Besos 
a ae 


‘Gait & 


7 NG 7 
aw. 
Wh + ba 
Ty Gi ae 
v 


i 
} 1. 


A 
Sind 


SMITHSONIAN MISCELLANEOUS COLLECTIONS , 


VOLUME 146, NO. 2 abd ob vara 


A CONTRIBUTION TOWARD AN 
ENCYCEOREDIA’ OF INSECT 
ANATOMY 


By 
ROBERT E. SNODGRASS 


Late Honorary Research Associate 
Smithsonian Institution 


(Pustication 4544) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
JULY, 12, 1963 
| 


4a 


f 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NO. 2 


A CONTRIBUTION TOWARD AN 
ENCYCLOPEDIA OF INSECT 
ANATOMY 


By 
ROBERT E. SNODGRASS 


Late Honorary Research Associate 
Smithsonian Institution 


(Pustication 4544) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
JULY. 12, 1963 


PORT CITY PRESS, INC. 
BALTIMORE, MD., U. S. A. 


FOREWORD 


At the time of his sudden death, on September 4, 1962, Robert 
E. Snodgrass was working on a book we might call “An Encyclopedia 
of Insect Anatomy.” His notes and correspondence suggest several 
possible titles, but this one seems most appropriate for the material. 
To judge from the list of terms he had compiled for the letters A to 
D, I would estimate that the work was only somewhere between 10 
and 20 percent completed. Most manuscripts would be unsalvageable 
when in such an early stage, but this one need not be thrown away. 
An encyclopedia may be considered as a dictionary in which definitions 
of maximum brevity are replaced by essays on the various terms. 
In this sense, each of the essays Dr. Snodgrass had written may be 
considered as complete—the work is incomplete only in the sense 
that he had progressed only a short way down the list of projected 
essays. Hence the title chosen for this publication. 

In consultation with Mrs. Snodgrass and others it was decided not 
to attempt completing the work, because who besides Snodgrass 
could write Snodgrass’s Encyclopedia? The essays are published 
almost word for word from the original manuscript. However, this 
was preliminary manuscript which did require some editorial emen- 
dations. No doubt, if he had lived, he would have done more revi- 
sion—such was his habit—but I have kept changes to a minimum 
in order not to alter the author’s meaning. Actually he had already 
done some rewriting, as shown by the fact that there were three 
versions of “Metamorphosis,” two of “Pleuron,” etc. In such cases 
the most extensive version is used here; in some cases additions to» 
it are taken from the less extensive versions. No attempt was made 
to make the several essays stylistically consistent with one another ; 
thus some begin with derivation of the word and/or a definition ; 
others do not. 

I presume that if this material had been completed it would have 
been assembled with the terms in alphabetical order. But, although 
Snodgrass had an aphabetical list from A to D, he was not writing 
simply by going down this list in 1, 2, 3 order. Rather he was writ- 
ing on series of related topics. Accordingly, in view of the limited 
amount that had been finished, it seemed preferable to assemble the 
finished articles into a natural rather than an alphabetical order. 
Perhaps this decision has one disadvantage. A certain degree of repe- 


ili 


1v SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


titiousness is inherent in a presentation of this sort in contrast to 
the presentation in a textbook of anatomy or morphology. To re- 
move the repetition would require so many cross references that the 
utility of the compilation would be seriously curtailed. Some of the 
repetition has been removed during editing this manuscript, but some 
of it has been left in for the same reason that the author put it there 
in the first place. With the subject-type of arrangement, instead of 
alphabetical, some of the repetitions are brought together in adja- 
cent articles where they become obvious in a way they would not 
have been were the manuscript complete and alphabetically arranged. 

Bibliographic references are limited to those he had written into 
the text. 

It is one of the lossés to entomology that this encyclopedia was not 
completed by the author. But even this group of essays is a contribu- 
tion. Unfortunately, it is his last contribution to entomology. 


A. GLENN RICHARDS 
Department of Entomology 
University of Minnesota 
St. Paul, Minn. 


List OF SUBJECTS TREATED 


Insect, entomology, Hexapoda.. 
Anatomical names 
Body segmentation 
Segments 
Segment areas and sclerotiza- 

tion 
Segmental plates 
Body regions and plates....... 
Tergum and notum........... 
12 Abra dchere Om SEA ODE eCeae 
Sternum 
External grooves of skeleton... 
—Ite 


aie ee 8 © 6,0 00 


@@ U6 6*6) 6 4S bieu ea) 6 b 4.00 600.6 


ey 


CC) Cire CPC ie er eC Ye Var UT WCC Me aC 


Pupa 
Metamorphosis 
Recapitulationt:,..-..s4006 fecceo: 
Moulting 
Ecdysis 


Gerdes = @ 616 = 0 © 6 be) < 06 000 6 6 8.6 


ee eee ee ee eesene 


ss sie ©.0 0 8/0 6.00 = as ¢ «6.0 6 


oa 6 S96 ue «eee 6706) 6).6 6).0)\6 0 


Page 


1 


OUOAANA AN & 


Alimentary canal 
Gastrula 


rea AC HCOMCEN aE 


Mesenteron 
Stomodaeum and proctodaeum. 
Head 
Epieranial sutires cc. cece cece 
Ecdysial cleavage line of head.. 
Antenna 
Neck 
Gulavess seer anicaecnine 
Thorax 
Spiracle 
Leg 


ee | 


ry 


ce) 


Siaijeliw ere) 0-6 \eijul/e’\ein\e/fe\/el'a) sits 1o.'0 se e).6 


ee 


CIC) Cay On uCUC nC nCUC CRC Ix ORNL) ECC i) 


Mleseenita laine sie cate 
Aedeagus 
Ovipositor 


[See also INDEX, page 47] 


AYCONTRIBUTION TOWARD AN ENCYCLO- 
PEDIA OMINSECT ANATOMY 


By Rosert E. SNopGRASS 


Late Honorary Research Associate 
Smithsonian Institution 


Insect, Entomology, Hexapoda: An insect, according to the 
composition of its Latin name (in + sectum, cut), is literally an 
“incut,” as it is also by its Greek name, entomon ( en + tomos, cut). 
The study of insects is entomology instead of insectology because 
the latter involves a combination from two languages. When arthro- 
pods came to be named according to the number of their legs, as 
decapods, myriapods, centipedes, etc., the 6-legged insects became 
hexapods and were classed as the Hexapoda (Gr. hexa, six, + pous, 
podos, leg). Hence we call them imsects, classify them as Hexapoda, 
call their study entomology, and call ourselves entomologists (= stu- 
dents of incuts). 


Anatomical names: The early zoologists who first studied the 
anatomy of invertebrate animals naturally carried over to what ap- 
peared to be functionally corresponding organs of the latter names 
that were long established in vertebrate anatomy. The anatomical 
names of insect parts, for example, except for a few applied on a 
basis of analogy, are almost wholly vertebrate names. It thus came 
about that the same names are applied to parts and organs in verte- 
brates and insects that can have no possible analogy. However, our 
whole anatomical terminology would be thrown into confusion if 
homology throughout the entire Animal Kingdom were made the 
basis of nomenclature. When organs are named on a functional basis, 
the same names are applicable to a worm, an arthropod, or a verte- 
brate. 

A food tract extending through the body, for example, is literally 
an alimentary canal in any animal in which it occurs. A blood-pump- 
ing organ is properly a heart regardless of its structure. An ap- 
pendage for walking is a leg. A head is a head whether on an insect, 
a snake, a man, or a snail. An organ of flight is a wing (pferon or 


SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 146, NO. 2 


2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


ala) whether on an insect, a bird, a bat, an angel, the devil, or an 
airplane. 

Of course, the early nomenclators made some mistakes in identi- 
fying organs of insects from comparison to vertebrates. For ex- 
ample, they called the cellular layer of the body wall below the 
cuticle the “hypoderm,” whereas it really corresponds with the epi- 
dermis of the vertebrate. The preoral space between the mouthparts, 
which are modified legs, they regarded as the mouth cavity of the 
insect and called the food pocket over the hypopharynx, now known 
as the cibarium, the “pharynx,” whereas a true pharynx is postoral 
and is an anterior part of the alimentary canal. Incidentally, they have 
left us the incongruous terms of epipharynx and hypopharynx for 
preoral structures which have no relation to the pharynx. 

A notable misnomer in insects is the term “suture” commonly given 
to the grooves of the exoskeleton that form strengthening internal 
ridges. The word suture can mean only a seam (sutura) or line of 
union between adjoining parts, and undoubtedly it was suggested to 
the early entomologists by the sutures of the vertebrate skull. The 
word suture has a specific meaning that could be applied to any line 
of union, but cannot be made to mean anything else. It is a distortion 
of its meaning to apply it to a surface groove formed by inflection 
of the cuticle. Of course, it is only in a figurative sense that any- 
thing in anatomy may be called a suture. The only true anatomical 
sutures are those made by surgeons. 

Another misnomer, now thoroughly established, is the application 
of the term chorion to the insect eggshell despite the fact that this 
shell is secreted by the ovarian follicle, whereas the vertebrate chorion 
is a cell layer proliferated by the embryo. 


It seems better to live with these incongruities than to attempt 
to rectify all of them. After all, everyone has some concept of the 
meaning of terms such as mouth, heart, leg, etc., and the only per- 
sons likely to be concerned with the differences between, for instance, 
vertebrate and invertebrate hearts are those who know the differ- 
ences. They will not be confused by using a term such as heart for 
several nonhomologous structures of different animal phyla. 


Body segmentation: The primary body segments of an adult 
insect are the annular sections of the integument marked by the lines 
of attachment of the longitudinal muscles. A body segment literally 
should be a somite (soma + ite), but preliminary to body segmen- 
tation there are formed corresponding pairs of cavities, the coelomic 
sacs, in the mesoderm. Some embryologists, as Manton (1949), de- 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 3 


fine the somites as the coelomic sacs and then contend that segmenta- 
tion begins in the mesoderm. This usage is confusing because the 
true mechanical segmentation of the body results from muscle attach- 
ments to the body wall. The muscles themselves, however, are de- 
rived from the walls of the mesodermal coelomic sacs. Since the 
coelomic sacs are typically connected with the exterior by coelomic 
ducts, their primary function was probably the collection of waste 
products to be excreted through these ducts. 

The primary segments of the body are established by the attach- 
ment of the longitudinal muscles to the cuticle. The lines of muscle 
attachment, as seen on the abdomen, are marked externally by trans- 
verse grooves which form internally submarginal ridges, the ante- 
costae, near the anterior edges of the terga and sterna. In a soft- 
bodied worm or insect larva the musculature, attached at the true seg- 
mental lines, brings about a shortening of the body and allows squirm- 
ing or flexing movements. In an animal with a fully sclerotized 
integument, however, such movements would be impossible. To give 
freedom of intersegmental movement, the posterior part of each 
segment remains membranous. The functional segments thus become 
the sclerotized annuli, and the connecting membranes are known as 
the intersegmental membranes. The definitive mechanism is thus a 
secondary segmentation. 


Segments (L. segmentum, from secare, sectum, cut off): The 
term applies to body segments or somites and also to leg segments or 
podites. 

The functional body segments are the sclerotized rings of the 
integument separated by flexible unsclerotized areas and movable on 
each other by intersegmental muscles. 

The true body segments are limited by the lines of attachment of 
the longitudinal muscles, marked externally by grooves of the cuticle 
forming anterior submarginal ridges or antecostae of the segmental 
plates on which the muscles are attached. This is the primary body 
segmentation which corresponds with the musculature. The func- 
tional segments represent a secondary segmentation since the so- 
called intersegmental membranes are the posterior of the primary 
segments. This secondary segmentation allows the consecutive seg- 
ments to be movable on each other because the connecting membranes 
can be infolded or extended according to the tension of the muscles. 
Where segments are united, as in the thorax, the membranes are 
either eliminated or themselves sclerotized as postnotal plates. 

The leg segments are movable by muscles arising in the proximal 


4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


segment, but the segmentation becomes confusing because the seg- 
ments are often divided into non-musculated subsegments. A true 
leg segment is thus best defined as a section of the limb provided 
with muscles (see Legs). In the same way the apical segment, the 
flagellum, of an antenna is commonly divided into subsegments (see 
Antennae). 


Segment areas and sclerotization: In an adult insect the cuticle 
of each segment is usually sclerotized in a definite pattern of plates, 
but the pattern may differ on different segments or on the same 
segment in different insects. There often results, therefore, some 
nomenclatorial confusion on the identification of the plates. 

In an unsclerotized wormlike animal, such as Peripatus, having 
a series of legs along each side of the under surface, the only dif- 
ferentiation of the body wall is its division by the legs into a dorsum 
above the legs and a venter between them. If the segmental body 
wall, as in some crustaceans, is completely sclerotized, the dorsal 
plate is a tergum or notum, the ventral plate a sternum. In some 
of the diplopods and crustaceans and in the prothorax of some in- 
sects the upper part of the tergal arch is produced on each side into 
a paranotal lobe. The sclerotized lateral parts of the segment are 
then called the pleura (sing. pleuron), and the name tergum or 
notum is restricted to the dorsal sclerotization above the lobes. In 
the winged insects the paranotal lobes of the mesothorax and the 
metathorax are extended as the wings. The pleura of these segments 
have to serve as supports for the wings as well as supports of the 
legs and are modified accordingly. Each is strengthened by a strong 
internal ridge formed by an external groove or sulcus from the leg 
base up to the wing base. The groove differentiates the pleuron 
into an anterior area called the episternum and a posterior area called 
the epimeron. At the wing base various small sclerites are formed 
which control the movements of the wings. Other modifications of 
the pleura are often present (see Pleuron), and the pleural area in 
wingless insects may be largely unsclerotized. The prevalent theory 
that a large part of the thoracic pleuron has been derived from a 
primitive “subcoxal segment” of the leg seems quite unnecessary 
from a comparative study. 

