts.

SHARP, Davin.

1910. [Enemies of Orthoptera.] Cambridge Nat. Hist., vol. 5, p. 201.
Cantharidae, Bombyliidae, and mites destroy eggs; birds and mam-
mals eat adults.

SmitH, Harrison E.

1915. The grasshopper outbreak in New Mexico during the summer of
1913. Bull. 293, U. S. Dep. Agr., 12 pp. 2 figs. Six species va
birds, several species of lizards, prairiedogs, a sarcophagid parasite,
and a wasp observed doing notable execution.

TREHERNE, R. C. AND BUCKELL, E. R.

1924. The grasshoppers of British Columbia. Bull. 39, Dominion of Canada
Dep. Agr., pp. 29-35, Oct. Enemies include: nematodes, Diptera,
Hymenoptera, Coleoptera, fungi and bacteria.

U. S. ENToMoLocicaAL CoMMISSION.

1878. First annual report .... for the year 1877 relating to the Rocky
Mountain locust, etc., pp. 477-++-[204], 111 figs., 5 pls. Invertebrate
enemies (pp. 284-334) include larvae of anthomyiid and sarcophagid
flies, ground beetles, blister beetles, click beetles, soldier beetles,
robber flies, and mites all attacking the eggs; and the following
preying upon the locusts after birth: mites, ground beetles, tiger
beetles, robber flies, wasps, tachinid and sarcophagid flies, ichneu-
monids and nematodes. The vertebrate enemies (pp. 334-350)
include birds, hogs, skunks, prairie squirrels, mice, and toads.
Appendix II [pp. 13-62], is devoted to an account of the food of
birds especially in relation to the locust.

NO. PROTECTIVE ADAPTATIONS—McATEE 161

N

PALEOPTERA

SHARP, DAVID.

1910. Blattidae—cockroaches. Cambridge Nat. Hist., vol. 5, pp. 220-241.
Food chiefly dead animal matter, but a great variety of refuse
also taken. Enemies include birds, rats, scorpions, spiders, and
wasps (Ampulicides).

DICTYOPTERA

SHaArp, DAvID.
1910. Mantidae—soothsayers or praying insects. Cambridge Nat. Hist.,
vol. 5, pp. 242-259. Voracious, eating insects of all kinds including
their own, and even small birds.

CORRODENTIA

Food animal and vegetable refuse, and fungi.

MALLOPHAGA

Externally parasitic on birds and mammals.

SIPHONAPTERA

External parasites on birds and mammals.

RHYNCHOTA
CiarK, L. B.

1928. Seasonal distribution and life history ot Notonecta undulata in the
Winnipeg Region, Canada. Ecology, vol. 9, no. 4, pp. 383-403, pl.
20, I fig., Oct. Summary of literature as to food and enemies, pp.
395-399. Food: eggs of giant water bug, water-boatman, eggs and
nymphs of dragonflies, ostracods, copepods, newly hatched fishes.
Enemies: giant water bug, water-scorpion, waterstrider, dragonfly
nymphs, fishes, and birds.

Curran, C. Howarp.

1920. Observations on the more common aphidophagous syrphid flies
(Dipt.). Can. Ent., vol. 52, no. 3, pp. 53-55, Mar. Larvae of five
species consumed on the average from 15 to 47 plant lice daily.

Distant, W. L.

1892. A monograph of the Oriental Cicadidae, pp. vii-vill. “ The Cicadidae
appear to be one of the most non-protected families of insects and
are the victims of most predacious creatures.” Mentions birds,
mantids, spiders, dragonflies, wasps, hymenopterous parasites and
fungi.

BEUKE, C,. 15,

1929. The known predacious and parasitic enemies of the pea aphid in
North America. Research Bull. 93, Wisconsin Agr. Exp. Sta.,
47 pp., 3 pls., 32 figs. Mites, spiders, crickets, lacewing flies,
Hemiptera, Coleoptera, Diptera, Hymenoptera, and birds. Lace-
wings, ladybirds, and Syrphidae appear to be most important.

162 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

FuLitaway, Davin T.

1920. Natural control of scale insects in Hawaii. Proc. Hawaiian Ent.
Soc. (1919), vol. 4, no. 2, pp. 237-246, June. Forty-four species
of scale insects, 87 parasites, 20 predators; varying up to 3 preda-
tors and 7 parasites to a species.

GarMAN, H.

1898. The chinch bug. Bull. 74, Kentucky Agr. Exp. Sta., pp. 45-70,
figs. 1-10, May. Coccinellids, toads, quail, and meadowlarks prey
upon it (p. 51); great fluctuations in abundance caused by disease.

GIBSON, EpmunpD H.

1916. The clover leafhopper and its control in the central States. Farmers’
Bull. 737, U. S. Dep. Agr., 8 pp., 5 figs., June. Agallia sanguino-
lenta. More than 100 species of birds, chickens, turkeys, and guinea
fowl prey upon leafhoppers (p. 5).

Hanrt, H.

1916. Beitrage zur Kenntniss der Cicadinenfeinde. Zeitschr. wiss. Insekten-
biol., vol. 12, pp. 200-204, 217-223, 274-279, figs. Stresiptera,
Dryinidae, Serphoidea, Pipunculidae, Neuroptera, Nematoda, mites,
fungi. Bibliography.

Huncerrorp, H. B.

1919. The biology and ecology of aquatic and semi-aquatic Hemiptera.
Kansas Univ. Bull., vol. 11, 265 pp., 30 pls. Gelastocoridae—grass-
hoppers, lace bugs, beetle larvae, capsids. Ochteridae—tabanid
larvae. Saldidae—drowned flies, etc. Hydrometra—Ostracoda,
culicid larvae and pupae. Mesovelia—springtail, crambus, chalcid,
Hydrometra, Ostracoda. Gerris remigis—midges, notonectid
nymphs; jassids, etc., falling in water, snails. Rheumatobates—
Ostracoda and fallen insects. Microvelia—Ostracoda and fallen
insects, waterfleas. Belostoma—fish, snails. Lethocerus—tfrog, fish.
Nepa—mayfly nymphs, Gyrinidae, Daphnia, Cyclops, fish eggs, fish,
tadpoles. anatra—ostracod, fish, mayfly nymphs. PlJea—Ostra-
coda and other small Crustacea. Notonecta—cannibalistic, Ostra-
coda and other small Crustacea, corixids. Buenoa—Entomostraca,
corixids. Corixid nymphs cannibalistic.

JENSEN-Haarup, A. C.

1924. Wasps preying on cicadas, Ent. Meddel, vol. 14, pp. 323-324. Also

birds, spiders, mantids, and dragonflies noted as their enemies.
JoHNSoN, Rosweti H.

1907. Economic notes on aphids and coccinellids. Ent. News, vol. 18, no.
5, pp. 171-174, May. Coccinellids, syrphids, spiders, and fungi as
aphid destroyers.

KirKaLpy, G. W.

1907. [Enemies of Aleyrodidae.] Bull. 2, Board Agr. and For. Terr.
Hawaii, pp. 80-84. Three species of flies, 4 of beetles, 1 of hemip-
tera, 22 of hymenoptera, I neuropteron, I thysanopteron, I mite, and
2 fungi.

Loranpo, N. T.

1929. A biological method for destroying bedbugs. Sci. Monthly, pp. 265-

268, Sept. Spiders, reduviid bugs, cockroaches, and ants as enemies.

NO: 7 PROTECTIVE ADAPTATIONS—McATEE 163

LuccER, OTTo.

1895. [Enemies of the chinch bug.] Bull. 37, Minnesota Agr. Exp. Sta.,

pp. 178-179. Birds, reptiles, frogs, toads; some specified.
MacAnprews, A. H.

1923. Some notes on the natural control of the pine bark aphid (Chermes
pinicorticis Fitch) in New Brunswick, 1922. Proc. Acadian Ent.
Soc., vol. 8, 1922, pp. 52-56. A coccinellid exerted from 75 to 90 per
cent of the natural control, and a syrphid fly and ant-lion the
remainder.

Martartt, C. L.

1907. The periodical cicada. Bull. 71, U. S. Bur. Ent., 181 pp., 6 pls.,
68 figs. Natural enemies include dipterous, hymenopterous, and
mite egg parasites, tachinid parasites of the adult, wasps, birds,
squirrels, fishes; in some cases birds ate the insects as fast as they
emerged.

McAtTEE, W. L.

1907. Birds that eat scale insects. U. S. Dep. Agr. Yearbook, 1906, pp.
189-108. Fifty-seven species of birds are recorded as feeding upon
scale insects.

1913. Relation of birds to [an outbreak of] grain aphides. U. S. Dep.
Agr., Yearbook 1912, pp. 397-404. Spring migrant birds on about
100 acres of grainfields in North Carolina destroyed about 1,000,000
grain aphids daily.

1918. Bird enemies of tree hoppers (Membracidae). The Auk, vol. 35,
no. 3, pp. 373-374, July. Treehoppers identified in the stomachs of
more than 120 species of birds, as many as 26 individuals being
found in a single stomach.

McGrecor, E. A.

1927. Lygus clisus: a pest of the cotton regions in Arizona and California.
Techn. Bull. 4, U. S. Dep. Agr., 14 pp., 7 figs., July. Bugs of genus
Lygus have been found in stomachs of 26 species of birds; Re-
duviidae and spiders also noted as enemies (p. 8).

MontTIzAMBERT, ERIC.

1908. Lampyrids and aphids. Can. Ent., vol. 40, no. I, p. 36, Jan. Tele-
phorus carolinus extirpating colonies of Siphonophora rudbeckiae
(a red aphis).

Moznette, Geo. F.

1915. Notes on the brown lace-wing (Hemerobius pacificus Bks.). Journ.
Econ. Ent., vol. 8, no. 3, pp. 350-354, pl. 15, June. Number of
aphids devoured daily by each of five jarvae varied from 24 to 27.
Captive.

Myers, J. G.

1927. The natural enemies of Dysdercus. Ann. Ent. Soc. Amer., vol. 20,
no. 3, pp. 290-204, Sept. In an article entitled ‘ Ethnological Obser-
vations on some Pyrrhocoridae of Cuba,’ the author reports on
observations, published records, and experiments with natural
enemies of Dysdercus. The actually observed enemies are spiders,
pseudoscorpions, thrips (the eggs), tachinid flies, reduviid and
other bugs, lizards, and birds.

164 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

REINHARD, EDWARD G.

1925. The wasp Hoplisus costalis, a hunter of tree-hoppers. Journ. Wash-
ington Acad. Sci., vol. 15, no. 5, pp. 107-110, Mar. 4. An exclusive
enemy of Membracidae; 12 species identified from nests.

SHERMAN, FRANKLIN, JR.

The harlequin cabbage bug. Bull. North Carolina Dep. Agr., vol. 32, no. 7,
pp. 17-24, July. Murgantia histrionca. The English sparrow is
quite an efficient aid in keeping this pest in check (p. 21).

SmitH, Harry S.

1917. Insect parasites and predators as adjuncts in the control of mealybugs.
Monthly Bull. California Comm. Hort., vol. 61, nos. 3-4, pp. 108-
114, Mar.-Apr. One chrysopid, seven coccinellids, one agromyzid,
and one syrphid as predators, and six Hymenoptera as parasites
upon species of Pseudococcus.

Surrace, H. A.

1907. [Enemies of plant lice.] Zool. Bull. Pennsylvania Dep. Agr., vol. 5,
no. 3, pp. 81-82, July. Warblers, wrens, titmice, kinglets, chickadees.

1907. [Enemies of the periodical cicada.] Zool. Bull. Pennsylvania Dep.
Agr., vol. 5, no. 3, p. 74, July. Skunks, squirrels, moles, chipmunks,
pigs, poultry, most birds, snakes (four species mentioned) and
turtles listed.

1907. Psyllidae. The jumping plant lice. Zool. Bull. Pennsylvania Dep.
Agr., vol. 5, no. 3, pp. 78-79, July. White-breasted nuthatches
practically freed an orchard from pear psyllids (Psylla pyricola).

THompson, W. L.

1928. The seasonal and ecological distribution of the common aphid pre-
dators of central Florida. Florida Ent., vol. 11, no. 4, pp. 49-52,
Feb. Cycloneda feeding on seven species of aphids; Hippodamia
on 5, the larvae averaging 56 and the adults 87 bean aphids per
day; Scymnus on 5; a few other species briefly mentioned, of
which Coccinella, Rodolia, and Chilocorus are said to be primarily
scale devourers.

Wesster, F. M.

1897. [Enemies of the periodical cicada.] Bull. 87, Ohio Agr. Exp. Sta.,
pp. 61-63, Noy. Birds, especially the English sparrow, parasitic
flies, dragonflies, soldier-bugs, ground beetles, hogs, and poultry.

1907. [Enemies of the chinch bug.] Bull. 60, U. S. Bur. Ent., pp. 58-59.
Birds, frogs, nematodes, ants, ladybeetles, predacious Hemiptera,
ground beetles, lacewing flies, spiders, and parasitic fungi.

1909. The Chinchbug (Blissus leucopterus Say.). Circ. 113, U. S. Bur.
Ent,, 27 pp., 8 figs., Nov. Natural enemies, bobwhite (100-400
eaten at a meal), 15 other birds, frogs, ants, ladybirds, anthocorids,
carabids, chrysopids, spiders, and diseases.

WILDERMUTH, V. L.

1915. Three-cornered alfalfa hopper. Journ. Agr. Research, vol. 3, no. 4,
PP. 343-362, pl. 43, fig. 1, Jan. Stictocephala festina. Spiders, ants,
mites, egg parasites, birds, and toads recorded as enemies (pp.
359-360).

NO. 7) PROTECTIVE ADAPTATIONS—McATEE 165

WIittiaMs, C. B.

1921. Report on the froghopper-blight of sugar-cane in Trinidad. Mem. 1,
Dep. Agr. Trinidad and Tobago, Jan. Tomaspis saccharina Dist-
ant. Enemies listed are: 2 species of hymenopterous parasites
and 2 species of thrips predatory upon the eggs; 3 kinds of birds,
1 syrphid larva, and 1 nematode upon the nymphs; and 17 species
of birds, 2 of grasshoppers, 7 of ants, 1 of Lampyridae and 5 of
Reduviidae predacious on the adults, besides sundry lizards, toads,
frogs, spiders, mites, and two fungoid diseases.

NEUROPTEROIDEA

Insects chiefly predacious in the larval state and often also as imagos. For note
on Trichoptera, see Muttkowski and Smith under Miscellaneous Insects.

McGrecor, E. A.

1914. Some notes on parasitism oi chrysopids in South Carolina. Can.
Ent, vol. 46, pp. 306-308, fig 1. Forty-eight out of 99 cocoons were
destroyed by hymenopterous parasites.

SMITH, Rocer C.

1922. The biology of the Chrysopidae. Mem. 58, Cornell Univ. Agr. Exp.
Sta., pp. 1287-1372, pls. 75-88, figs. 154-163, June. Parasitized in
all stages; ladybirds eat the eggs. Certain birds feed on adults.
Robber flies noted catching adults and some Hemiptera prey on
the larvae.

LEPIDOPTERA
AINSLIE, C. N.

1910. The New Mexico range caterpillar. Bull. 85, U. S. Dep. Agr., pt. 5,
pp. 59-96, figs. 1-53, June. HWemileuca oliviae—insect parasites,
mites, robber flies, and robins recorded as enemies (pp. 88-93).

ALLEN, J. A.

1894. On the mammals of Aransas County, Texas, with descriptions of
new forms of Lepus and Oryzomys. Bull. Amer. Mus. Nat. Hist.,
vol. 6, pp. 165-198. Onychomys longipes. Quotes a note from
H. P. A[ttwater] to the effect that he found several hundred
wings of Danais archippus, the bodies of which had been eaten
by the Onychomys. Allen adds “ This observation is of special
interest from the fact that this butterfly is supposed to be ‘pro-
tected’ by a nauseous odor or taste that renders it unpalatable
to animals” (p. 181).

BabEeNocH, L. N.

1899. [Enemies of the lictor moth.] True tales of the insects, p. 215. Notes
on hymenopterous and dipterous parasites. The caterpillars of a
species may be collected persistently for years for breeding and,
because of parasites, none of them reach the perfect stage.

Batt, E. D.

1904. The codling moth. Bull. 87, Utah Exp. Sta., pp. 119-120. Birds,
particularly the downy woodpecker and the chickadee, bats, ants,
spiders, and predacious insects, recorded as enemies of this pest.

166 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

BARBER, G. W.

1925. The efficiency of birds in destroying overwintering larvae of the
European corn borer in New England. Psyche, vol. 32, no. 1,
pp. 30-46. Birds credited with destroying an average of 61 per cent
of the larvae in test cases.

BEEBE, WM.

1924. Notes on Galapagos Lepidoptera. Zoologica, vol. 5, no. 3, pp. 51-50,
pl. A., Jan. 11. Birds feeding on butterflies (Callidryas eubule
and Agraulis vanillac) and moths. “ The relation between birds
and butterflies is quite a negligible factor in any lepidopterous theory
of evolution of pattern, color, form, or activity” (p. 57).

BELE Eee

1924. Notes on Asilus sericeus Say (Diptera, Asilidae). Journ. New York
Ent. Soc., vol. 32, no. 4, p. 219, Dec. Capturing Hesperiidae, in-
cluding Epargyreus tityrus.

BERGER, E. W.

1920. The semitropical army worm. Quart. Bull. Florida State Plant
Board, vol. 4, no. 2, pp. 17-34, figs. 4, Jan. Xylomyges eridania.
Birds, the spined soldier bug, a wasp, tiger beetles, ground beetles,
and parasitic insects recorded as enemies (pp. 26-28).

Birp, HENRY.

1909. [Enemies of Papaipema maritima.] Can. Ent., vol. 41, no. 2, pp. 67-8,
Feb. Field mice and skunks. “As very few pupae escape in any
locality these animals go over, they become an important factor
in the economy of the species.”

BrEAKEY, E. P.

1929. Notes on the natural enemies of the iris borer, Macronoctua onusia
Grote (Lepidoptera). Ann. Ent. Soc. Amer., vol. 22, no. 3, pp.
459-464, Sept. Six species of Diptera, two of Hymenoptera, one
beetle, one bird, and rodents.

Britton, W. E.

1906. The gypsy moth and the brown-tail moth. Bull. 153, Connecticut
Exp. Sta., p. 7. Several species of parasitic Hymenoptera, Diptera,
and predacious insects attack both the gipsy moth and brown-tail
moths in Massachusetts ; they are also devoured by birds, toads, and
other insectivorous animals.

Brooks, Frep E.

1907. The grapevine root-borer. Bull. 110, West Virginia Agr. Exp. Sta.,
pp. 19-30, 5 pls., Nov. Memythrus polistiformis family Sesiidae.
Great crested flycatcher feeding upon it (p. 28).

BrYANT, Harorp C.

1911. The relation of birds to an insect outbreak in northern California
during the spring and summer of 1911. The Condor, vol. 13,
pp. 195-208, Nov. Stomach examination revealed that four species
of birds fed upon the butterflies (Eugonia californica), which
formed an average of 32.8 per cent of their food. In addition a
western flycatcher was observed feeding upon them. Chickens and
ducks also reported as taking numbers of these insects.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 167

Burcess, A. F., AND CrossMAN, S. S.

1929. Imported insect enemies of the gipsy moth and the brown-tail moth.
U. S. Dep. Agr. Techn. Bull. 86, 147 pp., 6 pls., 55 figs. More
than 93 million parasites and predators liberated and a high degree
of control was obtained by 10924; parasite population fluctuates.
Heteroptera, dermestids, spiders, nematodes, mites, birds, and fungi
contribute to the mortality, as do starvation and severe weather.

CHITTENDEN, F. H.

1916. The common cabbage worm. Farmers’ Bull. 766, U. S. Dep. Agr.,
p. 9, Nov. The English sparrow, chipping sparrow, and house
wren are known to feed upon cabbageworms. It is certain that
other species eat them—one instance noted where the pupae were
reduced more than 90 per cent by birds.

1922, List of natural enemies of the celery leaf-tyer (Phlyctaenia rubigalis
Guen.). Can. Ent., vol. 64, no. 8, p. 174, Aug. Twelve hymenop-
terous parasites.

Crark, A. H.

1926. Carnivorous butterflies. Smithsonian Rep. 1025, pp. 439-508, figs.
1-5. Lycaenidae. feeding on ants or upon ant-tended insects, as
aphids, coccids, jassids, and membracids.

CocKERELL, T. D. A.

1898. Preliminary notes on the codling moth. Bull. 25, New Mexico Agr.
Exp. Sta., pp. 55-58. Woodpeckers, and quite likely the kinglet,
bats, toads, hymenopterous parasites, clerid beetle larvae, and
parasitic fungi are recorded as enemies of the codling moth. Men-
tion is also made of the house mouse.

Cottins, C. W.

1926. Observations on a recurring outbreak of Heterocampa guttivitta
Walker and natural enemies controlling it. Journ. Agr. Research,
vol. 32, no. 7, pp. 689-6900, Apr. 1. List of 15 hymenopterous,
dipterous, and nematode parasites, and 8 coleopterous, hemipterous,
and mammalian predators. Bibliography.

Comstock, J. H.

1879. Report upon cotton insects. U. S. Agr. Comm. 1879, 511 pp., 3 pls.,
77 figs. Enemies of the cottonworm (pp. 138-214) include hogs,
dogs, cats, raccoons, bats, wild birds, poultry, spiders, Chrysopa,
dragonflies, mantis, Hemiptera, robber flies, tiger beetles, ground
beetles, soldier beetles, ladybirds, boll worms, wasps, ants, chalcid,
ichneumonid, and tachinid parasites, flesh flies and phorids.

CooLey, R. A.

1908. An army cutworm (Chorizagrotis auxiliaris). Bull. 71, Montana
Exp. Sta. pp. 146-147. Several species of wild birds, domestic
fowls, parasitic flies and wasps, besides beetles, are named as foes.

1930. The codling moth. Bull. 42, Montana Exp. Sta., p. 7. Birds con-
sidered to be great destroyers of this insect.

CrIDDLE, NORMAN.

1920. Fragments in the life-habits of Manitoba insects—II. Can. Ent.,
vol. 52, no. 6, pp. 121-125, July. Proteopteryx oregonana checked
by parasites, Calosoma, and birds.

168 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Crug, S. E.

1926. Tobacco cutworms and their control. Farmers’ Bull. 1494, U. S.
Dep. Agr., 13 pp., 11 figs, Aug. Among the agencies of natural
control are mentioned spiders, ground beetles, birds, and toads.
Birds play an important part.

[Davis, W. T.]

1914. [Dragonflies eating butterflies.] Ent. News, vol. 25, no. 4, p. 191,
Apr. Mr. W. T. Davis said “ That the dragonflies, especially on
the west coast of Florida, were quite a nuisance to collectors on
account of their catching many of the smaller butterflies that were
disturbed.”

Dustan, ALAN G.

1923. The natural control of the white-marked tussock moth under city
and forest conditions. Proc. Acadian Ent. Soc., vol. 8, 1922,
pp. 109-126, pls. 15-16. In the city the principal enemies are para-
sites of the eggs and larvae, while in the forest, birds, ants, and
spiders assume that role. -

FELT, He ee

1912. [Green maple worm (Xylina antennata Walk.)]. 27th Rep. State
Entomologist, 1911, New York State Mus. Bull. 155, pp. 50-51,
Jan. Nine species of birds actually observed eating or carrying
away caterpillars, and nine others apparently associated in the work.

FLOERSHEIM, C.

1906. On some enemies of the diurnal Lepidoptera. Ent. Rec. vol. 18,
no. 2, pp. 36-39, Feb. Two cases of birds; predacious beetles very
abundant and get many sleeping butterflies. Spiders and coccinellid
larvae eat butterfly larvae; eggs destroyed by Hemiptera.

Forses, S. A.

1883. The regulative action of birds upon insect oscillations. Bull. Illinois
State Lab. Nat. Hist., vol. 1, no. 6, pp. 1-32, May. Chiefly on
birds in relation to an outbreak of cankerworms which made up
35 per cent of food of all birds in the affected orchard. Schedule
of all food items identified.

ForsusH, E. H.

1899. The destruction of hairy caterpillars by birds. Bull. 20, Div. Ent.,
U. S. Dep. Agr., pp. 85-93. List of 46 species with detailed notes
on feeding habits of some of them, especially in relation to gipsy
moth and brown-tail moth larvae, and tent caterpillars.

GarMAN, H.

1895. [Cutworm enemies]. Bull. 58, Kentucky Exp. Sta., p. 106, Nov.

Birds, chickens, turkeys, and pigs, besides insect parasites.
Gipson, ARTHUR.

1915. The army-worm. Dominion of Canada Dep. Agr., Ent. Branch,
Bull. 9, 34 pp., 19 figs. Cirphis unipuncta. Many species of wild
birds, large numbers of parasitic and predacious insects, domestic
poultry, toads, and skunks recorded as enemies. Bacterial and
fungous diseases also recorded attacking this worm (pp. 13-17).

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 169

GILL, JouHN B.

1913. The fruit-tree leaf roller. U. S. Dep. Agr. Bull. 116, pt. 5, pp. 91-110,
pls. 12-16, Mar. Archips argyrospila—eight species of birds, para-
sitic Hymenoptera and Diptera and predacious beetles and ants
recorded as enemies; a small mite also noted feeding upon the
eggs of the leaf-roller (p. 102).

GILLETTE, CLARENCE P,

1905. The beet webworm. Colorado Agr. Exp. Sta., Bull. 98, pp. 3-12, 2 pls.,
Mar. Lo-vostege sticticalis. Records of insect-eating birds devour-
ing these in quantities, mention being made of large flocks oi
blackbirds. Parasitic Hymenoptera also noted (pp. I0-I1).

HARDENBERG, C. B.

1912. The willow tree caterpillar (Angelica tyrrhea, Cramer). Agr.
Journ. South Africa, vol. 4, no. 3, pp. 397-418, Sept. Parasitic
flies and wasps attack the caterpillar and eggs. The larvae are
reported to be distasteful to birds although they are said to be
eaten by some tribes of Kaffirs. Guinea fowls and meerkats feed
upon the pupae, and moles probably do so. They are also attacked
by a fungus (pp. 412-416).

Haskin, J. R.

1916. Butterflies as food for squirrels. Ent. News, vol. 27, no. 8, p. 370, Oct.
Melitaea chalcedon, evidence of destruction of 25 or more by gray
squirrels; California.

Herrick, GLENN W.

1910. The snow-white linden moth. Bull. 286, Cornell Agr. Exp. Sta., pp.
51-64, figs. 54-58, Nov. English sparrow freed cities of this pest—
Ennomos subsignarius (p. 62).

Horton, J. R.

1922. A swallow-tail butterfly injurious to California orange trees (Papilio
zolicaon Boisd.). Monthly Bull. Dep. Agr. California, vol. 11,
no. 4, pp. 377-387, Apr. Larvae of Chrysopa californica sometimes
destroy 80-90 per cent of the eggs. Young caterpillars eaten by
same foe, also by Zelus renardi, and a hymenopterous parasite;
the pupa by Chalcis ovata (p. 385).

Howarp, L. O.

1904. The insect book. The Nature Library, vol. 8, pp. 56-57. Parasitic
Hymenoptera attacking the cotton caterpillar; an instance of de-
struction of 95 per cent of the eggs.

Howarp, L. O. AND CHITTENDEN, F. H.

1907. The catalpa sphinx (Ceratomia catalpae Bdy.). Cire. 96, U. S. Bur.
Ent., p. 6. Cuckoos, the catbird and the Baltimore oriole recorded
as enemies.

1909. The green-striped maple worm (Anisota rubicunda Fab.). Circ. 110,
U. S. Bur. Ent., p. 5. Domestic fowls and nine species of wild
birds recorded as enemies.

Jounson, E. E.

1926. Birds eating butterflies. The Field, London, vol. 147, p. 658, Apr. 15.

Observations made of a stonechat taking butterflies.

170 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

KERSHAW, JOHN C. W.
1905. Butterfly-destroyers in Southern China. Trans. Ent. Soc. London,

1905, pp. 5-8. Has seen only a dozen attacks by birds in five years; |

lizards destroy most adults. Notches in wings made by striking
twigs, etc., a cuckoo the worst enemy of larvae, taking even the
hairy and most conspicuous kinds. Also ants fastening on butterfly
tongues.

KirKLAND, A. H.

1896. The army worm. Massachusetts Crop Rep. 1896, pp. 34-36. Birds
are recognized as being the most important enemies of the army
worm. Toads, parasitic flies, and beetles also are recorded as being
enemies of this pest.

LamBorn, W. A.

1912. Butterflies a natural food of monkeys. Trans. Ent. Soc. London,

1912, p. iv. Mangabeys eating butterflies at mudholes.
Lawson, G.

1888. Insect injuries to field and orchard crops. Provincial Crop Rep.
Nova Scotia, 1888, p. 29. American and forest tent caterpillars
are eaten by the crow and cuckoo and to a lesser extent by the
linnet and swallow; cankerworms by the linnet (purple finch?).

LINTNER, J. A.

1888. Cutworms. Bull. 6, New York State Mus., 36 pp., 28 figs. Natural
enemies include wild birds, poultry, toads, ground beetles, preda-
cious bugs, mites, spiders, parasitic Diptera and Hymenoptera
(pp. 23-28).

LounssBury, C. P.

1895. Canker worms, army worms, etc. Bull. 28, Hatch Exp. Sta., p. 15.

Poultry, birds, frogs, toads, beetles, and parasitic flies listed as foes.
LucceEr, OTTOo.

1892. Tent-caterpillars. Ann. Rep. Minnesota Hort. Soc., p. 372. Cuckoos,

skunks, and Calosoma are enemies.
MaLioneg, A. M.

1916. Frogs catching butterflies. Science, n. s., vol. 43, pp. 386-387, Mar. 17.

Rana catesbiana eating a number of Papilio turnus.
MaALtty, F. W.

1893. Report on the boll worm of cotton (Heliothis armiger Hubn.). Bull.
29, U. S. Div. Ent., p. 26. Woodpeckers and sparrows reported as
enemies.

MAnopers, N.

1912. [Enemies of Danais chrysippus.] Trans. Ent. Soc. London, p. 446.
Trichogramma evanescens parasitizes large numbers of eggs; ants
eat the eggs; larvae are commonly parasitized; spiders and ants
eat them; they are cannibalistic; adults eaten by lizards, and
sometimes by birds.

Martatt, C. L. AND Orton, W. A.

1906. The control of the codling moth and apple scab. Farmers’ Bull. 247,
U. S. Dep. Agr., p. 9. Woodpeckers noted as preying upon the
codling moth.

INO. 7 PROTECTIVE ADAPTATIONS—McATEE 7

McAtTEE, W. L.

1912. Bird enemies of the codling moth. Yearbook U. S. Dep. Agr. 1911,
pp. 237-246. Birds recognized as most effective natural enemies—
from 66 to 85 per cent of the hibernating larvae recorded as being
destroyed. 36 species of birds known to prey upon this moth.

1923. Another insect birds should not eat. The Auk, vol. 40, no. 3, p. 560,
July. Red-humped apple caterpillar (Schizura concinna)—recorded
as preyed upon by six species of birds.

1924. Birds as factors in the control of the fall webworm. The Auk, vol.
41, no. 2, p. 372, Apr. Results of eight years’ study in Canada,
red-eyed vireo destroys 11.4 to 80.5 per cent of the broods, averag-
ing more than 68 per cent; birds “tremendously important”
in control of the insect.

1925. Economic ornithology. The Auk, vol. 42, no. 3, pp. 464-465, July.
European corn borer—This pest is recorded as preyed upon by
four species of birds.

1926. Birds feeding on the European corn borer. The Auk, vol. 43, no. 3,
p. 395, July. Red-wing blackbirds and downy woodpeckers re-
corded as feeding on the larvae.

1927. Economic ornithology in recent entomological publications. The
Auk, vol. 44, no. 3, pp. 458-450, July. European corn borer re-
corded as taken by six species of birds.

NEAL, H. V.

1912. Monkeys eating butterflies. Trans. Ent. Soc. London, 1912, pp. xvii-

xviii. Commonly do so in Lagos.
OTANES, F. O.

1925. The rice stem borer (Schoenobius incertellus Walker). Philippine
Agr. Rev., vol. 18, no. I, pp. 81-82. “The adult moths are said
to be preyed upon by birds, mudfish (dalag), spiders, frogs, and
mantids” (p. 82).

BACK, Hi. J.

1922. Toads in regulating insect outbreaks. Copeia, no. 107, pp. 46-47,
June 20. Feeding exclusively on sugar-beet webworms and taking
from 20 to 40 each.

PatcH, EpitH M.

1908. The saddled prominent, Hfeterocampa guttivilta (Walker). Maine
Agr. Exp. Sta., Bull. 161, pp. 311-350, figs. 14-40, Nov. Predacious
bugs and beetles, hymenopterous parasites, a fungus, skunks,
domestic fowls, and wild birds recorded as enemies. (Pp. 340-348.)

1921. A meadow caterpillar. Bull. 302, Maine Agr. Exp. Sta., pp. 300-320,
2 pls., Dec. Ctenucha virginica. One dipterous and 5 hymenopter-
ous parasites. “ Though covered with stiff hairs, the over-worked
theory that such caterpillars are thereby immune from birds’ attacks
cannot stand up against the testimony of my pet thrush which
whips these caterpillars vigorously against the floor of the cage
until, in a surprisingly short time, their bodies are beaten limp
and naked, whereupon they are swallowed in one gulp.”

berrit, I. EL.

1904. The codling moth in Michigan. Bull. 222, Michigan Agr. Exp. Sta.,
pp. 78 and 89-90. Birds most important; shrews, parasitic Hymen-
optera and fungi also mentioned.

12

172 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Puitiips, W. J. and Ktnc, Kennetu M.
1923. The corn earworm. Farmers’ Bull. 1310, U. S. Dep. Agr., 17 pp.,
18 figs., Jan. Heliothis obsolcta. Seventeen species of birds, egg
parasites, and other parasitic insects, certain ants, and spiders
recorded as foes. The earworms’ cannibalistic habit is the most
important factor in reducing its attacks on corn (pp. 10-12).
Purpps, C. R.
1927. The black army cutworm. Maine Agr..Exp. Sta., Bull. 340, pp. 201-
216, figs. 29-30, May. Agrotis fennica. Three kinds of flies and
one of wasp parasites, predators including bugs, beetles, wasps,

and birds recorded as enemies (pp. 212-213).
PRANK) Elke

1929. Natural enemies of the sugar cane moth stalkborer in Cuba. Ann.
Ent. Soc. Amer., vol. 22, pp. 621-640, 7 figs. Its burrowing habit
is “an admirable protection against direct, or artificial, but does
not seem to afford any great amount of protection against the
attack of a rather formidable array of natural enemies.” Tachinids,
sarcophagids, and Hymenoptera recorded as parasites, and earwigs,
beetles, and ants as predators. Bibliography.

Poos, F. W.

1928. An annotated list of some parasitic insects. Proc. Ent. Soc.
Washington, vol. 30, no. 8, pp. 145-150, Nov. Parasites varying
from r to 14 in number bred from each of 19 hosts, mostly
Lepidoptera.

Poutton, E. B.

1911. The attacks of tachinid flies upon the African Danaine genus Amauris.
Trans. Ent. Soc. London, tort, p. xcix. Twenty out of 25 pupae
of Amauris psyttalea parasitized; another lot all parasitized. This

a good indication enemies of adults are scarce; otherwise species
would be rare.

QUAINTANCE, A. L.