In the same way as the pleuron, the tergum and the sternum are 
usually differentiated into areas or distinct parts for mechanical 
reasons. ; 

On the abdominal segments, the terga and sterna are connected by 
membranes that may be regarded as pleural. But the small sclerites 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 5 


sometimes found in the pleural membrane of the abdomen appear to 
be detached parts of tergites or sternites and hence to be latero- 
tergites or laterosternites rather than true pleurites (see Abdomen). 


Segmental plates: Sclerotization of the body wall cuticle is highly 
variable in different parts of the insect according to the functional 
requirements. On the abdomen typically the sclerotization forms a 
back plate or tergum and a ventral plate or sternum separated on the 
sides by membranous areas to allow for the movements of respira- 
tion. On the thorax the support of the wings above and the legs 
below necessitates the presence of a strong lateral or pleural sclero- 
tization on each side. The head, though it includes at least four 
primary body segments, is continuously sclerotized above and on the 
sides to form a rigid cranium for the support of the antennae and 
the mouthparts. 

Since the skeleton of each section of the insect’s body is adapted 
to the functions of the particular part. it is difficult to deduce what 
the sclerotization of a primitive segment may have been. The centi- 
pedes with their undifferentiated bodies have on each segment a dis- 
tinct dorsal and a ventral plate with the legs arising from flexible 
pleural areas between. This condition, however, is simply an adap- 
tation to the centipede’s way of locomotion and is not necessarily 
primitive. On the other hand, in the lower Crustacea, such as Anas- 
pides, the back plates are continuous over the dorsum and down on 
the sides to the leg bases attached on the tergal margins. There are 
here no differentiated pleural plates. Among the Malacostraca, in 
the Mysidaceae the carapace covers only a part of the thorax, the 
segments behind it carry the legs on the lower margins of the terga, 
but where the carapace cuts through the back, the leg-carrying parts 
of the terga are cut off and are called pleural plates. The so-called 
pleural plates are here, therefore, only lateral parts of the tergal 
plates. Finally, in the diplopods the segments are continuous rings. 
It is clear, therefore, that there is no primitive basic pattern of seg- 
ment sclerotization, nothing comparable to the evolution of the bony 
skeleton of vertebrates, among the arthropods. An original wormlike 
creature probably had a soft cuticle which has been variously sclero- 
tized according to the needs in each group and according to the func- 
tional demands in each segment of the body. 

The sclerotized cuticle also becomes variously reinforced by linear 
inflections that form strengthening ridges on the inner surface. On 
the external surface these appear as narrow grooves or sulci, long 
erroneously called “sutures” in entomological terminology. The sulci 


6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


form characteristic lines on the head; on the thorax the pleuron is 
braced between the wings and the legs of the wing-bearing segments 
by a strong ridge-forming sulcus. Elsewhere, all over the body, simi- 
lar reinforcing grooves may be present. They differentiate the cutic- 
ula into areas known as sclerites, and have given the impression 
that the insect skeleton is composed of plates united along “sutures.” 


Body regions and plates: In describing the surface regions of 
the body or those of a body segment, we have in general three areas 
to distinguish and in each segment three corresponding sclerotiza- 
tions. To name these we have a choice of both Latin and Greek 
names for the body surface regions of an animal but no names for 
the segmental plates on the insects. Hence the available names have 
been used arbitrarily to fit the needs of insect anatomy without strict 
regard to the primary meaning of the words. 

The entire back of the insect or the back of any segment may be 
called the dorsum (L. for back), and from this we have the term 
dorsal. The back plate of a segment may then be given the name 
tergum, another Latin word for back. In the thorax, however, the 
Greek name notum is preferable in order to combine properly with 
the Greek prefixes pro-, meso- and meta- which designate the 
segments. 

For the sides of the animal we have no technical term in common 
use. Since, however, lateral refers to direction toward the side, it is 
to be assumed that the side itself is the Latin Jatus. Lateral sclero- 
tizations of the segments, when not a part of the dorsal or ventral 
plates, are termed the pleura (Gr. pleuron, a rib), and the pleural 
sclerites are properly pleurites. 

The whole underside of the animal is appropriately the ventral 
surface from the Latin word venter. The Latin word, however, 
meant specifically the belly (also the stomach or the abdominal 
cavity). A segmental sclerotization of the venter is a sternum (Gr. 
sternon, the breast or chest), whence sternutation or sneezing. 

The segmental tergal and sternal plates are often called “tergites” 
and “sternites.” The suffix ite, however, means “a part of” in 
anatomy, as in somite or podite. It is therefore incongruous to apply 
ite terms to whole plates, and, worse, it leaves us with no terms for 
parts of the terga and sterna when the latter are subdivided into 
true tergites and sternites. (It should be noted that tergite is pro- 
perly pronounced in English as tér’-jite.) 


Tergum and notum: Tergum is Latin for the back of men or 
animals, but, since we have also the Latin word dorsum for the whole 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 7) 


back (whence the adjective dorsal), it is useful to restrict the term 
tergum to a major plate of the dorsum. Many entomologists use 
“tergite” for a segmental back plate, but the suffix ite in biology means 
“a part of,” as in somite and podite. Properly, therefore, a tergite 
should be a division of a tergum; if the word tergite is used for the 
entire segmental plate we are left without a word for the parts of 
a subdivided tergum. 

Notum is the Latinized Greek equivalent of tergum (from Gr. 
noton). It is properly used for the back plates of the thorax in com- 
bination with the Greek prefixes pro-, meso-, and meta-. 


Pleuron: The term is derived from the Greek pleuron, pleura, 
a rib. The pleura in general may be defined as the lateral sclerotiza- 
tions of the body segments between the tergal and sternal plates. In 
insects such sclerotizations are present principally on the thoracic 
segments and are best developed in connection with the wings. 

The insect pleuron seems to have no prototype in the other arthro- 
pods. In the primitive crustacean Anaspides the back plates of the 
thoracic segments are continuous over the dorsum and down the 
sides, and they support the legs on their lower margins. In the 
Malacostraca the carapace cuts out the back of the dorsal plates, 
leaving the lateral parts as plates supporting the legs. These plates 
might be called “pleurites,” but they are simply remnants of the 
primitive terga. The diplopods likewise have no pleural plates sepa- 
rate from the terga. In the chilopods, plates in the pleural region 
above the coxae appear to be derivatives of the coxae. 

Among the insects, the pleural sclerotization of the thoracic seg- 
ments is never continuous with that of the dorsum. In the Protura 
and Thysanura, the terga and sterna are separated by wide mem- 
branous areas. The pleural sclerotization in each segment consists 
only of a pair of narrow sclerites concentrically arched over the base 
of the coxa; these are termed the anapleurite and the catapleurite. 
The same type of pleural sclerotization occurs in some larvae of the 
lower pterygotes and in adult termites. The presence of two supra- 
coxal pleural arches in the thoracic segments may be regarded as a 
primitive condition in the insects having no relation to anything in 
the other arthropods. 

In the pterygote insects the pleural sclerotization becomes more 
or less continuous over the sides of the thoracic segments but shows 
many modifications. Typically it is marked by a conspicuous groove, 
the pleural sulcus, extending upward from the leg base; this forms 
a strong ridge on the inner surface, on the lower end of which the 


8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


coxa of the leg is articulated. This sulcus and its ridge differentiate 
the pleuron into an anterior episternum and a posterior epimeron. 
Usually a triangular plate below the episternum, termed the trochan- 
tin, forms by its lower angle an anterior articular point for the coxa. 
The episternum itself may be variously subdivided, and often periph- 
eral parts of the pleural area remain membranous. In the wing- 
bearing segments the pleural sulcus extends up to the wing base, 
and its ridge forms the fulcral support of the wing. Before the wing 
fulcrum there is a small plate, the basalare, and behind it another, 
the subalare, that give attachment to the direct muscles of the wing. 
The pattern of the pleural sclerotization differs on the two alate 
segments according to the relative development of the wings and to 
the presence or absence of one of the pairs of wings. 

It is clear that the thoracic pleura of the pterygote insects are adap- 
tive developments, first for the support of the legs and then for the 
support of the wings as the latter were evolved from paranotal lobes. 
It has long been a popular theory that the pleura represent primitive 
subcoxal segments of the legs that have been incorporated into the 
thoracic wall. Yet a subcoxal segment is not present in any of the 
other arthropod groups; the coxa is always the functional base of 
the limb on which the principal motor muscles of the leg are attached. 
Differences in the leg segmentation among the arthropods are due 
principally to the presence of one or two segments in the trochanteral 
region of the leg. Most of the arthropods have a 7-segmented leg; 
the insect leg is 6-segmented by loss of the second trochanter (the 
crustacean basipodite). 


Sternum: The word is derived from the Greek sternon, which 
means the human chest or breast region. In the Latin languages the 
name was taken as the basis for words meaning sneezing, as in 
Latin sternuto and sternutatio, in Italian sternutare, in Spanish es- 
tornudar, and in Latin-English sternutation. In vertebrate anatomy, 
however, the name sternum was given to the breast bone (os pectoris 
in Latin). In arthropod anatomy it has been extended to any one 
of the segmental ventral plates of the skeleton. It is thus a curious 
coincidence that the word sternum as used in entomology is cognate 
with words signifying sneezing. 


External grooves of skeleton: Grooves on the surface of the 
integument, particularly those of the head and thorax, give the 
skeleton the appearance of being composed of sclerites united along 
these lines. The grooves, therefore, have long been called “sutures” 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 9 


(L. sutura, a seam). This was probably first suggested by the sutures 
in the vertebrate skull, which are formed by the coming together of 
bones growing out from centers of ossification. The analogy has 
given rise to the false impression that the insect skeleton with its 
“sutures” is formed by the union of parts developing from separate 
centers of sclerotization. 

Most of the grooves of the insect skeleton are actually lines of 
cuticular inflection forming internal ridges to strengthen the body 
wall in regions of mechanical stress. They are therefore not sutures 
in any literal sense, and for descriptive purposes are better termed 
sulci (L. sulcus, a groove or furrow). The Greek equivalent aulax 
has also been used. 

In a few cases grooves of the insect skeleton are lines of secondary 
union between sclerites. These might figuratively be called sutures. 

-Ite: A suffix used in biology to denote “a part of” some larger 
unit, as in somite, podite, sclerite, etc. Very commonly it is appended 
to tergum and sternum giving tergite and sternite for the major plates 
of the body segments. This usage, however, leaves us with no terms 
for subdivisions of the plates which properly would be the tergites 
and sternites. 

We encounter also the term gonoco.vite applied to what is evidently 
the coxa itself. The ite is here clearly unnecessary. The term co.ro- 
podite, however, is entirely correct since it means the coxal part of 
a leg. 


Larva: The word is derived from Latin and means a spectre, a 
ghost, a hobgoblin, or a mask. If we take the last meaning, a mask, a 
young insect is best defined as a larva if it differs so much in ap- 
pearance from its parents that it must be reared to determine its 
identity. When a young insect resembles its parents except for the 
full development of wings and reproductive capacity it is called a 
nymph or, in some aquatic orders, a naiad. [This distinction between 
and retention of the terms larva and nymph is not shared by many 
entomologists. Most embryologists and physiologists today do not 
make any distinction between the two; any immature insect is called 
a larva.—A. G. R.] 

Larvae of different species differ so much in the degree of de- 
parture from the adult form that it is evident they have under- 
gone various degrees of evolution diverging from the parental struc- 
ture. Larvae therefore can in no sense be regarded as representing 
ancestral adult forms of their species, nor can they be attributed to 
“early hatching” of the embryo—once a popular theory. We must 


Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


assume that at some time in the past history of the insects the young, 
as those of most other animal groups, resembled their parents except 
for immaturity, as does a modern young grasshopper or a young cock- 
roach. The question then is: Why have the young of some groups 
departed from the parental form along their own lines of evolution? 
The question is not so difficult to answer as it might seem, since some 
larvae are very similar to the adults and others depart in varying de- 
grees until they have lost all resemblance to the adults that produce 
them. 

As long as the young insect can live and feed in the same environ- 
ment as its parents, as the young grasshoppers and cockroaches do, 
there is no need of it having a special structure of its own. The 
adults of many insects, however, have taken advantage of their wings 
to explore other habitats for new sources of food, and in most cases 
they have been structurally modified for life on the wing and for 
feeding on some special kind of food. The flightless young, there- 
fore, could not possibly keep up with their parents. So, to insure 
the survival of the young, nature has fitted them for a way of living 
and feeding of their own. The young cicada affords a very simple 
example of juvenile metamorphosis since it is adapted merely for 
burrowing in the earth. The young mayfly and stonefly are supplied 
with gills for an aquatic life. More extreme cases are seen in the 
young of Lepidoptera, Diptera, and Hymenoptera. Caterpillars are 
adapted for climbing and feeding on vegetation, whereas the adults 
fly around and usually suck nectar. The young mosquito would 
starve if it had to feed on blood as does its mother or on nectar as 
does its father. Hence it has become strictly adapted to an aquatic 
life and equipped with a special feeding apparatus of its own. Young 
muscoid flies could not live the life of their winged parents and have 
become transformed into maggots fitted for other ways of living. 
The grubs of many Hymenoptera are fitted for living in cells where 
they would be completely helpless if not fed by the adult. 

In no case can the larva go over directly into the adult. It must 
at least discard its specialized larval structures, and the more it has 
departed from the parental form the more it has to discard. In ex- 
treme cases the larva is almost completely destroyed at the end of 
larval life. The modern adult represents the last stage of phylo- 
genetic evolution of its species; the larva is a temporary specialized 
form of the young insect. In ontogeny the larva develops first, but 
it must at last give way to development of the adult. (See Pupa.) 

Though the process of the destruction of the larval tissues and the 
resumption of imaginal development has commonly been called the 


NO: 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS II 


“metamorphosis” of the insect, the true metamorphosis is the change 
of form the larva has undergone in its independent evolution. (See 
Metamorphosis. ) 


Pupa: The term is taken over from the Latin word for young 
girl, puppet, baby, or doll. While there is no question as to the ap- 
plicability of the word, there has been much discussion as to the 
nature of the pupa. Does it represent the last nymphal instar of an 
insect without metamorphosis, or is it a preliminary form of the 
adult? Long arguments have been presented on each side of the 
question, but it seems that a few pertinent facts will give a sufficient 
answer. 

Naturally, since the pupa is formed inside the larva, when the 
larval cuticle is shed the pupa has the elongate form of the larva. 
On the other hand, the pupa has the imaginal compound eyes and 
the imaginal mouthparts, legs, and wings in a halfway stage of de- 
velopment. Clearly, therefore, the young pupa is a preliminary de- 
velopmental stage of the imago modeled in the larval cuticle. Within 
the larval cuticle it undergoes a stage of development and reconstruc- 
tion until when it finally casts off the larval skin it has the typical 
form of a pupa. Thereafter it does not change in external shape. 

The body of the mature pupa takes on the form of the imago. 
Thus it serves as a mold for the newly forming adult muscles and 
allows them to become attached properly on the imaginal cuticle. 
This alone has been proposed as a theory adequate to explain the 
pupa as a preliminary adult stage. On the other hand, it has been 
held that this theory of the pupa involves the unusual occurrence of 
a moult in the stage of holometabolous insects. But the mayflies 
moult once after attaining a fully winged condition, and the aptery- 
gote insects, as well as most other arthropods, moult successively 
throughout life. Still the pupal moult may be regarded as a second- 
ary one necessitated by the immaturity of the pupa. Moulting is 
determined by hormones, and hormones are powerful controlling 
agents in development. Insect endocrinologists have shown that they 
can make various adult insects moult again by transplanting into 
them the appropriate endocrine glands. 

The larval skin containing the young pupa has often been called 
the “prepupal stage of the larva,” but with the moulting of the larval 
cuticle, not yet cast off, the larval life is ended. The young pupa 
ensheathed in the larval cuticle has been called the “prepupa,” but 
it is simply a young pupa in a formative stage and still cloaked in 
the larval skin. It is not distinct from the mature pupa which is ex- 


I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


posed at ecdysis when the larval skin is shed. The young pupa still 
enclosed in the larval cuticle has, therefore, been more properly 
named by Hinton (1958) the pharate pupa (from the Greek word 
for hidden or concealed). The same term would apply to any larval 
stage still cloaked in the skin of the preceding instar, and to the adult 
when it is still cloaked in the pupal skin. Among the muscoid flies, 
the larva completes its growth, changes to the pupa, and finally to 
the adult, all inside the cuticle of the third larval instar. The cuticle 
of the third larval instar becomes greatly modified during this time 
and it is termed the puparium after this modification; from the 
puparium the fully formed adult emerges. 


Metamorphosis: The term is derived from the Greek words 
meta, a change, + morphe, form, + osis, a process of. Following 
its derivation the term metamorphosis means literally “a process of 
changing form,” and it should be emphasized that the implied change 
is one of form and not of substance. Thus it is comparable to the 
change of water to ice, not to the replacement of ice crystals by salt 
crystals or something else. The term, however, is widely used in 
zoology for almost any conspicuous change of form that an animal 
makes during its development regardless of how this is done. The 
tadpole is said to metamorphose into a frog, but it does so by a con- 
tinuous changing growth; and if this is metamorphosis then so is 
the embryonic development of any animal. The term probably origi- 
nated with the early writers of fiction who were fond of inventing 
tales about human beings who, at the whim of some offended god or 
goddess, were transformed into other animals or trees. It is, of course, 
to be supposed that in such imaginary cases the flesh and bones of 
the human were directly transformed into those of the animal. The 
early naturalists took over the word metamorphosis and applied it to 
the seemingly similar transformations of insects such as that of a 
caterpillar into a butterfly at a time when it was perhaps not known 
that the caterpillar was simply a young butterfly. Once established, 
the word metamorphosis became a standard part of our entomological 
nomenclature well before the true nature of the change from larva 
to adult was known. 

Modern studies on insect “metamorphosis” show that most of the 
larval tissues disintegrate and that the adult tissues and organs are 
newly built up in the pupa from cells that never formed an integral 
part of the larva. The adult cuticle is always a new secretion from 
the epidermal cells, which themselves may not change, though in some 
insects the larval epidermis itself is destroyed and replaced by an 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 13 


adult epidermis formed from islands of imaginal cells in the larval 
epidermis. The alimentary canal goes into dissolution, and the adult 
food tract is generated from replacement cells in the wall of the 
larval canal. The larval musculature may be completely destroyed 
and new muscles for the adult formed in the pupa. Some organs 
such as the tracheal and nervous systems may be simply remodeled 
to serve the needs of the adult. How much reuse versus remodeling 
versus replacement is involved for the cells within the nervous sys- 
tem has not yet been determined, but it is clear that the nervous 
system is not replaced in toto as some other systems are. Clearly, 
most of this process of change is not a metamorphosis of larval 
tissues into adult tissues but a replacement of larval organs by newly 
formed adult organs. The result is an entire transformation in the 
appearance of the insect between larva and adult. This is because 
the two stages are really two different animals—one stage is not 
transformed into the other. The egg simply has the potentiality of 
forming first the larva and then the adult, as was clearly expressed 
by Janet long ago (1909). 

The term metamorphosis has become so firmly established in 
entomological nomenclature that undoubtedly it will persist even if 
its erroneous implications become generally recognized. Insects have 
become famous for their metamorphoses. 

If the young insect in the form of a larva does not grow into the 
adult of its species, it may be of interest to speculate on its nature and 
how it came to differ from its parents. We must suppose that primi- 
tively the young of all insects resembled their parents except in mat- 
ters of immaturity, as do the young of a modern cockroach or grass- 
hopper. With most of the higher insects, however, the winged adults 
have become specialized for a life and ways of feeding that the 
flightless young could not follow. The young left behind were forced 
to adopt ways of living suitable to themselves and so have undergone 
a juvenile evolution quite independent of their elders; they have be- 
come specialized for their own various habitats and ways of feeding. 
Thus the larval stages have acquired many diverse forms in the 
several orders and have become as distinctive of their species as 
their parents (this is shown by the fact that taxonomists have been 
able to construct keys to larvae as well as to adults, and in some 
groups it is easier to identify larvae than the adults). Insect larvae, 
therefore, are not ancestral forms though many of them have taken 
on a wormlike shape. Structurally they remain insects. It must be 
clear that the evolution of young insects into their specialized modern 


14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


larval forms is the true metamorphosis of the insects. When the 
larva has served its purpose in the life of its species, it is practically 
destroyed and the developmental process reverts to the adult, which 
alone can perpetuate the species. The larval destruction and the 
adult reconstruction take place simultaneously in the pupa which is 
itself a preliminary stage of the adult. 

It is thus clear that the apparent change of the larval insect into 
the imago is not truly a metamorphosis. The term metamorphosis 
means literally “a change of form.” The change from caterpillar to 
butterfly, however, is a change of form only in the eye of the be- 
holder. Actually the change is a replacement of the larva by the 
butterfly. The writer has suggested the term retromorphosis for the 
reversion of morphogenesis to the adult line of development after 
the dissolution of the specialized larva (Snodgrass, 1961). The de- 
velopment of the adult and the destruction of the special larval tis- 
sues go on at the same time in the pupa, but the result is not a 
transformation of the larva into the adult. 


Recapitulation: This term as applied to individual development 
implies that an animal in its ontogeny goes through stages of de- 
velopment that represent successive adult forms in its phylogenetic 
history. Garstang (1922) has severely criticized this theory in its 
general concept, contending that the ontogenetic form in one gen- 
eration represents the ontogeny of preceding generations. This is 
particularly true of the larvae of holometabolous insects, which may 
be wholly adapted in their structure to their own way of living and 
feeding, and in no way represent adult ancestral forms of their 
species (see Larva, Metamorphosis). 

The development of the insect embryo also is in many ways an 
adaptation to its life in an eggshell in which it cannot possibly fol- 
low the evolution of its free-living ancestors. In minor ways, how- 
ever, the embryo is not necessarily prohibited from recapitulating 
the evolution of adult structures. It goes through a polypod stage, 
for example, when limb rudiments are present on all of the segments. 
At this stage it evidently represents a centipede-like ancestor. More- 
over, the growth and development of all the appendages from simple 
undifferentiated lobes should repeat a similar origin of the limbs in 
some remote wormlike progenitor of the arthropods. Likewise, the 
development of wings from simple outgrowths of the integument, 
whether they remain on the surface or are temporarily sunken into 
pouches, would appear to repeat the evolutionary development of 
wings from paranotal lobes. In other words, ontogenetic recapitula- 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS IS 


tion may take place in organs that are not affected by the living con- 
ditions imposed on the embryo or the larva. 


Moulting: The physiological process of separating the body cuti- 
cle from a new cuticle being formed beneath it by the epidermis. 
How this separation is accomplished is uncertain but it is soon made 
more obvious by the secretion from the epidermis of a moulting 
fluid which digests a greater or lesser amount of the old cuticle while 
the new cuticle is being secreted. The terms moulting and ecdysis 
have often been confused, with, as one result, the naming of the 
moulting hormone as “ecdyson.” Ecdysis, q.v., is literally the com- 
ing out of the insect from its moulted cuticle, and is not dependent 
on any hormone. 

The phonetic spelling “molting” which recently has become cur- 
rent in the U. S. A. is not justified by the derivation of the word 
from the Latin mutare. The “w” is clearly the essential vowel, and 
it is retained in other languages, as mudare in Italian and mudar in 
Spanish and Portuguese. 


Ecdysis: The word is derived from the Greek word meaning 
“coming out.” It is properly pronounced ék’-di-sis but is commonly 
heard as ek-dy’-sis. The word has commonly been defined as synony- 
mous with moulting, but the word can mean only the shedding or 
coming out of the moulted cuticle by the insect. Very commonly the 
newly moulted insect remains within the old cuticle for a variable 
length of time before emerging. With some insects, as the honey 
bee, the pupa goes through its preliminary change from the larval 
form within the last larval cuticle and comes out only when it has 
attained the final external form of the pupa. The most extreme 
case, however, is in the muscoid flies which form a puparium from 
the third larval cuticle, undergo the pupal stage within this, and then 
the adult emerges from it. 

The insect within the moulted cuticle of the previous instar is the 
pharate (cloaked) period of the larva, pupa, or adult (Hinton, 1958). 

Alimentary canal: The food tract of the mature insect always 
consists of three parts serially continuous but of different origins. 
The middle part, the mesenteron, represents the primitive endodermal 
stomach or archenteron of the gastrula; the anterior and posterior 
parts, the stomodaeum and proctodaeum respectively, are secondary 
ingrowths of the ectoderm. The origin and relations of the parts, 
however, are obscured in the development of the embryo by the 
various ways in which gastrulation takes place (see Gastrulation). 


16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


The two ends of the mesenteron must be interpreted as represent- 
ing the two ends of the blastopore, though no distinct blastopore is 
formed in insect development. The anterior end of the mesenteron 
is, therefore, the primitive mouth, and the posterior end the primitive 
anus. With the ingrowth of the stomodaeum, however, the primary 
mouth is carried inward, and the anterior opening of the stomodaeum 
becomes secondarily the functional mouth of the adult. This mouth 
of the adult becomes surrounded during development by the out- 
growing mouthparts and finally becomes enclosed by them in a space 
termed the preoral food cavity (often called the “mouth cavity” al- 
though, being outside the head, it is not truly a body cavity any more 
than is the space between the thoracic legs). 

The stomodaeum has a strong muscular sheath consisting of an 
outer layer of circular fibers and an inner layer of longitudinal fi- 
bers. In the head there are also numerous dilator muscles from the 
head wall and from the tentorium. The stomodaeum is commonly 
differentiated into several parts. That just within the mouth may be 
called the buccal cavity, next is the pharynx, of different form in 
different groups, and then the tubular oesophagus. The oesophagus 
is usually enlarged posteriorly as a crop, though in some insects the 
crop is a diverticulum of the oesophagus. Following the crop is a 
short division, the proventriculus, which opens into the mesenteron 
through a funnel-like infolding of the stomach wall, known as the 
stomodaeal valve. The proventriculus is commonly armed internally 
with cuticular teeth or other structures that presumably give it the 
function of a gizzard to grind the food, but in other cases it may 
possibly serve as a strainer, or merely to regulate the passage of food 
into the stomach. 

The mesenteron is usually a simple cylindrical sac, and it is the 
functional stomach or ventriculus of the insect. In some insects, 
however, it is divided into several parts (see Mesenteron). From 
the anterior end of the ventriculus there usually projects a circle of 
blind pouches, the gastric caccae, but tubular caecae may also be borne 
on other parts of the stomach in some species. 

The proctodaeum is divided into two principal regions, an anterior 
tubular part, which may be termed the anterior intestine, and an en- 
larged posterior part commonly called the rectum. The end of the 
anterior intestine adjoining the stomach is termed the pylorus (gate 
keeper) ; it gives off the excretory Malpighian tubules. The part 
following the pylorus is often differentiated into an anterior ileum 
and a posterior colon. The rectum consists of a large anterior rectal 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS v7, 


sac and a posterior narrow part or rectum proper. The terminal 
opening is the functional anus. 


Gastrula: The Greek word gaster really means the paunch or 
belly, but as used in anatomy the gaster is the stomach. Gastrulation, 
therefore, should mean stomach-formation regardless of the method 
of formation. 

Borradaile and Potts (Invertebrata, p. 127) state: “Every triplo- 
blastic animal passes through a stage—the gastrula—in which it con- 
sists only of ectoderm and endoderm. Save in this essential feature, 
the gastrulae of different animals may be extraordinarily unlike, and, 
especially when the animal is developed from a very yolky egg, they 
are sometimes very difficult to recognize as such; but where the 
gastrula is well formed, as in the familiar development of Amphioxrus 
or in that of a starfish, its two-layered wall may always be found to 
contain a cavity, the archenteron, which possesses a single opening, 
the blastopore.” 