1908. The apple-tree tent caterpillar (J/alacosoma americana). Circ. 08,
U. S. Bur. Ent., p. 6. Black-billed and yellow-billed cuckoos, blue-
jay, crow, chickadee, Baltimore oriole, red-eyed vireo, chipping
sparrow, and yellow warbler mentioned as enemies; also the com-
mon toad.

SANDERSON, E. D.

1903. The codling moth. Bull. 59, Delaware Agr. Exp. Sta., pp. 7-8.
Parasitic worms and insects, soldier beetles, named as enemies as
well as 10 or more species of birds which are the most efficient.

1905. The gipsy moth in New Hampshire. Bull. 121, New Hampshire
Agr. Exp. Sta. p. 99. Ground beetles, parasitic insects, and
several species of birds are recorded as preying upon the gipsy
moth.

1906. The brown-tailed moth in New Hampshire. Second report. Bull.
122, New Hampshire Agr. Exp. Sta., p. 127. Parasitic insects,
predacious bugs, toads, bats, and several species of birds, particu-
larly the English sparrow, recorded as enemies.

ae SSS

INO©:, 7 PROTECTIVE ADAPTATIONS—McATEE 173

1909. The codling moth and how to control it by spraying. Bull. 143,
New Hampshire Agr. Exp. Sta., pp. 64-82. Downy woodpeckers
and nuthatches are recorded as being the most important enemies
of the codling moth.

SANDERSON, E. D., HEApLEE, T. J., AND Brooks, CHARLES.

1907. Spraying the apple orchard. Bull. 131, New Hampshire Agr. Exp.
Sta., pp. 18-19 and 35. Woodpeckers and nuthatches are recorded as
feeding extensively on the codling moth.

SAUNDERS, ARETAS A.

1916. A note on the food of the western robin. The Condor, vol. 18, no. 9,
p. 81, Mar.-Apr. Robin feeding on Papilio rutulus, and a chipmunk
feeding upon the same species and also on P. eurymedon.

SHARP, D.

1910. [Parasites of winter moth.] Cambridge Nat. Hist., vol. 5, p. 521.
“The destructive winter moth—Chetmatobia brumata—is known
to be subject to the attacks of 63 species of Hymenopterous para-
sites. So abundant are these latter that late in the autumn it is
not infrequently the case that the majority of caterpillars contain
these destroyers.”

SHERMAN, F.

1921. Observations of natural enemies of the fall cankerworm (Alsophila
pometaria Peck) in forests of southern Alleghany Mountains in
1920. Journ. Econ. Ent., vol. 14, no. 6, pp. 478-481, Dec. Names
15 species of birds presumably of material help, five predacious
insects, and 3 parasites of which one destroyed from 25 to 4o per
cent of the eggs.

SHIRAS, GEO., 3RD.

1921. Frogs eating butterflies. Nat. Geogr. Mag., vol. 40, no. 2, p. 174,
Aug. Leopard frogs catching about 500 “blues” in a week; also
eating many Argynnis aphrodite.

SxkalrFe, S. H.

1921. Some factors in the natural control of the wattle bagworm. South
African Journ. Sci., vol. 17, nos. 3-4, pp. 291-301, July. Acan-
thopsyche junodi Heylaerts. Out of a total of 50,687 examined,
just over one per cent were destroyed by birds and rats, 19 per
cent by insect parasites, 16 per cent by fungous disease, and 17
per cent by other diseases. Only one-quarter of one per cent
survive the early perils of their life.

SmirTH, J. B.

1895. [Zeusera pyrina.] 15th Ann. Rep. New Jersey Exp. Sta. 1894, pp.
531-532. Almost all insectivorous birds, especially woodpeckers and
the sparrow, in addition to bats, and predacious insects, are re-
corded as enemies of this insect.

SPENCER, G. J. AND H. G. Crawrorp.

1923. The European corn borer in Ontario. Ontario Dep. Agr. Bull. 2095,
II pp., 10 figs., Mar. Ants, aphis-lions, ladybird beetles, ground
beetles, crickets, a parasitic fly, and several species of birds noted
as foes. One instance recorded of downy and hairy woodpeckers
destroying 60 per cent of the borers (pp. 7-8).

174. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

ToTHILL, JoHn D.

1922. The natural control of the fall webworm [Hyphantria cunea Drury]
in Canada together with an account of its several parasites. Dep.
Agr. Dominion of Canada, Bull. 3, n. s. (techn.), 107 pp., 6 pls.,
99 figs. Tabulations of the destruction by various enemies in differ-
ent localities and years; birds average most important, parasites
sometimes important, sometimes not.

1923. Notes on the outbreaks of spruce budworm, forest tent caterpillar,
and larch sawfly in New Brunswick. Proc. Acadian Ent. Soc.,
vol. 8, 1922, pp. 172-182. Spruce budworm. Natural checks effective
in New Brunswick were eggs, larvae, and pupal parasites, spiders,
and birds; and in British Columbia, parasites and birds. Nothing
of importance on enemies of the other forms.

TROUVELOT, LEOPOLD.

1868. The American silkworm. Amer. Nat., vol. 1, pp. 30-38, 85-94, 145-
149. Telea polyphemus. Thrushes, catbirds, orioles; 95 out of a
hundred worms become the prey of these feathered insect hunters.

U. S. ENtTomMoLocIcAL CoMMISSION.

1883. 3rd Rep. U. S. Ent. Comm., p. 125. All insectivorous birds, hogs,
chickens, turkeys, toads, and frogs prey upon the armyworm.
“The worms themselves, when hard pushed, will even devour each
other.”

1883. 3rd Rep. U. S. Ent. Comm., pp. 175-178. Forty or more species of
wild birds, notable mention being made of bluebirds, cedarbirds
and butcherbirds, and parasitic and predacious insects, also hogs,
are recorded as enemies of the cankerworm.

1885. 4th Rep. U. S. Ent. Comm., pp. 87-90. More than 20 species of wild
birds, poultry, hogs, raccoons, skunks, opossums, bats, tree frogs,
lizards, spiders, and numerous kinds of predacious insects are
recorded as preying upon the cottonworm.

Vickery, R. A. :

1929. Studies on the fall army worm in the Gulf Coast District of Texas.
Techn. Bull. 138, U. S. Dep. Agr., 63 pp. Numerous hymenopterous
parasites sometimes destroy 40-50 per cent of the caterpillars.

Warren, B. H.

1897. The army worm. Ann. Rep. Pennsylvania State Coll. 1806, pp. 164-
220, 16 pls. Much on natural enemies including tachinids and
ichneumonids, ground beetles, birds, mammals, and toads.

Wesster, R. L.

1909. The lesser apple leaf-folder. Iowa State Coll. Exp. Sta. Bull. 102,
pp. 181-212, figs. 1-13, Mar. Peronea minuta—tachinid and hymen-
opterous parasites recorded as the most important natural enemies
although birds and diseases are also important factors (pp.
206-211).

WEED, C. M.

1809. The forest tent caterpillar. Bull. 64, New Hampshire Agr. Exp.
Sta., pp. 77-98, figs. 20-33, Apr. Clisiocampa disstria—preyed upon
by insects, spiders, toads, and birds. Ten kinds of birds feeding
on larvae, one on the cocoons and four on the adults.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 175

1899. The spiny elm caterpillar. Bull. 67, New Hampshire Agr. Exp. Sta.,
pp. 125-141, figs. 40-51, Oct. Vanessa antiopa—egg parasites
chalcid, ichneumonid and tachinid flies on caterpillars; Calosoma,
wasps, cuckoos, and toads named as enemies (pp. 138-140).

1900. Insect record for 1899. Bull. 72, New Hampshire Exp. Sta., pp.

64-65. The Baltimore oriole and the cuckoos are especially noted

as enemies of tent caterpillars.

1900. The forest tent caterpillar. Second report. Bull. 75, New Hampshire
Exp. Sta., pp. 120-121. Eighteen or more species of birds are
recorded as preying upon these caterpillars.

1902. [Enemies of cankerworms.] Bull. 90, New Hampshire Agr. Exp.
Sta., 1902, p. 35, Mar. Robins, bluebirds, cedarbirds, and many
others feed freely upon the pests.

West, L. S.

1923. Immunity to parasitism in Samia cecropia Linn. (Lep.: Saturntidae;
Dip.: Tachinidae.) Ent. News, vol. 34, no. 1, pp. 23-25, Jan. Ineffec-
tiveness of attack of 35-40 tachinid larvae; nevertheless lists two
Tachinidae and eight species of Hymenoptera that do successfully
parasitize this host.

YorHers, M. A.

1913. Eugonia californica Bdy. in the Pacific Northwest. Can. Ent., vol.
45, no. 12, pp. 421-422, Dec. “I think that the total disappearance
of these caterpillars and chrysalids was no doubt due to birds”
(p. 422).

Youn, R. A.

1907. Insects affecting the poplar. Proc. Columbus Hort. Soc. 1906, pp.
68-82. Birds constitute an important agency in keeping the
Hemerocampa leucostigma in check (p. 74).

COLEOPTERA

Acassiz, L., AND Casot, J. ELtior.

1850. Lake Superior, etc., p. 72. Monochamus scutellaris preyed upon by

Canada jay and two species of Picoides.
BiackMaNn, M. W.

1915. Observations on the life history and habits of Pityogenes hopkinsi
Swaine. Techn. Publ. no. 2, New York State Coll. Forestry, pp.
11-66, 6 pls., Nov. Natural enemies include beetles, mites, and
parasitic Hymenoptera (pp. 53-560).

Britten, H.

1926. A pentatomid bug preying on beetle larvae. North Western Nat.,
vol. 1, p. 38. Rhacognathus punctatus found sucking larvae of
beetle (Hydrothassa marginella).

1927. Red ants and beetles. North Western Nat., vol. 2, p. 256. Myrmica
ruginodis killing beetles (Melandrya caraboides).

Brooks, F. E.

1919. The flat-headed apple-tree borer. Farmers’ Bull. 1065, U. S. Dep.
Agr., 12 pp., 13 figs. Chrysobothris femorata. Woodpeckers, and
other birds, ants, and six species of hymenopterous parasites re-
corded as enemies (p. 9).

176 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

Burcess, A. F. Anp CoLitns, C. W.

1917. The genus Calosoma, including studies of seasonal histories, habits,
and economic importance of American species north of Mexico and
of several introduced species. Bull. 417, U. S. Dep. Agr., 124 pp.,
19 pls. 5 figs. Caterpillars the principal food of these beetles;
cannibalism, and attacks of toads, skunks, and birds the chief
organic checks; predatory bugs, and parasitic flies also noted
(pp. 10-13).

CARTWRIGHT, OSCAR L.

1929. The maize billbug in South Carolina. Bull. 257, South Carolina
Agr. Exp. Sta., 35 pp., 13 figs. May. Natural enemies (p. 31)
include egg parasite, predacious beetles, ants, and wasps.

CHAMBERLIN, F. S., AND TENHET, J. N.

1923. The tobacco flea-beetle in the southern cigar-wrapper district.
Farmers’ Bull. 1352, U. S. Dep. Agr., 9 pp., 8 figs. Epitrix parvula.
A spider, lygaeid bug, and birds noted as enemies (p. 5).

CHITTENDEN, F. H.

1911. Notes on various truck-crop insects. Bull. 82, pt. 7, U. S. Bur. Ent.,
pp. 85-93, fig. 24. Natural enemies of Leptinotarsa decemlineata,
pp. 85-88; 1 beetle, 3 hemiptera, 16 wild birds, and guinea fowls.

1926. Notes on the behavior of Cotinis nitida L. and its bird enemies. Proc.
Biol. Soc. Washington, vol. 39, pp. 15-17, Feb. Starling and
cardinal eat it.

CHITTENDEN, F. H., anp Fink, D. E.

1922. The green June beetle. Bull. 891, U. S. Dep. Agr., 52 pp., Io pls.,
7 figs. Cotinis nitida. Natural enemies (pp. 31-37) include para-
sitic flies, digger wasps, ground beetles, mites, various mammals,
and birds; fungal and bacterial diseases also noted.

CHITTENDEN, F. H., anp Marsa, H. O.

1920. The western cabbage flea-beetle. U. S. Dep. Agr. Bull. 902, 21 pp.,
4 figs., 1 pl., Oct. Phyllotreta pusilla. Hymenopterous and worm
parasites recorded as well as 12 species of birds feeding on beetles
of this genus.

Davis, JoHN J.

1913. Common white grubs. U. S. Farmers’ Bull. 543, 20 pp., 12 figs.,
July. More than 60 species of birds, domestic fowls, skunks, a
number of predacious and parasitic insects recorded preying upon
white grubs at one stage or another (pp. 13-15).

FENTON, F. A., AND DunHAM, E. W.

1929. Biology of the cotton boll weevil at Florence, S. C. Techn. Bull.
112, U. S. Dep. Agr., 75 pp., 35 figs. Considerable variation exists
in mortality rate from parasitism (by three species of Hymen-
optera) from as low as 2.37 to as high as 51.52 per cent; predators,
heat, proliferation by the plant, disease and unknown causes take
their toll, the average total from all these causes being about 40
per cent; then from the number that go into hibernation only an
average Of 3.27 per cent survive.

i.
|
1
if
1
|
,@
|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 177

FrEYTAUD, JEAN.

1922. Le Doryphore, Chrysomcle nuisible a la pomme de terre (Leptino-
tarsa decemlineata Say). Rev. Zool. Agr. Appl., vol. 21, Numero
special, 48 pp., 13 figs., Aug. Natural checks include skunks, birds,
snakes, frogs, spiders, phalangids, mites, beetles, bugs, wasps, robber
flies, and parasitic flies (pp. 14-17).

Forses, S. A.

1880. Notes upon the food of predacious beetles. Bull. Illinois State Lab.
Nat. Hist., vol. I, no. 3, pp. 149-152, Nov. Both vegetable and
animal, the latter including beetles, larvae, and plant lice.

1880. Notes on insectivorous Coleoptera. Bull. Illinois State Lab. Nat.
Hist., vol. 1, no. 3, pp. 153-160, Nov. Carabidae, Lampyridae,
Coccinellidae, from stomach examination. Animal food, mites and
their eggs, ants, caterpillars, beetles and their larvae, plant lice,
and centipeds.

1883. The food relations of the Carabidae and Coccinellidae. Bull. Illinois
State Lab. Nat. Hist., vol. 1, no. 6, pp. 33-64, May. Report on
dissections of 175 Carabidae and 39 Coccinellidae. Animal food
included Hymenoptera, Lepidoptera, Diptera, Neuroptera, and
Coleoptera, spiders, mites, myriapods, mollusks. Notes on birds
as enemies of Cicindelidae and Carabidae.

1907. On the life history, habits, and economic relations of the white-grubs
and may beetles. Bull. 116, Illinois Agr. Exp. Sta., pp. 447-480,
Aug. Principal enemies, swine, crows, blackbirds, and Tiphia;
other parasites Macrophthalma, Sparnopolius, Pyrgota, and Ophion
(pp. 468-475).

ForsusH, E. H.

1912. 4th Ann. Rep. State Ornithologist Mass., 1911, 32 pp., 4 pls., 9 figs.
Galerucella luteola. Cedar waxwing clearing trees of infestations
of the elm leaf beetle (pp. 19-20).

Hess, WALTER N.

1920. The ribbed pine borer. Mem. 33, Cornell Agr. Exp. Sta., pp. 367-
381, pl. 8, figs. 61-66. Rhagium lineatum. Woodpeckers, most im-
portant; a parasite reared (pp. 378-379).

Hopkins, A. D.

1896. The relation of insects and birds to present forest conditions. Proc.
Amer. Forestry Assoc., vol. 11, pp. 175-176. Woodpeckers recorded
as enemies of bark and clerid beetles.

Hystop, JAMES A.

1912. The false wireworms of the Pacific Northwest. Bull. 95, pt. 5, U. S.
Bur. Ent., pp. 73-87, figs. 22-27. Numerous species of birds,
horned-toads, garden toads, skunks, parasites, and disease re-
corded as enemies (pp. 84-86).

1915. Wireworms attacking cereal and forage crops. Bull. 156, U. S.
Dep. Agr., 34 pp., 8 figs. Elateridae—a long list of bird enemies
given; horned-toads, mites, predacious flies, hymenopterous para-
sites, nematodes, fungi (pp. 25-20).

IncRAM, J. W.

1927. The striped blister beetle on soy beans. U. S. Dep. Agr. Leafl. 12,
5 pp., 3 figs. Epicauta lemniscata—three species of birds and a
robber fly named as enemies.

178 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

KALMBACH, E. R.

1914. Birds in relation to the alfalfa weevil. Bull. 107, U. S. Dep. Agr.,
64 pp., 5 pls. Forty-five species of birds found to feed on this
comparatively recently introduced pest; also domestic fowls, toads,
frogs, salamanders, horned-toads, snakes, and shrews.

McAteer, W. L.

1914. Bird enemies of Diabroticas. The Auk, vol. 31, no. 1, p. 120, Jan.
Southern corn root worm (Diabrotica duodecim-punctata) recorded
as preyed upon by 24 species of birds; western corn root worm
(Diabrotica longicornis) taken by the nighthawk and the wood
pewee.

1915. Bird enemies of two beetle pests. The Auk, vol. 32, no. 3, pp. 377-
378, July. Oncideres putator—it is believed that the southern downy
woodpecker and the Texas woodpecker attack the larvae of this
pest. Monocrepidius vespertinus—the records show that these
beetles are devoured by eight species of birds.

MaiL, G. ALLEN.

1930. Winter soil temperatures and their relation to subterranean insect
survival. Journ. Agr. Research, vol. 41, no. 8, pp. 572-592, Oct. 15.
Few parasites; mites, birds, fungal and bacterial diseases reduce
them, but climate a control factor of much importance.

Murr, F.

1917. The introduction of Scolia manilae Ashm. into the Hawaiian Islands.
Ann. Ent. Soc. Amer., vol. 10, no. 2, pp. 207-210, June. A parasite
of the beetles Anomala orientalis and Adoretus tenuimaculatus.

[NeEtson, E. W.]

1921. Report of chief of Bureau of Biological Survey, 34 pp. Bird enemies
of the Japanese beetle (Popillia japonica) mentioned (p. 14); five
species of birds and the toad listed.

QUAINTANCE, A. L., AND JENNE, E. L.

1912. The plum curculio. Bull. 103, U. S. Bur. Ent., 250 pp., 20 pls., 33
figs. Natural enemies (pp. 139-154) include an egg parasite,
hymenopterous and dipterous parasites of later stages, ants, chry-
sopids, carabids, lampyrids, fowls and wild birds; also the toad.

SATTERTHWAIT, A. F.

1919. How to control billbugs destructive to cereal and forage crops.
Farmers’ Bull. 1003, U. S. Dep. Agr., 23 pp., 24 figs. Insect,
worm, and fungus parasites, toad and bird predators mentioned,
the birds apparently most important (pp. 19-20).

SCHUSTER, W.

1909. [Beetles and their enemies in the bird world.] Ent. Blatt. Nurnberg,
vol. 5, no. 7, pp. 142-144, July 15. Birds the principal enemies of
beetles; notes on European bird foes of various families of beetles ;
similar notes for Lepidoptera.

SLINGERLAND, M. V.

1906. The bronze birch borer: an insect destroying the white birch. Bull.
234, Cornell Agr. Exp. Sta., pp. 65-78, figs. 31-38. Agrilus anxius.
Woodpeckers and chalcid parasites mentioned as foes.

|
|
|
|

a nn ee a aaa ar ae :

NO. 7

Tuomas, C.

PROTECTIVE ADAPTATIONS—McATEE 179

se

1931. The predatory enemies of Elateridae (Coleoptera). Ent. News, vol.

Wesp, J. L.

42, no. 5, pp. 137-140, May; no. 6, pp. 158-167, June. Mites,
pseudoscorpions, spiders, hemiptera, beetles, flies, hymenoptera,
amphibians, reptiles, birds, and mammals; birds most important ;
predators more effective than parasites. Bibliography.

1906. The western pine-destroying bark beetle. Bull. 58, pt. 2, U. S. Bur.

Wesster, F.

Ent., pp. 17-30, pls. 2-3, figs. 7-12. Dendroctonus brevicorms.
Woodpeckers destroyed large percentage in some trees (p. 27).
M.

1880. Notes upon the food of predacious beetles. Bull. Illinois State Lab.

Nat. Hist., vol. 1, no. 3, pp. 149-152, Nov. Chiefly on vegetarian
Carabidae but notes on carnivorous species of Carabidae, Staphy-
linidae, and Lampyridae; the prey mentioned includes plant lice,
cricket, grasshopper, and beetles.

1892. Underground insect destroyers of the wheat plant. Bull. Ohio Agr.

Exp. Sta., vol. 5, no. 9, pp. 221-247, 8 figs., Dec. Wireworms—
crows, thrushes, robins, blackbirds (p. 228). Whitegrubs—poultry,
crows, jays, nighthawks, robin, catbird, brown thrasher, wood
thrush, red-headed woodpecker ; swine, bats, badger, weasel, martin,
rat, skunk, raccoon, fox, mole, frogs, digger wasps, robber flies,
and fungi (pp. 236-237).

1913. The southern corn rootworm, or budworm. U. S. Dep. Agr. Bull. 5,

II pp. 2 figs., Sept. Diabrotica 12-punctata—12 species of birds
and parasitic flies (pp. 9-10).

1913. The western corn rootworm. Bull. 8, U. S. Dep. Agr., 8 pp., 5 figs.,

Sept. Diabrotica longicornis preyed upon by nighthawks, wood
pewees, a parasitic fly, and chinch bug fungus (p. 6).

WILDERMUTH, V. L.
1910. The clover-root curculio. Bull. 85, pt. 3, U. S. Bur. Ent., pp. 29-38,

WILson, C.

figs. 15-19. Sitones hispidulus—14 species of birds recorded as
enemies (p. 37). ;
B.

1923-1924. Life history of the scavenger water beetle Hydrous (Hydro-

philus) triangularis, and its economic relations to fish breeding.
Bull. Bur. Fisheries, vol. 39, pp. 9-38, 22 figs. Food of larvae,
snails, midge larvae, fishes, other water beetle larvae, tadpoles, and
several groups of insects and crustaceans in smaller quantity. Food
of adults, vegetable matter, fishes. Enemies of Hydrophilus in-
clude cannibalistic larvae, dragonfly nymphs, frogs, fishes, birds.
Bibliography.

1923-1924. Water beetles in relation to pondfish culture, with life histories

of those found in fishponds at Fairport, Iowa. Bull. U. S. Bur.
Fisheries, vol. 30, pp. 231-345, figs. 1-143. Larvae highly cannibal-
istic, dragonfly nymphs are enemies, as are also, mites, hydra,
ants, fishes, turtles, frogs, and snails; foes of pupae include hymen-
opterous parasites, horse fly larvae, and ants; of adults, turtles,
fishes, birds, toads, and frogs. Notes are given on the feeding

180 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

habits of the larvae and adults of a number of water beetles; fish
destruction not so apparent as would have been inferred from
previous literature. Bibliography.

MECAPTERA

Predacious.

DIPTERA
ALEXANDER, CHARLES P.

1929. The crane-flies of New York. Mem. 38, Cornell Univ. Agr. Exp.
Sta., pp. 699-1132, pls. 12-97, June. Ninety-one species of birds,
besides foxes, mice, shrews, moles, amphibians, fishes, mites, spiders, |
dragonflies, Diptera, Coleoptera, Hymenoptera, Protozoa recorded |
as enemies of crane flies in one stage or another (pp. 721-734).

BroMtey, S. W. |

1923. Observations on the feeding habits of robber flies. Part I. Psyche, |
vol. 30, no. 2, pp. 41-45, Apr. Tabulation of the prey of 26 Procta-
canthus rufus, all Hymenoptera and in 14 cases honey bees. Six
records for P. brevipennis include three of beetles, one ant, one |
blow fly, and one assassin bug.

1930. Bee-killing robber flies. Journ. New York Ent. Soc., vol. 38, no. 2,
pp. 159-176, pl. 10, June. Especially the honey bee; review of |
records from various countries; discussion of the U. S. species,
with notes on other kinds of prey taken by some of them. The
Dasypogoninae tend to favor Hymenoptera, the Laphriinae beetles,
while the Asilinae are more general feeders.

Borris, ALG: |

1913. Economic and biologic notes on the giant midge: Chironomus
(Tendipes) plumosus Meigen. Bull. Wisconsin Nat. Hist. Soc., |
vol. 10, nos. 3-4, pp. 124-163, Apr. Swallows, red-winged black-
birds as enemies (p. 146); other notes in annotated bibliography
refer to fishes, birds, Utricularia, and fungi as natural checks.

1913. Notes on Lake Michigan swarms of chironomids; quantitative notes
on spring insects. Bull. Wisconsin Nat. Hist. Soc., vol. 11, nos. 1-2,
pp. 52-69, June. Enemies of adults include mites, spiders, ants, and
birds (pp. 66-67).

CocKErRELL, T. D. A.

1894. On the habits of some Asilidae. Ent. News, vol. 5, no. 6, pp. 173-
174, June. Mallophora fautrix eating Odynerus sp.; Mallophora
sp. eating honey bee; Proctacanthus philadelphicus preying upon
Erax dubius, and butterfly, Synchloe lacinia var. crocale; Procta-
canthus milberti preying upon Bembex sp. and honey bee; Pro-
machus princeps preying upon Odynerus annulatus.

CUTHBERTSON, ALEXANDER.

1926. The trout as a natural enemy of crane-flies. Scottish Nat., 1926, pp.

85-88. Salmo fario an important consumer of crane flies in all
stages; earthworms, phalangids, and click beetles also in the
stomachs.

1926. Spiders as enemies of crane-flies. Scottish Nat., 1926, pp. 127-129.

List of species that eat crane flies, of which the names are given.
Special study of the prey found in webs of the wood spiders Zilla
atrica and Z. #-notata.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 181

HARSHBARGER, W. A.

1894. The bold robber-fly and the mantis. Ent. News, vol. 5, no. 6, p. 169,
June. Asilid attacked mantis (Stagmomantis carolina) but was itself
captured and partly eaten.

HILDEBRAND, S. F.

1919. Fishes in relation to mosquito control in ponds. Rep. U. S. Comm.

Fisheries 1918, App. 9, 15 pp., 18 figs.
Hine, JAMEs S.

1906. Habits and life histories of some flies of the family Tabanidae. Bur.
Ent., U. S. Dep. Agr., Techn. Bull., no. 12, pt. 2, pp. 10-38, 12
figs., Aug. Birds, hornets, and spiders noted as enemies.

Howarp, L. O.

1904. The insect book. The Nature Library, vol. 8, pp. 158-159. Outbreaks
of the armyworm sometimes completely controlled by tachina flies.
They also attack grasshoppers, bugs, and beetles, sawflies and saw-
fly larvae and bumble bees and wasps.

1910. Preventive and remedial work against mosquitoes. Bull. 88, U. S.
Bur. Ent., 126 pp., June 20. Use of natural enemies, salamanders,
dragonflies, predacious mosquitoes and fish (pp. 62-72).

Hystop, J. A.

1910. The smoky crane-fly. Bull. 85, pt. 7, U. S. Bur. Ent., pp. 119-132,
figs. 60-66. Tipula infuscata. Natural enemies include a tachinid
parasite, ground beetles, ants, mites, and birds; a long list is given
of birds that feed on crane flies; fungi also kill the insects.

LEATHERS, A. L.

1922. Ecological study of aquatic midges and some related insects with
special reference to feeding habits. Bull. U. S. Bur. Fisheries,
38, Doc. no. 915, 61 pp., 44 figs., May. Food includes Protozoa,
small Crustacea, diatoms, algae, and other vegetation.

MatTueson, Roperr.

1929. A handbook of the mosquitoes of North America. 268 pp., 25 pls.
Food; suck blood of mammals, birds, amphibians, and snakes (pp.
39-41); enemies, birds, bats, fishes, tadpoles, salamanders, and
insects (pp. 71-72).

SmitH, K. M.

1927. A study of Hylemyia (Chortophila) brassicae Bouche, the cabbage
root fly, and its parasites. With notes on some other dipterous
pests of cruciferous plants. Ann. Appl. Biol., vol. 14, pp. 312-330.
Description of life-history, enemies, and parasites. The larvae of
a small beetle (Aleochara bilineata) destroy the pupae of the fly ;
while a cynipid and a braconid parasitize the larvae, which are also
eaten by the carnivorous larva of an anthomyid fly. The larva
of the beetle is itself parasitized by a proctotrupid.

Twinn, C. R.

1931. Observations on some aquatic animal and plant enemies of mosquitoes.
Can. Ent. 63, no. 3, pp. 51-61, Mar. Other mosquito larvae, water
beetle larvae, dragonfly and damselfly nymphs, backswimmers,
water-scorpions, caddis larvae, salamanders, fishes, hydras, and
bladderworts. Bibliography.

182 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Van Dine, D. L.
1907. The introduction of top-minnows (natural enemies of mosquitos)
into the Hawaiian Islands. Press Bull. 20, Hawaiian Agr. Exp.
Stas) lO pps 3 tigss, July 2s:
WEED, C. M.
1902. [Enemies of Bibio albipennis.] Bull. 900, New Hampshire Agr. Exp.

Sta., pp. 32-33, Mar. Fishes eating those falling in lake; chief
food of robin in early spring.

HYMENOPTERA
BerLtawsky, A. G.

1927. [Enemies of bees.] Vragi Pchet, Leningrad, 204 pp., 2 pls., 148 figs.
Mammals, birds, reptiles, amphibians, insects, arachnids, worms,
and protozoans.

BEQUAERT, J.

1922. The predacious enemies of ants. Bull. Amer. Mus. Nat. Hist., vol.
45, Pp. 271-331, pls. 24-25, Oct. Spiders, beetles, ant-lions, Diptera,
Hymenoptera, amphibians, lizards, birds and mammals, including
man, discussed at length. ‘“ There is certainly little or no evidence
to show that, as the theory is often expressed, ants are unpalatable
to most insectivorous animals” (p. 271).

Bicetow, N. K.

1922. Insect food of the black bear (Ursus americanus.) Can. Ent., vol.
54, no. 3, pp. 40-50, Mar. Vespula diabolica, V. consobrina, and
ants; notes on observations of others.
Davis, Wo. T.

191g, A remarkable nest of Vespa maculata, with notes on some other
wasps’ nests. Bull. Brooklyn Ent. Soc., vol. 14, nos. 4-5, pp. 119-
123, Oct.-Dec. Notes on food habits of Vespa spp., cannibalistic,
eat flies and damselflies; robber flies are their enemies.

1924. Oak apple galls destroyed by gray squirrels. Bull. Brooklyn Ent.
Soc., vol. 19, no. 3, pp. 91-93, I fig., June. Amphibolips confluens
freely eaten.

GRAHAM, S. A.

1928. The influence of small mammals and other factors upon the larch
sawfly survival. Journ. Econ. Ent., vol. 21, no. 2, pp. 301-310, Apr.
Lygaeonematus erichsoni. Small mammals, probably Microtus
chiefly, destroy from 50 to 80 per cent of the hibernating cocoons;
parasites and fungi about Io per cent.

GRONEMAN, CArL F.

1923. Birds as destroyers of gall insects. Audubon Bull. (Illinois Audubon
Soc.), pp. 13-15, 6 figs., Fall issue. Birds and squirrels recorded
as enemies.

HEIKERTINGER, FRANZ.

1919. Die metoke Myrmekodie. Tatsachenmaterial zur Losung des Mimi-
kryproblems. Biol. Zentralbl., vol. 30, no. 2, pp. 65-102, Feb. Enemies
of ants (pp. 81-100), insects, spiders, amphibians, reptiles, mam-
mals, birds.

|
i

eR

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 183

Hersey, J. L.

1873. Bees and kingbirds. Can. Ent., vol. 5, pp. 159-160. Kingbirds and
purple martins feed on honey bees, mostly drones; kingbirds feed
freely on dragonflies also.

Howarp, L. O.

1904. [Prey of Proctotrypoidea.] The insect book, p. 51. Gall flies, gall
gnats, butterflies, moths, beetles, and the eggs of spiders, bugs,
butterflies, and moths.

Hunter, W. D.

1912. Two destructive Texas ants. U. S. Dep. Agr., Bur. Ent. Circe. 148,
6 pp., Apr. Pogonomyrmex barbatus molefaciens. Eight species of
birds and the horned lizard recorded as enemies.

IseLy, Dwicut.

1913. The biology of some Kansas Eumenidae. Kansas Univ. Sci. Bull.,
vol. 8, no. 7, pp. 235-3090, pls. 34-37, July. Prey includes larvae of
several families of Lepidoptera, of two of beetles, and of sawflies.
Bibliography.

Puitiirs, E. F.

1917. Beekeeping. Chap. 22, Bee diseases and enemies, pp. 397-416. Three
diseases of the brood, two of adults; enemies include two wax
moths, toads, birds, mice, rats, and other small mammals, certain
spiders and mites, dragonflies, various Hemiptera, the death’s head
moth, Mediterranean flour moth, a dipterous parasite (Braula
caeca), blister beetle (Meloe) and other beetles, wasps, hornets,
and ants. ‘‘ Dragonflies are so destructive to queens as to make
queen-rearing unprofitable in some places.”

Suarp, D.

1910. [Summary of the prey of Fossores.] Cambridge Nat. Hist., vol. 6,
pp. 92-93. General notes on prey of 16 families of wasps.

1910. [Prey of Ichneumonidae.] Cambridge Nat. Hist., vol. 5, p. 551.
“Most of the species, in the larval state, live inside the larvae of
Lepidoptera, and they thus keep the myriads of caterpillars within
bounds, the number of these destroyed by ichneumons being pro-
digious. Some of the family are, however, external parasites, and
some are known to attack spiders and insects of other Orders than
Lepidoptera.”

Swenk, M. H.

1910. A new sawfly enemy of the bull pine in Nebraska. Rep. Nebraska
Agr. Exp. Sta., pp. 3-33, 18 figs. Diprion n. sp.—natural checks
include ichneumonids, tachinids, chipmunks, birds, and a bacterial
disease.

Wit.iaMs, F. X.

1913. Monograph of the Larridae of Kansas. Kansas Univ. Sci. Bull.
vol. 8, no. 4, pp. 121-213, pls. 22-30, July. Prey includes Orthoptera
chiefly, but also Hemiptera, and spiders. Bibliography.

1913. Notes on the habits of some wasps that occur in Kansas. Kansas
Univ. Sci. Bull., vol. 8, no. 6, pp. 223-230, pl. 33, fig. 1, July.
Harpactus preying upon Gypona cinerea, Mimesa upon Athysanus
exitiosus and other jassids; Prionyx upon locusts.

184. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

ARACHNIDA
Bitsine, S. W.

1920. Quantitative studies in the food of spiders. Ohio Journ. Sci., vol. 20,
no. 7, pp. 215-260, May. Summarizes a large number of observa-
tions on prey actually seen eaten by spiders, and upon insects
found in their webs; gives also some experimental results.

CALveErRT, Puiip P.

1923. Studies on Costa Rican Odonata. XY. Megaloprepus, its distribution,
variation, habits, and food. Ent. News, vol. 34, no. 6, (Food),
pp. 171-174, June. Feeds on spiders.

Lincecum, G.

1867. The tarantula killers of Texas. Amer. Nat., vol. 1, no. 3, pp. 137-141,
May. Pompilus formosus Say feeds on Mygale hentzii and other
large spiders.

(COVELL) oat

1915. Insects captured by the Thomisidae. Can. Ent., vol. 47, no. 4, pp. 115-
116, pl. 2, Apr. Crab spiders prey upon butterflies, dragonflies,
wasps, bumble bees, honey bees, and large flies.

McAtTekr, W. L.

1911. Bird enemies of the Texas-fever tick and other ticks. The Auk, vol.
28, no. I, pp. 136-138, Jan. A résumé of seven publications on the
subject; of the birds mentioned, 12 species are inhabitants of the
United States.

Savory, THEO. H.

1928. The biology of spiders. 376 pp., 16 pls., 121 figs., London. Food
(pp. 116-125), flies, wasps, bees, ants, beetles, earwigs, butterflies,
moths, harvestmen, woodlice, and other spiders; more rarely cater-
pillars, worms, fish, birds. “‘ They show no trace of discrimination.”
Enemies (pp. 176-179) include birds, toads, lizards, mammals,
harvestmen, spiders, wasps, and ichneumon flies, and other parasites.

MOLLUSCA
Baker, F. C.

1916. The relations of mollusks to fish in Oneida Lake. Techn. Publ.
no. 4, New York State Coll. Forestry, 366 pp., 50 figs., one table,
one map, July. On pp. 154-218 is summarized information on food
of 54 species of fresh-water fishes, especially in relation to mollusks.

1918. The relation of shellfish to fish in Oneida Lake, New York. Cire. 21,
New York State Coll. Forestry, pp. 11-33, figs. 1-16, Aug. Some
snails carnivorous, eating other snails, leeches, and small fish;
shellfish form a large part of the food of many species of fishes;
other enemies of shellfish include flukes, dragonfly nymphs, horse
fly larvae, water bugs, water beetle larvae, leeches, crawfishes, frogs,
salamanders, turtles, ducks, other water birds, muskrats, mink and
otter.

SEQUAERT, J.

1925. The arthropod enemies of mollusks, with description of a new dip-
terous parasite from Brazil. Journ. Parasitol., vol. 11, pp. 201-212,
fig. 1. Carnivorous snails probably the most important predatory

a

|
|
|
|
|

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 185

enemies; predacious beetles, mites, and dipterous parasites also
numbered among their foes. Bibliography which is abstracted in
the paper.

BisHop, SHERMAN C,

1921. The map turtle, Graptemys geographica (Le Sueur) in New York.
Copeia, no. 100, pp. 80-81, Nov. 15. Feeding on Unio complanatus.

CHURCHILL, E. P., AND LEwis, SARA I.

1924. Food and feeding in fresh-water mussels. Bull. U. S. Bur. Fisheries,
vol. 39, 1923-1924, pp. 439-471, figs. I-26. Protozoa, diatoms, other
algae, organic detritus. Bibliography on the food of fresh-water
mussels, and upon that of lamellibranchs in general.

Cooke, A. H.

1895. [Enemies of mollusca.] Cambridge Nat. Hist., vol. 3, pp. 56-62.
Birds, rats, frogs, toads, beetles, mongooses, monkeys, walruses,
whales, fishes, other mollusks, trematodes, nematodes, and mites.

Dyeue, L. L.

1903. Notes on the food habits of California sea-lions (Zalophus californi-
anus Lesson.) Trans. Acad. Sci. Kansas, 1901-1902, pp. 179-182.
Food found in numerous stomachs, chiefly squids.

FEDERIGHI, HENry.

1930. Control of the common oyster drill. Econ. Circ. 70, U. S. Bur. Fish-
eries, 7 pp., 5 figs. Urosalpinx cinerea “destroys oysters to the
value of several million dollars annually in the United States”
(p. FE).

FIELp, IrvinG A.

1911. The food value of sea mussels. Bull. U. S. Bur. Fisheries, vol. 20,
1909, pp. 85-128, pls. 18-25. Food (pp. 92-95), mostly diatoms
and Protozoa. Enemies (pp. 95-97), are numerous, fishes, mollusks,
sea-stars, crows, rats, parasitic crabs.

Forrest, H. E.

1927. Fishes, Caradoc and Severn Valley Field Club, record of bare facts
for the year 1926, p. 19. Stomach of an eel (Anguilla vulgaris)
from the Severn was full of small bivalves (Sphaerium corneum).

HERRINGTON, Wo. C.

1930. The Pismo clam. Fish. Bull. 18, California Div. Fish and Game,
69 pp. 16 figs. Tivela stultorum—birds, rays, starfish, and marine
snails are enemies (pp. 52-54).

Moore, H. F.

1908. Volumetric studies of the food and feeding of oysters. Bull. U. S.
Bur. Fisheries, 28, pp. 1297-1308, pl. 125, 6 figs. Ninety-five per
cent diatoms; remainder of equally minute plants and animals.

Ritcutr, J.

1927. A remarkable whale invasion. Scottish Nat. 1927, pp. 161-163. A
school of false killers (Pseudorca crassidens) visited the Dornoch
Firth in October, 1927, and some of them ran aground there.
This whale is very rare, and had hardly been seen anywhere for
80 years (a few appeared off Western Europe in 1861 and 1862,
and it was also seen in Tasmania). Examination of these whales
showed they feed mainly on large cuttlefish.

°

186 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

STEVENSON, CHARLES H.
1892. A bibliography of publications in the English language relative to
oysters and the oyster industries. Extract from Rep. U. S. Comm.
Fish and Fisheries for 1892, art. 3, pp. 305-359. Some of the papers
referred to deal with the food and enemies of oysters.
StiIves, CH. WARDELL.
1902. Frogs, toads, and carp (Cyprinus carpio) as eradicators of fluke
disease. Ann. Rep. U. S. Bur. Animal Industry 1go1, pp. 220-
222, figs. 197-203. By feeding on snails the intermediate hosts.

PISCES

Apams, CHAs. C., AND HANKINSoN, T. L.

1928. The ecology and economics of Oneida Lake fish. Roosevelt Wild
Life Ann., vol. 1, nos. 3-4, pp. 242-548, I pl., figs. 179-244, Nov.
Notes on food and enemies of most of the species; full bibliography.

Auiin, A. E. .

1929. Seining records and food of the intermediate stages of Lake Erie
fishes. Suppl. 18th Ann. Rep. New York Conserv. Dep. 1928, pp.
95-106. Cyprinidae and Catostomidae feed on algae and diatoms;
the smaller Percidae on crustaceans and insect larvae, and the
larger Percidae, Esocidae, and Catostomidae (fish eggs) on smaller
fishes and fish eggs.

ANNIN, J.

1902. In Rhead, Louis, The speckled brook trout, pp. 129-140. Winged
enemies include night heron, kingfisher, ducks, loons, grebes, fish
hawk, bald eagle and barred owl.

Baker, F. C.

1918. The productivity of invertebrate fish food on the bottom of Oneida
Lake, with special reference to mollusks. Techn. Publ. no. 9, New
York State Coll. Forestry, 264 pp. Notes on food of five species
of fishes (pp. 214-216). Bibliography.

Bargour, T.

1921. Spiders feeding on small cyprinodonts. Psyche, vol. 28, no. 4, pp.

131-132, Aug. Dolomedes tenebrosus.
BiceLow, N. K.

1924. The food of young suckers (Catostomus commersonu) in Lake
Nipigon. Univ. Toronto Studies, no. 24, Publ. Ontario Fish Res.
Lab., no. 21, pp. 81-115. Rotifers, Cladocera, insects.

Breper, C. M., Jr.

1921. The food of Mustelus canis (Mitchill) in mid-summer. Copeia, no.
101, pp. 85-86, Dec. 20. Notes on contents of 102 stomachs. (Fish
in 10, crabs 44, Nereis sp. 1, univalves and vegetable matter 3.)

1922. Observations on young bluefish. Copeia, no. 106, pp. 34-36, May 20.
Contents of 15 stomachs listed; 86 per cent fishes.

Breper, C. M., Jr., AND CrAwrorp, D. R.

1922. The food of certain minnows. A study of the seasonal dietary cycle
of six cyprinoids with especial reference to fish culture. Zoologica,
vol. 2, no. 14, pp. 287-327, figs. 1ro1-128, Aug. Semotilus bullaris,
87 per cent insects, including larvae and adults of several orders
plus some worms, millipeds, crayfish; Leuciscus vandoisulus, 98

7"

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 187

per cent insects together with worms, spiders, crayfish and slug;
Notropis procne, 36 per cent insects and in addition some worms
and water mites; Notropis cornutus and Rhinichthys atronasus, 57
per cent insects plus same additional items as in last; Exoglossum
maxillingua, 35 per cent insects, plus worms and fish eggs. Most
of the insects were terrestrial species.

Carr, A. M.

1908. Food of fishes. Rep. Sci. Invest. Northumberland Sea Fisheries
Comm. 10907, pp. 68-71. Reports on stomach examinations of 10
species.

1909. The food and condition of fish obtained from the North-east coast.
Rep. Sci. Invest. Northumberland Sea Fisheries Comm. 1908-1909,
pp. 41-50. Stomach analyses of seven species of fishes (pp. 43-46).

CHAMBERLAIN, F. M.

1907. Some observations on salmon and trout in Alaska. U. S. Bur. Fisher-
ies, Doc. 627, 112 pp., 5 pls. Enemies (pp. 107-109) include trout,
sculpins, mergansers, golden-eyes, mallards. The trout feed on
other fishes, insects and their larvae, snails, and bivalves.

Cote, Leon J.

1905. The German carp in the United States. Rep. U. S. Comm. Fisheries
1904, pp. 523-641, pls. 1-3. Considerable on food and economic
relations. Bibliography.

DERYKE, WILLIS.

1922. The food of the fishes of Winona Lake. Indiana Dep. Conserv., 47
pp., 1 pl, 1 map. Notes on 17 species, 6 of which are treated
in some detail; yellow perch: young, midge larvae, Entomostraca,
amphipods; adults, chiefly fish; bluegill: young, chiefly midge
larvae and Entomostraca; older, the same plus caddis larvae,
insects, snails; large-mouth black bass: young, amphipods, Clado-
cera, mayfly and midge larvae; larger, chiefly fish; log perch:
amphipods, Cladocera, midge, caddis, and mayfly larvae, snails;
skipjack: chiefly nonaquatic insects; sunfish: snails, midge larvae,
insects; hogmolly: midge, and mayfly larvae, oligochaetes, Bibli-
ography.

Eaton, E. H.

1928, The Finger Lakes fish problem. Suppl. 17th Ann. Rep. New York
Conserv. Dep. 1927, pp. 40-46. Tabulation of food of some 30
species, 3 of which are almost exclusive fish-eaters, 7 others
largely so, 8 feed chiefly on larval, and 5 on flying insects. Six
species eat many scuds, and 4 even as adults, subsist more or less
on plankton Crustacea. Enemies of fish, besides their own kind,
include lampreys, turtles, snakes, loons, grebes, and mergansers.

Etmurirst, RICHARD.

1926. Notes on fishes from the Firth of Clyde. Scottish Nat., pp. 151-158,
and 179-186. Full notes on food of cod and briefer reference to
that of some other species. Nine kinds of fishes (including itself)
listed as predatory on herring.

13

188 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Exrop, M. J.

1929. The fishes of Flathead Lake. Montana Wild Life, vol. 2, no. 1, pp.
6-9, June. Notes on food of: Catostomus spp.: Insects, Ento-
mostraca; Ptychocheilus oregonensis: Mainly insects such as may-
fly and caddisfly larvae, grasshoppers, some fish and shrimps;
Mylocheilus caurinus: FEntomostraca and insects; Leuciscus gillu:
Entomostraca and insects; Salmo clarkti: Beetles, mayflies, grass-
hoppers; Salvelinus malma: Fishes including Coregonus and
Ptychocheilus; Coregonus williamsoni: Larvae of Tipulidae,
Simuliidae, Planorbis, Physa; Micropterus salmoides: Fish, insects.

FIELD, Irvine A.

1907. Unutilized fishes and their relation to the fishing industry. U. S.
Bur. Fisheries Doc. no. 622, 50 pp. Notes on the food of eight
species.

Forses, S. A.

1880. The food of fishes. Bull. Illinois State Lab. Nat. Hist., vol. 1, no. 3,
pp. 18-65, Nov. Notes on stomach examinations for numerous
species.

1883. The food of the smaller fresh-water fishes. Bull. Illinois State Lab.
Nat. Hist., vol. 1, no. 6, pp. 65-94, May. Examination of 319
stomachs representing 25 species; food chiefly neuropteroid and
chironomid larvae, and Entomostraca; other animal items, fishes,
mollusks, Hymenoptera, Diptera, Coleoptera, Hemiptera, Thysan-
ura, Arachnida, amphipods, isopods, worms, and protozoans.

1890. Studies of the food of fresh-water fishes. Bull. Illinois State Lab.
Nat. Hist. vol. 2, pp. 433-473. Many stomach examinations of 28
species; tabulation of items and percentages.

1890. On the food relations of fresh-water fishes. Bull. Illinois State Lab.
Nat. Hist., vol. 2, pp. 475-538. Summary of the preceding papers,
discussion of fishes as predators on other fishes, on mollusks,
insects, crustaceans, worms, fresh-water sponges, and protozoans.
Schedule of food items and the species taking them.

Futon, T. WEMyYSS.

1903. The distribution, growth, and food of the angler (Lophius pisca-
torius.) 21st Ann. Rep. Fishery Board Scotland 1902, pp. 186-217.
Analyses of 280 stomach contents; 269 containing fishes, 10 squids,
and I a crab.

Gupcer, E. W.

1927. Hydras as enemies of young fishes. Nat. Hist., vol. 27, pp. 270-274,
3 figs.

1929. Wide-Gab, the angler fish. Nat. Hist., vol. 20, no. 2, pp. 155-159,
illus., Mar.-Apr. Case of attempt to swallow a gull; review of
literature, showing that birds up to the size of the loon are eaten;
seven wild ducks from one stomach; the principal food, however,
is fishes.

HANKINSON, THomAsS L.

1908. A biological survey of Walnut Lake, Michigan. Rep. Biol. Surv.
Michigan Geol. Surv. 1907, pp. 158-288, pls. 13-75. Food of several
species of fishes noted from examination of stomachs (pp. 200-216).

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 189

HarNELL, J.. AND Nayupo, M. R.

1924. A contribution to the life history of the Indian sardine, with notes
on the plankton of the Malabar Coast. Madras Fisheries Bull. 17,
pp. 129-197, 10 pls. Food extensively treated; consists of diatoms,
peridineans, infusorians, Heliozoa, larval bivalves, and copepods.

HILDEBRAND, S. F., AND Towers, I. L.

1927. Food of trout in Fish Lake, Utah. Ecology, vol. 8, no. 4, pp. 389-397,
Oct. Contents of 181 stomachs tabulated, the more important items
being Daphnia, Gammarus, midges and vegetation; leeches, snails,
dragonfly nymphs, and fishes and their eggs are other items of
the food.

JOHANSEN, FRITs.

1912. The fishes of the Datimark Expedition. Danmark-Ekspeditionen
Gronl. Nordostkyst 1906-1908, vol. 5, no. 12, pp. 633-675, 5 figs.,
pls. 44-46. Notes on the food of Gadus and Salmo.

JouHNson, Ropert S., AND Stapleton, M. F.

1917. Fish ponds on farms. App. 2, Rep. U. S. Comm. Fisheries 1915,
29 pp. Cannibalistic and other predacious fishes, turtles, snakes,
birds, and minks are the principal foes.

Jupay, CHANCEY.

1906. A study of Twin Lakes, Colorado, with especial consideration of
the food of the trouts. Bull. U. S. Bur. Fisheries, vol. 26, pp. 147-
178, pl. 3. In addition to notes on contents of 370 trout stomachs
of six species this publication contains a good bibliography and a
digest of papers relating to Entomostraca as food of fishes.

KENDALL, WILLIAM C.,

1897. Notes on the food of four species of the cod family. Rep. U. S.
Fish Comm. 1896, App. 3, pp. 177-186. A long list of food items.
“Protective mimicry seems of little avail against these fishes.”

KENDALL, WILLIAM C., AND DENcE, W. A.

1927. A trout survey of the Allegany State Park in 1922. Roosevelt Wild
Life Bull., vol. 4, no. 3, pp. 291-482, figs. 54-86, tables, July. Notes
on 112 stomach contents (pp. 472-474, table 27): Midges, caddis-
flies, beetles, ants and other Hymenoptera, Diptera, grasshoppers,
plant lice, lacewing flies, stoneflies, mayflies, spiders, crustaceans,
and fish. Bibliography.

Knicut, A. P.

1927. Losses in speckled trout fry after distribution. Science, n. s., vol. 65,
pp. 525-526, Aug. Losses 71-08 per cent, mostly to natural enemies,
birds, trout and other fishes.

KrAAtz, WALTER C.

1923. A study of the food of the minnow Campostoma anomalum. Ohio

Journ. Sci., vol. 33, pp. 265-283. Diatoms, algae, etc.
Lezour, Marie V.

1924. The food of young herring. Journ. Marine Biol, Assoc. United
Kingdom, n. s., vol. 13, pp. 325-330. Among animal items, Infusoria,
larval mollusks, copepods.

190 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Lewis, RALPH C.

1929. The food habits of the California sardine in relation to the seasonal
distribution of microplankton. Bull. Scripps Inst. Oceanography,
Techn. Ser. 2, pp. 155-180, 2 figs. Items of animal food are schizo-
pods and copepods.

Linton, EpwIn.

1901. Fish parasites collected at Woods Hole in 1808. Bull. U. S. Fish
Comm., vol. 19, 1899, pp. 267-304, pls. 33-43. Notes on fish food,
Pp. 270-284.

1901. Parasites of fishes of the Woods Hole region. Bull. U. S. Fish
Comm., vol. 19, 1890, pp. 405-492, pls. 1-34. Summary of parasites,
pp. 425-488, contains many references to food of fishes.

1921. Food of young winter flounders. Rep. U. S. Comm. Fisheries,
App. 4, 14 pp. Pseudopleuronectes americanus, food of young
principally amphipods, other small Crustacea, and annelids; food
of adults, annelids, Crustacea, ascidians, fish, mollusks. Almost
as much on parasites (Sporozoa, trematodes, nematodes, and
Acanthocephala) as on food.

MacCoy, CLinTon V.

1929. The mackerel in New England. Bull. Boston Soc. Nat. Hist., vol.
53, PP. 3-7, Oct. Food, small fish, squids, pteropods, amphipods,
copepods. Enemies, whales, porpoises, sharks, dogfish, bluefish,
gannets, parasitic worms.

MarsHALL, W. S., AND GILBER, N. C.

1905. Notes on the food and parasites of some fresh-water fishes from
the lakes at Madison, Wis. Rep. U. S. Comm. Fisheries 1904,
App., pp. 513-522. Incidental notes on food.

Moore, J. Percy.

1922. Use of fishes for control of mosquitoes in northern fresh waters of
the United States. Rep. U. S. Comm. Fisheries, App. 4, 60 pp.,
7 pls. Food of roach: Entomostraca, insects, mites, Protozoa;
mudminnow: Insects, Crustacea, mollusks, Protozoa; killifish:
Oligochaetes, mollusks, Entomostraca; top minnow: cannibalistic;
blue-spotted sunfish: Midge larvae, Entomostraca, amphipods;
long-eared sunfish: Midge larvae, Entomostraca, oligochaetes;
common sunfish: Midge larvae, Entomostraca, snails, mites, tad-
poles. All eat mosquito larvae. Bibliography.

MurrKkowskI, RicHArpD A.

1925. The food of trout in Yellowstone National Park. Roosevelt Wild
Life Bull., vol. 2, no. 4, pp. 471-497, figs. 114-133, Feb. Stoneflies,
90 per cent; mayflies, caddisflies, adults and young of all; and
water-trapped land insects. :

1929. The ecology of trout streams in Yellowstone National Park. Roose-
velt Wild Life Ann., vol. 2, no. 2, pp. 155-240, figs. 53-116, Oct.
Food of trout, pp. 222-230, conclusions as in his 1925 paper on the
subject. Food of insects, pp. 230-233; see under Muttkowski and
Smith.

NEEDHAM, JAMES G.

1903. Food of brook trout in Bone Pond. Bull. 68, New York State Mus.,

pp. 204-217. Contents of 25 stomachs tabulated.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE IQI

NEEDHAM, JAMES G.; JupAY, CHANCEY; Moore, EMMELINE; SIBLE, CuAs. K.;
AND TiTcoms, JOHN W.

1922. A biological survey of Lake George, N. Y. N. Y. State Conserv.
Comm., 78 pp., 27 figs. Much on the food of fishes; the staples of
the diet of carnivorous fry are waterfleas, midges (all stages),
other insects, scuds (amphipods), and crayfishes; cannibalism
prevalent (p. 63); food of adults of eight species outlined on pp.
65-68. Lake trout: Principal food, lake smelt, other items yellow
perch, and caddisflies; black bass: Perch, crawfish, grasshoppers,
scuds; pike: Other fishes; yellow perch: Staples, midge larvae,
mayfly nymphs, scuds, snails, secondary, caddisworms and water-
fleas; bullhead: Scuds, midge larvae, mayfly nymphs, snails; long-
eared sunfish: Mayfly nymphs, midge larvae, ants, scuds, water-
fleas, miscellaneous insects, and crayfishes ; common sunfish: Snails,
mayfly nymphs, caddisworms, beetles, midge larvae, various insects ;
rock bass: Crayfish, fishes, insects.

NEEDHAM, P. R.

1929. Quantitative studies of the fish food supply in selected areas. Suppl.
18th Ann. Rep. New York Conserv. Dep. 1928, pp. 220-232. Ithaca,
N. Y., Erie-Niagara watershed. Foods consumed by trout in com-
parison with available foods; in the case of aquatic foods the re-
lation of consumption to availability is very clear. This is a
reworking of a similar paper in the 17th Ann. Rep. (1927) 1928,
Pp. 192-206.

New York CoNnsERVATION DEPARTMENT.

1928. A biological survey of the Oswego River System. Suppl. 17th Ann.
Rep. New York Conserv. Dep. 1927, 248 pp., 12 col. pls., text figs.,
maps. Much on fish food; in a tabulation of food items of adults
of 31 species, midges, mayflies, and minnows seem to be most
commonly used; and of young of eight species, copepods, Cladocera,
and midges.

Pace, Wo. F.

1895. Feeding and rearing fishes, particularly trout, under domestication.
Bull. U. S. Fish Comm., 1894, pp. 280-314. Some notes on natural
food, and an indexed bibliography.

Patterson, A. H.

1926-1927. Food of the Sturgeon. Trans. Norfolk and Norwich Nat. Soc.,
vol. 12, pp. 380-381. Stomach of one contained about 729 small
fish (lesser sandlaunces).

Pearse, A. S.

1915. On the food of the small shore fishes in the waters near Madison,
Wisconsin. Bull. Wisconsin Nat. Hist. Soc., vol. 13, no. I, pp. 7-22,
1 fig., Mar. Sixteen species, of which nine lived largely on insects
and their larvae, two on ostracods, two on copepods, and one on
Cladocera.

1918. The food of the shore fishes of certain Wisconsin lakes. Bull. U. S.
Bur. Fisheries, vol. 35 (1915-16), pp. 249-292. Report on more
than 1,600 specimens of 32 species, with bibliography.

1919. Habits of the black crappie in inland lakes of Wisconsin. Rep. U. S.
Comm. Fisheries 1918, app. 3, pp. 5-16. Tabulation of contents of
276 stomachs.

Ig2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

1921. Distribution and food of the fishes of Green Lake, Wis., in summer.
Bull. U. S. Bur. Fisheries, vol. 37, 1919-1920, pp. 255-272, I map.
Notes on 16 species; the food of all combined comprised insect
larvae 21.7 per cent, amphipods 16.5 per cent, fish 9.6 per cent, |
crayfishes 7.8 per cent, cladocerans 7.6 per cent, insect pupae 6.7 |
per cent, snails 4.4 per cent, bivalves 4.1 per cent, and the following |i
items in smaller proportions, adult insects, ostracods, oligochaetes, |
leeches, mites, Mysts, and copepods. Sixty-seven per cent is arthro- |
pods, composed of 31.7 insects and 35.6 crustaceans. Comparison |
is made with the fishes of Lake Mendota. Bibliography. |

1924. Amount of food eaten by four species of fresh-water fishes. Ecology, |
vol. 5, no. 3, pp. 254-258, July. Order of choice, minnows, earth- |
worms, amphipods, dragonfly nymphs, crayfishes, grasshoppers,
snails, and caddis larvae.

PearsE, A. S., AND ACHTENBERG, HENRIETTA.

1917-1918. Habits of yellow perch in Wisconsin Lakes. Bull. U. S. Bur. |
Fisheries, vol. 36, 1917-1918, pp. 297-366, pl. 83, figs. 1-35. Report |
on 1,147 stomach examinations of which the food as a whole was |
made up of 38.3 per cent insect larvae, 21.4 Entomostraca, 9.5 insect |
pupae and adults, 5.5 macroscopic crustaceans, 4.5 fishes, 2.4 mol- |
lusks, 1.4 oligochaetes, leeches and arachnids. Enemies of the perch

include pickerel, black bass, a number of birds, and a variety of |
parasites. Bibliography. |
PEARSON, JOHN C. |

1928. Natural history and conservation of the redfish and other commercial
sciaenids of the Texas coast. Bull. U. S. Bur. Fisheries, vol. 44,
pp. 129-214, 44 figs. Sciaenops ocellatus: Shrimps, crabs, mollusks,
fish; Pogonias cromis: Clams, mussels, oysters, crabs, shrimps,
fish, annelids ; Cynoscion nebulosus: Shrimps, crabs, fish; Micropo-
gon undulatus: Shrimps, crabs, annelids, fish.

Peck, JAMEs I.

1894. On the food of the menhaden. Bull. U. S. Fish Comm., vol. 13, 1893,
pp. 113-126, pls. 1-8. Food filtered from water by gill-raker
mechanism, consists chiefly of unicellular organisms, both animal
and vegetable. They also take ostracods, copepods, amphipods and
other small Crustacea, and young Nerets. Composition of food the
same as material filtered from water by mechanical contrivances:
Diatoms, rotifers, dinoflagellates, etc. The supply of such food
illimitable.

1806. The sources of marine food. Bull. U. S. Fish Comm., 1895, pp. 351-
308, pls. 64-71. Plankton, largely diatoms, the basis; notes on
the food of the squeteague, the bluefish, sea bass, scup, and tautog.

PETERSEN, C. G. J.

1894. On the biology of our flat-fishes. Rep. Danish Biol. Sta., vol. 4,
1893, pp. v-+146,2 pls., I map, 18 tables. Notes on food of young
and adults.

Scott, ANDREW.

1899. Observations on the habits and food of young fishes. Proc. & Trans.

Liverpool Biol. Soc., vol. 13, 1898-99, pp. 90-93.

INOS 7 PROTECTIVE ADAPTATIONS—McATEE 193

Scott, THOMAS.

1902. Observations on the food of fishes. 20th Ann. Rep. Fishery Board
Scotland 1901, pt. 3, pp. 486-538. Notes on 56 species.

1903. Some further observations on the food of fishes, with a note on the
food observed in the stomach of a common porpoise. 21st Ann.
Rep. Fishery Board Scotland 1902, pp. 218-227, 2 figs.

S1ptey, C. K.

1929. The food of certain fishes of the Lake Erie Drainage Basin. Suppl.
18th Ann. Rep. New York Conserv. Dep. 1928, pp. 180-188.
Thirty-four species feed mainly on immature aquatic insects, es-
pecially midge larvae, and Crustacea; eight species are pronounced
spawn-eaters; small fish are important food of the larger species;
food of young chiefly copepods and Cladocera.

SMALLWoop, W. M., AND STRUTHERS, P. H.

1927. Carp control studies in Oneida Lake. Suppl. 17th Ann. Rep., New
York Conserv. Dep., pp. 67-83. Much on food; animal matter taken
by adults includes fish, ostracods, phyllopods, copepods, crayfish,
midge and caddis larvae and other insects; by young, ostracods,
copepods, Cladocera, insect larvae, snails, worms, mites, eggs of
snails, insects and copepods, rotifers, and bivalves.

SmitH, HucH M.

1896. A review of the history and results of the attempts to acclimatize
fish and other water animals in the Pacific States. Bull. U. S.
Fish Comm., vol. 15, 1805, pp. 379-472, pls. 73-83. Notes on food
of a few species. Catfish, fish eggs and fry; carp, spawn; shad,
shrimps; striped bass, carp, catfish, crabs.

Smitu, W. Ramsay.
1889. On the food of fishes. 7th Ann. Rep. Fishery Board Scotland 1888,

pp. 222-258.

1890. On the food of fishes. 8th Ann. Rep. Fishery Board Scotland 1880,
pp. 230-256.

1891. On the food of fishes. 9th Ann. Rep., Fishery Board Scotland 1890,
pp. 222-242.

1892. On the food of fishes. toth Ann. Rep. Fishery Board Scotland 1801,
pp. 211-231. This and similar papers in three previous reports are
based on investigations of Thomas Scott.

STEWART, N. H.

1926. Development, growth, and food habits of the white sucker, Catosto-
mus commersoni Lesueur. Bull. U. S. Bur. Fisheries, vol. 42, pp.
147-184, 55 figs. Among animal food midge larvae are most impor-
tant at all ages; some rotifers, Entomostraca, and Protozoa are taken
at all stages also, but dragonfly, caddisfly, mayfly larvae, and
Mollusca are taken only by adults. Bibliography.

SrruTHERS, P. H.

1929. Carp control studies in the Erie Canal. Suppl. 18th Ann. Rep. New
York Conserv. Dep. 1928, pp. 208-219. Animal food (p. 214) in-
cludes insect larvae, snails, midge larvae, bivalves, ostracods,
Malacostraca, copepods, Cladocera, and decapods,

194 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

SUMNER, Francis B.; Ospurn, RAyMonp C.; AND Core, Leon J.

1911. A biological survey of the waters of Woods Hole and vicinity.
Bull. U. S. Bur. Fisheries, vol. 31, pt. 2. The catalogue of the
marine fauna, fishes, pp. 734-744, contains notes on the food
mainly quoted from Verrill, Goode, Linton, and Field.

TAVERNER, P. A.

1915. The double-crested cormorant (Phalacrocorax auritus) and its re-
lation to the salmon industries on the Gulf of St. Lawrence. Can.
Geol. Surv., Bull. 13, 24 pp., 1 pl. Food sculpins, herring, capelin,
eel, etc., no salmon; the salmon feeds on other fishes, and crusta-
ceans, and is cannibalistic.

TIFFANY, Lewis H.

1921. Algal food of the young gizzard shad. Ohio. Journ. Sci., vol. 21,
no. 4, pp. 113-122, Feb. Mentions several game fishes that prey
on this wholly vegetarian species.

TURNER, CLARENCE L.

1920. Distribution, food and fish associates of young perch in the Bass
Island region of Lake Erie. Ohio Journ. Sci., vol. 20, no. 5, pp.
137-152, Mar. Details of analyses of 138 stomach contents.

1921. Food of the common Ohio darters. Ohio Journ. Sci., vol. 22, pp.
41-62. Usually the food changes with age from Entomostraca to
midge larvae and similar organisms, and then with maturity, to
a varied diet in which ephemerid and other large insect larvae
predominate. |

1922. Notes on the food habits of young Cottus ictalops (miller’s thumb).
Ohio Journ. Sci., vol. 22, pp. 95-96. Midge and other insect larvae.

VerrILL, A. E.

1873. Report upon the invertebrate animals of Vineyard Sound and the
adjacent waters, with an account of the physical characters of the
region. Report on Sea Fisheries of New England, pt. 1, pp. 295-
778. Lists of species found in the stomach of fishes (pp. 514-521).

Warren, B. H.

1897. Fish-eating birds and mammals. Ann. Rep. Pennsylvania Dep. Agr.,
1896, pp. 297-303, I pl. Seventeen or more kinds of birds, wild
cats, raccoons, muskrats, mink, and the otter.

WELsH, WM., AND Breper, C. M., Jr.

1923-1924. Contributions to life histories of Sciaenidae of the eastern United
States coast. Bull. U. S. Bur. Fisheries, vol. 39, pp. 141-201, 60
figs. Notes on food of eight species; it is chiefly crustaceans, next
in order coming worms and fishes. Bibliography. Cynoscion re-
galis: Shrimps, schizopods, isopods, amphipods, worms when small,
fish when mature, but including shrimps and squids; Bairdiella
chrysura: Schizopods, isopods, amphipods, worms, fish; Stellifer
lanceolatus: Schizopods, copepods, decapods, ostracods, amphipods,
worms; Leiostomus xanthurus: Ostracods, copepods, amphipods,
worms, mollusks; Micropogon undulatus: Shrimps, echinoderms,
worms, mollusks, copepods, ostracods, amphipods; Menticirrhus
americanus: Crabs, shrimps, worms, fish; Menticirrhus saxatilis:
Shrimps, amphipods, schizopods, worms, fish; Pogonias cromis:
Mollusks including oysters; Eques pulcher: Small crustaceans.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 195

WICKLIFF, Epwarp L.

1920. Food of young small-mouth black bass in Lake Erie. Proc. Amer.
Fisheries Soc., pp. 364-371. Report on 313 specimens, the most
important items being copepods found in 61 per cent of the stomachs
and Cladocera in 39 per cent. Other commonly taken foods were
midge larvae and pupae, adult insects, fish, and mayfly nymphs.

AMPHIBIA
DRAKE, Caru J.

1914. The food of Rana pipiens Schreber. Ohio Naturalist, vol. 14, no. 5,
pp. 257-269, Mar. Detailed account of the contents of 209 stomachs
collected at Cedar Point, Ohio.

Frost, S. W.

1924. Frogs as insect collectors. Journ. New York Ent. Soc., vol. 32,
no. 4, pp. 174-185, pl. 14, Dec. Eat worms, snails, crayfishes,
spiders, mites, insects and frogs; insects most important. Larvae:
Lepidoptera 9; Coleoptera 24; Diptera 13; Neuroptera 1. Adults:
Orthoptera 1; Hemiptera 25; Neuroptera 3; Mecaptera 1; Diptera
33; Coleoptera 242; spiders 37; pseudoscorpions I.

GARMAN, H.

1901. The food of the toad. Bull. 91, Kentucky Agr. Exp. Sta., pp. 60-68,

fig. 16. Report on 20 stomach contents.
HAmILton, W. J., JR.

1930. Notes on the food of the American toad. Copeia, 1930, no. 2, June 30,
p. 45. Bufo americanus. Report on food of 400 young toads:
Diptera 22 per cent, mostly larvae; mites 15.5 per cent; ants
12.8; beetles and their larvae 11.8, the most abundant group being
Staphylinidae; thrips 10.1; Collembola 6.2; Lepidoptera, Hymen-
optera, aphids, sowbugs, spiders, worms, and snails, the remainder.

KIRKLAND, A. H.

1904. Usefulness of the American toad. Farmers’ Bull. no. 196, U. S. Dep.

Agr., 16 pp. Contents of 149 stomachs discussed.
Kiucu, A. Brooker.

1922. The economic value of the leopard frog. Copeia, no. 103, pp. 14-15,
Feb. 15. Contents of 25 stomachs; chiefly Melanoplus femur-rubrum
and Leptinotarsa 1o-lineata.

Muwnz, Pur A.

1920. A study of the food habits of the Ithacan species of Anura during
transformation. Pomona Coll. Journ. Ent. Zool., vol. 12, no. 2,
pp. 33-56, June. Report on 586 stomachs of eight species; sum-
maries of results of previous investigators.

SMALLWoop, W. M.