The first development of a metazoic egg commonly leads to a hol- 
low mass of cells known as the Ddlastula. If the blastula represents a 
free-living ancestral form it probably obtained its food from the 
water through its surface cells. If it commonly lived on the bottom 
it would be natural that the cells of the underside would become 
specialized for ingestion and digestion of food material. Then it 
would be a further advantage if this surface should sink into the 
blastula. Thus the animal would become a two-layered sac, the 
cavity of which would be the primitive stomach or archenteron, the 
opening of which is the blastopore. The outer cell layer becomes the 
ectoderm, the lining of the stomach becomes the endoderm. 

In a few of the metazoic animals the stomach is formed during 
embryonic development by introversion of the ventral wall of the 
blastula. In an elongate animal the blastopore becomes divided into 
a mouth and an anus. This stage in the ancestors of the arthropods 
has been thought to be well represented in the onychophoran embryo 
which presents a median ventral groove that closes between the two 
ends. However, Manton (1949) finds that this groove is not the 
elongated blastopore because its formation does not give rise to the 
endoderm ; the endoderm is proliferated internally from a generative 
area behind the groove. It is this generative area which thus repre- 
sents the true blastopore. The mouth-anus groove, therefore, is a 
secondary formation, but evidently it must be formed in some way 
from the blastopore. 

The method of endoderm formation by introversion is commonly 


18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


replaced by the internal proliferation of cells from the ventral sur- 
face of the blastoderm. It is bewildering to read the various con- 
flicting accounts of gastrulation in the insects as reported by differ- 
ent writers and summarized by Johannsen and Butt (1941) and by 
DuPorte (1960), but much of the confusion results from not recog- 
nizing that introversion and proliferation may be just two super- 
ficially different ways of forming the endoderm. 

Gastrulation: This is the process of formation of the stomach 
irrespective of the method by which it is accomplished. The word 
itself is derived from gastrula, the dimunitive form of the Greek 
word gaster, which is used in anatomy for the functional stomach of 
an animal. 

The first development of the egg commonly leads to the formation 
of a hollow mass of cells known as the blastula. If the blastula rep- 
resents a free-living ancestral form of the Metazoa, it probably 
obtained its food from the water through its surface cells. If it com- 
monly lived on the bottom of a body of water, the cells of the under- 
surface may be supposed to have become specialized for the ingestion 
and digestion of food material. It would then be an advantage if 
these cells should sink into the blastula forming a cavity in which the 
food could be carried about and more leisurely digested. This food 
cavity is the primitive stomach or archenteron, the opening of which 
is the blastopore. The outer cell layer of the body is the ectoderm, 
the wall of the stomach is the endoderm. 

In the embryogeny of a few of the lower Metazoa the stomach 
is formed by this method of introversion of the digestive cells of the 
blastoderm. However, when the egg, as in most insects, contains a 
large amount of yolk which becomes surrounded by the blastoderm, 
gastrulation by introversion becomes entirely impractical. Aside from 
the mechanical difficulties of invagination when the center of the 
blastula is filled with yolk, introversion would place the yolk (food) 
in the body cavity and outside the stomach. The insect embryo, there- 
fore, cannot recapitulate the primitive method of stomach forma- 
tion; it must adopt some other method. 

The ways by which the endoderm is formed by the insect embryo, 
as reviewed by Johannsen and Butt (1941) and by DuPorte (1960), 
are seemingly so various that it becomes bewildering to attempt to 
interpret them all as derived from introversion or some modifica- 
tion thereof. However, the process of infolding one wall of the 
blastula may be replaced by the immigration of single cells. Such cells 
are usually called yolk cells, but, since they presumably act as 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 19 


vitellophags (yolk eaters), they probably serve for digestion of yolk 
during early stages of development; they should therefore be con- 
sidered as endodermal. In some of the lower insects the yolk cells 
are said to form a stomach wall by investing the yolk. In such cases 
the yolk cells seem to demonstrate their endodermal nature. 

With most of the higher insects a narrow ventral strip of the 
blastoderm becomes differentiated from the lateral plates, and either 
sinks into the yolk or is overgrown by the lateral plates. A ventral 
groove is thus formed along the ventral side of the blastoderm; this 
ventral groove has been regarded as a remnant of the elongated 
blastopore closed between the mouth anteriorly and the anus pos- 
teriorly. The enclosed ventral plate spreads out and divides into 
an inner endodermal layer and an outer mesodermal layer. 

The early embryo of Onychophora presents a median ventral 
groove that eventually closes between the two ends which become 
the mouth and the anus. The onychophoran, therefore, has long 
been thought to give an example of the primitive blastopore of the 
arthropods. Manton (1949), however, has shown that this mouth- 
anus groove of the onychophoran is not the blastopore. The endo- 
derm, she says, is proliferated from an area behind the anus, and 
cells from this area form the complete stomach epithelium. The 
mouth-anus groove is thus a secondary formation, though perhaps in 
some way derived from the blastopore. 

Even in the insects it must be noted that the concavity of the 
ventral plate does not become the stomach lumen. The functional 
endoderm is proliferated from cell masses at the two ends of the 
endoderm, and in some cases also from the whole length of it. The 
growth of the definitive midgut epithelium from cell masses thus 
resembles the proliferation of endoderm in the Onychophora and 
of yolk cells in the lower insects. The business of endoderm is to 
surround the yolk in order to digest it. This is accomplished mostly 
by the anterior and posterior cell masses which send out ribbons or 
sheets of cells toward each other around the yolk; these eventually 
unite and form the stomach which thus comes to contain most or all 
of the remaining food (yolk) of the developing embryo. 

Since in some cases the endoderm appears to be proliferated from 
the inner ends of the stomodaeum and proctodaeum, some embryolo- 
gists have contended that the insect stomach is ectodermal. It is 
noted by Henson (1946), however, that the two ends of the stomach 
represent the extremities of the blastopore where naturally ectoderm 
should be generated externally and endoderm internally. It is evi- 


20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


dent, then, that while ancestral recapitulation plays no part in the 
formation of the insect’s stomach, the embryo has adopted another 
method of gastrulation, namely cell proliferation, and thus does not 
violate the germ layer theory. 


Mesenteron: As indicated by the origin of the word (Gr. meso, 
middle, + enteron, alimentary canal) this is the middle portion of 
the food tract. It extends from the ectodermal stomodaeum in front 
to the ectodermal proctodaeum behind, and it becomes the functional 
stomach of the insect, known as the ventriculus. It is probably always 
of endodermal origin though variously formed in the embyro (see 
Gastrulation ; also Snodgrass, 1935, and DuPorte, 1960). 

In form the mesenteron is typically an elongate cylindrical sac, but 
it may be a slender coiled tube, and in some insects it is differenti- 
ated by constrictions into several well-defined sections. The anterior 
part that surrounds the stomodaeal funnel is called the cardia, Blind 
tubular pouches, known as gastric caecae and varying in length and 
number, project from various parts of the stomach wall. Most 
commonly, however, they project from the anterior end of the 
stomach around the entrance of the stomodaeum. 

The wall of the mesenteron is a thick epithelium of columnar cells 
separated from the hemocoel by a distinct basement membrane. Ex- 
ternally, beyond the basement membrane, there is a muscular sheath 
of longitudinally and circularly arranged fibers, but the arrange- 
ment differs from that around the stomodaeum and proctodaeum in 
that the longitudinal fibers are external to the circular ones. The 
inner ends of the epithelial cells are somewhat irregular and, as 
seen in sections, present what is known as a striated border due to the 
presence of alternating dark and clear lines that give a brushlike 
appearance. All the epithelial cells probably function for both secre- 
tion and absorption. Simple secretions are discharged through the 
striated border, but the cells also go through a disruptive process 
that has commonly been regarded as a form of holocrine secretion. 
Globules filled with granular material are extruded into the stomach 
lumen ; these are then constricted and break off, followed by dissolu- 
tion of their walls and the scattering of their contents. Whether 
this is a process of secretion discharge or simply a degeneration 
and dissolution of the cells, or both, it results in such a destruction 
of cells that they must be continuously replaced. Replacement is 
effected by groups of regenerative cells intercalated between the 
bases of the epithelial cells; by mitotic division these regenerative 
cells give rise to new cells that replace the worn out or discarded 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 21 


ones of the epithelium. In some cases the regenerative cells are 
contained in crypts projecting on the outer surface of the stomach. 
In some insects the epithelium is completely regenerated at each 
moult, and the larval epithelium is always replaced by an adult 
epithelium at the last moult of holometabolous insects, usually form- 
ing for the imago an entirely new type of stomach adapted to the 
special food of the adult. 

Usually there is one or more very thin sheets of secreted material 
separating the food from the surface of the midgut cells. This is 
the peritrophic membrane. In some cases (e.g., Diptera) it is clearly 
produced by a ring of cells at the anterior end of the mesenteron; 
in other cases (e.g., honey bee) it is delaminated from the surface 
of the mesenteron. The peritrophic membrane is composed of chitin 
and protein, and obviously must be sufficiently permeable to permit 
the ready passage through it of digestive enzymes and of digested 
products from the food. 


Stomodaeum and proctodaeum: The primary mouth (Gr. 
stoma) of the arthropods represents the enclosed anterior end of the 
blastopore, but it is carried inward by the tubular ingrowth of the 
ectoderm known as the stomodaeum. Thereby the primary mouth 
becomes the opening of the stomodaeum into the stomach, and the 
functional mouth of the insect is the external opening of the 
stomodaeum. 

Likewise, the primary anus (G. proktos) represents the open pos- 
terior end of the blastopore, but it is carried inward by an ingrowth 
of the ectoderm that forms the proctodaeum. Thereby the primary 
anus becomes the opening from the mesenteron into the proctodaeum, 
and the functional anus of the insect is the external opening of the 
proctodaeum. 

The words stomodaeum and proctodaeum mean literally “on the 
way to the mouth” and “on the way to the anus,” respectively. And 
this is just what they are! The -daewm part of these words is 
taken from the Greek hodaios meaning “belonging to a way” (from 
hodos, a way or path). By eliding the ho and latinizing the rest 
of the word, daeum is obtained. 


Head: The insect head is a continuously sclerotized cranium- 
like capsule. Its simpler or more generalized form and structure are 
best seen in the head of an insect such as a grasshopper or its rela- 
tives. In these the face is directed forward and the mouthparts hang 
downward. The compound eyes then have a lateral position, and the 


22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


antennae arise from the upper part of the face. The mouthparts 
(mandibles, maxillae, and labium) are suspended from the lower 
cranial margins, as the legs are from the thorax. The hypognathous 
head, therefore, should be primitive since the mouthparts represent 
appendages serially homologous with the legs. The ventral wall of 
the head is completely concealed by the mouthparts; it contains the 
mouth opening into the alimentary canal, and supports below the 
mouth a large median tonguelike organ known as the hypopharynx. 
The back of the head is perforated by a large opening into the neck; it 
is analagous to the foramen magnum of the vertebrate skull but is 
called the occipital foramen in insects. 

The cranial areas are given specific names. The top of the head 
is the vertex; the facial area between the antennae and compound 
eyes is the frons; below the frons is an area known as the clypeus, 
from which is suspended the broad, free anterior lip called the 
labrum; the sides of the head are the genae; and the back is the 
occiput. These head areas are merely topographical regions, though 
some may be separated by grooves or sulci of the cranial wall. Most 
commonly present is a prominent frontoclypeal or epistomal sulcus 
separating the clypeus from the frons and forming a strong internal 
ridge between the mandibles. Even this sulcus and ridge, however, 
may be absent, as in the cockroach in which the frontoclypeal region 
is continuous. In some insects a vertical groove below each compound 
eye separates the gena from the frons. A groove near the lower 
edge of the gena may set off a narrow subgenal area; internally it 
forms a ridge that strengthens the genal margin for the support of 
the mandible and the maxilla. The subgenal sulcus is usually con- 
tinuous anteriorly with the epistomal sulcus. The occiput may be 
separated from the vertex and genae by an occipital sulcus, but this 
sulcus is not commonly present. 

The head includes at least four primitive body segments united 
with an anterior protocephalic part bearing the eyes and antennae. 
The four known head segments are a premandibular segment bear- 
ing in some insects a pair of vestigial appendages, a mandibular seg- 
ment, a maxillary segment, and a second maxillary segment the 
appendages of which unite to form the labium. None of the head 
grooves mentioned in the preceding paragraph represent lines of 
segment union; they are merely cuticular inflections forming inter- 
nal ridges for the strengthening of the head wall along lines of 
mechanical stress. On the back of the head, however, there is a 
groove of a different nature. It closely surrounds the occipital 
foramen dorsally and laterally, setting off a narrow postocciput be- 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 23 


hind it and forming internally a deep ridge on which are attached 
muscles from the thorax that move the head. The fact that the 
basal angles of the labium are suspended from the postocciput indi- 
cates that the latter is a sclerotic remnant of the labial segment and 
that the postoccipital sulcus represents the intersegmental line be- 
tween the first and second maxillary segments. This intersegmental 
line alone has been retained on the head to provide for muscle at- 
tachments from the thorax. Anterior to this line there are no 
somatic muscles in the head; there are only muscles connected with 
the appendages and with the proctodaeum. It is not to be concluded, 
however, that the narrow postoccipital flange of the head represents 
the entire labial segment; the segment may well include a part of 
the membranous neck too. 