1928. Notes on the food of some Onondaga Urodela. Copeia, no. 160,
pp. 89-98, Oct. 25. Ambystoma maculatum: Centipeds, earth-
worms, snails, sowbugs, crickets, grasshoppers, beetles; Plethodon
cinereus: Centipeds, earthworms, snails, sowbugs, ants, beetles,
mites, spiders, phalangids, caterpillars, grasshoppers, flies spring-
tails; Eurycea bislineata: Earthworms, caterpillars, and beetle,
fly, and caddisfly larvae; Triturus viridescens: Snails, water-
boatmen, fish, earthworms, beetle larvae; bivalves, daphnia, cater-

196 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

pillars, amphibian eggs (including its own), water bugs, mosquito
and other fly larvae, slugs, snails, leeches, spiders, springtails,
beetles, mites.
SurFAceE, H. A., [Ep.]
1913. First report on the economic features of the amphibians of Pennsyl-
vania. Zool. Bull. Div. Zool. Pennsylvania Dep. Agr., vol. 3, nos.
3-4, pp. 67-152, figs. I-25, pls. 1-11, May-July. General discussion
of the subject, including report on examination of stomachs of 14
species of salamanders, two of toads, and nine of frogs.
WricutT, A. H.
1920. Frogs: Their natural history and utilization. App. 6, Rep. U. S. |
Comm. Fisheries 1919, 44 pp. Notes on the food of various species,
pp. 38-42. Enemies, pp. 42-44; invertebrates, fishes, amphibians, |

reptiles, birds, and mammals discussed.
WricHt, A. H., AnD Haaser, Juria M.
1922. The carnivorous habits of the purple salamander. Copeia, no. 105,
pp. 31-32, April 15. Feed on aquatic insects; in captivity take
frogs and salamanders.

Burt, Cuas. E.

1928. Insect food of Kansas lizards with notes on feeding habits. Journ.
Kansas Ent. Soc., vol. I, no. 3, pp. 50-68, July. Notes on stomach
contents of seven species with compiled information on others. Of
the total food in all lizard stomachs examined 51.92 per cent was
Orthoptera, 11.65 Lepidoptera, 9.35 Arachnida, 8.900 Hymenoptera,
and 6.00 Coleoptera; Diptera, Hemiptera, Trichoptera, and Mol-
lusca in smaller amounts.

KELLocG, REMINGTON.

1929. The habits and economic importance of alligators. Techn. Bull. 147,
U. S. Dep. Agr., 36 pp., 2 pls. Dec. Food (pp. 21-32), nearly
half is crabs, crawfishes, and shrimps; spiders, insects of various
orders, toads, smaller alligators, lizards, turtles, snakes, birds and
mammals also eaten.

LyYDEKKER, R.; CUNNINGHAM, J. T.; BoULENGER, G. A.; AND THomson, J. A.

1912. Food and growth [of reptiles], reptiles, Amphibia, fishes, and lower
Chordata, pp. 47-61, London. ‘“ The food of reptiles is very vari-
ous,” a dictum which shows distribution of predation is as charac-
teristic of this phylum as of others. Details in many cases.

PACK Elead):

1921. Food habits of Sceloporus graciosus graciosus (Baird and Girard).
Proc. Biol. Soc. Washington, vol. 34, pp. 63-66, Mar. Report on
the contents of 71 stomachs.

1922. Food habits of Crotaphytus wislizenit Baird and Girard. Proc. Biol.
Soc. Washington, vol. 35, pp. 1-3, Mar. 20. Report on the con-
tents of 18 stomachs.

1923. Food habits of Callisaurus ventralis ventralis (Hallowell). Proc.
Biol. Soc. Washington, vol. 36, pp. 79-81, Mar. Twenty stomachs;
disclosing caterpillars, coccinellids, meloids, erotylids, chrysomelids,
weevils, grasshoppers, mantids, Hemiptera, ant-lions, Diptera, and
spiders.

(
REPTILIA |
{

|

INO: 7. PROTECTIVE ADAPTATIONS—McATEE 197

1923. Food habits of Crotaphytus collaris baileyi (Stejneger). Proc. Biol.
Soc. Washington vol. 36, pp. 83-84, Mar. Report on 16 stomach
examinations; Orthoptera the principal animal food, caterpillars,
wasps, bugs, leaf-hoppers, and ant-lions also being taken.

1923. The food habits of Cnemidophorus tessellatus tessellatus (Say). Proc.
Biol. Soc. Washington, vol. 36, pp. 85-80, Mar. Sixty-three
stomachs containing caterpillars, 37.7 per cent of the food, grass-
hoppers 14.4; beetles 14.2; other insects 14.27; and arachnids 8.2.

Surrace, H. A. [Ep.]

1906. The serpents of Pennsylvania. Monthly Bull. Div. Zool. Pennsyl-
vania Dep. Agr., vol. 4, nos. 4-5, pp. 115-202, pls. 14-52, figs. 5-23,
Aug.-Sept. Includes data on contents of stomachs of snakes ot
14 species.

1907. The lizards of Pennsylvania. Bull. Div. Zool. Pennsylvania Dep.
Agr., vol. 5, no. 8, pp. 235-258, pls. 30-33, figs. 26-28, Dec. 1.
Notes on food of five species, in the case of two of them based on
examinations of stomachs.

1908. First report on the economic features of turtles of Pennsylvania.
Bull. Div. Zool. Pennsylvania Dep. Agr., vol. 6, nos. 4-5, pp. 107-
195, pls. 4-12, 16 figs., Aug.-Sept. Includes report on stomach
contents of representatives of nine species.

WIntTon, W. M. .

1915. A preliminary note on the food habits and distribution of the Texas
horned lizards. Science, n. s., vol. 41, pp. 707-8, May 28. Brief
summary of the results of examination of 485 stomachs; agricul-
tural ants found in 80 per cent and stink bugs in 60 per cent of
the stomachs.

Wricut, A. H.; FUNKHousErR, W. D.; AND BrsHop, S. C.

1915. A biological reconnaissance of the Okefinokee Swamp in Georgia.
Turtles, lizards, and alligators, by Wright and Funkhouser, pp.
108-139; snakes by Wright and Bishop, pp. 139-192. Proc. Acad.
Nat. Sci. Philadelphia, pp. 107-192, pls. I-III, figs. 1-14, Mar.
(Apr.). Notes on food of many of the species.

AVES

The entries under food of birds are chiefly general papers in which bibliogra-
phies introductory to the very extensive literature of the subject can be
found.

CLELAND, J. B.

1922. The parasites of Australian birds. Trans. and Proc. Roy. Soc. South
Australia, vol. 46, pp. 85-118. Cestodes in 86 species, adult nema-
todes in 28, Microfilaria in 34, Acanthocephala in 25, trematodes
in 38, fleas on 3, Hippoboscidae on 4, Mallophaga on 107, ticks
on 4, mites on 38, Haemosporidia in 47, and haemoflagellates in 12.

Cram, ELoIse B.

1927. Bird parasites of the nematode suborders Strongylata, Ascaridata,
and Spirurata. U. S. Nat. Mus. Bull. 140, 465 pp., 444 figs. About
500 species.

198 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

ForsusH, E. H.

1904. Special report on the decrease of certain birds, and its causes with
suggestions for bird protection. 52nd Ann. Rep. Massachusetts
State Board Agr., pp. 420-543, 2 pls. Chief causes, gunners, trap-
pers, egg collectors, destruction of environment, natural enemies,
and the elements.

1907. Useful birds and their protection. Massachusetts State Board Agr.,
437 pp., 56 pls., 171 figs. Capacity of birds for destroying pests,
birds as enemies of insects, and mammals, hairy caterpillars, plant
lice, also on natural checks upon bird life.

1916. The natural enemies of birds. Econ. Biol. Bull. 3, Massachusetts
State Board Agr., 58 pp., 7 pls., figs. A thorough review of the
subject, treating enemies among domesticated animals and among
wild mammals, birds, reptiles, amphibians, fishes, and insects.

Gross, A. O.

1928. The heath hen, pp. 525-526. Marsh hawk, Cooper’s hawk, sharp-
shinned hawk, and goshawk, the snowy owl, and crows enemies
of this species. Domestic cat the worst.

HENDERSON, JUNIUS.

1927. The practical value of birds, 342 pp. An exhaustive review of litera-
ture on the economics of American birds, with a long bibliography.
Chapters on birds as enemies of injurious insects, mammals, and
plants; birds as scavengers, and on the destruction of birds.

Hersey, L. J.

1907. A naturalist’s notes on birds and snakes. Outdoor Life, pp. 481-483,

Nov. Snakes eating birds and their eggs.
Lewis, ELIsHA J.

1857. [Enemies of the partridge]. The American sportsman, 3rd ed.,
Enemies of the partridge (pp. 102-4) : fox raccoon, weasel, polecat,
serpent, hawk, crow (p. 102); sparrow hawk, pigeon hawk,
goshawk (p. 103).

1857. [Enemies of the wild turkey]. The American sportsman, 3rd ed.
Wolf, fox, lynx, cougar, opossum, and wild cat. Also the larger
hawks and owls (p. 141).

1857. [Enemies of the ruffed grouse]. The American sportsman, 3rd ed.
Polecats, weasels, raccoons, opossums, foxes, crows, larger hawks
(p. 150).

Linton, E. F

1927. Notes on cestode parasites of birds. Proc. U. S. Nat. Mus., vol. 70,
art. 7, 73 pp. 15 pls. Thirty-four species.

1928. Notes on trematode parasites of birds. Proc. U. S. Nat. Mus., vol.
73, 30 pp., 11 pls. Twenty-two species.

LoncstaFF, T. G.

1927. Bird’s-nesting mice and insects. British birds, vol. 20, pp. 198-1090.
Notes certain insects (e.g. ants) attacking nestlings, and mice
destroying eggs.

McAter, W. L.

1913. Index to papers relating to the food of birds by members of the
Biological Survey in publications of the United States Department
of Agriculture, 1885-1911. U. S. Biol. Surv. Bull. 43, 1913, 69 pp.
Bibliography with subject index.

INOS) 7, PROTECTIVE ADAPTATIONS—McATEE 199

PuiatH, O. E.

1919. Parasitism of nestling birds by fly larvae. The Condor, vol. 21, pp.
30-38. Protocalliphora azurea in 39 out of 63 nests; parasites and
scavengers on this dipteron.

Ransom, B. H.

1909. The taenioid cestodes of North American birds. U. S. Nat. Mus.

Bull. 60, 141 pp., 42 figs. About 140 species; bibliography.
RusseEL, J. F.

1926. Predatory bass. Outdoor Life, vol. 57, no. 2, pp. 146-147, Feb. Black

bass with swallow in its stomach. San Diego Co., Calif.
Tucker, B. W.

1926. Bird’s-nesting bank voles. British birds, vol. 20, pp. 158-160. Evi-
dence was given that bank voles destroy birds’ eggs. This note
was followed by a number of other communications by various
authors, in the same journal (vol. 20, pp. 180-181, 198-199, 207,
230, 255), which showed that various species of mice commonly
attack birds’ eggs.

Weep, C. M., AND DEARBORN, NED.

1903. Birds in their relations to man, 380 pp., illus. Extensive chronological
bibliography; chapters on birds as regulators of outbreaks of in-
jurious animals, relations of birds to predacious and _ parasitic
insects.

Wairp, OF A.
1927. Wasps destroying young birds. British birds, vol. 20, pp. 254-255.

MAMMALIA
Bascock, H. L.

1914. Some observations on the food habits of the short-tailed shrew
(Blarina brevicauda). Science, n. s., vol. 40, pp. 526-530, Oct. 0.
Review of literature, chiefly about observations on captive animals.

BaILey, VERNON, AND SPERRY, CHAS. C.

1929. Life history and habits of grasshopper mice, genus Onychomys.
Techn. Bull. 145, U. S. Dep. Agr., I9 pp., 4 pls., Nov. Animal
food (pp. 10-19), nearly 90 per cent of the whole, largely grass-
hoppers, crickets, caterpillars, moths, and beetles; insects of other
orders, spiders, and mice also taken.

Brooks, Frep E.

1908. Notes on the habits of mice, moles, and shrews. Bull. 113, West
Virginia Agr. Exp. Sta., pp. 89-133, 10 pls., 2 figs., Jan. Con-
siderable on food; review of previous writings.

Bruce, JAY.

1925. The problem of mountain lion control in California. California Fish
and Game, vol. 2, no. I, pp. I-17, figs. 1-5, Jan. Each mountain-
lion costs the State $1,000 a year in deer meat, or about $15,000
to maintain the animal during its natural existence.

CrIDDLE, NORMAN.

1917. Varying hares of the prairie provinces. Agr. Gaz. Canada, vol. 4,
no. 4, p. 262, Apr. Goshawk, golden and bald eagles, and great
horned owls serious enemies.

200 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

Drxon, JosEPH.

1925. Food predilections of predatory and fur-bearing mammals. Journ.
Mamm., vol. 6, no. I, pp. 34-46, pl. 4, Feb. Wild cat: Mammals,
birds, fish; coyote: Game, stock, rodents, insects, mammals, birds;
mountain-lion: Deer, stock, small wild mammals; skunks: Insects,
rodents, birds, mammals.

DYCHE We. le:

1903. Food habits of the common garden mole (Scalops aquaticus ma-
chrinus Rafinesque). Trans. Acad. Sci. Kansas 1901-1902, pp. 183-
186. Report on the stomach contents of 50 specimens.

ForsusH, E. H.

1916. The domestic cat. Bird killer, mouser, and destroyer of wild life.
Means of utilizing and controlling it. Econ. Biol. Bull. 2, Massa-
chusetts State Board Agr., 112 pp., 20 pls., figs. The most compre-
hensive review of the subject; cats kill millions of birds annually;
destructive also to moles, shrews, toads, field mice, wood mice,
insects.

GARMAN, H.

1895. The food of the common mole. 7th Ann. Rep. Kentucky Agr. Exp.

Sta. 1894, pp. xli-xlv. Notes on contents of 14 stomachs.
Hamitton, W. J., Jr.

1930. The food of the Soricidae. Journ. Mamm., vol. 11, no. I, pp. 26-39,
Feb. Over 300 stomachs representing four species; food is insects,
annelids, Crustacea, snails, mice, salamanders, arachnids, centipeds,
and millipeds. Bibliography.

JOHANSEN, FRITs.

1910. Observations on seals (Pinnipedia) and whales (Cetaceae) made on
the “Danmark Expedition” 1906-1908. Danmark Eksp. Gronl.
Nordéstkyst, 1906-1908, vol. 5, no. 2, pp. 203-224, 9 figs. Includes
some notes on food.

Jounson, Cuas. E.

1925. The muskrat in New York; its natural history and economics.
Roosevelt Wild Life Bull., vol. 3, no. 2, pp. 205-320, pl. 5, figs. 48-
87, Mar. Animal food includes bivalves, snails, crayfish, insects,
fishes, turtles, and birds; enemies include minks, foxes, weasels,
otters, hawks, and owls.

Lantz, D. E.

1905. Kansas mammals in their relations to agriculture. Bull. 129, Kansas
Agr. Exp. Sta., Dec., 1904, pp. 331-404, I pl., 1 fig. Notes on the
food habits of most of the groups.

1906. Meadow mice in relation to agriculture and horticulture. U. S. Dep.
Agr. Yearbook 1905, pp. 363-376, pls. 38-41, fig. 89. Natural
enemies (pp. 370-373) include wolves, lynxes, foxes, badgers,
raccoons, opossums, skunks, minks, weasels, shrews, hawks, owls,
crows, shrikes, cranes, herons, bitterns, snakes, and domestic cats
and dogs.

1918. The house rat the most destructive animal in the world. U. S. Dep.
Agr. Yearbook 10917, pp. 235-251, pls. 41-44. Natural enemies (pp.
248-249) include domestic dog, cat, and ferret, as well as snakes,
storks, herons, owls, hawks, skunks, weasels.

NO. 7 PROTECTIVE ADAPTATIONS—McATEE 201

1923. Economic value of North American skunks. Farmers’ Bull. 587,
U. S. Dep. Agr., 24 pp., 10 figs. Food (pp. 9-14), poultry, game,
mice, and armyworms, tobacco worms, whitegrubs, hop grubs,
grasshoppers, potato beetles and other insects.

L.[ucas], F. A.

1905. The Newfoundland whale fisheries. Science, n. s., vol. 21, p. 713,
May 5. Large whales feed almost exclusively on Euphausia; fin-
backs upon caplin.

Piper, S. E.

1909. Mouse plagues, their control and prevention. U. S. Dep. Agr. Year-
book 1908, pp. 301-310, pls. 21-25. During a plague near Humboldt
Lake, Nevada, 2,000 predatory birds and 1,000 mammals put in
their appearance and together consumed about 1,350,000 mice per
month.

1928. The mouse infestation of Buena Vista Lake Basin, Kern County,
California, September, 1926, to February, 1927. Monthly Bull.
California Dep. Agr., vol. 17, no. 10, pp. 538-560, figs. 91-102, Oct.
Ring-billed gulls, short-eared owls, barn owls, various hawks,
ravens, great blue herons, road-runners, shrikes, coyotes, skunks,
and house cats noted as predators (pp. 550-552).

RAtnzsow, W. J.

1913. Food, medicines, and charms of savage man. Abstract in Rep.
Trustees Australian Mus. 1913, p. 9. Humans feeding on spiders,
beetle larvae, caterpillars, grasshoppers, ants, bees, wasps, termites,
and scorpions.

SCHEFFER, THEO. H.

1910. The common mole. Bull. 168, Kansas Agr. Exp. Sta., 36 pp., figs.
Natural enemies (pp. 20-21) include hawks, owls, coyotes, domestic
dogs. On the whole has few foes.

1927. American moles as agricultural pests and as fur producers. Farmers’
Bull. 1247, U. S. Dep. Agr., 20 pp., 18 figs. Animal food (pp. 7-8),
earthworms, beetles and their larvae, spiders, centipeds, ants,
caterpillars.

West, JAMES A.

1910. A study of the food of moles in Illinois. Bull. Illinois State Lab.
Nat. Hist., vol. 9, pp. 14-22, Oct. Details of contents of 56
stomachs; references to previous literature.

SMITHSONIAN MISCELLANEOUS COLLECTIONS
VOLUME 85, NUMBER 8

MODERN SQUARE GROUNDS OF THE
CREEK INDIANS

(WITH Five PLATES)

BY
JOHN R. SWANTON

Ethnologist, Bureau of American Ethnology

(PUBLICATION 3126)

| CITY OF WASHINGTON
| PUBLISHED BY THE SMITHSONIAN INSTITUTION
NOVEMBER 11, 1931

The Lord Battimore Press

BALTIMORE, MD., U. 8. As

MODERN SQUARE GROUNDS OF THE CREEK INDIANS
By JOHN R. SWANTON,

ETHNOLOGIST, BUREAU OF AMERICAN ETHNOLOGY
(WirH Five PLateEs)

The writer has already published descriptions of many of the
square grounds of the Creek Indians, the sacred areas where their
busks and other annual ceremonies took place." In collecting this
material, however, my endeavor was to learn the most ancient
arrangement of the several grounds and the arrangement of those
grounds no longer in use. During the summer of 1929 I visited the
Creek country again to secure information regarding the organization
of the extant grounds. This work duplicates and supplements the
earlier to a considerable extent, but the main purpose was somewhat
different.

Besides the three Yuchi grounds, now in the vicinity of Kellyville,
Bixby, and Depew, respectively, with which I did not concern myself,
there are, or were in 1929, 17 square grounds, as follows: Abihka
and Otciapofa near Henryetta ; Nuyaka north of Okemah ; Latogalga °
or Fish Pond and Asilanabi west of Okemah; two Tulsa grounds
near Holdenville; Tukabahchee at Yeager; Laptako near Wetumka ;
Alabama east of Alabama Station on the Frisco Railroad; Eufaula
west of Eufaula; Kasihta east of Okmulgee; and Hilibi, Kealedji,
Okchai, Pakan Tallahassee, and Wiogufki about Hanna. Abihka,
Otciapofa, Nuyaka, Latogalga, Kasihta, Hilibi, Pakan Tallahassee,
and Wiogufki were visited, and new information obtained regarding
all of the others except Eufaula, of which I secured very good
descriptions 17 years before. The Eufaula ground, that of Asilanabi,
one of the Tulsa grounds, Tukabahchee, Alabama, and Okchai were
visited in the winter of tg11-1912. Kasihta is the only square ground
representing the Lower Creeks now maintained. It was not in existence
during my earlier work in the Creek Nation, nor were Laplako or
Kealedji... These two last and the Yuchi grounds are the only ones
that I have not seen. I was present at part of the busks at Otciapofa,
Nuyaka, and Pakan Tallahassee. Not much attention was devoted

*42d Ann. Rep., Bur. Amer. Ethnol.. 1924-25, pp. 204-206, 1028.

*In the present paper, a indicates the obscure a in such a word as ability, and
t or £. is a surd 1 approximating thl in English.

*But see p. 35 regarding the former.

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 8

Zz SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to the Seminole grounds, but information was obtained regarding one
of these, Ochesee Seminole, which had been discontinued in 1912,
when I visited the Seminole squares, but was afterwards revived.

In order to make the material obtained in 1929 intelligible, it will
be necessary to give a brief outline of the Creek political, social, and
ceremonial organization.

The name Creek is a shortened form of Ochesee Creek Indians,
a name which English traders from South Carolina came to apply
to that part of the Indians of the Creek Confederation who were
living upon Ocmulgee River in the closing decades of the seventeenth
century and the opening years of the eighteenth. The word Ochesee
signifies “people of a different speech” in the language of the
Hitchiti, one of the minor constituents of the Creek Confederacy,
being equivalent to the word Tciloki in the Creek or Muskogee
language. It was applied to the Creeks proper or Muskogee by the
Hitchiti along with many other tribes, but came in some way to be
particularly associated with the Muskogee and the river upon which
they were then living.

Anciently there seems not to have been a single term applicable
to all of the Muskogee, the latter name having been unknown to
the Spaniards who first entered this section. It does not make its
appearance until the English had settled in the Carolinas. The origin
of the word is uncertain, but there are indications that it was derived
from Shawnee, since a band of Shawnee lived for a time near what
is now Augusta, Ga., and from a very early period occupied an inter-
mediate position between South Carolina and Georgia on the one hand
and the Creek Nation on the other. It is probable that there were
originally several tribes speaking the same language but having sepa-
rate names and that the necessity for a distinguishing term for all
did not present itself until the number of non-Muskogee tribes in
the Confederation came to be considerable. As to the names of
these Muskogee tribes, we seem to have indications of the following:
Abihka, Coosa, Okchai, Pakana, Tukabahchee, Hilibi, Eufaula,
Kasihta and Coweta (or perhaps an original tribe of which the
Kasihta and Coweta were sections). There were some other groups
on the lower course of Tallapoosa River, such as the Atasi, Kealedji,
and Kolomi, which cannot be definitely placed and may have been
independent of these or early subdivisions of them. Of course some
may represent people of wholly different connections who had become
assimilated to the Muskogee and had lost their own language and
customs. This is rendered probable from the fact that we have

|
.

No. 8 CREEK SQUARE GROUNDS—SWANTON

ww

several actual cases of such assimilation in later times. However,
so far as the tribes enumerated are concerned, this must always
remain in doubt.

When the tribes of the Confederation first became known to
white people, they were distributed geographically into two main
sections to which the names Upper Creeks and Lower Creeks have
become attached. The former were on the Coosa and Tallapoosa
rivers and the upper course of Alabama River in the present state
of Alabama, the latter on the lower and middle courses of the
Chattahoochee, which now forms part of the boundary between
Alabama and Georgia. It was this latter division principally which
lived upon Ocmulgee River for a time and thus gave rise accidentally
to the popular English name for the entire people. A minor division
also existed between those Creeks living on the middle course of
Coosa River and those centering about the lower Tallapoosa, the
two being sometimes designated as Upper and Middle Creeks,
respectively. In the distribution of the original Muskogee tribes,
the Abihka and Coosa constituted the greater part of the Upper
Creeks, while the Kasihta and Coweta were the dominating element
among the Lower Creeks. The Okchai, Pakana, Tukabahchee,
Atasi, Kealedji, Liwahali, Laptako, Kolomi, and a number of towns
descended from the Coosa, including Otciapofa, the Tulsa towns,
and the Okfuskee towns, besides several minor groups, formed the
bulk of the Middle Creeks. The Eufaula had the distinction of
being connected with all three. Their oldest seat seems to have been
in the Upper Creek country ; later they established themselves among
the Middle Creeks and about the period of first white contact they
formed a colony well down Chattahoochee River, among the Lower
Creeks. To complete the story of their migrant habits, we may
add that they seem to have furnished the first true Muskogee
contingent to the Florida Seminole in the Red House or Tcuko Tcati
Indians north of Tampa Bay.

Tradition seems to be borne out by circumstantial evidence in
pointing to the Lower Creek country as that region in which the
tribes in question began their federation. According to the story
this had to do on the one hand with the division of the Muskogee
into the Kasihta and Coweta and on the other with the relations
between them and the non-Muskogee elements, particularly the
Apalachicola. The relations of the Kasihta and Coweta to each
other are somewhat uncertain, for while it is at times implied that
they resulted from the fission of a single body of people, the most
popular traditions speak of them as having come from the west

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

as two distinct tribes and it is possible that the one-tribe idea may
be the result of later rationalizing. The story goes that, having
defeated all of their enemies, the Kasihta and Coweta instituted
periodical ball games as a kind of ‘moral equivalent for war,” and
afterwards, when either of them established relations of friendship
with other Indians, whether Muskogee or not, these Indians entered
on the same side as their friends so that this dual system soon
became general.

One of these sides, that of Kasihta, came to be known as the
White or Peace side, though it did not receive that specific name;
while the side of Coweta was the Red or War side.

At the same early period the Muskogee entered into intimate
relations with the Apalachicola Indians, who spoke a dialect related
to Hitchiti. This was the outgrowth of a treaty of peace following
upon hostilities, or to avert threatened hostilities. The Apalachicola
were then taken into the Confederation on the same side as Kasihta.
In some particulars, however, they are held to have been more
representative of the White towns than Kasihta and for that reason
their settlement came to be called Talwa Lako, “ Big Town.” Indeed,
the migration legend related to Oglethorpe by Tchikilli implies that
Kasihta was at least partly Red, their hearts being “ red on one side
and white on the other.” However, in all later times Kasihta
assumes the leadership of the White towns among the Lower Creeks,
as does Coweta the leadership of the Red towns. Four having
been the sacred number—the sacred formulae being gone through
four times, four arbors or beds constituting the ceremonial buildings
in the square ground and four sticks the number employed in the
ceremonial fires—it is not surprising that the Creeks should select
two towns from the Upper Creeks, taken collectively, to add to
these two leading Lower Creek towns. The White towns of the
Upper Creeks were represented by Abihka, the Red towns by
Tukabahchee, the second being from that group I have called Middle
Creeks, the other from the northernmost bounds of the Nation.
These four towns were the “back sticks”? of the Confederation,
and each had a special ceremonial name, viz., Kasihta Lako (“ Big
Kasihta”’), Coweta Mahmayi (“Tall Coweta’), Tukabahchee
Ispokogi, and Abihka Nagi. Ispokogi was the name of the culture
heroes of the Tukabahchee and it may be a Shawnee term. It bears
a suspicious resemblance to that of the Kispokotha band of Shawnee.
I do not know the meaning of Nagi. The Abihka were also called
specifically ‘“ the door-shutters ” because they protected the northern
frontier of the Confederation.

no. 8 CREEK SQUARE GROUNDS—SWANTON 5

In course of time the non-Muskogee element represented by the
Apalachicola Indians was increased, first by other groups related
to the last mentioned—such as the Hitchiti, Okmulgee, Sawokli, and
Tamati—who spoke closely related languages and called themselves
Atcik-hata, a term said to have some reference to the ashes of the
ceremonial fire in the square grounds. These Indians formed the
greater part of the first Creek invaders of Florida who presently
constituted the Seminole nation. The leading town in this southward
movement was Oconee, almost certainly affiliated with the Atcik-hata,
and the titular leadership among the Seminole remained with them
until after the Seminole War. However, the complexion of the
Seminole as a whole was changed from Atcik-hata or Hitchiti, to
Muskogee by the multitudes of refugees which fled to Florida after
the Creek War of 1813-14. The later removal to Oklahoma seems
to have reversed the situation since more than two-thirds of the
Indians now in Florida speak a language of the Hitchiti group.

There is strong evidence that the Chiaha Indians originally spoke
Hitchiti and that the Mikasuki of Florida branched off from them,
but some early event in their history separated them from the other
Atcik-hata and made them allies of the Coweta. This friendship
they shared with the Osotci who seem originally to have belonged
to the Timucua linguistic group of Florida. To the Upper Creeks
were added the distinct but dialectically related Alabama, Koasati,
and Tuskegee, while bands of Yamasi and Apalachee were
temporarily connected with both Upper and Lower Creeks. The
Alabama town of Tawasa seems to have had an origin similar to
that of the Osotci. At a very late date the wholly alien Yuchi
population was admitted into the Confederation, most of them
making their home among the Lower Creeks though there was a
small body also among the Upper Creeks. And more divergent still
were the Shawnee, from among whom two towns made their homes
in Creek territory for several decades during the eighteenth century.
One of these probably continued on into the early years of the
nineteenth century.

It may be added that towns are known to have changed from
one side to the other. Alabama was once a White town closely
associated with the Okchai, but later they were affiliated with the
Tukabahchee and came to be reckoned as Red. Wiogufki, Hilibi,
and Wiwohka are also said to have shifted from one side to the
other. In the case of the two last this may be partially explained by the
fact that, if we may trust native tradition, they were built up of
refugees from other settlements.

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

In former times a certain aloofness was maintained by the towns
of one moiety toward those of the other. They did not encourage
intermarriage and did not attend each other’s annual ceremonies.
This latter inhibition is now breaking down and it is claimed that
men of all towns attend the busk of Otciapofa. Otciapofa, however,
has long occupied an exceptional position. A chief belonging to the
Bird clan of this town always delivered the principal speech when
a new chief of the Confederation was installed. This town was also
the residence of the Creek dictator Alexander McGillivray, and it
was here that Crazy Snake, leader of the Creek conservatives,
called his important councils. Evidently the functions of the White
and Red sides in maintaining peace or bringing on hostilities were
formerly of great importance and some White towns, certainly
Apalachicola and Coosa, were places of refuge for murderers. The
“regular ” ball games, as distinguished from practice games, always
took place between towns of different sides and the supporters of
each town marched to the encounter in much the same spirit as if
they were going to war.

The principal White towns were: Kasihta, Apalachicola, Hitchiti,
Okmulgee, Sawokli, Yuchi, Abihka, Coosa, Otciapofa, Tulsa,
Okfuskee, Okchai (including Latogalga and Asilanabi), Pakana,
Koasati, Tuskegee, and Wiogufki.

The principal Red towns were: Coweta (including Likatcka),
Eufaula, Chiaha, Osochi, Tukabahchee, Liwahali, Laptako, Atasi,
Kealedji, and Hilibi. Alabama changed from White to Red in the
manner described.

The people of each town were subdivided into clans which were
usually named after animals and were invariably perpetuated in the
female line. The only clan of importance not named for an animal
was the Wind clan and with this the Skunk was closely associated,
the Skunk clan having always been linked with it. I obtained the
names of over 50 clans but some of these were known to only one
or two informants, and a number of others were small and bound
into phratral associations with clans of greater prominence. Some
clans were considered as equivalents throughout the entire nation.
The Skunk, Fish, Rabbit, Otter, and Turtle seem always to have
been united in one phratry with the Wind; the Wolf and Salt with
the Bear; the Pahosa with the Deer; the Wildcat with the Panther ;
and the Turkey and Tami with the Alligator. In the same way the
Snake, Kapitca, and Woksi were almost invariably counted in with
the Aktayatci; the Mole, Toad, and Tcikote always went together
and were generally allied with the Deer and Pahosa; and the

No. 8 CREEK SQUARE GROUNDS—SWANTON A,

Lidjami, Eagle, Hickory nut, Fox, Cane, and Muskrat * were usually
placed with the Raccoon.

Some phratral associations, however, were confined to one or a
few towns and did not extend throughout the nation. Thus the
Potato was commonly placed in one phratry with the Raccoon but
in Tukabahchee it was separated. The Beaver was usually placed
with the Bird, but in Alabama it was quite distinct. On the other
hand it was sometimes classed with the Alligator. Occasionally the
Aktayatci formed one phratry with the Raccoon, and much more
rarely the Deer and Panther were found together. Differences of this
kind were due in some measure to the council system. Every
important clan in a given town, or every group of related clans,
held meetings during the annual ceremony known as the busk and
each listened to an address by its oldest capable male member or
“uncle.” If an individual came to live in a town in which his own
clan or his phratral group was not represented, he would elect to
affiliate with one of those already in existence. It was usual for all
of the children of each group of this kind to consider themselves
brothers and_ sisters between whom marriage was ordinarily
prohibited. However, the information I received shows plainly that
sexual intimacy between individuals of linked clans was not considered
as serious as between members of the same clan. It is specifically
stated regarding some of these clans that “they were kin” up to
noon, or up to midnight, and separate the rest of the time, 7. e., a
limited taboo. was maintained against them. It is also said that a
man would sometimes pretend that a woman whom he wished to
marry was of a certain clan, for which he would manufacture a name,
although she was in fact of his own, and that, if he were a man of
influence, he often “ put over” this new creation of his. On the
other hand, I have been told that, even though children of certain
of the primary clans were brought up together, they would never
be regarded as brothers and sisters. It is quite plain that all sorts
of variations had grown up in response to unpredictable situations.

When one eliminates the obscure and the constantly linked clans,
about nine are left of something like major importance. These are
the Wind, Bear, Bird, Beaver, Alligator, Raccoon, Aktayatci, Deer,
and Panther. We should perhaps add the Potato. The Beaver,
however, has importance mainly in one group of towns, and the
Aktayatci appear to have been rather closely associated with the
Hitchiti and the Seminole, but also with Hilibi, Wiogufki, and
Eufaula.

“ee

*In the 42d Ann. Rep., Bur. Amer. Ethnol. (p. 116) I erroneously called this
the Mink.

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The question naturally arises whether some of these clans may
not have been brought in with formerly independent tribes. All we
can say is that certain clans are more prominent in some of these
tribes than among the true Muskogee but whether they were brought
in by them we do not know. Thus, as just mentioned, the Aktayatci
was particularly prominent among the Hitchiti, as were the Snake,
Kapitca and Woksi, and in a more pronounced manner the Toad,
Mole, and Tcikote. The Daddy-long-legs and Salt were similarly
associated with the Alabama, the former, indeed, being hardly known
outside of that tribe.

Besides this division into phratral groups all of the clans were
ranged in two moieties called respectively, Hathagalgi, ‘‘ White
People,” and Tcilokogalgi, “People of a different speech.” The
Wind and Bear with their phratral associates were almost invariably
White, and the Raccoon and Aktayatci and their allies almost
invariably Tciloki. The Bird is usually White but among the
Alabama and Koasati it is Tciloki. The Beaver is also White usually,
but when it is associated with the Alligator and when the Alligator
is not a White clan, the Beaver often becomes Tciloki. The Alligator
is most often Tciloki but in a number of towns it is White. The
Deer is usually Tciloki but it is White in a few towns. Today the
Panther is almost always Tciloki but some of the oldest myths and
some of my best informants assert that it was anciently White.

When I first went among the Creeks, I was told that in one or
two towns the clan moieties were exogamous, but the greater number
of my informants held the contrary opinion. I was much surprised,
therefore, during my last visit to have most of my informants
maintain that they were exogamous. This much is certain, that there
were striking exceptions to this law in comparatively early times,
for instance, in the case of the famous Creek speaker Hobohit Yahola.
Probably it will never be possible to say whether this phratral
exogamy was breaking down or growing in times known to us. In
recent years the principal function performed by these moieties has
been to determine the line-up of the players in practice games within
the town. The important bearing the mere character of a name
may have in social evolution is shown by the fact that, on account
of the name, persons of European blood were usually reckoned as
“friends ” of the Hathagas, and in consequence the latter acquired
a reputation as “ progressives,’ while the Tcilokis were considered,

oe : 93
conservatives.

and acted like,
Besides the clans, phratries, and moieties there were certain groups

in each town which had official functions. Some of these were

no. 8 CREEK SQUARE GROUNDS—SWANTON 9

determined by the individual ability—usually military—of those who
belonged to them. Thus a man might start his. public career as a
common warrior or tasikaia, a word now often translated “ citizen,”
be promoted to the position of an imata labotski, or “ Little Imata,”’
then to that of an imata tako, or “* Big Imata,” and finally become a
tastanagi, or “‘ war leader.” If sufficiently prominent he might be
made a tastanagi tako or hohbonaia, “‘ war speaker,” though of these
there was never more than one in a town at any one time. There
were ‘“ beds ”’ or seats in the square grounds for each of these classes,
but not all were promoted into them. Men who belonged to the clan
of the chief (miko) would be given seats in his section and form
the mikalgi, ‘‘ chiefs,” who acted as a kind of special executive
council. If they belonged to a certain clan known as henthalgi, they
would be given seats in another place. The functions of the henthalgi
are somewhat uncertain but they were concerned largely with the
maintenance of peace and charged themselves with the internal
prosperity of the tribe. The henthalgi were almost always formed
of the Wind clan, and if, for any reason, the Wind clan could not
be used, the Bird or Beaver, or at least some clan considered White,
would take their place. There was also a class of men called ist-
atcagagi, the old, experienced men from all tribes, retired from
active service but keepers of the tribal lore. It seems fairly evident
that a correlation existed anciently between White towns, White
clans, and the henthalgi, and that the miko of a White town was
generally chosen from a White clan. It is even possible that the
miko of a Red town was formerly chosen from a White clan.
Certainly there is a marked tendency to choose chiefs from the Bear
clan, even in Red towns, though many of them are also from the
Raccoon and the Aktayatei.

A word must now be said regarding the ceremonial grounds.
Originally every Creek town had such a ground which at a still earlier
period was probably the ceremonial ground of a small tribe. As a
tribe increased in numbers, however, the ground often became too
small to accommodate all of its members comfortably and so it split
into two or more. Undoubtedly, before the organization of the
Creek Confederation, there was great diversity among these grounds
and a‘ certain element of diversity has persisted until the present day,
but on the other hand considerable standardization has undoubtedly
taken place. Creek legend asserts that the first ceremonial ground
was given to the Coweta, the Kasihta, or the Tukabahchee by the
Breath Controller or by other supernatural beings, and copied by the
remaining towns from them. While this represents a modern

‘

LO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

rationalization, there can be no doubt that the earlier ceremonial
grounds of the constituent members were altered in many particulars
in conformity with the prevailing pattern.

All the grounds known to us originally consisted of three elements,
a tcokofa or community hot house used in bad weather or for secret
ceremonies, a “square ground,” and a “chunkey yard,’ or ball
ground. The name “ chunkey yard ” is derived from an old pastime
which consisted in rolling a stone disk along a level plot of land
and throwing certain long poles after it, the game turning on the
relative nearness of the poles to the roller after all had come to
rest. There was a single pole in the middle of this yard surmounted
by a cow or horse skull or by a wooden figure, and about this men
and women played against each other in a kind of ball game. This
game was mainly confined to people of the town and was social in
character while the great ball game, similar to our game of lacrosse,
was played by men only and was highly ceremonial. The fact that
the “chunkey yard” was a part of the ceremonial ground may
indicate that the single pole game formerly had more religious
significance than was the case in later times.

The tcokofa has long been out of use, though at Tukabahchee fire
was until recently lighted in the middle of a circular offset of the
ceremonial ground where this structure would stand if it were still in
existence, and one such building was put up at Pakan Tallahassee
after the Civil War.

The most important part of the ceremonial area today is the
“square ground,” so called because in the largest towns there were
on it four long cabins or arbors, in native parlance “ beds,” forming
four sides of a square. Partly from tribal idiosyncrasy and still
more on account of failing numbers, several of these grounds now
lack one cabin, and the Alabama ground lacks two. Today the
cabins consist merely of two or three rows of split logs to serve
as seats and an arbor of boughs to shield their occupants from the
direct rays of the sun, but anciently the seats consisted of mats
woven out of cane raised upon short posts and the cabins were
provided with a back and roof of wattle or split shingles plastered
with clay. The arbors in the largest modern towns are supported
on eight posts, four in front and four behind, but some have only
six, and most of the Seminole towns only four. On the other hand,
a sketch of one of these cabins made by a Frenchman early in the
eighteenth century shows ten posts, five in front and five at the
back. Today, however, the eight post arrangement seems to be
considered orthodox, and the three sections marked off by these

No. 8 CREEK SQUARE GROUNDS—SWANTON It

posts are used for the seating of as many clans, groups of clans
or related officials.

Considerable variation in the ancient and intended plan has been
brought about by the attrition which the tribe and its several
divisions have undergone, loss of the keepers of the sacred lore,
and other factors, but it is plain that normally one of the four cabins
was mainly devoted to the miko and his clan. Hence it was called
mikalgi (or mikagi) intupa, the “ Chiefs’ bed.” Another was devoted
mainly to the henihalgi and was named from them, another to the
higher class of warriors, the tastanagalgi, and so received their name,
and still another to the novitiate warriors or youths from whom it
was called tasikaialgi intupa or tcibanagalgi intupa. The positions
of these in the square ground varies considerably. It should be
stated in the first place that the cabins are placed normally toward
the four points of the compass, but that for some unexplained reason
in the square of Tukabahchee the entrances are toward the cardinal
points. In the old Kasihta square, which seems to have set the
fashion for many other towns, the mikos’ cabin lay west, the henihas’
south, the tastanagis’ north, and the tasikaias’ east. In the Okfuskee
towns, of which Nuyaka is an example, the mikos’ cabin lay north,
the henihas’ east, the tastanagis’ west, and the tasikaias’ south. In
Pakan Tallahassee the mikos’ cabin is north, the henihas’ south, the
tastanagis’ west, and there is no east cabin. In Eufaula the mikos’
cabin is north, the henihas’ east and the tastanagis’ south, the west
cabin being missing. In Alabama, where there are but two cabins,
the mikos’ cabin is east.

The other variations in the arrangement of the squares will come
out in the subjoined material, but before presenting it something
must be said regarding town officials.

The miko, as already stated, was chief of the town, its head
presiding officer and responsible executive. Like many of the other
officers he had a special companion or heniha taken from the Wind
clan or whatever clan occupied the position of henithalgi. This heniha
is sometimes called miko apokta, “ second chief,” but the latter name
is also given to still another functionary who shares the burdens
of state with his superior. The chief had one or more yatikas or
interpreters who also bore the name asimbonaia, ‘‘ speakers.” Unless
wanted to make an announcement or to send upon an errand, they
sat with their clan or other group to which they normally belonged.
The tastanagi tako and holibonaia have been mentioned. The position
of hotibonaia, “ war speaker ” was the most exalted military position,
and it is possible that there was only one such official at a time in

WZ SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the entire Creek Nation. It was the position occupied by Hobohit
Yahola, famous leader ot the Creeks during their removal west and
all of their subsequent troubled history until the Civil War. He
was not a miko even of one town, but his influence was actually
greater than that of any miko, or any number of them. There were
two ta‘pala, whose chief function was to act as messengers before
and during the women’s dance, and there were two singers for the
women (inyahaikalgi, “ singers for them”), who sat at the end of
one of the cabins just in front of the spot where the women began
dancing. They were usually selected for their knowledge of songs
rather than on account of their clan affiliations. The hilis haya was
the head priestly functionary. He supervised the preparation of the
medicines and gave them their final potency and he ordered
everything in any way touching upon the supernatural. He was
assisted by one or more men called hilis tcalaba or “ medicine
mixers,’ whose functions are defined by the name, and by two or
more young men called hilis hoboia, ‘ medicine gatherers,’ who
collected the red root, pasa, and other plants that went into the
sacred medicine. Some towns seem to have had a separate official
called tutka didja, “ fire builder,” to start the fire, but in others the
hilis haya did that, and there was instead a tutka oktididja, “‘ gatherer
of wood” for the fire. A number of boys known as oidjawalgi
brought water for the medicines. In one town we learn of ahaga
haiyalgi, “‘law makers,” who are said to carry out the instructions
of the tastanagis and may be identical with the imatas elsewhere
mentioned. There were also officers called simiabaia, or “ leaders.”
The hoktagi immiko, “
have been identical with the ta‘palas. Anciently the tastanagis and
their assistants acted as town police, but nowadays three or four
light horsemen are selected at random to police the square ground
during ceremonies, and there are boys called “‘ dog whippers ”’ with
long whips stationed at each opening into the square during the
women’s dance to drive away dogs. Many of these officers were chosen
for four years only. If one died before the expiration of his term,
a substitute was selected from the same clan, which seems to indicate
that the position was something of a clan prerogative.

Of the ceremonies which took place on these grounds, only two
have survived. One, called by the popular local name of “ stomp
dance,”’ was confined to the people of the town and was simple in
character, the miko hoyanidja (red willow) being ordinarily the only

chief of the women,” and his henitha, may

medicine used during it. There were no dances other than the common
and relatively secular ones, no ceremonial lighting of the fires, no

no. 8 CREEK SQUARE GROUNDS—SWANTON 13
ceremonial complications of any sort. It seems to have been usual
to hold three of these dances in the spring and early summer, a
month apart, the series paving the way for, and leading up to, the
second ceremonial, the “ busk,” which was the great annual ritual.
This last is usually considered as lasting four days, though in that
four are included the days of assembling and departure. The
principal event on the second day is the women’s dance. On the
third day the men fast, take medicine four times, and near its close
march down to the creek and bathe. After they return, they are
dismissed to their camps and break their fast. Later they are
summoned to the square again where they dance four times, and
then the dance becomes “‘ common,” visitors from friendly towns
being admitted to it. A fire is kept up all night in the center of the
square and dancing continues about it until it is nearly day. In
former times many, perhaps all, of the towns extended their busks
over eight days, but from what can be learned of these longer
ceremonials they seem to have been in the main a simple doubling
of the shorter ceremony, except for a few features like the kindling
of the new fire which took place but once annually.

Let us now turn to the new material regarding the square grounds
and the ceremonies conducted there. Under each heading I give the
notes obtained from native informants belonging to the square or
town in question.

ABIHKA

Figure I gives the general arrangement of the square of Abihka,
‘or Talladega, and Plate 1, Figure 1, gives a view of the ground itself
near Henryetta, Okla., as it appeared in 1912. The exposure was
made from the southwest.

The medicines were taken first by those in the Bears’ bed, then
in succession by those in the beds of the Raccoon on the west side,
the Raccoon on the east side, the Deer, and those in the south bed.
The asimbonaia in the northern section of the east cabin acted more
particularly as the chief’s messenger; the one in the western
compartment of the south cabin called first the women and then
the children to come to take their medicine from the vessel at the
north end of the east cabin. The ta‘palas were changed every four
years. The women gathered preparatory to their dance at a tree on
the edge of the tadjo (the ridge of sweepings that makes the edge
of the ceremonial ground) and entered’ the section about the fire at
its northwestern corner. Five pots of medicine were prepared at
the north end of the west cabin (fig. 1, 11). Afterward one pot

14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

was placed a little farther north and west for the boys (12) and
another was carried around to the north end of the east cabin for
the women and children (13). The ingredients of this medicine

N

*
5

Fic. 1.—General arrangement of the square of Abihka.

A. Chieis’ Bed (mikalgi or mikagi intupa); I, miko (Bear); 2, heniha
(Raccoon) ; 3, hilis haya (Raccoon) ; 4, hilis tcalaba; 5, yahaikas (any clan).

B. Henihas’ Bed (henihalgi intupa): 6, asimbonaia (Alligator); 7, ta‘pala
(Panther) ; 8, ta‘pala (Wind).

C. Warriors’ Bed (tastanagalgi intupa) : 9, asimbonaia (Raccoon) ; 10, tutka
didja (Deer).

II, medicine pots (Ist position) ; 12, medicine pot for boys (2d position) ;
13, medicine pot for women (2d position); 14, place where medicines were
piled immediately after they were brought in; 15, ball post (pokabi).
It is to be noted that, in all of these diagrams, the ball post was actually much
farther from the center of the square than is indicated.

were miko hoyanidja (“red root”), pasa (“ buttonsnake-root ’’),
wilana (“ wormseed”), and hobaga (“maypop’’); tcato hatki
(‘‘ white stones”) were added. After the ceremony was about over and
the fasters were ready to go down to bathe in the creek, what was left

no. 8 CREEK SQUARE GROUNDS—SWANTON T5

of the medicine was poured on the fire. It is said that one gallon of
spring water was brought for all five pots. This must mean a gallon
for each. Following are the busk names of the present officers :

ATUL Oey cote erenck retort ce neeh iors ate a one aie ors Tctktcat Heniha

eMMiliah By: Ae-pert sensi eee cise keccte eine ae Itchas Hadjo

nilis*tcalabay ac access oe. saiseles Konip Yahola

tals ail dea ragetenete baeyeters witiehe se Sherer tiece sic Katca Tastanagi
el NEST Ser as steiodte octet weaxcyeus.e eens etene he Kona Yahola

EUTECRGIC |i ae sepa re atniaie Sere *.... Tastanakutci

ASI DOMlal awe ete ie eieioe es cleyen is aietere Wotko Fiksiko

“

Bia RPM SiGe TELS Kapitca Hadjo

The Hathagalgi of this town are Wind and Bear; the Tcilokogalgi
are Panther, Raccoon, Deer, Beaver, Alligator, and Bird.

My information regarding this ground was obtained mainly from
Jim Star who described the Talladega ground to me in 1912, the
plan of which is in the Forty-second Annual Report of the Bureau
of American Ethnology, page 205. The different aspect of the east
cabin is mainly due to the fact that the earlier account gives a more
ancient arrangement, when the warriors were graded into tastanagis,
and big and little imafas. The position assigned to the hilis haya
in the earlier plan is probably erroneous. The other differences are
due mainly to the more extensive information obtained on my last
trip. My new information disagrees with the older, however, in
assigning the Bird, Beaver, and Alligator clans to the Tciloki side
as was said to be the case at Abihka-in-the-West instead of to the
White side as was given me for Talladega and the old Abihka town
near Eufaula. It is probable that the new information is correct
since the last mentioned square was given up when the man from
whom I obtained data regarding it was a boy. However, it must be
remembered that these allocations are not invariable and probably
changed at times even within the same town.

OTCIAPOFA, OR HICKORY GROUND

Figure 2 shows the square ground of Otciapofa, popularly known
as Hickory Ground.

The hilis tcalaba was changed every four years and was not taken
invariably from the same clan.

At the southernmost front post of the east cabin were fastened
two poles with black feathers tied to the ends and at every other
front post were three similar poles but with white feathers. These
were carried by the men in the “ feather dance.”

2?

16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

In the women’s dance there is but one leader who carries a notched
stick called atasa, the old name of the war club, from the middle
of which depends an eagle feather.

Besides the invariable miko hoyanidja or “red root,’ the busk
medicine contained tutka hiliswa (“fire medicine’), wilana
(““ wormseed ’’), tutka-tcok-hissi (a place on the ground where wood
has been burned and moss has sprung up), and hobaga (‘‘ maypop”’).

‘

< AKTAYATC/
iK AND

BEAVER
AND
ALLIGATOR

°

\
‘
‘
'
‘
‘
‘
t
t
t
t
‘
'
t
t
'
v
'
’
‘
‘

Fic. 2—The square ground of Otciapofa, or Hickory Ground.

A. Chiefs’ Bed: 1, miko (Beaver); 2, miko apokta (Beaver); 3, hilis haya
(any clan; at present Wind); 4, hilis tealaba (changed every 4 years; now
Beaver).

B. Henihas’ Bed.

C. Warriors’ Bed.

5, medicine pots (Ist position); 6, medicine pot for boys (2d position) ; 7,
medicine pot for woman (2d position); 8, log on which medicines were laid
and macerated with a small wooden pounder; 9, place where medicines were kept;
10, point where women gathered preparatory to their dance; 11, point where
women made a final stop before entering to dance; 12, where the ashes from
the central fire were deposited every year; 13, sweepings from the square ground
(tadjo) ; 14, ball post.

As usual, the doctor perfected this medicine by blowing into it
through a hollow reed. Two men took medicine at the same time,
using gourd dippers. After all were through a dipperful was poured
On the fire:

For a drum they use a stout jar and there are two coconut-shell
rattles.

There is a line of tadjo.(sweepings) around the four cabins but
it does not include the ball post. In playing the single pole game a

no. 8 CREEK SQUARE GROUNDS—SWANTON 7

hit on the skull at the top counts five and a hit on the pole above
a certain mark counts two when it is struck twice in succession.

The Hathagas consisted of the Beaver, Alligator, Bird, Bear,
Skunk, Wind, and Rabbit; the Tcilokis of the Raccoon, Deer,
Panther, and Aktayatci. The Rabbit, Wind, and Skunk formed one
phratry.

The information regarding this ground which I obtained in 1912
was particularly incomplete. It is therefore gratifying to find
that there are no serious discrepancies between the plan based on
that (42d Ann. Rep., Bur. Amer. Ethnol., p. 211) and the present
information. Some clans are given in one and omitted from the
other, but where the same ones appear in both they have practically
the same positions except for the Bear, which, according to the
earlier description, sat in the west cabin and according to the later
at the east end of the north cabin; and the Tami, which the former
places at the west end of the north cabin and the latter at the back
of the northern section of the west cabin. The location of the
miko, miko apokta or heniha, and hilis haya is probably more exact
in the later plan which also adds many more details. The informants
differed somewhat regarding the Hathagas and Tcilokis, the earlier
authority placing the Beaver among the Tciloki and the Panther
among the Whites, allocations exactly reversed by my later informant.

Dies. PU Ss

Next we come to one of the two divisions into which the Tulsa
have recently split, this being known as Little Tulsa (fig. 3). Plate 1,
Figure 2, shows the old Tulsa ground in 1911 from the southeast,
before the fission had occurred.

It is said that all of the offices are filled from particular clans.
The following is a list of the present officers; exclusive of the miko
and his heniha:

ASTANA CAN AAO 2 --cer eh cisiseclerse seaman Ispani Tastanagi
RUG frUllal Clete enero mites ree mene et set arora Kapitca Tastanagi

A NES RMR tate o<as Toate gues ’s, ah areas. as aie Lata Miko (controls the two above)
ahaga haiyalgi (“law makers”)...... Kapitca Tastanagi (Aktayatci), and

(messengers for the tastanagis ) Tami Tastanagi (Tami)
SUIT UT cl Del AR ea er eyeiae rele iehe exert hae iercnare wise 30/6 Kantcati Miko (busk name) or
Nokos Hadjo (common name)

IITs iach Vicia Meet ates res os cin iat chee once Miko Tcapko (Beaver)
Inilisiatcalalawesvere s skteeie nes «arenes one Tami Yahola (Tami)

faticao atic ticle vemta sites: soa satus ties se. Tamatakutci (Tami)

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

YOUTHS

\ARTAYATC /

Oo

B

Fic. 3.—Little Tulsa square ground.

A. Chiefs’ Bed: 1, miko; 2, heniha; 3, simiabaia.
B. Henihas’ Bed.

C. Warriors’ Bed.

D. Youths’ Bed.

4, old location of medicine pots; 5, modern location of medicine pots.

No. 8 CREEK SQUARE GROUNDS—SWANTON TQ

AED all AMR versyareye.sls nxcpnlehanciahes creas stereiae Nokos Fiksiko (Bear)
SMU EGER, a 0ss te of cys aig Ae. olde di ovayah ote leheue stots Tami Hatkutci (Tami)
AAU AST ave cars. eesctie ox Bis Sreisieie! s oie auetaa ave Kapitca Fiksiko (Aktayatci), and
Pin Hadjo (Tami)
Captain of the Light Horsemen........ Yahola Tcapko (Tami)
Oktcan Hadjo (Bear)
To hit MAOUSEMIe! ee sree nee le eileen .. Fas Yaholute (Beaver)

Kapitcutci (Aktayatci)

The ta‘palas functioned at the women’s dance, the yahaikas sang
at the women’s dance and at the feather dance.

My original Tulsa data, published in the Forty-second Annual
Report of the Bureau of American Ethnology (p. 213), was obtained
from an old man and was intended to reflect the most ancient
arrangement he could remember. Since that time the Tulsa Indians
who used to meet at the Little River ground have divided and
maintain two distinct squares. The general agreement between the
older and the later plan is therefore surprisingly close. The principal
difference seems to be in the position given the Aktayatci who appear
in the north bed in the earlier plan and in the south bed in the later
one. The earlier plan may also be in error in the position given the
medicine pots but this was subject to change from town to town and
during the ceremony itself. The Eagle clan, which appears on the
older plan, died out so long ago that it can barely be remembered
by any living Creeks.

NUYAKA

The plan of this ground is given in Figure 4 and a view of it as
it appeared some years ago in Plate 2, Figure 1.

The positions of hilis haya, hilis tealaba, and tutka didja were held
for four years when the man and clan were changed so as to teach
others the duties of these offices. The tastanagis and imatas had
become confined to one or two clans. The hoktagi immiko and his
heniha controlled the women’s dance and were called istatcagas. The
term hola‘ta was applied to a certain class at the square ground
in some towns, sometimes to the Tcilokis, but its application here is
not explained. The first ta‘pala acted under the women’s chief. He
was taken from the Bear clan, or, failing that, from the following
in order of preference: Wind, Raccoon, Tami.

In taking the medicines they drank of the miko hoyanidja first
and pasa second. :

A rock was placed under the miko’s seat “‘ to make the seat heavy.”
Anciently there was a tcokofa northeast of the square. Nowadays
the ground is hoed off only once a year.

20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

B/RD

AND

BEAVER

TASTAIVVAG/S

HOLA TAS | ALLIGATOR

o1i3
Fic. 4.—Nuyaka square ground.

A. Chiefs’ Bed: 1, miko (Bear); 2, heniha (usually Wind; Deer in 1929) ;
3, hilis haya (changed every 4 years; Turkey in 1929); 4, hilis tealaba (Alli-
gator); 5, tutka didja (Raccoon).

B. Henihas’ Bed.

C. Warriors’ Bed; 6, two tastanagis (or law makers).

D. Tcilokis’ Bed (Tcilokogalgi intupa), or Tasikaias’ Bed (Tasikaialgi
intupa): 7, hoktagi immiko (Aktayatci) ; 8, heniha for hoktagi immiko (Alli-
gator); 9, Ist ta‘pala (Bear preferentially) ; 10, 2d ta‘pala (Alligator).

11, medicine pots; 12, point at which women enter; 13, ball post.

No. 8 CREEK SQUARE GROUNDS—-SWANTON 2]

In this town the Turkey and Alligator clans belong in one phratry
and so do the Wind and Skunk.
The present officers are as follows:

5100) COMES rey oan yh con tree ee Nokos Miko
Weta Meat ects ciciereveletersia sretersee? ofetenttciers Miko Tcapko
HntlsS Una yaar eese he araee ates al rotons coats. ohs. crepe alas Tastanakutci

EUG ae © cts C11Cl ici parapet eas ore elles sewtedate ete vee 2) Hotalgi Hadjudji
Hig tSpec Gall lly daeareteuste exe rstons coleasietivein ie teins Lodja Yahola
Moktaei 1mMMiIkOWs. a. .<cessis es seen. Wotko Yahola

When the present Nuyaka data are compared with that which |
obtained in 1912 for Nuyaka and the related towns Okfuskee,
Abihkutci, Talmutcasi, and Tcatoksofa, the agreement is found to
be close except in the cases of the two last where the square grounds
had long been given up and were described by individuals from
memories of their early years. The main correction is in locating
the miko, heniha, and hilis haya and the difference here is not great.

PAKAN TALLAHASSEE

Figure 5 gives the plan and Plate 2, Figure 2, and Plate 3, Figure 1,
views of the ground, one showing the three cabins, or arbors, and the
other the chunk yard and ball post.

The tutka oktididja, hilis tealaba, hilis haya, and oidjawas, were
appointed every four years from any clan. The ta‘palas and hilis
hoboia were appointed every four years from the same clan. The
miko and asimbonaias held their positions for life.

Here we seem to meet some strange innovations. The Bears’
section of the south cabin receives one name connected with war,
tasikaialgi intupa, and the section of the Birds, Beavers, and
Alligators another, tastanagalgi intupa, while, at the same time, they
are White or Hathaga clans and their cabin is called hathagalgi
intupa. Yet one section of the west cabin is called tastanagalgi intupa
also, and the whole cabin receives the unusual name of laksafaskalgi
intupa, “bed of the Blacks,’ the Blacks being evidently the clans
elsewhere called tcilokogalgi.

The tcokofa was to the northwest and this was the last town to
put up such a structure.

The Birds’ section was called istatcagagi intupa. The Deer and
Pahosalgi were formerly called the imatagalgi.

Three poles with white feathers attached were fastened to each
of the front posts for use in the feather dance.

Back of the tastanagis’ section of the west cabin was a little
structure in which to inclose the medicine pots when they were not
in use.

No
bo

SMITHSONIAN MISCELLANEOUS COLLECTIONS

voL. 85
N
oT TO O
196
©
©
48
©
©
1 23
Mt |
e |
15
14% a |
12

%
x |
©
Co
een tt

Fic. 5.—Square ground of Pakan Tallahassee.

A. Chiefs’ Bed (mikalgi intupa): 1, miko (Bear); 2, hilis haya (Panther
in 1929) ; 3, yahaikas (singers for women) ; 4, dog whipper.

B. Whites’ Bed (hathagalgi intupa): 5, asimbonaia (Bird); 6, ta'‘pala
(Bear) ; 7, hilis hoboia (Bird) ; 8, oidjawa (Bird) ; 9, oidjawa (Bear) ; 10, dog
whipper.

C. Blacks’ Bed (laksafaskalgi intupa): 11, asimbonaia (Deer); 12, ta‘pala
(Bird; father Raccoon) ; 13, hilis tealaba (Deer in 1929); 14, tutka oktididja

(Raccoon in 1929); 15, hilis hoboia (Raccoon in 1929) ; 16, oidjawa (Panther
in 1929); 17, dog whipper.

18, medicine pots (Ist position) ; 19, medicine pot for boys (2d position) ;
20, medicine pot for women (2d position); 21, drum; 22, box for tobacco; 23,
place where medicine was laid before being used; 24, woodpile; 25, ball post.
The dotted line marks the course pursued by the women when they entered to
dance.

no. 8 CREEK SQUARE GROUNDS—SWANTON

to
©

In this town the Hathagas were the Bird, Alligator, Beaver, Bear,
Tami, and Wind. The Tcilokis were the Deer, Raccoon, Panther,
and perhaps Aktayatci. The two moieties were anciently exogamous.

The particular opponents of this town in regular ball games were
the people of Atasi but they also played against [ufaula Hopai,
Alabama, Hilibi, and Upper Eufaula. The Koasati Indians are said
to have divided up in the ball games, some playing on each side.

The names of the present officers are:

TTT <Ouseiewls em eis es Nokos Miko (‘“ Bear Miko”) (Bird)
ASiMVONAIA® “Vaan Tastanak Imata (Deer)
or. Tf apeta. dhe Tastanak Hadjo (Bird)
Ihilis haya >....:. Katcutci (‘‘ Little Panther”) (Panther)
hilis tcalaba ..... Nokos Hadjutci (‘“ Little Bear Hadjo”) (Deer)
oidjawalgi ....... 1. Fus Yaholotci (“Little Bird Yahola”) (Bird)

No

. Hotalgutci (‘Little Wind”) (Bear, father Wind)

3. Halak Hopai (“ Potato Hopai”’) (Panther, father Rac-
coon, Raccoon and Potato belonging to the same
phratry )

hilis hoboia ...... 1. Itco Ilutci (‘‘ Little Deer Foot’’) (Raccoon, father Deer)
2. Talsi Yahola (Bird)
Easel Mieteteveseteee eee 1. Halak Yahofla (“Potato Yahola”) (Bird, father Rac-
coon)
* weeeeeeeee 2 Hotalgi Hadjutci (“Little Wind Hadjo”) (Bear, father
Wind)
tutka oktididja ...Itco-ili Imata (“Deer Foot Imata’’) (Raccoon, father
Deer)
Valiaikal Sipe meresre rie 1. Pahos Fiksiko (Raccoon, father Deer with which the
Pahosa is affiliated)
Be) Reece. 2. Miko Yahola (Bear)

If one town wished to play a match game with another, said my
Pakan Tallahassee informant, they sent a man to that town with a
ball stick, and when thé people of town number two had reached an
agreement they sent the ball stick back. My informant said that he
then had a ball stick hanging up in his house which had been sent
by the Alabama.

In the women’s dance, the atasa held by the leading woman is
red and has an eagle feather attached to it; that of the second is
white and has a feather of the fus hatki (“‘ white bird’), a bird
found down by the creeks, attached to it. They used from 12 to
14 terrapin rattles. During the women’s dance the two ta’palas stand
about where the two pots nearest the west bed are in the plan. Each
carries a wand called si’dik-kika having a white feather fastened
to the end.

The dog whippers used in this dance are taken from any clan.

24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

About 54 men were present at the last preceding busk and 25
women and girls participated in the dance.

At the top of the pokabi in this town was a horse skull. In scoring
for this game they draw a line from the ball post to the nearest
corner of the south cabin. When the skull is struck, 4 are scored;
when the pole under the skull is struck, it counts 2 on the way out
and 1 on the way back. They may count it as a game, however, by
agreement when they reach the corner post. The women’s tallies are
marked on one side of the line, the men’s on the other.

There is little difference except in detail between the above plan
of Pakan Tallahassee and the several I recorded in 1912.

WIOGUFKI

The plan of Wiogufki is given in Figure 6 and a view of the ground
from the southwest in Plate 3, Figure 2.

There is a little log house on the grounds in which the pots are
stored when not in use to keep them from being broken. There
never was a north cabin so far as my informant knew. On the upper
end of the ball post is a cow skull.

The women walk four times around the fire; then their leaders
stop opposite the singers and they begin to dance.

The hilis tcalaba holds office for four years. In this town the place
of the miko’s heniha is taken by a tastanagi. Indeed all of the
tastanagis are considered the same as the henihas. They are called
“the people who are named” and are of the nature of lawmakers
and assistants to the miko. The miko’s tastanagi is also the same
thing as the yatika. There are no istatcagagis (retired leaders who
acted as councillors), and no Creek town now has a holibonaia. There
are five water boys picked at random. At the front posts of each cabin
are four poles with feathers tied to the ends for use in the “ feather
dance.”

The two leading women in the women’s dance carry atasa. The
principal function of the ta‘palas is to call the women up for their
dance which they do four times. Each has a wand with a little white
feather at the end. Their official positions do not end with the
women’s dance but continue to the end of the busk.

No medicine is now put on the firesticks but it was formerly done.
The only medicine they use at the busk is the miko hoyanidja, to
which nothing is added.

The Hathagas and Tcilokis were the same as in Hilibi. They
were exogamous, and if the exogamic law was violated the ears of
the culprits were cut off.

“=

no. 8 CREEK SQUARE GROUNDS—SWANTON 25

The towns of Wakoka1, Tukpafka, and Talahasutci were all one
with this. In regular ball games they always played against Alabama.
Comparison between the plan given above and the two obtained
in 1912 shows considerable differences, but since one of the former

N
e
4S “il
we
er we
ae cai RS3y
: eo x Gm
ead oe * af SN oc
OGIHOS) v r © 2 3 .
2 / \ 52 =
| x2 1 0 : S822
— a SSTEHO
e So iy ( ! SQA
TASTANAGIS a A aN
%5 eee cw
“ S88
0 : 0 a So
© (
A @1° Y ©6 c
Roo hes 10
2% :
BiRPD own-: BEAR AND
£D THIS \ALLIGATOR
13 SIRE BUT USUALLY SAT

SAT WEITHER, HE RE

B

Fic. 6—Square ground of Wiogutki.

A. Chiefs’ Bed: 1, miko (Alligator); 2, hilis haya; 3, hilis tealaba (Bear
and Deer); 4, hilis hoboia (Panther); 5, a tastanagi (instead of heniha)
(Raccoon).

B. Whites’ Bed: 6, yahaikas (Deer).

C. Warriors’ Bed: 7, ta‘pala (Raccoon); 8, hilis hoboia (Deer); 0, tutka
oktididja (Deer).

10, medicine pots (near bed A for men; near bed C for women and boys) ;
II, point where women enter to dance; 12, point where women begin dancing;
13, woodpile; 14, ashes of old fires; 15, ball post.

was obtained from the same man, I think the explanation lies in
the smallness of the town and the weakness of many of the clans
which has resulted in many changes within a comparatively short
time. Nevertheless there is a general correspondence and _ the
allocation of clans to the moieties also agrees except in the case of

26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the Panther clan. This is explained by the fact that the Panther
was anciently considered a White clan and later came to be regarded
as Tciloki.

OKCHAI

Figure 7 shows the arrangement of this square and Plate 4, Figure
1, shows a view of it taken in winter.

SIDYNULS UL

SILL 4 ILV7 M/
‘SUSOH YJ

O | SIDYNYLSUL

B14
©
V/SITORS
0D £XCEPT O
6
19

Fic. 7—Arrangement of square ground of Okchai.

A. Chiefs’ Bed: 1, miko (Raccoon formerly; Bear in 1929); 2, heniha
(Deer); 3, yatika (Deer formerly; Wind in 1929); 4, hilis haya; 5, ta‘pala
(any clan) ; 6, hilis tcalaba (Bird) ; 7, tutka oktididja (Wind).

B. Henihas’ Bed: 8, ta‘pala (any clan); 9, hilis hoboia; 10, yahaikas (any
clan).

C. Warriors’ Bed.

D. Youths’ Bed: 11, oidjawa (any clan).

12, medicine pots; 13, medicine pot for boys; 14, medicine pot for women;
15, point where women assemble preparatory to their dance; 16, point where
women start dancing; 17, ashes of previous fires; 18, woodpile; 10, ball posts;
20, a spring lies in this direction.

The hilis tcdlaba, ta‘pala, and tutka oktididja changed every four
years. The hilis haya was reappointed every four years.

Wiley Buckner, an old informant of mine, was the former yatika.
The tutka oktididja in 1929 was Hotalgi Miko.

no. 8 CREEK SQUARE GROUNDS—-SWANTON 27

The medicine was pounded up just in front of the pots at the
north end of the chiefs’ bed.
_ The fire was brought from Alabama in the great migration and
new fire was lighted from it. They dug up the earth as deep as the
arm would reach, put the fire in there, and made the new fire on
top. This is the very spot at which they first placed their square
in Oklahoma and it has not been moved. The squares of the kindred
towns, Latogalga and Asilanabi, are not so old.

There were two poles with feathers on the ends at each of the
front posts of the cabins all the way round.

Atasa were borne by the two women who led in the women’s dance.

The tastanagis were law makers. They had charge of the rules
governing the taking of medicine and if anyone broke one of these
regulations they made him stay in the square ground all night without
eating instead of breaking his fast that evening as was usual.

In the various beds of this square ground the “‘ sons of the clan’
can sit with their fathers.

The Hathagas and Tcilokis were the same as at Hilibi.

Their principal opponents in the ball games years ago were the

Hilibi. \ 9

This arrangement agrees substantially with that obtained by me
in 1912 (42d Ann. Rep., Bur. Amer. Ethnol., p. 234), the principal
difference being in the location of the Bear and Bird clans. My
earlier informants placed the Bird and Aktayatci in the middle of
the south cabin and the Bear at the east end of the same, while the
later ones said nothing of the Aktayatci but placed the Bird clan in
the southern section of the west cabin and the Bear clan in the center
of the south cabin, the sons of the Bear being located next to them
on the east. The arrangement of this square must be taken in
connection with the plans of Latogalga and Asilanabi.