The mouth of the insect, as already mentioned, is in the concealed 
ventral wall of the head just above the base of the hypopharynx. 
Before the mouth, however, there is a large preoral cavity shut in 
by the mouthparts. Its anterior wall is the inner surface of the 
labrum and the clypeal region known as the epipharynx. Between 
the epipharynx and the base of the hypopharynx there is a food 
pocket, the cibarium, just before the mouth; the masticated food 
is deposited here before being taken into the mouth. The cibarium 
can be dilated by muscles from the clypeus, and contracted by trans- 
verse muscles in its anterior wall. In liquid-feeding insects, the 
cibarium becomes a sucking pump by the partial union of the edges 
of its epipharyngeal and hypopharyngeal walls. The duct of the 
thoracic salivary glands commonly opens into the preoral cavity above 
the base of the labium, but in some insects it enters the hypopharynx 
to open on it. In either case the saliva mixes with the food in the 
preoral cavity so that the food is all ready to be swallowed when 
taken into the mouth. 

The cibarium was long regarded as the “pharynx” of the insect; 
hence we have the incongruous terms “epipharynx’’ and “hypo- 
pharynx” for parts outside the mouth and having no relation at all 
to the true pharynx, which is a part of the alimentary canal within 
the head. The misapplied terms are still in current tisage because 
we have no appropriate substitutes for them. “Palatum” and “‘lingua” 
have been suggested, but both the palate and the tongue are properly 
intraoral. 

The cranial walls are braced by an internal skeletal structure 
known as the tentorium. It consists essentially of two pairs of 
apodemal processes. A pair of posterior arms arises from pits at 


, 


24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


the lower ends of the postoccipital sulcus; these grow transversely 
and unite into a posterior tentorial bridge just within the occipital 
foramen. A pair of anterior arms arises from the subgenal sulci 
above the mandibles, or more frequently in the epistomal sulcus; 
these grow posteriorly and unite with the posterior bridge. Pri- 
marily the tentorium therefore comes to have the shape of the 
Greek letter x, but often the space between the arms is partly filled 
by a central sclerotization giving the structure a resemblance to a 
canopy suspended by four stays. It is probable that the structure 
got its name from the latter situation because tentorium is the Latin 
word for “tent.” In some of the apterygote insects the anterior arms 
are not yet united with the posterior bridge, and there is evidence 
that these arms were primitively ventral head apodemes. On the 
other hand, in some of the higher insects modifications take place 
resulting in either enlargement or reduction of the anterior arms, 
and in some cases an obliteration of the middle part of the bridge. 
Such modifications, however, are clearly secondary. 

In immature insects the frontal region of the head is commonly 
marked by an inverted Y-shaped line the stem of which continues 
back over the vertex to the postoccipital margin. This line has long 
been called the “epicranial suture,” and supposed to be an important 
structural feature of the head. It is now known, however, to be a 
preformed line of weakness in the cuticle where the head wall will 
split at ecdysis (Snodgrass, 1947). The line on the vertex is con- 
tinuous with the splitting line on the back of the thorax, and the 
arms diverge downward on the face at various angles from the 
compound eyes to the clypeus. Only rarely is a remnant of this 
ecdysial cleavage line preserved on the head of an adult insect. 

This account of an orthopteroid head will give the student a pic- 
ture of the fundamental structure of the head in a pterygote insect. 
Numerous modifications, however, will be found in other orders ac- 
cording to the position the head takes on the neck and its adaptations 
to different feeding habits on the part of the insects. The orthopteran 
is said to be hypognathous because the mouthparts hang downward 
from the lower margins of the cranium. In a prognathous beetle 
with forwardly directed mouth parts, the change in position of the 
head on the neck has involved various alterations in the head struc- 
ture, particularly in its lower parts. A third type of head is 
opisthognathous, as in the Hemiptera, in which the sucking beak 
projects backward beneath the thorax and so causes adaptive changes 
in the head structure. These and other derived types of head struc- 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 25 
ture cannot be fully described here; the student must refer to special 
papers on the subject or to more general texts for wider information. 
In the study of any insect head, however, an attempt must always 
be made to homologize the special features encountered with the 
fundamental head structure from which the specialized types pre- 
sumably have been derived. To correlate structural evolution with 
changes in function is the essence of morphology. 

To understand fully the nature of the insect head it would be neces- 
sary to know its phylogenetic evolution. This we cannot know, but we 
can infer something about it from embryonic development. The pri- 
mary embryonic head in all the arthropod groups is a large lobe at 
the anterior end of the body on which the eyes and antennae are 
developed, and which contains the primitive brain ganglia. This 
protocephalon or first head, therefore, is entirely a sensory region. 
The mouth is formed by ingrowth of the stomodaeum at the base 
of its under surface. If the protocephalon truly represents the primi- 
tive head of arthropods it might well be termed the archicephalon. 
But as the head of the embryo, without any phylogenetic implica- 
tions, it has been well named the Ddlastocephalon by DuPorte 
(G. blastos, a bud or sprout, generally in embryology for the first 
recognizable beginnings of something, as in blastoderm, blastopore, 
ectoblast, etc.). 

Behind the protocephalon of the early embryo is a region of four 
body segments in front of the thorax. The first of these segments 
in some insects bears a pair of minute, transient limb vestiges which 
correspond to the second antennae of Crustacea, the second is the 
segment bearing the mandibles, the third is the segment of the first 
maxillae, and the fourth is that of the second maxillae which unite 
with each other in insects to form the labium. These four segments 
are eventually consolidated with the protocephalon to form the defin- 
itive head. The ganglia of the first of these segments are drawn 
forward and unite with the protocephalic brain to become the trito- 
cerebral lobes of the definitive brain. The ganglia of the other three 
segments combine to become the suboesophageal ganglion of the 
mature head. These are the visible facts of the embryonic develop- 
ment of the head. Theories on head segmentation are not so simple. 

Inasmuch as the embryonic head lobe lies in front of the mouth, 
bears the antennae and the eyes, and contains the primitive brain, 
it has been interpreted as representing the prostomium of the annelid 
worms (Holmgren and Hanstrom). This idea gives a very simple 
concept of the relation of the arthropods to the annelids. More 
recent embryological studies on the arthropods have, however, re- 


26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


vealed the presence of small, paired, temporary cavities in the meso- 
derm of the antennal region of the head, and another pair in the 
preantennal region. Some embryologists insist that any pair of 
coelomic sacs must represent a segment. They contend, therefore, 
that primitively the embryonic head lobe of insects contained both 
an antennal segment and a preantennal segment, thereby making 
six primary segments in the definitive head in addition to a small 
anterior prostomial region bearing the labrum. [The maximum num- 
ber of segments in the insect head, based on these coelomic sacs, is 
nine according to Janet; four of these would be in front of the 
antennal segment, if the antenna does indeed respresent a segment. | 
The contention is logical if we accept the premises. Coelomic sacs 
are spaces in the mesoderm for the accumulation of waste products 
of metabolism, and most of them have ducts leading to the exterior. 
The outer walls of the sacs form the longitudinal muscles that deter- 
mine the segmentation of the ectoderm. Where there is no ectoder- 
mal segmentation, as in the embryonic head of modern arthropod 
embryos may we not question that coelomic sacs are always ac- 
companied by ectodermal segments? Those of the embryonic head 
may have purely a physiological function (and a transitory one since 
they are not carried over to later stages of the insect). The reported 
presence of coelomic sacs in the labrum is particularly difficult to 
account for since few morphologists regard the labrum as repre- 
senting a segment. The actuality of an antennal segment and a 
preantennal segment in the primitive head of insects may, therefore, 
be doubted, but not outright denied. 

A theory of head segmentation promulgated a few years ago 
caused much confusion by its sensational claim that the tritocerebral 
lobes of the brain are the ganglia of a “labral segment,” because the 
labral nerves are connected with them (Ferris). This idea was based 
on observation that the endings of body nerves remain in the segment 
of their origin even after their central ganglion has been transposed 
to another segment, and that thus one may identify the segment of 
the ganglion. This generalization is true for motor nerves which 
arise from the ganglia and grow outwardly to the part which they 
will innervate. The labral nerves in question, however, are sensory 
nerves arising in the labral epidermis and growing in to the trito- 
cerebral ganglia. The origins of sensory nerves do not identify the 
segment of the ganglion to which the nerves go, and nothing indi- 
cates that the labrum is a head segment. We have here an excellent 
example of how revolutionary ideas can be drawn from logical 
reasoning based on false premises. 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 27 


Epicranial suture: See Ecdysial Cleavage Line of Head. 


Ecdysial cleavage line of head: This is the familiar inverted 
Y-shaped line on the front of the head of young insects. The stem 
of this line on the top of the head is continuous with the ecdysial 
line on the back of the thorax. The arms of the cleavage line ordi- 
narily diverge downward on the face at various angles, but in some 
hymenopterous larvae the stem is unbranched and continues straight 
down the face to the labrum. Though the cleavage line on the head 
has long been known as the “epicranial suture,” and regarded as an 
important structural feature of the cranium, it is in no sense a 
“suture.” It is merely a preestablished line of weakness where the 
head cuticle will split at ecdysis. A remnant of the line is rarely 
retained as a groove on the adult head (Snodgrass, 1947). 


Antenna: The antennae are paired segmented appendages of the 
head of the trilobites and most of the mandibulate arthropods. They 
are absent only in the Protura, the chelicerates, Limulus, the 
arachnids, and some insect larvae. In the diplopods, chilopods, 
symphylans, and the entognathous hexapods, the antennae consist 
of a variable number of divisions each of which is provided with 
muscles inserted on its base and arising in the proximal divi- 
sion. Such antennae, therefore, are fully segmented, and probably 
represent the primitive antennal structure. In the Thysanura and the 
pterygote insects, however, the antenna consists of a basal stalk or 
scape, a small intermediate pedicel, and a distal flagellum which is 
usually subdivided into a variable number of annuli. The only 
muscles in an antenna of this type arise on the scape and are in- 
serted on the base of the pedicel. The flagellar annuli have no 
muscles and vary in number from one to many; evidently the flagel- 
lum is a single subdivided segment. The pedicel contains a large 
sensory organ known as the Organ of Johnson. If the pedicel is a 
separate segment then it must have lost its muscles. The thysanuran- 
pterygote antenna consists of not more than three segments. 

Each antenna is movable as a whole by muscles arising in the head 
and inserted on its base. These muscles usually arise on the anterior 
arms of the tentorium. The whole antenna is set into a membranous 
socket of the head wall, and is pivoted on a point on the lower rim 
of this socket. Thereby it is freely movable in all directions. 

The antennae serve as delicate organs of touch and are the prin- 
cipal seat of the olfactory sense of insects. In addition, some in- 
sects, such as the male mosquito, hear at least the tone of the female’s 
wing vibrations by means of sensory hairs on the antennae. 


28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Whether the antennae are segmental appendages serially homolo- 
gous with the mouthparts and the thoracic legs is a question bound 
up with that of the segmentation of the head (q. v.). When an 
antenna is amputated, the flagellum may be regenerated in a form 
resembling the distal part of a leg, but the significance of this phe- 
nomenon is uncertain. 


Neck: The neck of the insect is a cylindrical membranous con- 
nection between the head and the prothorax. It varies somewhat 
in length in different insects. There are usually various plates, called 
cervical sclerites, in its walls; a lateral pair of these may form a 
support for the head by articulating on the postoccipital margin of 
the latter. The flexible neck allows for movement of the head in 
various directions by muscles arising in the prothorax and inserting 
on the postocciput or the postoccipital ridge. 

The morphology of the neck is difficult to understand from its 
musculature. The principal longitudinal muscles are dorsal muscles 
from the intersegmental phragma between the pronotum and the 
mesonotum and extending to the postoccipital ridge of the head, 
and ventral muscles from the prosternal apodemes to the cross bar 
of the tentorium. The extent of the muscles, therefore, might suggest 
that the neck is a part of the prothorax. In this case the postoccipital 
ridge of the head would be the intersegmental line between the labial 
segment and the prothorax, but this ridge is evidently the line be- 
tween the maxillary and labial segments. Otherwise we have to 
assume that the muscles are those of two primary segments, the 
labial and the prothoracic, that have become continuous with the 
obliteration of the intersegmental fold between labial and prothoracic 
segments. The embryo gives no clue to this problem because the 
labial segment before it is added to the head is followed directly by 
the prothoracic segment. Unfortunately, the development of the 
neck musculature has not yet been followed in the embryo. A larva 
has no appreciable neck. 

The number of head-neck muscles is variable in different insects. 
In the locust head, muscles arise from the pronotum as well as from 
the following phragma. When lateral neck plates are present they 
usually support the head on anterior processes, but when neck plates 
are absent the head may be supported on anterior processes of the 
prothoracic epineura. When a pair of lateral neck sclerites on each 
side are angularly articulated end to end, attached muscles, by re- 
ducing the angle, serve to protract the head. The variable structure 
of the neck contrasts with the standardized structure of a thoracic 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY 


SNODGRASS 29 


segment, and suggests that the neck mechanism has been secondarily 
developed in the different insect orders. 


Gula: The term is derived from the Latin word for gullet, wind- 
pipe, and neck, and in vertebrate anatomy it is used for the upper 
part of the ventral side of the neck next to the chin. In insect 
anatomy the gula refers to a ventral plate of the neck behind the 
base of the labium. It is commonly continuous with the postocciput 
of the cranium and may become united with the submentum of the 
labium behind the posterior tentorial pits. Since the anterior part 
of the neck is probably a membranous posterior part of the labial 
segment, both the postocciput and the gula appear to belong to the 
labial segment. The cervical sclerites lie behind the gula. A review 
of the literature is given by DuPorte (1962). 


Thorax: The term is derived from the Greek word thorax, a 
breastplate of ancient Grecian armor; in anatomy it refers to the 
part of the human body covered by a breastplate. The thorax of in- 
sects is the locomotor section of the body between the head and the 
abdomen. It consists of three segments, the prothorax, mesothorax, 
and metathorax, as a result of the reduction of the number of walk- 
ing legs to three pairs. Once established as the locomotor center, the 
thorax also became the site of wing development in the winged in- 
sects. Wings, however, are present only on the mesothorax and the 
metathorax, but either one of these pairs may be transformed into 
nonflight organs. 