,

LALOGALGA, OR FISH POND

For a plan of this ground, see Figure 8.

The tutka oktididja is appointed every four years but always from
the Aktayatci clan. The hilis hoboia are chosen every four years and
ordinarily from the Aktayatci and Bear clans but this is not necessarily
the case. However, if one dies before his four years have expired
he is replaced by someone of his clan. The hilis hoboia also act as
the hilis tealaba.

Ta‘palas are appointed only when needed. There are no special
water carriers, the hilis hoboia calling upon any boys for this purpose
whenever water is required.

28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The two leaders among the women carry atasa, that held by the
first being colored red, that by the second white.

One of the two medicine pots contains pasa; the other the miko
hoyanidja along with wilana, tutka hiliswa (“ fire medicine”), which

Be chars

s a

‘

eucan: N

se”
" TASTANAGIS
‘ AND
ed JIMALAS
~ iS o of o
uy
Sods Ge
= ss
S x} PSs. aes
Q Sy @©° . Q
Ss Vy OS ee ~ Q
S ra al} §
a “ q tN t
ye 2 ON Sen ’
Ss 0 « \ Soo ' 4
x WK 1 Vee il ' &
SoS eek / Ol) § {JO
WSS i a Ki &
zk 1 teas x
9 Ae ' %
Ss x
us :
4

Re nt

>
U

es
6
ALLIGATOR S8EAR

Oe
B

Fic. 8—The square ground of Latogalga, or Fish Pond.

A. Chiefs’ Bed: 1, miko (Raccoon); 2, heniha (or 2d chief) (Wind); 3,
asimbonaia or yatika (head speaker, drawn from any clan); 4, hilis haya (any
clan) ; 5, hilis hoboia.

B. Citizens’ Bed (Tasikaialgi intupa) ; 6, yahaikas (no fixed clan).

C. Warriors’ Bed: 7, tutka oktididja (Aktayatci).

8, medicine pots; 9, boys’ medicine; 10, point where women assembled pre-
paratory to the dance.

grows in wet places in swamps and has red flowers, tcato hatkutci
(four “little white rocks”’), and tutka-tcok-hissi (green moss from
an old fireplace).

If one has eaten new roasting ears he is given a little pasa root
to chew and then a little pasa in cold water of which he must take
four drinks. Then he can take all of the medicines like the others.

no. 8 CREEK SQUARE GROUNDS—SWANTON 29

Some of the medicine is swallowed, the rest spit out. The doctor sees
that the medicines are taken and fines those who neglect to do so.

There is a town policeman called istikona’ha who carries out the
orders of the miko against those who have refused to obey him, and
collects fines from them. The incumbent in 1929 was Maxcy Alakotci.
When I interviewed them they were using the tastanagi as the
town miko.

The Hathagas are the Bear, Wind, Bird, and Alligator; the
Tcilokis are the Raccoon, Aktayatci, Deer, and Potato.

The Aktayatci are said to have formed one phratry with the
Raccoon.

In match games, they played against Tukabahchee, Atasi, Laplako,
Eufaula, Hilibi, and Kealedji. Alabama was formerly of the same
fire but later drew away.

This year (1929) they did not use the pasa. The Asilanabi square
ground is arranged just like this one, but the Okchai differ from these
two a little in the use of their medicines. It is thought that Asilanabi
is older than Latogalga and that the latter branched off in order to
get the extra money that was paid to its six representatives in the
national assembly.

Except that there is more detail, the arrangement given here differs
only slightly from that which I recorded in 1912 (42d Ann. Rep.,
Bur. Amer. Ethnol., p. 236). The only noteworthy divergence is
in the position assigned to the Deer clan by my earlier informants,
but this may be attributed to the fact that the miko apokta was then
a Deer and his clan was probably brought over to the west cabin
for that reason. As we should expect, the agreement is also close
with the arrangement of the Asilanabi ground though there are
minor divergencies in the allocation of clans to the south cabin. All
of my authorities agreed well in assigning clans to the two moieties,
but the oldest of all of them thought that the Beaver and Alligator
were probably Tciloki. This may have been the ancient arrangement.

TURABAHCHEER

The plan of Tukabahchee square appears in ligure 9 and a view of
it as it looked in 1912 is given in Plate 5, Figure 1.

The number of tastanagis is indeterminate. The toba mawidine
were officers not otherwise named who always remained in the cabins.
There is one in each of the 12 beds except a part of the southeast
cabin as indicated. They were selected from any clan, given names
taken from the father’s clan, and seated with the latter. Thus, if

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

a man were a Deer and his father a Raccoon, they would give him
a name from the Raccoon clan’s names and seat him with them.
After being so seated these officers were not obliged to do any further
work connected with the busk.

9

“wy
t
~
Q
<
S

-OL DTSITOM/LV

SNY7I SIOIYA GA
ASO SHAGLILNW

Oo

IKK KE XK

a ae

RACCOON, BEAR
vax Se

BIFfPD 5

'
'
|
'

Fic. 9—Tukabahchee square ground.

A. Chiefs’ Bed: 1, miko (Raccoon); 2, mikos (substitutes) (Raccoon) ;
3, henthas (Wind); 4, henihas and yatikas; 5, hilis haya; 6, ta‘pala.

B. Potato Bed; sometimes also called Youths’ Bed: 7, 7, 7, toba mawidine.

C. Warriors’ Bed: 8, ta‘pala; 9, tastanagi takalgi.

D. Youths’ Bed (Tcibanagalgi intupa): 10, yahaikas; 11, point where women
assemble preparatory to the dance; 12 point where the women start dancing.

The istatcagagis, most of whom sat in the southeast bed, were like
a jury, or like a committee, to settle matters of routine. They com-
prised both men and women. The admittance of women to this select
group is something which I heard from no other informant.

no. 8 CREEK SQUARE GROUNDS—-SWANTON 31

A man was picked out of the bed of the henihas to speak for
others, 7. ¢., the yatikas were selected from the henihas. None was
used in the stomp dance. As payment for his services the yatika
was given a deer hide and a ribbon. When he was wanted by the miko
he was summoned four times, but he did not start until he had
received the fourth call. He set a basket on the ground into which
all of the tasikaias threw bits of tobacco to be taken to the singers
for the women as payment for their services.

The hotibonaia was a special speaker used in the ball game and
in war. The term asimbonaia was about equivalent to that of yatika.
He was a head man chosen from among the tastanagis for almost
any purpose. He listened to any message which the miko wished to
give out and then repeated it aloud.

There were two hilis tcalaba who sat with their clans and were
summoned whenever needed. One was a Raccoon and one a Bear.

There were two oktididja. who sat with their clans or wherever
they belonged until summoned to attend to the fire.

Four hilis hoboia, selected from any clans, were sent to gather
the medicine.

All of the tasikaias shared in getting water.

The famous Tukabahchee plates were taken out and cleaned four
years in succession and then left covered for four years.

The feather dance was discontinued at Tukabahchee when my
informant was a small boy.

The three leading women in the women’s dance carry atasa. The
ieading woman has an eagle feather on her atasa, the second a white
crane feather, and the other the feather of a third bird, perhaps
a goose. They start one at a time and when the third moves all the
rest follow. After they have walked round the fire four times they
begin to dance and circle the fire again six times.

The Hathagas were Bird, Wind, Bear, and Beaver; the Tcilokis
were Turkey, Alligator, Raccoon, Deer, Panther, and any others.

In olden times the moieties were exogamous. The Raccoon and
Potato were then brothers, or rather half brothers, but intermarriage
between them is now common. Even today, however, the Beaver
and Bird will seldom intermarry. The Bird was the “uncle” of
the Beaver and the Beaver the former’s “ nephew.”

A man’s children called his father’s clan “ fathers.” One can say
anything he wishes to, however disrespectful, about his own clan,
but he must not speak against his father’s clan or permit anyone
else to.

32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The general arrangement is the same as that obtained from the
father of my informant in 1911-12 (42d Ann. Rep., Bur. Amer.
Ethnol., p. 244), but the stations here assigned to some of the indi-
vidual officers are probably more nearly correct, particularly the
locations of the mikos.

KEALEDJI

Figure 10 shows the Kealedji square.

The hoktagi immiko, “‘ women’s chief,’ supervised the women’s
dance. There are now no regular water carriers. The imatas
formerly sat in the north cabin, but now there are no officers so
called. The beds are called by the names of the clans which occupy
them. If the asimbonaia is wanted to deliver a speech, he is called
over to the miko’s seat for that purpose. For tastanagis the best
men are chosen. The ta‘palas are changed every four years. In
former times they had a regular four days’ busk, but now it lasts
for but one day and there is no feather dance.

In the match games they played against Okchai, Wiogufki, and
Tulsa.

The Hathagas are Wind, Bear, Panther, and Turkey; the Tcilokis
are Raccoon, Deer, and Aktayatci. In ancient times these moieties
were exogamous.

The names of the present officers are as follows:

TIKOwese Ae na arene orate Kasihta Yahola (Raccoon)
henthaaprorinc Sema eccr ts Heniha Imatutci (Wind)
miilkcoOuapoktamermc cece Kayomulgi (Raccoon)
hoktagi*immikor 2.22. cces =» sok Wiwohka Yahola (Raccoon)
henihaicesaec een eine ere Oikas Hadjo (Wind)
asin bonalay eee nee ee occ Kosa Fiksiko (Bear)
taipalae ac PR ascent: one Talmutcas Hadjo (Bear)

Se he | wc Ce PrN Ror RT Alak Hadjutci (Turkey)
lnilbis: Wemleloey skoknodsecodoucuane Kan Tcati (Wind)

See RY. antares el eae: Itco Imata Fiksiko (Deer)
tutka oktididj¥ ........0000.0.004 Ahali Imata
lapbicy 1oVO\OR), sceacassobcosaonnc Tatk6na Hadjo (Wind)

- Tee Rave ere oe et Hotalgi Fiksiko (Wind)
tAStANACIN ces cnka cree erties Tastanak Tcapko

SN eA eet COM ORIN OO Lotete Tastanakutci

“

(ces ete ieee eR Inheniha (= heniha)
Sse EERE neha eeoeeioererte Tatkona Hadjo

The plan of the revived square ground of the Kealedji agrees
in all essentials with the one given me in 1912 from memory by a
very old man, except that in recent times the Deer clan seems to
have lost its importance or died out. It is interesting to find that old

no. 8 CREEK SQUARE GROUNDS—SWANTON a2
and recent informants agree in assigning the Panther clan to the
White moiety. They differ, however, regarding the position of the

RACCOON

xX
10

AMTAYATC!' TURKEY 617

Fic. 10.—The Kealedji square ground.
A. Chiefs’ Bed:

I, miko (Raccoon); 2, heniha (Wind); 3, miko apokta
(Raccoon) ; 4, hoktagi immiko (Raccoon); 5, heniha (Wind); 6, hilis haya
(usually Raccoon; Wind in 1929); 7, hilis tealaba (Deer and Wind) ; 8, hilis
hoboia (Wind).

B. Youths’ Bed: 9, ta‘pala (Turkey) ; 10, yahaikas.

C. Warriors’ Bed.

D. White Bed: 11, asimbonaia (Bear) ; 12, ta‘pala (Bear).

13, medicine pots (Ist position); 14, medicine pots for those fasting (2d
position) ; 15, medicine pot for boys (2d position) ; 16, medicine pot for girls

(2d position) ; 17, medicine pot for women (2d position) ; 18, point where women
assemble preparatory to the dance; I9, point where women start dancing;
20, ball post.

Turkey, which the earlier informant considered Tciloki and the later
White. But this clan, together with the Alligator and Tami, is known
to have been on one side in some towns and on the opposite in others.

34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

EAPLAKO

Figure 11 shows the plan of this ground.
The two pots of medicine were prepared where they are shown
(nos. 4 and 5) and remained there all of the time. The Bird clan sat

N
©
4

©
J

Fic. 11.—Square ground of Laptako.

A. Chiefs’ Bed: 1, miko (Bear) ; 2, heniha or miko apokta (Bird) ; 3, yatika
(sometimes seated here).

ie Warriors’ Bed. This was usually occupied by middle aged men of various
clans.

C. Citizens’ Bed (Tasikaialgi intupa). Occupied by various classes, includ-
ing persons not properly citizens of the town but taking medicine there, men
married in the town or adopted into it, and men whose fathers belonged to it.

4, a large pot of medicine for the fasters; 5, a small pot of medicine for the
women and children; 6, rack for ball sticks.
anywhere. The yatika belonged to no regular clan, and at the present
time there is no such officer. There are two hilis tcalaba who occupy
no special seats, and one hilis haya who is taken from any clan of,a
friendly town. Towns belonging to the other town moiety do not

no. 8 CREEK SQUARE GROUNDS—SWANTON 35

take medicine with them and in fact are not invited except in the
case of individuals married into the town.

The Hathagas are the Bear (the most important of them), Bird,
Beaver, and Wind; the Tcilokis are the Raccoon, Deer, Potato,
Aktayatci, Panther, Alligator, etc. Here the Alligator, Turkey, and
3eaver can intermarry; in Eufaula they could not. But probably this
means that the Beaver could intermarry with the other two, as the
Alligator and Turkey were seldom allowed to marry under any cir-
cumstances. The Bird and Beaver were classed together in Laplako,
but this group does not include the Bear or Wind. The Hathaga clans
had the reputation of being progressive while the Tcilokis were full-
bloods and reactionaries. This characterization probably followed the
coming of the white people.

The towns of the opposite group were called Talipota, which means
“ foreign but not unfriendly.”

The word Laptako indicates a place where there are many marshes
filled with canes. This town square, which had been discontinued,
was revived in the year 1903 in this way. They had to prepare a
ground in order to take medicine before a game with the Nuyaka
Indians next year. Later this was improved with regular cabins, but
it must have been inconspicuous or have been considered unimpor-
tant, as I heard nothing about it in 1912.

Lapltako and Atasi are now nearly fused on account of the number
of marriages between individuals belonging to them. Before the
Civil War the Laplako had an Atasi Indian named Hotalgi Hadjo
married among them as their hilis haya. One of the great men of
Laptako in former times was Jim Boy (Tastanagi Imata) whom my
informant remembers to have seen. He thinks he died just before
the Civil War broke out. McKenney and Hall ( History of the Indian
Tribes of North America, vol. 2, pp. 71-74) give a portrait of this
chief and a considerable account of his life. He was born in what is
now Alabama in 1793 and accompanied the warriors of his town
during the Creek War of 1813-14, but was too young for active
participation. In the war with the Seminole he was one of the leaders
of the Creek contingent which aided the Americans. The exact date
of his death seems to be unknown.

The arrangement of the modern ground differs more from those
described to me in 1912 (42d Ann. Rep., Bur. Amer. Ethnol., pp.
254 and 255) than any of the others. The first of the latter was
obtained from a very old man who should have known the ancient
arrangement well, but of course my interpreter and I may have
misunderstood him. The cabins are at different points of the compass

30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

and more confusion is shown in the later organization, but they
agree in stating that the miko belonged to the Bear clan and in
placing that clan in the south cabin. My older informant allocated
this clan with the Tciloki. This was probably an error on his part
but he seemed to insist upon it.

They have had no busk since the Civil War and no women’s dance,
and the pasa is no longer used, only the miko hoyanidja. The dances
in the square are three stomp dances. Seven days before one of
these the “broken days” are sent out and on the day when the
sixth stick is thrown away they are all to be at the ground, while
on the seventh they are supposed to be taking medicine. Seven days
before the dance they also meet and pick out four clean young
men, called hilis hoboia, men whose wives are not pregnant, who
are not given to intoxicants, and who have not attended to the
digging of a grave during the preceding month. These men gather
four bundles of medicine (miko hoyanidja) which they lay down
with their tip ends toward the west. First they spread out a bed
of leaves called lodja issi, “ turtle leaves,’ which should be taken
from hickory trees. The medicine is laid on top of these and more
leaves are spread over it. On the morning of the fast day all of
those who are to take medicine are supposed to present themselves
at the square ground. The fire is built up so that it will not go out
all day. Early in the morning the two hilis tcalaba prepare the
medicine, first the medicine for the women and children and then
that for the adult men. This is taken four times during the day,
the fourth time between one and two o’clock. Before they take the
medicine the yatika announces, on behalf of the chief, that each of
those intending to take it is to get a stick and throw it into a blanket.
These sticks are counted and the yatika announces the number. Four
times (or sometimes twice) during the day the men who are to
take medicine with the exception of the officials (the mikalgi, yatika,
hilis haya, and the two ta‘pala) go out to get firewood so as to be
ready for the dance that night. Four men are selected to keep watch
of the fasters during the night, to see that no one sleeps or breaks
his fast, or drinks or goes with a woman. That is why the numbers
are taken. These four men are called istikona’ha, “men taking
away,” because they take away the hat of anyone found sleeping
(and treat similarly anyone who breaks the taboos in other ways).
The hat they carry to the miko and, when the owner comes to get it,
he is fined.

After they have taken medicine for the fourth time, they go to
the creek for a bath and then return to the square. The yatika

no. & CREEK SQUARE GROUNDS—SWANTON . 37
talks to them telling them to take care of themselves all of that
night, and then they scatter to the camps to eat. Before sundown,
however, they are supposed to be back on the ground. ‘The yatika
calls to them four times and by the fourth time they are to be in
their seats. When all are in their places, the two ta‘pala are selected.
The yatika makes a speech on behalf of the miko, calling upon his
hearers not to act in an unfriendly way toward the outside friends
who are about to be admitted to the dance, not to take liquor, to
behave themselves all through the night, not to fight, and so on.
This speech is addressed to the town people and outsiders alike.
After it the fasters dance four times and then the visitors are
admitted. That “kills the fast.” All through the day the square is
to be kept clean with the idea that the fasters will in consequence
be clean. They go up to take the medicine two by two, and those
who are ceremonially unclean take the medicine last. They dance
until about daybreak. Then all belonging to the town go down to
the branch and bathe, after which they return and sit in their
respective beds. Then a collection may be taken to defray their
expenses and they settle other matters. The night before, the chief
of a friendly town may have announced a dance, and, if so, the
announcement is now made and advice given as to how they are
to help their friends. This advice is uttered by the yatika, speaking
for the miko, and he then gives a general talk, advising his people
not to use liquor, not to break the laws, and to be good citizens in
every sense of the word. Then they disband for the year.

HILIBI

A plan of Hilibi square is shown in Figure 12 and a general view
of the ground as it appeared in the winter of Ig11-12 in Plate 5,
Figure 2.

Two poles with white feathers attached were at each front post.
These were used in the “ feather dance ” and were called ‘ the path,”
because the path was to be white.

When there are visitors the owners of the east cabin move
elsewhere in order to make room for them. When the south cabin
is overcrowded, some of its occupants move into the east cabin. In
this town the clans were always considerably mixed up in the beds.
The ta‘palas can sit anywhere. The tutka oktididja does not have
a particular seat on account of his official position. He is appointed
every four years. The tutka didja, who starts the fire, is identical
with the hilis haya.

38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL. 85

If a man whose father is of the Aktayatci, the miko’s clan, is given
a busk name, he is brought to the mikos’ bed and given a seat there.

The Hathagas are the Bear, Bird, Alligator, Beaver, Turkey, and
Wind; the Tcilokis are the Panther, Potato, Deer, Aktayatci, and
evidently the Raccoon. The Alligator and Bird are considered as
practically one clan.

The two leaders in the women’s dance carry atasa.

drs

Fic. 12.—Square ground of Hilibi.

A. Chiefs’ Bed: 1, miko (Aktayatci); 2, heniha (Alligator); 3, hilis haya
(Raccoon in 1929); 4, ta‘pala (Bear); 5, hilis tcalaba (Bear in 1929).

B. Raccoons’ Bed: 6, hilis hoboia; 7, tutka oktididja (Raccoon in 1929) ;
8, yahaikas (any clan).

C. Cabin used by visitors: 9, ta‘pala (Alligator) ; 10, hilis hoboia (Alligator).

II, medicine pots; 12, medicine pot for boys; 13, medicine pot for women;
14, woodpile; 15, ball post.

The history of this town is as follows. It was founded by a man
of the Aktayatci clan who went off to live by himself and then put
up a ball post. Many women belonged to his family and he had
numerous visitors, some of whom married these women so that it
soon grew into a large band of which he was probably the first miko.
Anyhow that is the way the Aktayatci came to have the town. After
stomp dances had been held there for some time, more visitors came

no. 8 CREEK SQUARE GROUNDS—SWANTON 39

to join them and it grew still larger. Because it was built up very
rapidly, its founder called it Hilibi, which means “ hurry ” (hila’pkis,
‘“T make haste”). Since it was an “ illegal” band, a talwa fatcasigo
(“town deviating from correctness’”’), all the clans do not have
regular places, having been drawn from so many other bands.

The following notes on some personal names contained in the town
roster give an interesting insight into the manner in which totemic
names were bestowed :

Fos Hatki Imata (“ White Bird Imata’’), so named because he
belonged to a White clan, the Wind.

Halak Hopaie. He belonged to the Bear clan, but his father
belonged to the Raccoon, hence the name Halak or Ahalak (“ Po-
tato’’), since both these clans are Tciloki.

Pahos Fiksiko. He belonged to the Wind clan, but his father was
a Deer and the Pahosa is of the same phratry as the Deer.

As shown by the native story above given this town was not
supposed to be ancient or to have a firmly fixed town organization,
and, while it was older than the Hilibi people themselves believed,
it seems to have preserved the irregularities which might naturally
be associated with a new town. Not improbably the tradition of
irregularity preserved the fact. At the same time there is a general
agreement between the plan here given and that which I obtained in
Toi2 (42d Ann: Reép., Bur. Amer. Ethnol., p» 258): The mikos’
cabin is to the west, and the mikalki and henihalgi were of the same
clans, Aktayatci and Alligator respectively. The Alligator and Tur-
key were classed as Whites from association with the Bird by
my later informants, but the earlier ones gave them as Tciloki.

ALABAMA

This was a very simple square of exceptional arrangement as shown
in Figure 13.

The Alabama were one of the incorporated tribes with a language
distinct from Creek. The clan names in Alabama are: Mahaleha
(Wind), Sawaha (Raccoon), Aktayatciha (Aktayatci), Hatcuntco-
baha (Alligator), Konoha (Skunk), Nitaha (Bear), Kotha
(Panther), Fociha (Bird), Fitoha (Turkey), Ofataha (Beaver).

All of the officers were brought over to the chiefs’ bed. Men of
the Alligator, Bear, and Aktayatci have been mikos and a man of
the Skunk clan was once the heniha. There is no regular rule for
either position.

The Hathagas were the Wind, Bear, Panther, Skunk, and Raccoon ;
the Tcilokis were the Bird, Beaver, Turkey, Alligator, and Aktayatci.
The Wildcat was the same as the Panther.

4O SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

One of the greatest Alabama Indians now remembered was
Kantcati Yahola (Alabama name Tcisoki), who was the hilis haya.
He was born in Alabama and came west with his tribe. He lived
until about 1866.

The above plan and the three others I obtained in 1912 and the
years immediately following (42d Ann. Rep., Bur. Amer. Ethnol.,
pp. 263-264) show considerable minor variations but all agree in
locating the chiefs’ bed in the east and the warriors’ bed in the west.
Most of them also place the Bear and Panther clans in the latter
and the Wind, Aktayatci, and Deer in the former, where they are
noted at all. In allocating the clans the above informant agrees with

Oo
PANTHER

BIRD,

TORREY
AND

BLAVER

[THE WORTH ii
END OFRIGI-

NALLY BELON
GEO TO THE
BEAR CLAN)

oO

B

Fic. 13.—Alabama square ground.

A. Chiefs’ Bed: 1, miko (Wind); 2, heniha (Raccoon).
B. Warriors’ Bed.

the older ones except regarding the Panther which the men first
consulted asserted was Tciloki while it is here given as a White clan,
but this is a clan which has been placed on both sides.

KASIHTA

For the plan of Kashita square ground, see Figure 14.

There should be two hilis tcalaba, drawn from the Alligator clan,
but they are not employed now. Four hilis hoboia for the pasa, four
for the miko hoyanidja, and one tutka oktididja are chosen by the
miko without reference to clan. There is no definite body of water
carriers. Two ta‘palas are selected from any clan to serve just for
the night. They carry sticks called reels, and their function is to
invite the dance leaders to lead dances and see that all take part.
There is only one singer for the women’s dance. He sits behind
the miko.

no. 8 CREEK SQUARE GROUNDS—-SWANTON Al

In rgor the old square ground was given up and, the new one was
established in June, 1920. Because the new generation was weak
one cabin was cut out. The Coweta square is said to have been the

same as that of Kasihta.
Every time anything is brought in or anything repaired they dance

ll night because thus the two things are joined together, just as

0 — SS? oO
TASTANAGIS RACCOON,
Ano AMTAYATC!
MIALAS AND PANTHER
{= 1
Cc 5

SEAR
a TREES

o ‘
‘

Bien f POTATO, s

‘
‘
t
t
' ‘
'
'
'
; i
266
3
: iW
E LLIGATOR ))
\ St * ‘
: F |
; t
'
‘
'
0
'
‘
' tf WIND ANS B/FAD BEAVER 4° DEER 5

12
®

Fic. 14.—Kasihta square ground.
A. Chiefs’ Bed: 1, miko (Bear); 2, heniha (Alligator); 3, hilis haya (seat
Ww Fa on duty) ; 4, asimbonaia.
Whites’ Bed (Henihalgi intupa), or “ Cabin of the Greater Kings.”

c Warriors’ Bed: 5, asimbonaia (Aktayatci).

6, medicine pot; 7, point where women assemble before dancing; 8, trash
heap; 9, where ashes of old fires were placed; 10, split log where are seated
those young men who have broken the rules and are in consequence placed
here as a punishment before being allowed to mix with the rest of the people;
11, line of sweepings (tadjo) marking limit of square ground; 12, ball post.
two days are united by the night on which they dance. Formerly
dances were held every Saturday night, but hard times have put an

end to them. The dance which was being held when I paid my visit

was because they were then putting up a ball post.
The Hathagas are the Wind, Beaver, Bird, and Deer; the Tcilokis
are the remaining clans, yet it is said that the Bear, Alligator, and

Mae

42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85
Wind were of the same “ class.”
marry outside of third cousins.

The Kasihta defeated the Coweta three times in the ball game,
the last time in 1878, and after that they took them under their
jurisdiction and they have played on the same side.

Comparison of the plan of the new Kasihta with the plans of the
earlier grounds secured by Gatschet and myself (42d Ann. Rep., Bur.
Amer. Ethnol., pp. 266-268) shows that the old order has been fairly
well maintained allowing for the disappearance of some clans,
particularly the Fish clan. Incidentally I wish to correct an erroneous
statement in my report in which I misquoted Gatschet to the effect
that the Kasihta miko belonged to the Alligator clan. While the
Alligator clan occupied half of the chiefs’ cabin, the miko himself
has always been taken from the Bear and was stated to have been
so by Gatschet. The allocation of clans to the moieties by my recent
informants contains a number of difficulties, for they seem to have
placed the Bear among the Tciloki, which is unlikely, especially as
it was said to belong to the same “class ” as the Alligator and Wind.
My own earlier authorities also classed the Beaver and Deer as
Tciloki.

The following information will be interesting to those who wish
to study acculturation processes in intangibles. It was told me by
the Indian considered best able to speak for the town.

Under the present law they can all

The original four cabins represented the New Jerusalem with its 12 gates.
The busk goes back to the time when Jacob set up the altar at Bethel and is
traced from him and his 12 sons. All of the Indians in America entered in two
migrations, one at the time of Jacob (1500-2000 B. C.) and the second 600 years
later, at the time when Jerusalem was destroyed by Nebuchadnezzar. Then
they talked face to face with the Great Spirit because they were more obedient
to God than any other tribe, but about 700 years after the Messiah they got
away from the original law on account of desire for riches. Then they lost the
old law and asked for a new government, and by holding a ceremony in mid-
summer, in the month of July, it was given to them. That new law taught
them to tell the truth and be honest with their fellowmen and to raise their chil-
dren in such obedience, not to touch anything that did not belong to them, not
to make a false statement. That is the law which we are trying to follow.

He added:

We have a hard time because the white men have failed to fulfill their part
of the agreement. They have strong laws that we can’t begin to understand and
our customs are about choked out through grafters who claim to have bought the
claims of the allottees on which our squares are located. In order to hold their
grounds several towns have to pay rent year after year. If the law makers
would cooperate with us and give us full privileges we would raise more sub-
stantial, law-abiding young men and young women. That was the custom and
the wish of our forefathers.

no. 8 CREEK SQUARE GROUNDS—SWANTON 43

OCHESEE SEMINOLE

This is the only Seminole ground from which I obtained informa-
tion during my recent trip. The plan of it is shown in Figure 15.

There are two pots of medicine for the miko hoyanidja and pasa
respectively, but the latter is introduced only at the busk. The women,

N

©
7

DEER, PANTHER,
POTATO ALLIGATOR

(/NCLUDING TURKEY)

ee oe Ne Ss
” . Gre
ve ‘ q =
‘ ‘ S Q\
° aa | Cee
\ ‘ i aw
i : Ah SN
eee a7 x a &
- k ~
~ aK
\ ‘
ae \

AKTAYATC/,
WIND, SNAKE

AND HAPITCA
(2)

B

Fic. 15.—Ochesee Seminole square ground.
A. Chiefs’ Bed: 1, miko (Bear) ; 2, heniha (Bird) ; 3, yatika (Deer in 1929) ;
4, hilis haya (Wind in 1929).
B. Warriors’ Bed: 5, yahaikas.
C. Warriors’ Bed.
D. Youths’ Bed.

6, medicine pots; 7, medicine pot for boys (containing red root) ; 8, point where
women gather before dancing; 9, point where women start to dance; 10, ashes
of old fires; 11, ball post.

however, use wormseed (“ wilana”’). After they are through with the
medicines, whatever is left is poured out at the place where they have
been taken.

The miko is chosen if possible from the Bear clan, but if they do
not find a suitable man of that clan, they select someone from the
Beaver clan.

44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL.. 85

The yatika and hilis haya may be selected from any clan, the former
being chosen for his oratorical gifts.

There is now no heniha. The last they had went to live with the
Creeks and did not come back.

Two ta‘palas are used in the women’s dance and hold their position
for four years. Other ta‘palas are chosen temporarily for the other
dances. In the stomp dances they change these ta‘palas several times
during the night.

There are four hilis hoboia taken from the west, north, and south
cabins but from any clan. They keep their positions as long as they
choose to serve.

There is one tutka didja who can be of any clan.

Five or six boys bring water to the ground.

The two leaders among the women carry atasa which are painted
with white clay annually just before the dance. Women do not take
the wilana internally; they merely wet their faces and other parts
of their bodies with it. The boys use only the miko hoyanidja.

The ashes of the old fire are removed from the square and the
new fire lighted on the morning of the fast day.

The Hathagas are the Bear, Bird, Beaver, Wind, Otter and Skunk;
the rest are Tcilokis. The Bird and Beaver belonged in one phratry,
and so did the Wind, Otter and Skunk; the Alligator and Turkey ;
and the Aktayatci, Kapitca, and Snake.

The name of the present miko is Nokos Miko, and his father
belonged to the Deer clan. The last heniha was named Henitha Miko,
and his father was of the Bear clan. The busk name of the yatika
(my informant) is Pahosa Tastanagi; his father belonged to the
Aktayatci, and his father’s father to the Bear.

They have a ball post surmounted by a wooden fish. A hit on this
fish counts 4, and on the post above a certain mark 2. There have
been no regular match games between towns in the lifetime of my
informant, but about two years ago the old men and the young men
played against each other.

The above plan of the Ochesee ground agrees closely, naturally
enough, with that which I obtained from the man whose advice was
particularly resorted to in reestablishing it some years ago (42d Ann.
Rep., Bur. Amer. Ethnol., p. 283), for in 1912 it had been given up.
The main differences are in the seating of the Potato and Alligator
clans. In the matter of the moieties the only change is in the case
of the Raccoon clan, which I previously set down as White. This is so
exceptional, however, that I have always believed that I must have
misunderstood my informant.

No. 8 CREEK SQUARE GROUNDS—SWANTON 4

wm

GENEALOGIES

The two brief genealogies which follow will illustrate in some
measure the influence of the clan system on marriage. The first is
the genealogy of Jeff Canard of Laplako; the second that of Earnest
Gouge of Hanna, the latter of especial interest because it includes
the famous orator Hobohit Yahola.

paternal paternal maternal maternal
grandfather grandmother grandfather grandmother
(Bear) (Alligator ) (Wind) (Beaver )
father mother
(Alligator ) (Beaver )
a
self
(Beaver)
paternal
grandfather maternal
(this was paternal grandfather maternal
Hobohit Yahola) grandmother (Potato of grandmother
(Potato) (Raccoon) Abihka) (Bird)
a
father’s = father mother
second (Raccoon) (Bird of Otciapofa )
wife __— = = x -
(Raccoon?) self wife’s wife’s
(Bird of father mother
Otciapofa) (Bear) (Alligator )
ee
wife
(Alligator)
CONCLUSION

The busk undoubtedly represents a long period of development,
and as the stages through which it passed and the elements entering
into it have been lost, we can never understand it fully. Its main
purpose, however, was evident. It was to restore the connections
of the tribe with the universe which a year of civil or profane living
had tended to rupture. Hence the new fire, extracted from its
abiding place in the wood and not as yet sullied by contact with
humanity. Hence the rigorous fast accompanied by administration
of medicines divinely revealed to the ancestors, the general pardon
of offenders, the sabbath calm prescribed for that period of
regeneration, the use of white paint, and the employment of the
term white—‘ the white day,” “the white smoke,” “the white
drink ”—in various parts of the ritual. The one discordant note
seems to have been provided by the women’s dance, since the leaders

46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

of that dance carried representations of war clubs, which there is
every reason to believe were anciently adorned with scalps, and some
of these were painted red. But I suspect that this dance was an
attempt to represent war as a protective institution and to thank the
being or beings who preside over human destiny for having so well
defended them against assaults of the—as usual—perfidious foe.
Possibly some element of propitiation also entered into this dance.

The universe with which the Creeks sought reconciliation was not,
however, a material one. What they had in mind was rather the
mind or minds believed to be operant there. While we know of some
supernatural beings connected primarily with the busk and numerous
spirits associated with natural objects were anciently believed in,
it seems fairly certain that the peculiar patron of the ceremony was
a solar, or rather celestial, being generally called Hisagita-immisi,
“the breath controller,’ and also Ibofanga, “the one above,” and
that the busk fire was in some way an earthly representation of the
great solar fire overhead. While it is probable that Hisagita-immisi was
not in ancient times the monotheistic deity he has now become, there
is every reason to think he was already, before White contact, the
supreme being of the Creeks.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL... 85, NO. 8; PEs 1

1. Abihka or Talladega square ground.

2. Tulsa square ground from the southeast.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL.|85, NO=8, Pie

1. Nuyaka square ground.

2. The three beds of Pakan Tallahassee, from the southwest.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, (NO.-8; PL. 3

1. Pakan Tallahassee square ground looking between west and south beds, show-
ing chunkey yard and ball post.

2. Wiogutki square ground from the southwest.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 8, PL. 4

1. Okchai square ground in winter from the southwest.

2. Fish Pond square ground.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLE. 85; NO. 8. -PE. 5

1. Tukabahchee square ground.

—

2. Hilibi square ground in winter from the southwest.

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VOLUME 85, NUMBER 9

Roebling JFrund

THE DETERMINATION OF OZONE
BY SPECTROBOLOMETRIC
MEASUREMENTS

(WiTH THREE PLATES)

BY
OLIVER R. WULF
Smithsonian Institution and U. S. Bureau of Chemistry and Soils

(PUBLICATION 3127)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
NOVEMBER 30, 1931

Tbe Lord Waltimore Dress

BALTIMORE, MD., U. S. A.