The thoracic wall of pterygote insects is necessarily well sclero- 
tized and the presence of both legs and wings differentiates the cir- 
cumference of the segments into tergal, pleural, and sternal. The 
tergal plates are termed nota (Greek) in order to combine properly 
with the Greek prefixes pro-, meso-, and meta-. The notum of the 
prothorax is relatively simple because of the absence of wings on 
this segment. In the winged segments, however, the notum becomes 
the essential lever for the wing movement since its lateral margins 
must vibrate up and down to give the vertical movement of the 
wings in flight. The notal movements result from an alternating 
longitudinal upward curvature and flattening of the notum produced 
by constriction of longitudinal and vertical muscles. In adaptation to 
its function, the wing-bearing notum must be properly flexible. It 
is typically divided by a V-shaped ridge-forming groove in the pos- 
terior part (the apex of the V is forward) ; this apparently controls 
the bending of the notum so that the principal lateral movements 


30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


occur at the bases of the wings. Anteriorly it is commonly marked 
by a transverse groove, and the resulting three areas of the notum are 
termed the prescutum, scutum, and scutellum. Close to the anterior 
margin is another transverse groove that forms a deep internal ridge, 
the prephragma, for the attachment of the longitudinal dorsal 
muscles. These muscles, however, are intersegmental, their posterior 
attachment being on the phragma of the following segment. The two 
segments, therefore, must be solidly attached, and this is accomplished 
by a lengthening of the precostal part of the following segment as a 
postnotal plate firmly joined to the scutellar margin of the preceding 
notum. The contracting muscles thus give a strong upward curvature 
to the wing-bearing notum which effects the depression of the wings 
in flight. Flattening of the notum results from the contraction of 
vertical notosternal muscles, and produces the upward movement of 
the wings. 

In the 4-winged insects, the two winged segments are essentially 
alike, the postnotal plate of the second being derived from the first 
abdominal segment, and the third phragma likewise. In the clistogas- 
trous Hymenoptera, the first abdominal segment is so thoroughly in- 
corporated into the thorax during early pupal development that it 
becomes virtually a part of the thorax in the adult stage. The ab- 
dominal pedicel is then formed from the second abdominal segment. 

The pleuron of a winged segment is marked by a deep vertical or 
inclined groove from the leg base to the wing base. This forms a 
strong internal pleural ridge to strengthen the pleural wall in its 
double duty of supporting both the leg and the wing (see Pleuron). 
The ridge forms the coxal articular process at its lower end, and 
the wing fulcrum at its upper end. In the prothorax, the ridge sup- 
ports the leg but usually does not extend on to the back. The pleural 
ridge also usually gives off a strong apodemal arm directed inwardly. 
The pleural sulcus divides the pleural surface into an anterior 
episternum and a posterior epimeron, but these parts may themselves 
be further differentiated into areas by sulci or by desclerotization. 

The sternal region of the thorax is generally continuously sclero- 
tized in each segment except for small membranous areas between the 
major plates. The latter is often differentiated by a transverse groove 
into an anterior basisternum and a posterior sternellum. Between the 
two parts at the ends of the groove arises a pair of sternal apodemes 
which are commonly united at their bases to form a Y-shaped process 
known as the furca. The arms of the furca turn outward and are 
closely associated with the inner ends of the pleural process, the 
two being usually connected by short muscle fibers. The pleural 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 31 


plates are then braced against the sternum. The small intersegmen- 
tal sternites are known as the spinasterna because each usually sup- 
ports a pair of internal processes for muscle attachments. There is, 
of course, much diversity in the relative size of the sternal parts in 
different insects, and a large part of the venter in each segment may 
be membranous. If the sternal sclerotization extends to the base 
of the leg on each side it forms a ventral articular process for the 
coxa. 

While the above description applies to the thorax of most winged 
insects, there are structure and function variations in the different 
orders. In the honey bee, for example, a line of flexion in the 
mesonotum cuts across the posterior part of the scutum and scutel- 
lum between the bases of the wings. In two-winged insects the 
metathorax is generally much reduced in size but retains the funda- 
mental thoracic structure, showing that the 4-winged condition is 
primitive for these groups. 


Spiracle: The breathing apertures should be called spiracles 
from the Latin word spiro, to breathe, and spiraculum, a breathing 
hole. The term stigmata formerly given to them means “spots,” and 
probably reflects the ignorance of the early entomologists regarding 
their function, A “stigma” was also a brand with which slaves were 
marked, and hence a blemish. The word should be discontinued as 
an entomological term. 

A spiracle is more than simply a hole into a trachea; it is usually 
a depression of the cuticle into which the trachea opens, forming 
thus a spiracular atrium. Most spiracles have a special closing ap- 
paratus which may be the outer lips of the atrium but more com- 
monly is a valvelike structure at the opening of the trachea operated 
by muscles. 


Leg: The Latin word for leg is crus, cruris, but this word and 
its Greek equivalent skelos have not been adopted into anatomical 
terminology for the legs of an animal though we have the crura 
cerebri of the brain and crural nerves of the legs. On the other 
hand, the Latin pes, pedis, and the Greek pous, podos, each strictly 
meaning the foot, have become the basis of most leg names, as in 
biped, centipede, milliped, arthropod, diplopod, hexapod, etc. 

In insect anatomy the word Jeg is used in a functional sense rather 
than a morphological one, since it is applied to the thoracic legs and 
to the abdominal prolegs of larvae though the two sets of organs 
have no homology. The thoracic legs represent the embryonic leg 


32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


rudiments which reach their full development as organs of locomo- 
tion only on the thorax. The appendages of the head became feed- 
ing organs; those of the abdomen are represented by vestiges in the 
embryo which disappear unless some of the external genital organs 
are derived from them. 

The thoracic legs of the insects are 6-segmented; those of most 
other arthropods usually have seven segments. The segments, be- 
ginning at the leg base, are named coxa, trochanter, femur, tibia, 
tarsus, and pretarsus. In a 7-segmented arthropod leg there are two 
segments in the trochanteral region, the basipodite and the ischiop- 
odite. A leg segment is best defined as a section of the limb inde- 
pendently movable by muscles. The tarsus may be a single segment, 
but it is commonly divided into as many as five small parts which, 
though they are frequently called “tarsal segments” are really sub- 
segments or tarsomeres since the only muscle of the tarsus are those 
of its base arising in the tibia. The pretarsus bears the terminal 
claws of the leg (often called tarsal claws). The pretarsus, however, 
is clearly an end segment of the leg corresponding with the 
crustacean dactylopodite. In some insects it is a simple clawlike seg- 
ment; in others it becomes 3-clawed by the development of a pair 
of lateral claws, but generally the median claw is lost and the typical 
insect foot has only the pair of lateral claws. 

The pretarsus has only a single ventral muscle of several branches 
arising in the more proximal segments of the leg; these attach on 
it by a long tendon traversing the tarsus. In this feature the insects 
resemble the centipedes (in the Crustacea there are both levators 
and depressors of the dactylopodite). 

The leg segments are connected by short membranous areas that 
allow them movement on each other. These flexible areas are the true 
joints of the limbs (from the French word joindre, to join), and 
this term should not be used for the segments themselves. Move- 
ments of the segments are controlled by articulations between them 
which are sclerotic extensions through the joint membranes from 
the opposing ends of the segments. The movement of the distal seg- 
ments at a joint depend on the nature of the articulation, some are 
dicondylic, others monocondylic. 

The so-called prolegs of insect larvae are short, unsclerotized, 
cylindrical outgrowths of the body. They have no structural re- 
semblance to the thoracic legs, and there is no proof that they 
originate from the abdominal leg vestiges usually present in the 
embryo. The best-known examples are those of the caterpillar, each 
of which end in a claw-bearing foot pad. Body muscles are attached 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 33 


on the base of the proleg, but the principal muscle is a long bundle 
of fibers arising on the lateral wall of the body and attached distally 
in the foot pad. The prolegs of the caterpillar serve principally for 
the support of the long and heavy abdomen and for grasping a stem 
or twig when climbing. 


Wings: Of all the animals that fly, the insects alone have wings 
developed independently as organs of flight. The others have con- 
verted a pair of legs into wings. To be sure, the winged creatures 
of fiction imitate the insects in having wings specially created for 
flight, but it is doubtful that any of them could really fly if alive. 

The insect wings grow out in immature stages of the nonmetab- 
olous orders as small flat lobes from the edges of the back on the 
mesothorax and metathorax. Some fossil insects have similar 
lobes on the prothorax, suggesting that the ancestors of the winged 
insects had three pairs of paranotal lobes. Since at this stage they 
could not have served for flight, it is postulated that at first the lobes 
enabled the insects to glide through the air a longer distance than 
they could jump (in the manner of flying squirrels, etc.). If the 
second and third lobes then became lengthened and flexible at their 
bases, they might have been able to flap up and down, and thus sus- 
tain the insect in the air longer. 

From some such early stage of wing development it seems to have 
required some evolutionary experimenting to produce an efficient 
mechanism of flight. The dragonflies have the simplest way of mov- 
ing the wings. Each wing is pivoted on a process of the pleuron 
and is moved by antagonistic muscles inserted on the wing base at 
opposite sides of the fulcrum. Yet the dragonflies even today are 
among the most efficient of flying insects. The cockroaches, mantids, 
and termites are weak flyers compared with the dragonflies, and it 
is not well understood just how they move their wings. They have 
neither the dragonfly mechanism nor the typical thoracic muscula- 
ture of the higher insects. The wings in these groups are supported 
on pleural fulcra, and muscles acting on the wing base before and 
behind the fulcrum probably effect a depression of the wings, while 
it is possible that the numerous leg muscles attached on the back 
sufficiently flatten the notum to raise the wings. Though the thoracic 
musculature of these insects is well known, no real study has been 
made of its action on the wings. 

The typical insect wing mechanism is found in the mayflies and 
in all the higher orders. It provides for the up-and-down wing 
movements and for a partial rotation of each wing on its long axis, 


34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


the latter being necessary for directed flight. Both sets of move- 
ments depend on a strong support of the wing on a pleural fulcrum. 

The vertical wing movements result from an alternating upward 
curvature and flattening of the wing-bearing notum, the margins of 
which thus depress or elevate the attached wings on the fulcral sup- 
ports. The notal movements are produced by the antagonistic con- 
traction of the dorsal longitudinal muscles and the notosternal verti- 
cal muscles of the wing-bearing segments. But this action of the 
muscles involved some radical alterations in the thoracic structure. 
The dorsal muscles are intersegmental, and ordinarily serve to pull 
the two tergal plates together, accompanied by an infolding of the 
intersegmental membranes. To effect a dorsal curvature of the 
notum, therefore, the intersegmental membrane has to be replaced 
by a sclerotization that would solidly unite the consecutive notal 
plates. The sclerotized membrane forms the so-called postnotum of 
the back. The tension produced by the muscles must now effect an 
upward curvature of the notum, giving the downstroke of the wings. 
The upstroke then follows from the depression of the notum by 
the vertical notosternal muscles attached on it. With the develop- 
ment of the wings the dorsal muscles became enormously increased 
in size, and to accommodate them the notal ridges of their attach- 
ments have been expanded into large plates, the intersegmental 
phragmata of modern insects. 

The wing being a flat fold of the body wall, its upper layer is 
continuous with the supporting notum, and its lower layer is reflected 
into the pleural wall. The basal region of the wing is largely mem- 
branous to allow for flexibility, but to maintain a hinge movement on 
the notum small sclerites are present in the membrane that articulate 
with specific wing processes of the notal margin. These sclerites, 
though on the upper surface of the wing, are known as the avillaries. 
A first axillary sclerite articulates with an anterior wing process of 
the notum, a third axillary with a posterior notal process. An inter- 
mediate second axillary loosely connects the other two and forms 
ventrally the pivotal point of the wing on the pleural wing fulcrum. 

The mechanism that converts the flapping wing into an organ of 
flight pertains to the pleuron. The under surface of the wing is con- 
tinued into the pleuron by wide membranous areas before and behind 
the pleural fulcrum. The wing, therefore, rocks freely on the ful- 
crum. In the membranes before and behind the latter are small 
sclerites, the basalare and subalare respectively, on each of which 
are attached vertical muscles. Since the sclerites are closely connected 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 35 


with the wing base, contraction of the basalar muscles turns the wing 
somewhat forward and deflects its margin during the downstroke. 
With the upstroke, the subalar muscles turn the wing posteriorly and 
deflect its posterior margin. The wing thus acts as a propeller, and 
with the downstroke it exerts a backward pressure on the air that 
drives the insect forward. 

This account of the flight mechanism must be understood to give 
only its fundamentals and the action of the muscles that is the basis 
of the wing action with most insects. Other factors, however, may 
complicate the picture, and variations in the thoracic structure in the 
different insect orders involve modifications in the wing and its mecha- 
nism. A notable example is the conversion of the hind wings of 
Diptera into small knobbed oscillating stalks called haltcres, so called 
because they were first regarded as balancers. It has now been shown 
that they have a gyroscopic action in some way stabilizing the in- 
sect’s flight. The development of the halteres leaves no doubt that 
they are the reduced metathoracic wings. Another example is the 
modification of the forewings of beetles into elytra which seem to 
be good protective shields but are of no help in flying. 

For a fuller discussion of the principles of flight mechanics and 
aerodynamics than can be given here the student is referred to the 
book “Insect Flight” by Pringle (1957). 

Returning to the anatomy of the wings, the primitive wing lobes 
have been lengthened and properly shaped during their evolution 
into organs of flight. Necessarily, the wings must be as thin and light 
as possible, and at the same time stiff enough to withstand air pres- 
sure. This involved the development of lines of rigid, branching thick- 
enings known as the wing veins. An old idea is that the veins were 
formed around tracheae, but this is not supported by recent critical 
studies (see Whitten, 1962). Another old idea is that the wings 
were first gills, but this idea has been superseded by the glider theory 
of wing origin. The venational pattern must have been established 
early in the evolution of wings since the wing veins seem to con- 
form to a fundamental pattern which permits a generally uniform 
nomenclature. 

In the holometabolous insects the wing buds of the embryo are 
sunken into pockets of the epidermis. These pockets become closed 
off externally, and remain thus concealed throughout larval life. The 
larva is thus not encumbered with externally growing wings, and 
none of the thoracic modifications related to the wings are developed 
until the pupal stage. Throughout the whole span of its life, then, 


36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


the larva preserves the larval simplicity of its thoracic segments. The 
wing rudiments continue to grow in their pockets, and at the moult 
to the pupa they become evaginated as well developed lobes resem- 
bling the wing pads of hemimetabolous nymphs. Their final develop- 
ment, together with that of the adult thoracic structure is then de- 
veloped within the pupa, so that the emerging adult is fully able to 
fly. In most cases the adult has to expand the wings and allow them 
to dry and harden, but in some aquatic groups (e.g., mayflies) the 
adults emerge and fly immediately. 

It is clear that the evolution of the wings and of the thoracic 
modifications that enabled the glider lobes of the early insects to be- 
come organs of flight must have been a long and complex process. 
The winged insects, however, owe almost all that they are today to 
their wings. Note what simple creatures by comparison are the ap- 
terygotes, which probably have changed little since the time they first 
became hexapods. The wings of higher insects freed the adults from 
a ground existence, and many of them have taken advantage of their 
freedom to adopt new kinds of food and new ways of feeding, for 
which they have developed new types of mouthparts. The young in 
such cases could not lead the lives of their parents, and have become 
adapted to habitats and ways of living of their own. Thus it has 
come about that the young of these insects have been specialized to 
such an extent that they have lost all resemblance to their parents. 
The adult development is then delayed to the end of the larval life 
when the special larval tissues are destroyed in the pupa. This change 
from larva to adult is commonly known as metamorphosis, but really 
it is largely a replacement of the larva by the adult. 


Abdomen: The name “abdomen” for the third section of the 
insect body does not clearly follow from its derivation; however, 
the insect abdomen does contain the principal viscera, and thus it may 
be likened to the vertebrate abdomen. 

The primitive abdomen probably had 12 segments, this number 
being present in some embryos and in adults of the Protura. The 
terminal segment bearing the anus is probably a telson, since the 
last embryonic appendages in the embryo, which are retained as the 
cerci, pertain to the penultimate or eleventh segment. In most adult 
insects, however, there are only 10 abdominal segments, and the 
cerci are carried by the tenth segment and the anus is contained in 
an apical lobe. 

The base of the abdomen may be broadly joined to the thorax or 
narrowed to a petiole. In the Hymenoptera the first abdominal seg- 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY-——SNODGRASS 37 


ment is incorporated into the thorax as the propodium, and the sec- 
ond segment forms a petiole. The petiolate part of the abdomen is 
sometimes called the “gaster,” but this is an inappropriate name for 
it, since the gaster is properly the stomach. 


Male genitalia: The external genital equipment of male insects 
includes primarily organs for the insemination of the female and 
secondarily copulatory organs for holding her. To the taxonomist 
the male genitalia offer the best characters he has for the separation 
of species because of their highly diversified structure. This very 
fact, however, makes the study of the genital homologies difficult 
and has given rise to a great deal of confusion in current termi- 
nology. Recent studies on the development of the organs have given 
a better understanding of their basic structure and the homologies 
of the parts. Thus there has been made possible the adoption of a 
more uniform nomenclature. That confusion still persists is due 
largely to the fact that specialists in each order of insects insist on 
retaining their traditional ordinal nomenclature. 

Since the inner organs of reproduction are duplicated on opposite 
sides of the body, and each has its own outlet duct, it is probable 
that primitively the ducts opened through paired external apertures, 
as they still do in some arthropods other than the insects. For effi- 
cient insemination of the female it became more practical to have the 
openings of the male ducts carried out to the ends of simple tubular 
outgrowths. Thus we find paired penes present in Crustacea and 
Diplopoda and, among the insects, in Ephemerida and Dermaptera. 
In most of the crustaceans and diplopods the penes arise on the bases 
of a pair of legs and are not themselves intromittent in function. 
The sperm is transferred from the male to the receptacle of the fe- 
male by one or two pairs of modified legs (gonopods) of a following 
body segment. The insects have not adopted this indirect method 
of insemination, but they have developed a great variety of struc- 
tures in the copulatory organs, associated with the organ of insemina- 
tion, for grasping and holding the female. 

The penes of Ephemeroptera arise from a small ventral plate or 
a pair of plates above the stylus-bearing plates of the ninth segment ; 
these evidently belong to the much reduced tenth segment of the 
abdomen. The penes vary in form in different species, in some that 
are armed with subterminal prongs, and rarely they are united basally 
in a single organ, but the ducts are always separate. 

In the Dermaptera the two penes are united basally on a long 
apodemal plate and are variously developed in their distal parts, so 


38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


that they have little resemblance to the simple organs of the Ephem- 
eroptera. Distally each organ splits into a median tube containing 
the exit duct and a strong outer lobe. In some species the two penes 
are united in a single organ with three terminal lobes, the median 
one giving exit to the two ducts, or to only a single duct if one duct 
is reduced and nonfunctional. The Dermaptera thus give an example 
of the modification potential of paired penes. 

The division of each penis in the Dermaptera into a mesal lobe 
containing the duct and into a clasperlike lateral lobe is suggestive 
of the division of each primary phallic lobe of the higher insects into 
a mesomere and a paramere. In the latter, however, the ducts never 
enter the mesomeres but unite with a common duct formed between 
their bases. 

The Thysanura in the adult stage have a single median penis aris- 
ing between the bases of the ninth segment stylus-bearing plates ; 
into the base of this the two genital ducts open by a common orifice, 
The penis, however, is developed from two small primary lobes that 
become concave on their opposed surfaces and unite to form the 
tubular organ of the adult. The penis lobes have no connection with 
the stylus-bearing plates of the ninth segment, and thus would ap- 
pear to belong to the tenth segment. Heymons (1899) has noted that 
in fact the embryonic ducts of Lepisma end in the tenth abdominal 
segment. 

In the Orthoptera and the higher insect orders, a new type of 
genital apparatus appears. It likewise begins as a pair of simple 
lobes, but the lobes do not give exit to the genital ducts. The paired 
ducts open into a single duct that grows inward between the lobes 
and becomes the unpaired ejaculatory duct of the adult. It is perhaps 
possible that the two lobes in this case are the paired penes of the 
lower insects from which the ducts have withdrawn, but there is no 
direct evidence for such. A common idea has been that the primary 
genital lobes are rudiments of appendages serially homologous with 
the embryonic limb vestiges of the pregenital abdominal segments 
and the thoracic legs. They arise, however, close together on the 
venter of the ninth segment in nymphal or late larval stages after 
the disappearance of the embryonic limb vestiges. Their position 
on the ninth segment is always behind the area of the sternal plate. 
Possibly, therefore, they belong to the tenth segment as do the penes 
of the lower insects, and have been moved forward into the “inter- 
segmental” membrane of the ninth segment. In any case, the future 
history of these primary genital lobes has no counterpart in the lower 
insects. 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 39 


Each primary lobe divides distally into two secondary lobes, a 
median mesomere, and a lateral paramere. Then in most cases the 
mesomeres unite by their edges to form a hollow median organ, the 
aedeagus, with the gonopore in its base; the lateral parameres be- 
come the claspers of the adult. The genital organs of the Orthoptera 
begin their development in this way as a pair of lobes that divide, 
but in their later growth they take on such a diversity of structure 
in the different families that the adult organs have little or no like- 
ness to the genitalia of the other orders. 

It is in the Hemiptera that we first encounter the genitalic struc- 
ture typical of the higher orders. The mesomeres of the primary 
genital rudiments unite to form the median aedeagus, the parameres 
become the adult genital claspers. This same genitalic complex, 
though with many modifications, can be followed through to the most 
highly evolved orders. The parameres may be simple lobes, as in 
Hemiptera and Coleoptera, but in the other orders each is usually 
differentiated into a basal part and a muscularly movable distal part. 

The known development of the genitalia leaves little doubt of the 
homologies of the major parts in the different orders. A great diver- 
sity in the genitalic nomenclature from one order to another, how- 
ever, has grown up because most taxonomists are ordinal specialists 
and hence are interested in maintaining an established set of descrip- 
tive names handed down from their predecessors. This contrasts with 
their proclivity at changing the Latin names of the species and genera 
with which they deal. But a uniform nomenclature is desirable, as 
well as now being possible, and the lack of a uniform genital termi- 
nology is highly inconvenient to the nonspecialist and must be dis- 
tressing to teachers and their students. 

The common assumption that the primary genital lobes represent 
a pair of former legs has given the parameres the theoretical status 
of “gonopods.” When divided, the basal part is identified as the 
“gonocoxite,” the distal part as the “gonostylus.” The aedeagus is 
then supposed to have been formed by the union of endite lobes of 
the “coxites.” All this does very well as a basis for a practical no- 
menclature, but a leg origin of the genitalia has never been demon- 
strated or even supported by any concrete evidence. The postem- 
bryonic origin of the primary lobes and their median position behind 
the sternum of the ninth abdominal segment are in strong contrast 
to the true limb vestiges seen in the embryonic abdomen. There is 
no evidence that the mesomeres are endite lobes of the parameres. 
The primary genital lobes arise on the venter of the ninth abdominal 
segment, but always behind the region of the sternal plate. The adult 


40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


organ, aedeagus and parameres, therefore, is supported on the poste- 
rior margin of the ninth sternum. When the parameres are sepa- 
rated from the aedeagus, they appear to be independent appendages 
of the ninth sternum, and have been regarded as such. The fact, 
however, that the parameres always originate as lateral branches of 
the genital rudiments shows that their lateral position results from 
a secondary displacement giving them a mechanical advantage as in- 
dependently movable clasping organs. The subsequent division of the 
parameres into “‘coxites” and “styli” occurs secondarily and only in 
the higher orders. It seems desirable, therefore, to adopt a nomen- 
clature free from unproven hypothetical assumptions of dubious 
validity. 

The genital claspers need no other name than that of parameres 
(side parts), a name first given to them in the Coleoptera. The 
two segments have been called the “basimere” and the “telomere,” 
which terms to be specific should be basiparamere and teloparamere. 
When the distal segments resemble grappling hooks they have ap- 
propriately been called harpagones (sing. harpago). The median or- 
gan that gives exit to the ejaculatory duct is best termed the aedea- 
gus (qg. v.) because the word means simply the principal genital part. 
By the dipterists it has been called the mesosome, by others the phallus 
or phallosome, or more generally the penis. Since the functional 
intromittent organ is usually the everted membranous inner wall 
of the aedeagus with the gonopore at its tip, this structure is more 
literally a penis. 

The name phallus is not inappropriate for the medium genital 
organ alone, but the latter has for so long been known to entomolo- 
gists as the aedeagus that this term has entomological priority, and 
the insect organ has no homologue even in other arthropods. The 
word “phallus” in ancient Greek was a vertebrate term and was spe- 
cifically applied to an artifact, symbolical of generation, carried in 
certain processions. 

Since we seem to lack a good general name for the genitalia of 
the insects, the writer (1941) has suggested that the term phallus 
might, consistent with its original meaning, be applied to the entire 
genital structure developed from the primary rudiments. The word 
combines euphoniously with prefixes and suffixes. The aedeagus and 
parameres are often not separated at their bases, in which case the 
three parts form a phallic unit with a common phallobase. The 
eversible inner tube of the aedeagus which serves as the functional 
penis of the insect may then be termed the endophallus (more 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 4I 


euphonious than “endoaedeagus’’) in distinction to the outer genital 
parts as ectophallic structures. The primary genital lobes thus become 
the phallic rudiments and their branches phallomeres. The distal 
aperture of the aedeagus is the phallotreme. 

This suggested nomenclature applies to only the major parts of 
the male genitalia which can be consistently named on a basis of 
homology. In all the insect orders, however, there are numerous 
secondary developments, and these structures must be given special 
names by workers in each order. 


Aedeagus (variously written also aedaeagus, aedegus, aedoeagus, 
oedagus, edoegus, from Greek pl. ozdoia, the genitalia + agos, chief 
or leader): The median organ of the male genitalia characteristic 
of pterygote insects from Hemiptera to Hymenoptera. The ejacu- 
latory duct opens into it. The aedeagus is thus, as the name implies, 
the principal member of the genital complex. It is formed by the 
union of the two mesal branches (mesomeres) of the primary geni- 
tal lobes at the sides of the gonopore. The gonopore, therefore, 
opens into the base of the aedeagal lumen, which latter is thus a 
secondarily added part of the genital exit passage, not a continuation 
of the ejaculatory duct, and its distal opening, the phallotreme, is not 
the gonopore. 

The aedeagus is commonly known as the penis of the insect, but 
usually the whole organ does not serve for sperm intromission. The 
spermatazoa are discharged from the duct into the lumen of the 
aedeagus and then introduced into the female by eversion of the 
membranous inner wall of the aedeagus as a vesicle, or a long slender 
tube with the gonopore at its tip. This eversible tube, or endophallus, 
thus becomes the functional penis. In some insects the spermatozoa 
are incapsulated in a spermatophore, which during coition is attached 
to the opening of the spermathecal duct of the female. With others 
they are freely discharged either into the genital chamber of the 
female, from which they may make their way into the spermatheca, 
or they are introduced directly into the sperm receptacle. The en- 
dophallus may become a highly developed complex organ in itself, 
as in the honey bee, in which the outer part of the aedeagus is re- 
duced to a pair of small plates guarding the phallotreme. 

Though the aedeagus is fundamentally a tubular organ, it takes on 
various forms in different orders and families. In some Hymenoptera 
the lateral parts of the aedeagus become separated as a pair of free 
prongs (sagittae) from a median penis tube. Among the Diptera 
long rodlike processes (paraphyses) grow out from the aedeagal base. 