Roebling Fund

THE DETERMINATION OF OZONE BY SPECTROBOLO-
METRIC MEASUREMENTS

By OLIVER R. WULF
SMITHSONIAN INSTITUTION AND U. S. BUREAU OF CHEMISTRY AND SOILS

(With THREE PLateEs)

In the fall of 1930, at the suggestion of Dr. C. G. Abbot, measure-
ments of the transmission of visible light by ozone were made on the
solar spectrometer of the Smithsonian Institution at Table Mountain,
Calif. This determination of the absorption of ozone in the region
of its very weak absorption, practical in the laboratory only by the
method of photographic photometry, by which it has been done by
Colange,’ can be accomplished by direct energy measurements on the
spectrobolometer because of its great sensitiveness and the extreme
intensity of the source. At the same time fluctuations in weather
conditions are likewise registered with great sensitiveness and consti-
tute a serious source of error in the measurements. But on the other
hand it is favorable that observations may be made of the absorption
of controlled amounts of ozone placed in the sun’s beam with all other
conditions of operation identical with those of the regular solar
observations. One of the useful results of this work has been the
selecting of a large number of points on the solar bologram which may
be satisfactorily used for ozone determination. The present paper has
to do for the most part with the results of these measurements as
forming a basis for the determination of atmospheric ozone from
solar bolograms and not with the results of their application.

The essential apparatus used in this work auxiliary to the spec-
trometer is shown in Plate 1. A cylindrical glass cell, 20.0 cm. long
and approximately 13 cm. in diameter, was mounted in the front
room of the spectrometer tunnel in such a way that it could be easily
moved in or out of the solar beam as it came from the coelostat mir-
rors on its way to the first slit of the instrument. The circular windows
of this cell were made of high-grade plane plate glass. These windows

*Colange, G., Journ. Phys. et le Rad., ser. 6, vol. 8, p. 254, 1927.

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 9

2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

of the cell were covered with diaphragms having vertical rectangular
openings sufficient to pass a beam that considerably more than covered
the first slit of the spectrometer. The absorption cell was sealed with
all glass seals in connection on its inlet side with the ozonizers and
on its outlet side with an analytical apparatus for ozone. No altera-
tions of any sort were made in the regular observing conditions except
for the interposition of the cell. Tank oxygen was supplied to the
ozonizers under a small constant pressure through the capillary of a
glass flowmeter having nujol as the manometric liquid, this flow-
meter being calibrated in a series of independent measurements. The
exit ozone was analyzed by the method described by Wulf and Tol-
man. The samples, whose volumes were known from the time and
rate of gas flow, were collected over potassium iodide solution, set
aside, and subsequently analyzed. It is evident that, in the filling of
such a cell by sweeping at the low rates of gas flow necessary in such
a system for producing ozone of relatively high concentration, a con-
siderable amount of time will elapse before the exit gas attains practi-
cally the concentration of the entering gas. To study this circumstance
a tube was also brought to the analytical apparatus from before the
cell as the gas came from the ozonizers allowing a sample of the inlet
gas to be taken, and by means of this it was possible to determine the
time sufficient for the exit gas to rise nearly to the concentration of
the inlet gas. The concentration of the exit gas was taken, in view
of the processes of diffusion and mixing going on in the cell, as
representative of the ozone concentration in the cell. The ozonizers
were of the familiar silent discharge type. For part of the work one
alone was used, while for the rest of the work two were used in series
and both water-cooled, these giving the largest concentrations em-
ployed. Ordinarily over the period of the taking of four ozone bolo-
grams six analytical samples of the ozone were collected.

Knowledge gained from earlier work on ozone permits a descrip-
tion of the character and position of the absorption.” A small fraction

*Wulf, Oliver R., and Tolman, Richard C., Journ. Amer. Chem. Soc., vol. 40,
p. 1650, 1927.

“a. Colange, G., Journ. Phys. et le Rad., ser. 6, vol. 8, p. 254, 1927.

b. Wulf, Oliver R., Proc. Nat. Acad. Sci., vol. 16, p. 507, 1930.

c. Ladenburg, Erich, and Lehmann, Erich, Ann. Phys., ser. 4, vol. 21, p. 305,
1906; Verh. Deutsch. Phys. Ges., vol. 8, p. 125, 1906.

d. Schoene, E., Journ. Russ. Phys.-Chem. Soc., vol. 16, pt. 9, p. 250, 1884;
Journ. Chem. Soc., vol 48, pt. 2, abstracts, p. 713, 1885; Chem. News, vol. 69,
p. 289, 18094.

e. Chappuis, J., Ann. l’Ecole Norm. Sup., ser. 2, vol. 11, p. 137, 1882; Compt.
Rend., vol. 91, p. 085, 1880; Compt. Rend., vol. 94, p. 858, 1882.

NO. Q DETERMINATION OF OZONE—WULF 3

of the solar energy will be cut out by the ozone as is illustrated in
Figure 1, which represents a normal spectral energy curve roughly
similar to that of the sun showing the approximate area removed by
the atmospheric ozone in the visible.

However, the actual observing of this reduction of intensity must
be made on the complicated solar curve as shown in Plate 2, which is
composed of a series of typical bolograms notched by many Fraun-
hofer absorption lines. The discontinuities in the curves are due to
the insertion of rotating sectors in the path of the solar beam to cut
down the intensity in the regions of great intensity, to values such
that the galvanometer deflections will still fall upon the photographic

A

Fic. 1.—Spectral energy curve of black radiator, approximating that of
the sun, showing the atmospheric ozone absorption in the visible.

plate on which these deflections are being continuously recorded in
the form of these curves. In addition, at two points a shutter is
inserted for the purpose of determining the base line, that is, the line
of zero deflection. Owing to the scattering of the rays by the earth’s
atmosphere the apparent maximum of the sun’s intensity is shifted
to longer wave lengths. The strong atmospheric absorption in the
deep red is conspicuous, while over the region of ozone absorption
chiefly the Fraunhofer lines are in evidence.

In spite of these complexities, however, comparison of two bolo-
grams, one without ozone in the cell and one with ozone, should show
the reduction in intensity caused by the ozone, providing weather
conditions remained sufficiently constant between the two. If, for
example, we were to take the ratio of ordinates at corresponding
points on two such bolograms under ideal conditions this ratio should

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

be unity in regions not influenced by ozone, and somewhat greater
than this where the intensity was reduced by the ozone absorption.
In view of the exponential decrease in intensity of light passing
through an absorbing medium, it is the logarithm of the ratio of
ordinates which is proportional to the amount of ozone in the path.
This will be zero evidently for those cases where the ratio is unity,
and positive or negative as the ratio is greater or less than unity. This
proportionality is based on the assumption that Beer’s law may be
used here. There appear to be no causes concerned with the apparatus
and structure of the spectrum which could bring apparent false devia-
tions from this law, since the spectrum is composed of broad diffuse
bands,’ or perhaps more accurately fluctuations in the absorption co-
efficient without any apparent discontinuities. The question as to
whether the ozone absorption actually obeys Beer’s law is an impor-
tant one and, so far as the author is aware, has not as yet been
satisfactorily answered, although attention has been called to it by
Ladenburg.” However, this assumption is contained in all previous
determinations of ozone such as we are employing here and must be
similarly contained in the present work. Actually, instead of a single
bologram in these measurements, four consecutive bolograms were
taken with the cell containing no ozone and four more with the cell
containing ozone. The average values of the ordinates at correspond-
ing points were compared.

A difficulty enters, however, because of the time which elapses
between the taking of the ozone bolograms and the uninfluenced ones
which we will call the oxygen bolograms, since a very appreciable
change in air mass, or amount of atmosphere traversed, occurs be-
tween the two as well as possible weather changes. This was unavoid-
able with the apparatus at our disposal at the top of Table Mountain
because of the nature of the ozone technique, and is not easy to avoid
under any circumstances. Under these conditions, the ratio of
ordinates outside of ozone absorption would be far from unity and
the logarithm far from zero. Standard correction for air mass was,
therefore, applied to the average ozone ordinates to bring them to the
air mass of the oxygen observations. It is hardly possible, however,
to make correction for all differences between the two sets with
accuracy sufficient to bring the logarithm of the ratio of ordinates to
zero within a quantity small compared to the small ozone effect, and
it is not necessary because it is a difference in the logarithm for
different wave lengths that is of interest.

* See footnote 2) on page 2.
*Ladenburg, R., Gerlands Beitr. Geophys., vol. 24, p. 40, 1920.

NO. 9 DETERMINATION OF OZONE—WULF 5

The points shown on the bolograms were chosen as far as possible
with respect to the known position of the ozone bands. These bands
are shown in Plate 3 as they were obtained in the work described in
footnote 2b, on page 2. The attempt was made to find points on

1000 5000 6000 7000 8000 9000

Fic. 2—Diagrammatic example of ozone area.

the solar bologram suitable for measurement which lay in every ozone
absorption maximum and minimum.

From the above considerations, if a plot of the logarithm of the
ratio of ordinates is made against wave length, the type of result to
be expected is something of the form shown in Figure 2. The failure
to make accurate correction for changes in atmospheric transparency
or sensitiveness of apparatus will result in the points outside of ozone

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

absorption being something greater or less than zero, here illustrated
as less than zero, but this will not obscure the increased value of the
ratio in the region of ozone absorption showing as an increase above

4000 5000 6000 7000 8000 3000

Fic. 3.—Typical ozone area.

the smooth curve passing through the values of the points on both
sides of the absorption. It is the area under this curve above the
smooth curve which is closely proportional to the amount of ozone
in the path.

Figure 3 shows the results of one typical day’s observations. For
clearness of illustration Figure 2 was made similar to this. In this case

NO: 9 DETERMINATION OF OZONE—WULF Wi

an amount of ozone approximately equal to the average atmospheric
quantity was present in the cell. Failure to correct exactly for all
differences between the two sets of bolograms led to the smooth curve
of the points in the absence of ozone absorption being below zero, but
the area due to ozone appears clearly and can be measured with
sufficient accuracy to be of much use. Several days’ determinations
were made with larger quantities of ozone in the path, giving areas
which can be measured with greater percentage accuracy. On other
days smaller amounts were used in order to observe how well such
areas could be determined. Included in figure 3 are also three points
lying far off the curve which are illustrative of unsuitable points.
They lie in weak atmospheric water absorption in the red, which ab-
sorption may vary considerably over short intervals of time, rendering
the points evidently unsuitable for aiding in ozone determination.

The area under the observed curve should evidently be limited
between ordinates whose values are still large compared with the un-
certainty in placing the base line. In the blue this has been taken as
the value at 4750 A, while in the red it was necessary to terminate
the area at 6135 A because of the uncertainty in the point at 6335 A.
This limitation was caused by instrumental circumstances which can
be altered in the future to include a greater area.

These areas were determined for 10 independent sets of observa-
tions, the amount of ozone in the cell being known in each case from
the analytical work carried on at the time of the measurements. From
these results, shown graphically in Figure 4, a value sufficiently accu-
rate to be useful can be had for the amount of area per unit path
length of pure ozone at o°C and one atmosphere pressure, the
common meteorological form of expressing atmospheric ozone. The
least-squares solution, assuming the ozone concentration values essen-
tially free from error compared to values for the areas, which method
automatically weights the individual values in proportion to the area,
leads to the result 21.5 sq. cm. of area per mm. of ozone at standard
conditions of temperature and pressure when a plot is made to the
scale ordinates 0.001 per cm., abscissa 200 A per cm. This area can
thus be stated as 4.30 A independent of the scale to which it is plotted.
Plotting the data and planimetering the area is a procedure which has
the decided advantage of giving a visual record of the amount of
ozone which can be judged approximately at a glance. From the
results of Colange’s data on the absorption coefficient of ozone one
can compute this same area, and it may be estimated directly from his
published curve of the absorption coefficient. One finds thereby that
the value obtained in the present work is about 4 per cent higher than

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

that given by Colange’s results, a difference which lies within the
limits of error of the present work. With a larger number of sets of
observations than the 10 of the present work, and particularly at high
ozone values, the accuracy of the determination could, of course, be
greatly increased.

AREA

/ 2 3 4 5 6 7 8
PATH mm 0,

Fic. 4.—Character of the results of the absorption of ozone.

Anarea such as that illustrated in Figure 3, and pertaining to atmos-
pheric ozone only, can be obtained from atmospheric transmission
coefficients, if combined with the knowledge that there is no important
atmospheric absorption except ozone across this spectral region. That
atmospheric transmission coefficients show unmistakably the ozone
absorption has been pointed out and used by Fowle and others.’

*Fowle, F. E., Smithsonian Misc. Coll., vol. 81, no. 11, pp. 1-27, 1929. Caban-
nes, J., and Dufay, J., Journ. Phys. et le Rad., ser. 6, vol. 7, p. 257, 1026.

NO. 9 DETERMINATION OF O0ZONE—WULF 9
Figure 5 shows an area analogous to the one shown earlier, deter-

mined from the transmission coefficients for March 24, 1929, as a
typical day used simply as an illustration. The logarithms of the

0.025

0.0/0

0.005

°

0 ; 7
°

co 4
|

4000 5000 6000 7000 8000 5000
Fic. 5.—Ozone area for March 24, 1920.

transmission coefficients for this day, read for all the wave lengths
selected in this work, were plotted against wave length and a smooth
curve passed through the values lying to both sides of the ozone
absorption, and the differences of the points lying in the ozone region
from the smooth curve were read. These differences were then
plotted against wave length to the scale used above, yielding the area

IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

shown in Figure 5. Thus the ordinates are the logarithms of the ratio
of the intensities before and after passing through the atmospheric
ozone. Referring this area so obtained to the results of these present
measurements pictured in Figure 4 affords a method of determining
the atmospheric ozone based on direct intensity measurements, and
yields for this day over Table Mountain, Calif., an ozone value of
3.8 mm., while the value given by Dobson for Table Mountain on this
same day was 3.42 mm. A somewhat more satisfactory extension of
this method of determining the ozone transmission may be employed,
whereby the curved base is reduced to a straight line. Instead of
plotting the logarithm of the atmospheric transmission coefficients
against wave length and determining the difference of the observed
points from the smooth curve in the ozone region, the logarithm of
the logarithm of the transmission coefficients may be plotted against
the logarithm of the wave length, yielding very closely a straight line,
except for points in the ozone region. This fortunate circumstance
is due to the approximate inverse A" dependence of the logarithm of
the transmission coefficients on the wave length. The ozone trans-
mission can be determined from this plot in a similar way from the
difference of observed points from the straight line. The trans-
mission coefficients for March 24 were treated independently in this
second way. The area resulting was very closely the same as that
shown in Figure 5, yielding 3.9 mm. ozone path. It is believed that the
application of this method to data of days which give good trans-
mission coefficients affords a satisfactory method for determining
atmospheric ozone from direct intensity measurements.

In order to make a somewhat more extensive comparison of ozone
values determined by this method with those previously known from
observations made with the Dobson apparatus, a series of eight days in
1928 and 1929, for which values by the Dobson method have been
obtained at Table Mountain, was treated according to a somewhat
abbreviated form of the above method. For these days there were
available the atmospheric transmission coefficients as regularly read
and already computed from the “ long method ” observations at Table
Mountain. These values are not given at all points used in the deter-
mination of the area defined in this present work, but at the regular
spectrum points ordinarily determined in the solar-constant work.
These relatively few points are scattered over the spectrum in such
a way as to outline, somewhat less accurately to be sure, essentially the
same area as that defined in the present work. In particular there were
but four ordinates lying in the ozone region, but if these values were
sufficiently accurate the area of ozone absorption outlined by them,
and lying between the wave lengths specified above, would be suf-

NO. 9 DETERMINATION OF OZONE—WULF ll

ficiently close to the area defined in the above work to yield a satis-
factory determination of the ozone.

For these eight days, then, the logarithms of the logarithms of the
transmission coefficients were plotted against the logarithm of the
wave length and a straight line passed through the values lying to
both sides of the ozone absorption. The differences of the anti-
logarithms of those points lying in the ozone region from the anti-
logarithms of the corresponding points on the straight line were read.
These differences were then plotted against wave length to the same
scale as that described above and the areas planimetered and divided
by the area corresponding to 1 mm. of ozone path, determined as the
result of the work described in this paper. Thus an approximate
determination of the ozone on these days was afforded, utilizing an
area practically the same as that described above, defined by the

TasLe I.—Atmospheric Ozone for Eight Days

Path mm. o»

Date Present work Dobson
PICA LOZ Oe sarees seen see Blin atewarete tists acute eee 27
OCE sha LO2S rere eee ereeie fh tere DB Oiaasste Sut teat s7k 22520
INOvaedaelO2Sretertem ine erro ccs PAE rowed onar eaheyeites oh orers 1.97
DECI Ga LOZO Memes oo eiecs TPO Vass siete erect eae 1.08
Dece 18; 1028 seataecere gece e TeGyaiicemem eines 5s
TD CGE TO LOD Siprcacrat hacrateyela, Hees 2A cate So 60 vivartye eo. 2.43
Mary 245 O20 nies aoe ee ees oe Oe eee ee 3.42
NDT 23 A TO2O mrrensveieacske oe ia/e vec DEAN e re ee erie 3.08

MICA TI raceeecrer eerie se eee ge 2.35 2.49

positions in which these points lay. The results are shown in Table 1.
For these eight days the values of the ozone over Table Mountain as
determined by the Dobson method were also available and they are
given in this table for comparison. From these results it appears that
the average amount of ozone given by the two methods is essentially
the same. The independent values, however, are not in good accord,
which may be entirely due to the uncertainties in the present determi-
nations. It must be emphasized that this is due to the insufficiency of
the data used to make such a determination, which had been collected
from solar-constant work, and not to the method employed. It is to
be noted especially that differences in the relative values for the eight
days obtained by the two methods of Dobson and of Wulf cannot be
due to the method of the ozone absorption measurements described in
this paper, but must be contained in the roughness of the data. For
quite independent of the evaluation of the areas in terms of ozone,
these areas should be closely proportional to the ozone values on these
days. That they are not is actually due largely to the uncertainties in

I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the transmission coefficients, which in this case define the ozone area.
This is shown by a comparison of the March 24 value in this set of
eight with the value for this same day as given above, which was
determined from the transmission coefficients at all the points chosen
in the above work. The former value is 2.8 mm., while the latter
3-9 mm., and the latter is, as stated above, believed to be a satisfactory
determination. The cause of this discrepancy appears to lie chiefly in
the fluctuations in the few observed transmission coefficients. To
summarize: The use of the few single values of the transmission
coefficients as regularly determined in the solar-constant work only
suffices to define an area which gives the approximate amount of ozone
and is not ordinarily competent to show the fluctuations from day to
day within an error small compared to the fluctuations.

SUMMARY

The transmission of ozone for visible light has been determined
from spectrobolometric data using the solar spectrometer of the
Smithsonian Institution at Table Mountain, Calif., with the sun as
the source and introducing chemically determined quantities of ozone
in the path of its rays. The results are in close accord with the labora-
tory results of Colange. Using the results of this study the amount
of ozone over Table Mountain for one typical day as an illustration
has been determined. By an abbreviated method, using only the trans-
mission coefficients normally measured in the regular solar work, the
value for the ozone over Table Mountain has been determined for a
series of eight days. The mean of these eight days presumably gives
a good value for the mean amount of ozone, but this abbreviated
method is not ordinarily sufficient to show the fluctuations in the ozone,
since the possible error in a single determination is of the order of
the fluctuations. The mean value for the eight days differs but about
6 per cent from the mean value for the same days determined by the
method of Dobson. It is very interesting that the bolographic method
depends on ozone absorption in the yellow, while Dobson’s photo-
graphic method employs the ultraviolet ozone absorption.

The author wishes to express his sincere thanks to Dr. C. G. Abbot
for suggesting the problem and for his continued interest and help
throughout the work, and to Mr. J. A. Roebling for a financial grant
which made the work possible. The efforts of a number of people
have contributed directly to the completion of this work, especially
Mr. Alfred F. Moore, Mrs. Beatrice J. Wulf, Mr. Fred Greeley, and
Mr. George Cox. The author is grateful to the members of the
Smithsonian Astrophysical Observatory for their frequent kind
assistance.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL, 85, NO. 9, PL. 1

Two views of absorption cell and auxiliary apparatus.

Z “1d ‘6 ‘ON ‘G8 “10A

vat
idl ,
; A
a y
a
= 2 ¥
ra, ¥
oD]
Apa? %
Te eee
Cee Ue y)
rt ny ry}

SNOILO31100 SNOANVITSOSIN NVINOSHLIWS

“QTISIA OY} UT 9U0zZO Jo WINAJads UOdIOSqY

pi T
O'OT

6 JOO OT

0002

O00F

€ “Id ‘6 “ON ‘S8 “10A SNOILO3S11090 SNOANVTIISOSIN NVINOSHLIWS

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VOLUME 85, NUMBER 10

HUMAN HAIR AND PRIMATE
PATTERNING

(WITH FIvE PLATEs)

BY
GERRIT S. MILLER, JR.

Curator, Division of Mammals, U.S. National Museum

(PUBLICATION 3130)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION

DECEMBER 19, 1931

The Lord Gattimore Press
BALTIMORE, MD., U. S. A.

HUMAN HAIR AND PRIMATE PATTERNING

By GERRIT S. MILLER, JR.,
CURATOR, DIVISION OF MAMMALS, U. S. NATIONAL MUSEUM

(WitH Five PLates)

Few problems have caused more perplexity to anthropologists,
physicians, and zoologists than those presented by human hair. Why
is it that only some relatively small areas of the human skin are
normally capable of bearing a hair growth dense enough to be in any
way comparable with the fur of other mammals? Why do men have
beards and women not? Why are beards better developed in some
races than in others? What is the cause of baldness, and why is there
no certain cure for it? Why does baldness commonly occur on the
crown and rarely on the sides of the head? Why do we turn gray?
Why does grayness usually show itself first on the temples or in
the beard? Why does the moustache often remain dark after the
beard has turned gray? Why do we have hairy eyebrows, and why,
when there is a difference in color between the hair of the eyebrows
and that of the head, are the eyebrows usually the darker of the two?
Why is the hair of the scalp often different in quality from that of
other parts of the body? Why are there several types of hair—kinky,
curly and straight?

To all of these questions so many and such unsatisfying answers
have been suggested that it would be a huge and useless task to try
to list them. Variable and inconclusive though they are, most of the
answers possess one quality in common, namely, they have in their
background the tacit assumption that all these peculiarities of human
hair are things that arise from man’s special constitution and its
reaction to the natural environment or to the artificial conditions that
man has imposed upon himself. It has, for instance, been urged that
the general bareness of the human skin comes from the widely preva-
lent habit of wearing clothes; that baldness comes from barbers and
tight-fitting hats; that women have less baldness than men because
women have for centuries taken better care of their scalps than men
have; that graying hair is the result of a lessening bodily energy
supposed to go with increasing civilization or “ domestication ;”’ that
the axillary and pubic tufts of hair were once useful for babies to

SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 10

.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

to

cling to; that eyebrows exist for the purpose of keeping sweat from
running down into the eyes; that men are bearded to protect their
throats from cold weather; that women are beardless because they
look better that way. All of which gives evidence of ingenuity if of
nothing else.

But not one of the explanations that I have been able to find in
print has taken into consideration the zoological possibility that many
features of the human hair system may be generalized primate
traits instead of specifically human developments. By this I mean
the possibility that they may be characteristics that are forced on
man because they are common property of the Primates, the animal
group to which man pertains. Their explanation, in that event, would
have to be made less in terms of human activities and requirements
than in terms of the great heritage of characteristics that man shares
with all his primate relatives. Each one of these creatures has modi-
fied his portion of this heritage in such a way as to make it his own;
or, in more technical language, each one of the 800-odd kinds of
living apes, monkeys, and lemurs has developed “ specific ” characters
by which it can be distinguished from all the others while remaining
none the less a primate among primates. That.man should have done
the same thing would be far from strange.

This paper is a brief summary of a study on which I have been
engaged for several years with the result that I have become con-
vinced that the chief peculiarities of human hair are best and most
simply explained as special examples of primate ‘ patterning.”

WHAT IS MEANT. BY “PATTERNING:

Patterning is familiar to every systematic zoologist because it is
seen in every group of animals. It consists in the arrangement of
(a) contrasted colored areas on the surface of the body, or (b)
contrasted long and short outgrowths from the surface of the body,
or (c) combinations of colors and outgrowths, in such a manner that
the resulting patterns of color or form are sufficient to distinguish
one related species from another.

Familiar examples of patterning are furnished by the color mark-
ings of butterflies, or of the American wood warblers, by the minute
surface sculpture on the shells of some mollusks, and by the spiny
outgrowths on the back and head of the different species of iguana
and horned-toad. Among mammals, striking instances are provided by
cats and squirrels with their diverse stripes, spots, mottlings, and plain
colors, and by African antelopes with their stripes and spots as well
as their maned necks, fringed throats and briskets, and tufted tails.

NO. IO HUMAN HAIR AND PRIMATE PATTERNING—MILLER 3

The making of patterns appears to be a process quite distinct from
that by which a general harmonizing of animals with their natural
surroundings has been effected. Nearly allied, pallid, desert species,
for instance, may be distinguished from each other by details of indi
vidual pattern as obvious as those that serve to mark richly colored
species living in humid forests. General types of color and surface
may have their relations to the surroundings in which animals pass
their existence; but the special patterns of the species that conform
to any one type cannot be shown to have such relations. It may be
plausibly argued that the blotched and spotted color schemes of
arboreal warblers and the streaked color schemes of grass-living
finches have something to do with the unlike surroundings in which
warblers and finches pass their lives. But this argument would not
apply to the differences between the patterns of Blackburnian and
black-throated green warblers nesting together among the same ever-
greens, nor to those between savannah and grasshopper sparrows
living in one meadow. Still less would it be possible to explain, on
grounds of special needs, why species of horned-toad differ from
each other in the number and form of the spiny outgrowths on the
head, or why one species of gnu has a fringe on the brisket and
another has not. Patterning, therefore, seems to be something physio-
logically inherent in animals rather than something that the environ-
ment has imposed upon them.

PATTERNING IN PRIMATES

Though patterning occurs in all groups of mammals—even in
rhinoceros, hippopotamus, elephant, and cetacean—it is among the
primates that the tendency attains its greatest development. In no
other group does it make such full use of its chief materials, namely,
the color of the skin, the color of the hair, and the contrasts that
can be obtained from differences in quality and length of hair. No
better example of this process could be given than the one furnished
by the head markings of monkeys grouped on Plate 1. The animals
there represented are nearly related species that live under essentially
uniform surroundings in the great African forest belt. No two of them
have the same arrangement of dark and light areas on the head; three
have conspicuous white stripes over the eye; one has a black stripe in
the same place ; in five the cheeks are white, while in three they are not
white ; one (fig. 2) is bearded, while seven are not; one (fig. 4) has a
moustachelike mark of white in the skin of the upper lip; another
has a boldly contrasted spot of fine white hairs on the nose. Other
patterns in primates come from lengthening, shortening, and varying

4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the direction of growth of the hair on different.parts of the crown,
also from varying the length and quality of the hair on the chest,
shoulders, tail, and legs, and from making contrasts, often more
striking than the one seen in the white-lipped guenon shown on
Plate 1, in the color of different portions of the skin itself. All of
these elements of color and hair growth are combined and recombined
in a variety that seems to be without end.

On no part of the primate form is patterning so conspicuously
developed as on the head, where strikingly marked color designs of
both hair and skin are profusely exhibited, and where tufts, beards,
moustaches, whiskers, and crests are brought into varied contrast with
areas of short hair and bare skin.

PATTERNING ON THE HUMAN HEAD

In conformity with this universal primate trait human patterning
shows itself more conspicuously on the head than on the body or limbs.

The human head pattern is not exactly duplicated by any other
primate, but all the elements that enter into it can be easily found in
nonhuman members of the order. The usual head pattern of the young
adult Caucasian is shown in Plate 2, Figures 1 and 2. Characteristics
that both sexes have in common are the completely haired cranium,
the bald forehead, nose, and upper median part of the cheeks, and
the presence of a narrow transverse hairy strip on the forehead over
each eye. The female’s pattern differs from the male’s in an extension
of the bare area downward over the entire lower part of the face and
sideways to the ears.

In most primates the forehead and face, except the region im-
mediately bordering the eyes, nose and mouth, are thickly haired.
The first step in the process of baring the forehead is shown by one
of the Celebean macaques, Magus hecki (pl. 2, fig. 3). Other steps
have been taken by some of the South American monkeys; while
an essentially human forehead can be found in the orang (pl. 2, figs.
4, 6). The bare or nearly bare lower part of the face seen in the
females of all human races, and in the males of those races in which
the beard is slightly developed, is presaged by the very common
occurrence among other primates of a short-haired, nearly bare area
around the mouth (shown by all of the monkeys represented on pl. 1).
Extensions of this bare area on the cheeks may be seen in the great
apes. It is carried farther in some of the South American monkeys,
culminating, apparently, in the “ cotton head,” Oedipomidas oedipus
(pl. 2, fig. 5), which has reached a stage slightly more advanced than
that of the human female.

NOs 1O HUMAN HAIR AND PRIMATE PATTERNING—MILLER 5

Though a bare or nearly bare condition of the mouth area is the
usual condition among nonhuman primates, it is not universal. Beards
like those of the male Caucasian or Australian occur in the orang
(pl. 2, fig. 6), in the bearded African guenon shown at the top of
Plate 1, and in a South American monkey (Pithecia chiropotes) of
which I have not been able to obtain a photograph. Moustaches are
not common among primates. Even that of the full-bearded orang
is poorly developed. But the South American Mystax imperator
(pl. 2, fig. 7) goes far to make up for this deficiency. While neither
moustache nor beard is peculiar to man the strong development of
both together appears to be a human specialty.

Nothing exactly like the human eyebrows is known in other pri-
mates, but the brow region is one where patterns are made in great
profusion. Sometimes these brow patterns take the form of light
or dark stripes (as shown in pl. 1 and in pl. 4, figs. 10 and 13) ; some-
times they are made by lines of hair differing in quality and direction
from that of the head (pl. 2, fig. 8), thus showing a near approach
to the condition found in man. Human eyebrow hair, as is well known,
often differs in color from the hair of the crown. In such cases
(pl. 4, figs. 11, 12) it is usually darker than the crown hair, after
the manner of the gray-cheeked mangabey (pl. 2, fig. 8) or the
Himalayan langur (pl. 4, fig. 10) ; rarely if ever is it conspicuously
lighter, after that of the white-browed gibbon (pl. 4, fig. 13).

This normal human pattern does not always remain constant
throughout life. Changes of two kinds usually take place; and the
courses of both kinds tend to follow lines that can be traced through
the group of primates at large.

TURNING BALD

With arrival at full maturity a considerable percentage of human
males undergo a modification of their hair pattern that serves to
differentiate them still further from the females. The forehead line
begins to rise, either uniformly along its entire extent (pl. 3, fig. 1),
or, more commonly, by pushing back a blunt reentrant wedge on each
side (pl. 3, fig. 4). Frequently a bare spot begins to form at the
same time on the top of the crown (pl. 3, fig. 6), and the hair of the
entire median part of the crown becomes sparse. These changes may
continue until the bald forehead area has been carried back over the
dome of the head, leaving a well-haired border extending around the
sides and across the nape (pl. 3, fig. 8).

This series of maturity changes in the hair covering of the human
male head has been the subject of endless speculation. By a few

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

writers it has been recognized to be, like the beard, a secondary sexual
character,’ but, so far as I am aware, no one has hitherto shown that
it follows the lines laid down by the patternings of other primates.
The uniform raising of the forehead line can be found as a specific
character in the bald chimpanzee ; it is exactly paralleled by the pattern
of short and long hairs on the head of one South American monkey,
Pithecia monachus (pl. 3, fig. 2), and by the color pattern of another,
Cebus hypoleucus (pl. 3, fig. 3). The development of the two blunt
wedges results in a pattern much like the one present in the Celebean
black ape, Cynopithecus niger (pl. 3, fig. 5). The bald spot on top
of the crown is an occasional character of the toque macaque,
Macaca pileata (pl. 3, fig. 7). The completely developed human
bald area (pl. 3, fig. 8) is perfectly outlined in the South American
monkey known as Cacajao rubicundus (pl. 3, figs. 9 and 10). The
long dark hair at the side and back of the head of this animal occupies
the area that remains haired in normal human baldness, while the
light hair on the median area corresponding with the human bald spot
is so short and sparse that it does not conceal the skin of the scalp
in the living animal (two have recently been on exhibition in the
National Zoological Park). Finally it is to be noticed that the human
bald area follows the outline of the dark cap of the West African
gorilla (pl. 5, fig. 5) as well as that of a color pattern not infre-
quently seen in blond men. This pattern, (which appears to occur
in women also, but is obscured by long hair) is produced by an
obviously paler tint of the hair that grows on the bald-spot area.
It is visible as a faint but accurate picture of the color pattern made
by a bare scalp contrasted with dark side hair. At the Harvard com-
mencement exercises of 1930 I saw it on the heads of seven of the
young men awaiting the conferring of their degrees.

TURNING GRAY

Another change in the human hair that begins at or slightly after
the attainment of full maturity is seen in the familiar process of turn-
ing gray ; this may lead in the end to a stage when all the pigmented
hairs of the entire body have been replaced by colorless ones.

This loss of color, like baldness, has given rise to conjecture with-
out end. And, as in the case of baldness, its near relation to primate
patterning seems to have passed unnoticed. Nevertheless, it can be

*The examination of many hundreds of photographs makes it appear prob-
able that the males of races with strong beards tend to show the highest per-
centage of baldness, thus differentiating themselves most fully from the females.

NO. 10 HUMAN HAIR AND PRIMATE PATTERNING—MILLER 7
shown to have the same tendency to follow the main lines of primate
pattern making.

When gray hairs begin to replace the pigmented ones they do not
appear uniformly all over the body. “ A vigorous man just beginning
to show a touch of gray on the temples ”’ is an often-heard phrase that
unconsciously recognizes this fact. When beards were common among
us everyone knew how usual it was for them to turn gray before
the scalp.

As they increase in numbers the gray hairs tend to form patterns.
These are sometimes nothing more than faint sketches or suggestions.
Often, however, they develop into striking color contrasts. The
faint and fugitive human patterns are not always easy to correlate
with the patterns of other primates, but the definite ones rarely pre-
sent any such difficulty.

Eight of these well defined human color patterns with their primate
homologues are shown on Plates 4 and 5.

The first and second (pl. 4, figs. 1 and 3), consisting of a white
beard contrasted with a dark crown, are frequently seen. In the first
the mouth area is white. In the second it is dark. Both occur in
many species of monkey, two of which, the African Erythrocebus
pyrrhonotus and Cercopithecus Uhoesti, are shown in Figures 2 and 4.
The identity is so obvious that it requires no comment.

The third pattern (pl. 4, fig. 5), consisting of a white chin beard
sharply contrasted with dark whiskers and head, is less common.
Sometimes the white involves the moustache. It is then exactly the
same as the white area in the African monkey Cercopithecus brazzae
(pl. 4, fig. 6). I have seen several examples of this human pattern
with white moustache, but have not yet secured a photograph.

The fourth pattern (pl. 4, figs. 7, 8, 9) is merely a dark mark on
the cheek margin of a gray beard accompanying a gray or bald head.
Insignificant though this marking may seem, it is surprisingly com-
mon. On April 21, 1930, I visited the Jewish pushcart market district
in New York City, one of the few convenient places where many
full beards can now be seen, to look for this mark. I found it in no
less than 47 out of 55 men with gray or white beards. The same
dark line at the edge of the longer hair on the cheeks is found in
many of the monkeys that have a partly bare median facial area. An
example is shown in Plate 4, Figure 10, the Himalayan langur
(Pygathrix schistacea). It may be easily observed in immature
Japanese macaques, animals that are often exhibited in zoological
gardens.

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

The gray human temple spot (pl. 5, fig. 1) is a common feature
of primate color patterning. It is particularly well developed in the
gelada baboon (pl. 5, fig. 2).

As they increase in size the temple spots often extend backward
along the sides of the head until they cover the entire area that re-
mains haired in normal baldness. The pattern thus formed—dark cap
contrasted with grizzled sides and back of head (pl. 5, figs. 3, 4)—
is a common one among non-human primates. It is particularly well
developed in the West African gorilla (pl. 5, fig. 5. The specimen
represented by this photograph is an unmounted skin with the head
not filled out to natural form). Occasionally this pattern may be
seen reversed. The grizzling is then confined to the area of the normal
bald spot, while the hairs at the sides and back of the head remain
dark. When this happens the color scheme of the cacajao monkey
(pl. 3, fig. 10) is reproduced.

White locks situated on or near the forehead line (pl. 5, figs. 6, 7)
are not uncommon, but on other parts of the head they are rare. They
may be present without other signs of the graying process (as in fig. 6)
or they may appear as a step in that process (as in fig. 7). In either
event they are usually confined to some part of an area where pattern-
ing occurs in nonhuman primates (pl. 4, fig. 6, pl. 5, fig. 87).

PATTERNING ON OTHER PARTS OF THE HUMAN BODY

The process of turning gray usually begins on the head and extends
gradually downward over the body. As it advances it often passes
through a stage, particularly well represented in Figure 9 of Plate 5,
in which the gray area ends abruptly at the middle of the chest, leav-
ing the hair of the arms and lower part of the body dark. The general
lines of a pattern found in an African colobus monkey, Colobus poly-
comos, and in an Asiatic macaque, Macaca albibarbata (pl. 5, fig. 10),
are then closely followed.

Turning to other parts of the human body we find that the same
correspondence with widely distributed primate tendencies holds good.

The pubic region is an area of pattern formation in widely sepa-
rated nonhuman primates. Young chimpanzees have a white pubic
patch contrasted with the black surrounding hair. It disappears by
becoming black before the animals reach full maturity. Some species
of gibbon have no pubic mark whatever. Others display a black spot

* The spider monkey represented in fig. 8 of pl. 5 has a band of white extend-
ing along the entire frontal border of the true head-hair. The forehead-hair
is also white, but it differs from the head-hair in quality and in direction of
growth. Before photographing this skin I darkened the forehead-hair with ink.

NO. I0 HUMAN HAIR AND PRIMATE PATTERNING—MILLER 9

that stands out against its pale surroundings. Still another gibbon
has the hair of this region so greatly lengthened that, in adult males,
it may form a tassel reaching almost to the knees. A South American
monkey (Oreonax hendeet) has a long yellow hair tuft in the male
and two shorter tufts in the female. In both sexes the tufts are
rendered very conspicuous by contrast with the dark belly and thighs.
By specializing the hair of this region man has, therefore, merely
followed one tendency of his tribe.

The last conspicuous hair-pattern feature of man is the tuft in the
arm pit. This, perhaps, comes the nearest of all the patterns to being
an exclusively human trait; I have not yet seen an exactly similar
development in any other primate. But, on gently blowing the hair
of the axillary region of a freshly dead African monkey, Cercopithe-
cus aethiops, I once found that the hairs growing in the deepest part
of the pit tended to separate themselves from the surrounding fur
by a slight difference in quality and in the direction of growth. More
recently I have been able to see, in several adult chimpanzees,’ that
these animals have a definitely specialized axillary tuft confined to
the region of greatest glandular activity. To produce the human
condition it would merely be necessary to suppress the long surround-
ing hair.

SOME OTHER FEATURES OF HUMAN PATTERNING

A few other points about human patterning require brief mention.

(a) The general bareness of the human body.

Why the human body lacks a protecting general coat of fur is a
question that has been often asked and variously answered. A final
explanation seems to be as remote now as ever; but it is possible
to recognize the fact that human bareness is only an exaggeration
of a tendency that is found in other primates,’ and that it is no more
essentially mysterious than the bare face of one tropical American
monkey (pl. 2, fig. 5) when compared with the fully haired face of
another (pl. 2, fig. 7). In neither instance can it be shown that a
special need of the species is served by the bare skin; but in both it
is evident that the tendency found throughout the primate group to
form patterns by contrasting long-haired areas with short-haired areas
has been carried to an extreme.

2 At the Yale University Anthropoid Experiment Station, Orange Park, Fla.,
an opportunity for which I have to thank Professor Yerkes and Dr. Tinklepaugh.

?On this subject see Schultz, Human Biology, vol. 3, pp. 303-321, September,
1931, and Sci. Monthly, vol. 33, pp. 392-393, November, 1031.

IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The general distribution of longer and shorter hair on the body of
gorillas rather closely coincides with the human scheme. By continu-
ing the process along the lines marked out in this great ape a stage
would eventually be reached in which the body would become bare
while the arms and legs retained traces of their original coat.

(b) The different face pattern of men and women.

The sexual hair pattern on the human face is another subject of
age-long speculation. No one has ever been able to show that its
presence has aided man’s career as a species. Equally impossible would
it be to show that the analogous sexual patterns in other primates have
given these species any advantage over their relatives that lack them.
But it seems clear that in this respect man has developed in the same
general way as the white-cheeked gibbon of Siam, the orangs of
Borneo and Sumatra, the black howler monkey of South America,
and the macaco lemur of Madagascar, all of which have sexes that
differ from each other in appearance. That is to say, man and these
other primates have followed a tendency that may crop out anywhere
in the group of animals to which they all belong.

(c) Racial differences in hair pattern and in general color of the

hair.

It is well known that not all races of man are exactly alike in hair
pattern. Some have better developed eyebrows, beards, pubic patches,
or axillary tufts than others; some appear to be not as subject as
others to grayness and baldness. Racial tendencies toward darker or
lighter colored hair are also well known. These racial characteristics
have never been satisfactorily explained on the basis of the special
needs of different peoples. On the other hand, as examples of the
slight differences that are everywhere found among races of primates
nearly related to each other they are readily understood.

The differences between the two races of orang, for instance, are
of this nature. The United States National Museum contains 6 males
and 6 females of the Sumatran orang, 6 males and 10 females of the
Bornean race, all adult or nearly adult. These two series show the
same kind of differences that are shown by races of men. In the first
place, the beards of the males are much better developed than those
of the females. Then, when the beards of the Sumatrans are compared
with those of the Borneans they are at once seen to be larger, so much
so that an adult male from either island can usually be recognized
at once by this feature alone. Finally there is a general difference
in the color of the hair on body and head, this being more tawny in
the Sumatran race, more mahogany brown in the Bornean.

NO. 10 HUMAN HAIR AND PRIMATE PATTERNING—MILLER Ld

These two races of orangs inhabit separate parts of one climatic
zone, exactly as Caucasians and Mongolians inhabit opposite ends of
another. Therefore the differences in hair growth can be no more
attributed to the influence of unlike natural surroundings in the second
instance than in the first. But it seems clear that the two races of
orang and the two races of men are both in early stages of species
differentiation, and that the manner of their differentiating is one that
is common to the whole primate group.

(d) Total graying.

Often, though not invariably, the process of turning gray culmi-
nates in a stage of complete whiteness. But even when a human being
has turned gray over the entire body or even has lost all hair color
he has done nothing that is essentially new or peculiar for a primate.
Light gray or nearly white species of primates have arisen in both
Asia and South America. These animals are not albinistic nor in any
way individually abnormal. Their near relatives, living in the same
regions, are richly colored; and there is nothing to indicate that either
light or dark has any advantage over the other. General graying and
whitening in man seems likely to be nothing more than another ex-
ample of human submission to a rule that some other primates have
followed. Therefore the strong tendency present in the “ white race ”
of man for the hair to lose its color at an early age may be part of
a racial process of depigmentation that has already almost whitened
the skin and that may be destined, in the future, to bring about
permanent whitening of the hair as well.

CONCLUSION

For the present I wish to avoid detailed discussion of the published
attempts to explain those peculiarities of human hair that have just
been passed in review. Most of the authors who have considered the
subject have done so from the view-point that these peculiarities must
have originated from conditions (pathological or cultural) or needs
(physiological or esthetic) that pertain exclusively to man. That this
view-point is wrong seems to be sufficiently indicated by the evidence
here selected from the large mass that I have assembled. This evidence
points to the probability that man has these characteristics because,
as a primate, he cannot avoid them. They are common property of
the great group of mammals to which he pertains, and neither he nor
any other member of this group can wholly escape from the tenden-
cies imposed on all of them by their primate heritage. With regard
to no nonhuman primate can it be shown that the possession of any

I2 , SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

special assortment of these characteristics makes a species better,
more efficient, or more at ease in the world than one that has another
assortment. So also with man.

The similarities that I have shown to exist between some hair
characteristics of man and those of particular monkeys and apes must
not be supposed to indicate any special relationship between man and
these other primates. When superficial features of this kind are
common to a whole group they will often appear in almost identical
form in two animals whose relationship is shown by their anatomical
structure to be remote.

EXPLANATION OF PLATES
All figures greatly reduced, not to scale
PLATE I

Color patterns on the heads of eight species of African guenon (Cercopithe-
cus). From Elliot, after Pocock.

PLATE 2

The human head-hair pattern and its characteristics as they occur in other
primates.

Fics. 1, 2. Young adult Caucasian.

Fic. 3. Partly bare forehead of a Celebean macaque (Magus hecki).
Fic. 4. Bare forehead of an orang.

Fic. 5. Bare face of a “cotton head” (Oedipomidas oedipus).

Fic. 6. Beard of an orang.

Fic. 7. Moustache of a marmoset (Mystax imperator).

Fics. 8, 8a. Eyebrows of a mangaby (Cercocebus albigena).

PLATE 3

Types of human baldness and the corresponding conditions in other primates.

Fic. 1. Raised human forehead line.

Fic. 2. A South American monkey (Pithecia monachus) with hair pattern
corresponding with the raised human forehead line.

Fic. 3. A South American monkey with color pattern corresponding with the
raised human forehead line.

Fic. 4. The two reentrant forehead wedges in man.

Fic. 5. The two reentrant forehead wedges in the Celebean crested macaque
(Cynopithecus niger).

Fic. 6. Bald spot at middle of crown—human.

Fic. 7. Bald spot at middle of crown—toque macaque (Macaca pileata).

Fic. 8. Complete, normal, human bald crown area.

Fics. 9, 10. Nearly bald crown area in a South American monkey (Cacajao
rubicundus ).

NO. I0 HUMAN HAIR AND PRIMATE PATTERNING—MILLER 13

PLATE 4

(a) Human color patterns formed during the process of turning gray and
the corresponding patterns in other primates.

Fic. 1. White face contrasted with dark head in man.

Fic. 2. White face contrasted with dark head in an African monkey (Eryth-
rocebus pyrrhonotus).

Fic. 3. White face contrasted with dark mouth area and dark crown in man.

Fic. 4. White face contrasted with dark mouth area and dark crown in an
African monkey (Cercopithecus l’hoesti).

Fic. 5. White chin and lower lip contrasted with dark face and head in man,

Fic. 6. White chin and mouth area contrasted with dark face and head in an
African monkey (Cercopithecus brazzae).

Fics. 7, 8, 9. Dark area at edge of light cheek hair in man.

Fic. 10. Dark area at edge of light cheek hair in an Asiatic monkey (Pyga-
thrix schistacea).

(b) Eyebrow patterns, human and simian.

Fics. 10, 12. Dark eyebrows contrasted with light head hair.

Fic. 11. Dark eyebrows contrasted with hair that has turned white.

Fic. 13. White eyebrows contrasted with black head (white-browed gibbon,
Hylobates hoolok).

PLATE 5

Human color patterns formed during the process of turning gray and the cor-

responding patterns in other primates (continued). :

Fic. 1. The human gray temple area.

Fic. 2. The gray temple area in the gelada baboon (Theropithecus gelada).

Fics. 3, 4. The human gray temple area extended around the head.

Fic. 5. Color pattern on the head of the East African gorilla.

Fics. 6, 7. White locks on the human forehead line.

Fic. 8. White stripe along the forehead line in a South American spider monkeys
(Ateles hybridus).

Fic. 9. Gray area extending downward from head to middle of chest in man,

Fic. 10. Gray area extending downward from head to middle of chest in an
Asiatic monkey (Macaca albibarbata).

7

ih

Te 4
mm. oT ‘ : ?
ai 1 - 4 a - ele a To
b ee

. s f
' Z > ; ;

i hee : mie ; a
Bye on las i
J it ; .
(ek

ul

i
riay
;
, if
' A ,
il a
. wit? O47
ave
4
ai Pi
‘
* ;
*
‘
.
'

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOE85;, NO. 107,"REt

Patterning on the heads of eight species of African monkey.

(For explanation see p. 12.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. (85; NOs 10); PR?

Human face patterns compared with the analogous patterns of five
kinds of nonhuman primates.

(For explanation see p. 12.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85, NO. 10, PL. 3

Human baldness patterns compared with the analogous patterns of five kinds of
nonhuman primates.

(For explanation see p. 12.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85; NO 10; (Piet

Fics. 1-10.—Human erayness patterns compared with the analogous patterns
of four kinds of nonhuman primates.
Fics. 10-13.—Eyebrow patterns, human and simian.

(For explanation see p. 13.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOET 65. (NO. 10, (PE.

)

Human grayness patterns compared with analogous patterns of four kinds of
nonhuman primates.

(For explanation see p 3.)

SMITHSONIAN MISCELLANEOUS COLLECTIONS

VOLUME 85, NUMBER 11

(End of Volume)

Hodgkins JFund

SUPPLEMENTARY NOTES ON
BODY RADIATION

BY
L. B. ALDRICH

(PUBLICATION 3131)

CITY OF WASHINGTON
PUBLISHED BY THE SMITHSONIAN INSTITUTION
FEBRUARY 2, 1932

The Lord Baltimore Dress

BALTIMORE, MD., U. 8 A.

Hodgkins Fund

SUPPLEMENTARY NOTES ON BODY RADIATION ®
By L. B. ALDRICH
WALL TEMPERATURES AND BODY RADIATION

In present-day ventilation, three basic factors are considered:
(1) air temperature, (2) relative humidity, and (3) air movement.
The results of my previous report indicate that, in addition, considera-
tion should be given to a fourth factor, the temperature of the walls
and surrounding objects.

For normal indoor conditions, with the surrounding objects all at
the temperature of the air in the room and with the subject clothed
and at rest, the radiation loss of a human subject is nearly one half
of his total heat loss. This radiation emitted from skin and clothing
has been shown to be nearly that of a “ black body.” We may assume
that, by virtue of repeated reflections from the other walls and sur-
rounding objects ina closed room, the radiation from the walls to the
subject is also nearly “ black.’ Then the radiation loss of the sub-
ject is proportional to the difference of the fourth powers of the
absolute temperatures of the subject and the surroundings, in accor-
dance with the Stefan-Boltzmann law. Suppose the mean surface
temperature of a clothed subject to be 32° C. and the mean tempera-
ture of surrounding walls and objects to be the same as the air
temperature, 23°C. The difference of the fourth powers of the
absolute temperatures is 977 < 10°. Now imagine the air temperature,
humidity, and air movement to stay constant and the wall tempera-
ture to be lowered 10°. The difference of the fourth powers becomes
1963 X 10°, an increase of 100 per cent in the radiation loss. On a
winter day the temperature of exposed walls might easily be IO”
below air temperature, and the inner surface of window panes prob-
ably would be considerably more than 10° below air temperature.
Thus a subject, particularly if on the exposed side of the room, would
radiate at least twice as much on one side as on the other, and his
total loss of heat would be increased some 25 per cent or more.

Of even greater importance is the consideration of surrounding
objects which are at higher than air temperature. As before, suppose

See A study of body radiation, Smithsonian Misc. Coll., vol. 81, no. 6,
1928.
SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 85, No. 11

2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

the air temperature, relative humidity, and air movement to remain
normal but let the surrounding walls be raised to a temperature of
32° C. Radiation loss from a human subject would now be negligible,
the normal balance between heat produced and heat lost would be
destroyed, and until readjustment is made, a condition of discomfort
results. In actual schoolroom conditions, a student near an unshielded
steam radiator or other artificial heat source is exposed to a tempera-
ture much higher than his surface temperature. In classes, the student
is surrounded by other students and the summation of the solid angles
subtended at a point on one student by the other students may be
very appreciable.

As a rough example, assume a class of students placed in rows,
with spaces of 2 feet between students in a row, and the same dis-
tance between rows. To simplify matters, imagine each student to
be cylindrical, 1 foot in diameter and 4 feet high. The four students
nearest to a given student would occupy roughly 1o per cent of the
total space to which the central student is radiating. The four next
nearest students exposed to the given student would occupy an addi-
tional 5 per cent, and the eight next nearest another 4 per cent. Sum-
ming up, the amount of space occupied by surrounding students would
be about 20 per cent of the total space to which the central students
radiate. If we reduce the space between students to only 1 foot
instead of 2 and proceed to sum up in a similar manner, the area
occupied by the other students increases to about 35 per cent of the
whole. For a spacing of 3 feet between students it reduces to only
10 per cent. In other words, when students are spaced 1 foot apart,
the total radiation loss of each student is some 35 per cent less than
if he were alone in the room. When the spacing is 2 feet between
students the radiation loss is 20 per cent less than if he were alone,
and when the spacing is 3 feet the radiation loss is 10 per cent less.
These rough figures serve in a general way to show the relationship
between the spacing of students and the radiation loss of individual
students.

For a given wall temperature, what air conditions produce maxi-
mum comfort? Evidently if the walls are cold an increased air tem-
perature is indicated, and vice versa. A further study of the effect on
a subject of various wall temperatures under controlled air conditions
is needed. Such a study should tell us to what extent one’s radiation
loss may be altered without producing discomfort and should furnish
evidence as to the minimum spacing advisable in classrooms without
injurious reaction resulting from decreased heat loss.

NO. II BODY RADIATION—ALDRICH 3

ACCURACY OF SKIN TEMPERATURE MEASUREMENTS

In the study of body radiation above referred to, skin and clothing
temperatures were measured by a special thermoelement device sug-
gested by Dr. C. G. Abbot. For convenience I quote the following
illustrated description of the instrument from my previous publi-
cation:

For the direct measurement of skin and clothing temperatures, a special device
was prepared with the help of Mr. Kramer, the Observatory mechanician, and
embodying Dr. Abbot’s suggestions. The device is shown in Figure 12 It con-
sists of a specially mounted copper-nickel thermoelement of fine drawn wire.
A frame of German silver is bent as shown in the figure and fastened in a
wooden handle, 1”. Two silk threads are stretched to form a cross between the

Fic. 1—Thermoelement device for measuring surface temperatures.

F—Fibre rings. P—Spring steel projection.
S—German silver frame. U—Silk thread.
W—W ooden handle. T—Thermoelement.

four spring-wire posts, p. The thermoelement wires are fastened symmetrically
to these silk threads with the junction straddling the lengthwise thread. The
wires lead out through fibre rings, Ff, and through the wooden handle. The
copper wire (see fig. 2) leads through a switch to a sensitive type Leeds and
Northrup D’Arsonval galvanometer and thence to the constant temperature
junction in a stirred kerosene bath as shown in Figure 3. The Cu-Ni wires are
sufficiently long so that all desired positions can be reached without moving
the constant temperature bath. Holding the device by the wooden handle, one
presses lightly the four prongs of spring wire f/ upon the surface whose tem-
perature is desired. This places the junction in excellent contact with the sur-
face. There is no backing to the junction save a single silk thread, and thus no
possibility of heat piling up and causing too high temperatures. For about 4 cm.
on each side of the junction, the wire also touches the surface and assumes
the surface temperature, thus eliminating error due to cooling of the junction
by conduction along the wires.

*The figure numbers of the original publication have here been changed to
accord with the arrangement in this paper.

SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

T C

Fic. 2——Diagram of electrical connections of copper-nickel thermoelement.

G—Galvanometer. ;
T—Thermoelement junction. 4
C—Constant temperature junction.

y q
Ni cd N V
N Mea aN
Nit Che N
Nit LIN
Te
NEEUT X
N teas N
NN Wain f spo N
N Ciny toa NN
N Ht aN
Nt) te N
Nita §
N voy a) N
N : Ls N
N

7

FF

Fic. 3—Bath for constant temperature junction.

Th—Thermometer.
D—Stirring device.
K—Kerosene bath.
V—Vacuum flask.

NOw LL BODY RADIATION—ALDRICH 5

The instrument has recently been recalibrated. In mounting for
calibration and comparison, the constant temperature junction was
fastened against the bulb of a mercury thermometer. The ther-
mometer was then inserted in a metal tube and lowered into a stirred
kerosene bath, surrounded by a vacuum flask. In calibrating, the
thermoelement device was placed in a well-stirred kerosene bath whose
temperature was measured with a second mercury thermometer. A
sensitive-type Leeds and Northrup D’Arsonval galvanometer was used
with the instrument. The calibration curve, plotting galvanometer
deflections against temperature differences, is nearly a straight line.

As certain systematic discrepancies had been noted between skin
temperatures observed with this thermoelement and corresponding
ones computed from observations of body radiation with the me-
likeron, it was desired to ascertain whether the thermoelement was in
any considerable degree influenced by air temperatures in making such
measurements. The instrument was accordingly tested in the follow-
ing manner by measurements on a skinlike membrane of known
temperature.

In the vertical copper calorimeter previously used (see Smithsonian
Misc. Coll. vol. 81, no. 6, p. 15) three holes were made in the side at
equal altitude, each 6 cm. in diameter. These holes were closed with
rubber diaphragms, cemented in with waterproof cement. The thick-
ness of the diaphragms was as follows (determined with micrometer
gauge) :

0.18 mm. (thinnest dental dam)

0.36 mm. (sheet rubber)
1.20 mm. (composite sheet rubber used for gaskets):

Rubber was chosen because it is pliable, simulating the surface pre-
sented by the skin or clothing. The calorimeter as before was filled
with water kept thoroughly stirred and a record of its temperature
determined by a mercury thermometer.

It is evident that the surface of the thickest diaphragm will be
appreciably lower in temperature than the water in the calorimeter,
and that the thinner the diaphragm the more closely the surface tem-
perature approaches the temperature of the water. By obtaining a
series of surface temperatures of the various diaphragms, a curve may
be plotted and extrapolated to zero thickness. The more nearly cor-
rect the thermoelement temperatures, the more closely the zero dia-
phragm value will approach the calorimeter water temperature.

A series of comparisons is summarized in Table 1. Each value in
the table is the mean of three separate determinations. Air motion was
produced by a fan in the same manner as in the body-radiation experi-

6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VoL, 85

ments above referred to, and velocities (given in feet per minute)
determined as before with the Hill katathermometer. Before drawing
conclusions from the data in Table I it appeared advisable to obtain
more comparisons with other thicknesses of diaphragm. Pieces of
rubber of the 0.18 and 0.36 mm. thickness were stretched for several
days and. then cemented into the calorimeter holes previously filled by
0.18 and 0.36 diaphragms. The new thicknesses measured 0.12 and
0.27 mm. A series of comparisons with these new diaphragms is sum-
marized in Table 2.

TABLE I.
Calorimeter Water temp. minus thermoelement Air
Date Room Water temp. at diaphragm thickness of motion
1929 temp. Temp. oo (in ft. per
Aug. C. Gc .18 mm. -36 mm. 1.20 mm. min.)
3 20°50 31°82 °92 1°32 2°46 0
21.40 32.47 1.03 1.50 1.80
4 2510 31.50 73 TevLO 1.43
26.83 28 . 32 .40 .40 357
5 20.70 35.63 1271 2.43 3.87
6 26.20 35.73 .97 1.70 2.20
24.77 33-30 .80 1.33 Bos,
10 231.27 34.00 27; 2.73 3.07 80
24.00 30.80 73 1.67 2.57
27.80 29.83 257 .93 1.40
7 22.57 32.73 e277 Bn27, aly, 180
22.43 31.07 1.20 2.90 4.63
8 23.73 30.60 1.00 2.07 3.00
23.63 29.27 1.00 Tey 2567
24.17 26.97 | 270 1.03 1.53
9 24.97 25507 2.00 3.80 5.63 280
22.30 28.30 1.20 2.20 35.07)
2297, 260.50 -47 Tgp 1.43
23.93 20.37 1.40 2.73 3:27

From Tables 1 and 2, preliminary plots were made of the differences
calorimeter temperature minus room temperature and calorimeter
temperature minus thermoelement temperature, for each thickness of
diaphragm, and for the four conditions of air velocity, viz, 0, 80, 180,
and 280 feet per minute. As would be expected, the difference
between the calorimeter temperature and the surface temperature
determined by the thermoelement appeared to be a linear function
of the difference between the calorimeter temperature and the sur-
rounding room temperature. For each of the plots the best straight
line was drawn through the points and the origin. From each of the
plots values of the calorimeter temperature minus thermoelement
temperature were read off at two places, 5° and 10° calorimeter tem-
perature minus room temperature. These values were then replotted

NOo it BODY RADIATION—ALDRICH 7

as shown in Figures 4 and 5, using thickness of diaphragm as abscissae
and calorimeter temperature minus thermoelement as ordinates.
Partly from experimental error and partly because of differences
in conductivity of the various diaphragms, the individual points in
Figures 4 and 5 do not all lie on the curves. Smooth curves are drawn
however with fair certainty. In each case the extrapolation to zero
thickness yields a zero value of the difference calorimeter temperature
minus thermoelement temperature. This result is gratifying since it
indicates that the thermoelement device measures correctly the surface

TABLE 2.
Water temp. minus thermoele-
Calorimeter ment temp. at diaphragm Air
Date Room water thickness of motion
1929 temp. temp. SE (in ft. per
Sept. Cc c .I2 mm, .27 mm, min.)
22 17°00 24°00 247 293 oO
26 29.60 37 .97 -43 “73
28 22.87 28.47 33 .57
24.63 27.70 20, -40
25.97 36.67 57 -97
30 20.50 32.70 .93 2.00 80
Oct. 2 23.30 27.43 .40 1.30
5 22.77 33.50 .87 2507,
22.30 29.63 .60 1.77
Sept. 30 21637 28.10 “73 1.83 180
Oct, 2 23.43 ZA, eh 1-2
5 22.90 33.13 1.10 2.90
22330 20.37 Ted 2.33
2 23.50 26.87 -43 1.33 380 *
5 23.07 B25c3 1.10 2.07
22.30 29.03 1.33 2.80

1 This velocity was intended to be 280 ft. but through an error was, found to be 100 ft.
per minute too great. In Figures 4 and 5, the calorimeter minus thermoelement temperature
was adjusted to an air velocity of 280 ft. per minute.
temperature. It confirms satisfactorily the substantial accuracy of the
skin temperature measurements reported in my previous paper cited
above.

The following conclusions also are drawn from the surface tem-
perature measurements summarized in Tables 1 and 2:

(1) With the Smithsonian thermoelectric device, the flexibility of
the surface measured is an important factor as the air motion in-
creases. On a soft, flexible surface the instrument appears to give
nearly correct temperatures for all air motions. On a stiff surface it
probably reads nearly correctly for zero air motion but increasingly
too low as air motion increases.

(2) In appreciable air motions, the device shows large, irregular
drift, making readings difficult.

8 SMITHSONIAN MISCELLANEOUS COLLECTIONS voL. 85

IN

W,

°o

NM

°o

DEGREES C. BELOW CALORIMETER TEMPERATURE

DIAPHRAGM THICKNESS —MM.

Fic. 4—Calorimeter minus Room Temperature = 10° C.

NOSE

DEGREES Cc. BELOW CALORIMETER TEMPERATURE

BODY RADIATION—ALDRICH

DIAPHRAGM THICKNESS — MM.

Fic. 5.—Calorimeter minus Room Temperature = 5° C.

IO SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

TRANSMISSION OF RADIATION THROUGH THE SKIN

In measurements previously made comparing temperatures by the
thermoelement device with temperatures computed from melikeron
radiation measurements, it was noted (see Smithsonian Misc. Coll.,
vol. 81, no. 6, p. 19) that in measurements on the uncovered skin the
computed temperatures were about 1° C. higher than those measured
by the thermoelement. In measurements on clothing and calorimeter
this difference appeared to be much smaller. It was thought that pos-
sibly the skin was sufficiently transparent to long-wave radiation so
that the melikeron in reality received radiation from a warmer layer
below the outer surface. To test the transparency of the skin the
following arrangement was prepared:

Pyranometer S. I. 8 (for description and use of pyranometer see
Smithsonian Misc. Coll., vol. 66, nos. 7 and 11) was mounted without
glass hemisphere and with the absorbing strip vertical. A grid, cut
from platinum foil and blackened, served as a source of low tempera-
ture radiation. The resistance of the grid at room temperature
(22.5° C.) was 2.68 ohms. A voltmeter measured the potential fall
across the grid, and an ammeter measured the current flowing. The
temperature of the grid was roughly determined from its increase in
resistance as computed from the voltmeter and ammeter readings. A
doublewalled screen close to the grid exposed 8 sq. cm. of grid sur-
face. The distance from grid to pyranometer was 10 cm., which per-
mitted the interposition of two filters and a double-walled shutter.

The accepted procedure with the pyranometer is to use the first
swing of the galvanometer as proportional to the incident radiation.
When the shutter is opened, exposing radiation to the pyranometer
strip, the galvanometer spot immediately starts to move and, if the
radiation remains constant, swings to its maximum deflection in a
definite time. In the galvanometer used (Leeds and Northrup Type
R) this first swing required 3.53 seconds (mean of many trials). It
was noticed that when certain more or less opaque filters were inter-
posed the galvanometer spot did not start to move immediately and
took appreciably longer than 3.5 seconds to reach maximum deflection.
This delayed deflection was due to a combination of the direct radia-
tion transmitted by the filter and of the radiation from the filter
itself due to its increased temperature when exposed to the grid. To
minimize this indirect heating effect, a stop watch was used and only
those readings retained in which the maximum deflection was reached
within 4 second of 3.5 seconds. Temperatures of the grid source
were varied in the range 75° to 170° C.

NO. LT BODY RADIATION—ALDRICH II

deflection with a filter interposed

- : z is a measure of the
deflection without the filter

The ratio:

direct transmission of the filter, plus a small quantity diffusely trans-
mitted. Tests of the transparency of various screens were made.
These are summarized in Table 3.

TABLE 3.

Transmission of various substances. Temperature of source between
. 75° and 170° C.

Material Thickness % Transmitted

Rock-salt ......:...% (\Quaatiiiti eae seers terete fore tye 8s.

BIUOFIte. 2465.2 0 de ins CCE WW ch whee is oon enue: 44.

WhiGal sara cees 1 hee 03 ee pe ocr hee 50.

Tissue paper ........ [OSs aa Je det oea sees DOut, 45:

Blotting paper ...... TAY Mos Pedi oss Sawai ee Negligible.

Hard rabber 20.24 BEA ge. Weare tis ata Seatee Very small.

Rubber: dam J. ses. 17, A We sorte eh ciner ys Gus Less than 10.

Lampblack ..sssss07. One coat, painted on 6 (partly due to pin holes)
rock-salt. of rays transmitted by

R. S.

Weampolack NG. oteas es Two coats, painted at Less than 3% of R. S. rays.
right angles, on R. S.

Camphor smoke ....Smoked on R. S. plate, 20% of rays transmitted by

so thick a lamp filament R. S. plate.
is invisible through it.

Camphor smoke .... Very thick coat, flaking 6% of R. S. rays.
off.
Skin, freshly removed. About 2 mm. ............- Negligible.

Through the interest of a surgeon in a local hospital, a piece of
human skin was obtained immediately after removal from the body.
Its transmissibility was measured before it had materially lost its
moisture. The piece obtained was about 2 mm. in thickness, with some
fatty tissue adhering to it. When inserted as a screen in the arrange-
ment described above, its transmissibility was found to be wholly
negligible. Bazett and McGlone in a paper entitled ‘“* Temperature
gradients in the tissues in man” (Amer. Journ. Physiol., vol. 82,
no. 2, p. 415, 1927) have shown that in general an increase of 1° above
surface temperature is found at a depth of something over 3 mm.
below the skin. Forsythe and Christison (General Electric Rev.,
vol. 34, no. 7, p. 440, 1931) and others have pointed out that flesh,
since it consists largely of water, would be practically opaque to the
longer wave lengths, just as water is. It seems evident then that the
higher melikeron skin-temperature values are not due to the instru-
ment receiving radiation from deeper and warmer layers beneath the

surface.

I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 85

The melikeron is an instrument which responds sluggishly and is
rather difficult to manipulate. Furthermore, temperatures computed
from its readings depend upon the Stefan radiation constant and upon
the assumption that the radiation measured is similar to that of a
black body. For these reasons the melikeron-computed temperatures
should not be given equal weight with those measured by the thermo-
element, and the 1° difference noted may not be entirely real. There
are, however, three factors each of which tends to make the melikeron
skin temperature higher than the thermoelement values on the skin,
namely :

(1) Due to the ridges and roughness of the skin surface, the
thermoelement touches the outer and cooler parts of the ridges,
whereas the melikeron views both ridges and hollows.

(2) As shown by Bazett and McGlone (loc. cit., p. 433) the tem-
perature I mm. below the surface of the skin may be as much as
.6° C. higher than the surface temperature. Since the outer layer of
skin is scaly and comparatively dry, it may well transmit a small but
appreciable amount of radiation coming from the moist and warmer
layer below. —

(3) Each measurement with the melikeron requires several min-
utes. The involuntary, psychological reaction resulting from so long
an exposure of skin near the instrument aperture may tend to raise
the temperature of the exposed skin.

Our conclusion then is that the 1° higher temperatures on the skin
resulting in the mean from the melikeron observations would probably
be reduced to about 4° if all experimental error were removed. Due
to the combination of the three tendencies just mentioned, tempera-
tures at least several tenths of a degree higher than those measured
by the thermoelement appear to result from the melikeron readings
on the skin.

hae 231, fag Me es 9 . |
: me, iy, nie

- ali iv
iy

oy, viCnaneg Pah eank ff Nf a)
oe mae

, hg = bern 7 ane! remit e - a

i eb J
: j wii ul Bd Uh VE) es ee

8
2 bee ' ;
j 4 i ou
4} 1 ¥ Gs 4 - ; iW Ve Ae
ive a th 7 nee
f Ee ne, ty
asea Sb
j ; : as \ ; 7 : ey iy
; a Hl os ¢
7?
I ae, oy
Wey leans ro
i, eu aq i
sf 7.
Y ' | ,
j j
fi ‘
se 4)
: +
7 ' 1 t
5 dj -
ie ics 7
r ‘
ne
\ t =
* 7
oa i*
'
{ J
‘
j
> i
; ij
1 ‘
/ om i
+ \
Me
i
i
- os
: =
»
j
in
t
'
1
i
i
*
i
‘x aa

anes lien J

: ica ; a ; a i

Viagra on

Hr oie ae i i
ay oy j ; a i
ONE an p

von f ce a ae

May st eee, ae
Na vere

ua

ae

Z,

| | |
3 9088

TL

i

014

y
RSMo

eNArEve Y

See ee nea eT ee