42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


When the aedeagus and the parameres are not separated at their 
bases, the three parts arise from a comman phallobase, as among 
Coleoptera. In some cases the base of the aedeagus is connected with 
the parameres by a pair of small basal plates, or again the parameres 
may be entirely separated from the aedeagus. 

Among the dipterists the aedeagus is known as the “mesosome,” but 
this term (middle body) is not in itself specific for a genital organ. 
Others use the name phallus, which is from the Greek word for the 
male vertebrate organ with which the insect aedeagus has no possible 
homology (and in its ancient usage the “phallus” was particularly 
an artificial symbol of generation). Aedeagus is specifically an 
entomological term since the organ has no homologue even in other 
arthropods. Under the dissertation on Male Genitalia (q. v.) the 
writer has proposed that the term phallus is a convenient name for 
the whole genital complex developed from the primary genital 
rudiments. 


Ovipositor: According to its derivation the word ovipositor 
should be applicable to any organ used for placing eggs. Among the 
insects, then, there are two types of ovipositors, one being the ex- 
tensible abdomen itself, the other special pronglike outgrowths of 
the abdomen. 

An ovipositor of the first type is present in the tubuliferous 
Thysanoptera, the Mecoptera, the Lepidoptera, the Coleoptera, and 
the Diptera. In these insects the distal part of the abdomen can be 
extended as a tapering, telescopic tube, near the end of which is the 
opening of the oviduct. Some of these insects deposit their eggs on 
exposed surfaces and protect them with a covering of glandular 
secretion. Others use the extended abdomen for inserting the eggs 
under the edges of loose bark, or into crevices, or for depositing them 
in rafts on the surface of water. In some of the fruit flies the 
greatly elongated abdomen has a sharp terminal point that enables 
the female to pierce the skin of fruit and insert their eggs into the 
flesh. A similar piercing tip is found in some primitive moths. 

An ovipositor of the second type composed of movable sclerotic 
prongs is the organ usually referred to as the insect ovipositor. It 
is present in a very simple form in the Thysanura but is developed 
as a complex organ in some Odonata, in the Orthoptera, in the 
Hemiptera, and in the Hymenoptera. In these insects the ovipositor 
consists of two or three pairs of closely associated processes sup- 
ported on two pairs of ventrolateral plates of the eighth and ninth 
abdominal segments. 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 43 


An outgrowing genital process is in general a gonapophysis, a 
term applicable to the male as well as to the female. Specifically the 
prongs of the ovipositor are called the valvulae and the supporting 
plates the valvifers (valve carriers). The word valva in Latin was 
the name for one of a pair of folding doors; in modern mechanics a 
valve is a device for shutting off the flow of gas or water through a 
pipe, and in anatomy a valve is a fold in a blood vessel or the heart 
wall that regulates the flow of blood. Clearly, then, the use of the 
term valves, or the dimunitive valvulae, for the prongs of the insect 
ovipositor has no justification from the original meaning of the word. 
The ovipositor is not a closing apparatus but a conducting organ. 
However, since we cannot well describe objects or anatomical parts 
without having names for them, the terms valvulae and valvifers 
will be used in the following description for the lack of appropriate 
substitutes. 

In the typical pterygote ovipositor the free part of the organ is 
usually a tapering shaft composed of the first and second valvulae 
which enclose a narrow passageway for the eggs discharged from the 
opening of the oviduct between their bases. The ventral first valvulae 
slide back and forth on the second valvulae by interlocking ridges 
and grooves, and the second valvulae have a similar movement of 
their own alternating with that of the first valvulae. The movements 
of the valvulae are produced by muscles of the supporting valvifers, 
since the valvulae arise from the anterior ventral angles of their re- 
spective valvifers. The second valvifers are rocked on the lower 
edges, or sometimes on pivots, of the ninth tergum by strong antago- 
nistic anterior and posterior muscles arising dorsally on the tergum. 
This imparts a back-and-forth movement to the second valvulae, 
which often united, giving a stronger support for the first valvulae. 
The first valvifers are small plates articulated on the anterior ends 
of the second valvifers, and each is provided with a muscle from 
the eighth tergum. The up-and-down movements of the first valvi- 
fers give a back-and-forth movement to the first valvulae on the 
second valvulae. The movements of the valvulae on each other carry 
the eggs through the channel of the ovipositor. The so-called third 
valvulae, when present, are usually either slender styluslike processes 
projecting from the rear ends of the second valvifers, or flat lobes 
that ensheath the distal end of the ovipositor shaft, but in some 
Orthoptera they are broad lateral plates of the shaft. 

There are, of course, many variations in the size and shape of the 
ovipositor in the several orders of pterygote insects, but the gen- 
eral structure and mechanism of the organ are essentially the same 


44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


in all. Insects that possess this type of ovipositor are able to deposit 
their eggs in the ground, to insert them into the stems and twigs of 
trees, or into the bodies of other insects. The ovipositor of some 
Odonata is a well-developed piercing organ by which the female 
inserts her eggs into the stems of underwater plants. In other 
Odonata the ovipositor has become greatly reduced and nonfunc- 
tional ; such species merely drop their eggs on the water during flight. 
In the wasps and the bees the ovipositor has been remodeled into a 
stinging organ for the injection of poison from glands opening into 
its base. The eggs of these insects are discharged directly from the 
opening of the oviduct at the base of the sting. 

Theoretically the ovipositor has been interpreted as a development 
from primitive legs of the eighth and ninth abdominal segments. 
The valvifers are supposed to be the coxae, the first and second 
valvulae to be coxal outgrowths, or gonapophyses, and the third 
valvulae perhaps coxal styli of the second valvifers. Superficially 
this interpretation looks plausible since the valvifers are moved by 
muscles arising on the terga of their respective segments, and a com- 
parison with the simple ovipositor of Thysanura at first sight appears 
to bear out the suggested homologies. The two long slender gona- 
pophyses of each genital segment of the Thysanura appear to arise 
from the anterior mesal angles of the stylus-bearing coxal plates. 
However, they are only closely attached to these plates, and their 
basal muscles arise on the sternal area or a sternal plate between 
them. The gonapophyses have no musculature from the coxal plates, 
as they should have if they are either telopodites of the limbs or 
gonapophyses of the coxae. 

Matsuda (1957, 1958) has given a critical historical review of 
work on the structure of the insect ovipositor and of opinions that 
have been held on the homologies of its parts. Among the earlier 
writers, Heymons (1899 and earlier papers) was the foremost 
proponent of the concept that the gonapophyses (prongs of the ovi- 
positor) are secondary ectodermal outgrowths of the eighth and ninth 
abdominal sterna in no way related to the transient embryonic limb 
vestiges on the other segments of pterygote insects, or to the stylus- 
bearing coxal plates of Thysanura. Tillyard (1917) notes that in 
the Odonata the rudiments of the ovipositor develop early in the 
larval life, “but have nothing to do with the primitive paired seg- 
mental appendages of the abdomen, which are lost during embryonic 
life.” On the contrary, most subsequent writers down to the present 
time have held to the theory that the oviposior represents a pair of 
abdominal limbs of which the valvifers are the coxae. From a com- 


NO. 2 ENCYCLOPEDIA OF INSECT ANATOMY—SNODGRASS 45 


parative study of the ovipositor muscles, however, Matsuda concludes 
that the musculature does not support the idea that the ovipositor 
has been derived from a pair of abdominal limbs. He concludes that 
the valvifers are sternal in origin, as claimed by Heymons. Certainly 
the ontogenetic origin of the ovipositor in the larva does not suggest 
that its rudiments represent legs. In the higher insects, as the 
writer (1933) has shown in the honey bee, the ovipositor is de- 
veloped from two slender median processes on the venter of the 
eighth abdominal segment and two pairs on the ninth. These 
processes no more suggest a homology with legs than do the rudi- 
ments of the male genitalia. 

Hence, if we do not wish to discard the idea of the leg origin of 
the ovipositor, the subject must remain doubtful until substantiated 
by better evidence than is at present available. The ovipositor is an 
organ peculiar to the insects that possess it. In no other arthropod 
is there any such structure, either anatomical or functional, associated 
with the opening of the oviduct. 

Morphological generalizations are mental products of morphol- 
ogists, but they are always intriguing in that they bring a lot of 
seemingly unrelated facts into a single concept. In this way they may 
be more convincing by the mental peace and satisfaction they give 
than by the evidence from the facts on which they are based. We must 
be cautious, then, not to accept a generalization on its mental appeal 
alone. 


LITERATURE CITED 


BorraDAILe, L. A., and Ports, F. A. 
1958. The Invertebrata, ed. 3 (edited by G. A. Kerkut), 795 pp. Cam- 
bridge University Press. 
DuPortet, FE. M. 
1960. Gastrulation and the endoderm problem in insects. Ann. Ent. Soc. 
Quebec, vol. 6, pp. 45-52. 
1962. Origin of the gula insects. Can. Journ. Zool., vol. 40, pp. 381-384. 
GARSTANG, W. 
1922. The theory of recapitulation: A critical restatement of the biogenetic 
law. Journ. Linn. Soc. London, Zool., vol. 35, pp. 81-101. 
Henson, H. 
1946. The theoretical aspect of insect metamorphosis. Biol. Rev., vol. 21, 
pp. 1-14. 
Heymons, R. 
1899. Der morphologische Bau der Insektenabdomens. Eine kritische 
Zussamenstellung der wesentlichen Forschungsergebnisse auf ana- 
tomischen und embryologischen Gebiete. Zool. Centralb., vol. 6, 
pp. 537-556. 


46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Hinton, H. E. 
1947. The insect cranium and the “epicranial suture.” Smithsonian Misc. 
Coll., vol. 107, No. 7, 52 pp. 
1958. Concealed phases in the metamorphosis of insects. Sci. Progress, 
No. 182, pp. 260-275. 
JANET, C. 
1909. Sur l’ontogénése de l’insecte, 129 pp. Limoges. 
JoHANNSEN, O. A., and Butt, F. H. 
1941. The embryology of insects and myriapods, 462 pp. McGraw-Hill 
Book Co., New York. 
Manton, S. M. 
1949. Studies on the Onychophora, VII: The early embryonic stages 
of Peripatopsis. Philos. Trans. Roy. Soc. London, ser. B, vol. 233, 
pp. 483-580. 
Matsupa, R. 
1957. Comparative morphology of the abdomen of a machilid and a 
raphidiid. Trans. Amer. Ent. Soc., vol. 83, pp. 39-63. 
1958. On the origin of the external genitalia of insects. Ann. Ent. Soc. 
America, vol. 51, pp. 84-94. 
PRINGLE, J. W. S. 
1957. Insect flight. Cambridge University Press. 
Snoperass, R. E. 
1933. Morphology of the insect abdomen, II: The genital ducts and the 
ovipositor. Smithsonian Misc. Coll., vol. 89, No. 8, 148 pp. 
1935. The history of an insect’s abdomen. Ann. Rep. Smithsonian Inst. 
for 1933, pp. 263-287. 
1936. Morphology of the insect abdomen, III: The male genitalia. Smith- 
sonian Misc. Coll., vol. 95, No. 14, 96 pp. 
1941. The male genitalia of Hymenoptera. Smithsonian Misc. Coll. 
vol. 99, No. 14, 119 pp. 
1954. Insect metamorphosis. Smithsonian Misc. Coll., vol. 122, No. 9, 
124 pp. 
1960. Facts and theories concerning the insect head. Smithsonian Misc. 
Coll., vol. 142, No. 1, 61 pp. 
1961. Insect metamorphosis and retromorphosis. Trans. Amer. Ent. Soc., 
vol. 87, pp. 273-280. 
TILLYARD, R. J. 
1917. The biology of dragonflies, 396 pp. Cambridge University Press. 
WuHittTEN, J. M. 
1962. Homology and development of insect wing tracheae. Ann. Ent. Soc. 
America, vol. 55, pp. 288-295. 


Abdomen, 36 

Aedeagus, 39, 40, 41, 42 

Alimentary canal, 15 

Anapleurite, 7 

Anatomical names, 1 

Annulus (annuli), 3, 27 

Antecostae, 3 

Antenna (antennae), 22, 
27 

Anus, 21 

Archenteron, 17, 18 

Archicephalon, 25 

Articulations, 32 

Axillaries, 34 


Basalare, 8, 34 
Basiparamere, 40 
Basipodite, 32 
Basisternum, 30 
Blastocephalon, 25 
Blastopore, 16, 17, 19 
Blastula, 18 

Body segmentation, 2 
Buccal cavity, 16 


Caecae, 16, 20 
Cardia, 20 
Catapleurite, 7 
Cervical sclerites, 28 
Chorion, 2 
Cibarium, 2, 23 
Clypeus, 22 
Coelomic sacs, 3, 26 
Colon, 16 
Coxa (coxae), 7, 8, 32, 
44 


Coxites, 39, 40 
Coxopodite, 9 
Crop, 16 


Dorsum, 4, 6 


Ecdysial cleavage line of 
head, 24, 27 

Ecdysis, 15, 24 

Ectoderm, 17, 18 

Embryo, 25 

Endoderm, 17, 18 

Endophallus, 40, 41 


INDEX 


Entomology, 1 
Epicranial suture, 24, 27 
Epimeron, 4, 8, 30 
Epipharynx, 2, 23 
Episternum, 4, 8, 30 
Epistomal sulcus, 22 
Eyes, compound, 21 


Face, 21 

Femur, 32 

Flagellum, 4, 27 

Frons, 22 
Frontoclypeal sulcus, 22 
Furca, 30 


Gaster, 37 

Gastric caecae, 16, 20 
Gastrula, 17, 18 
Gastrulation, 18 
Genae, 22 

Genitalia, male, 37 
Gonocoxite, 9, 39 
Gonopods, 37, 39 
Gonopophyses, 43, 44 
Gonopore, 41 
Gonostylus, 39 
Grooves, external, 8 
Gula, 29 


Halteres, 35 
Harpago (harpagones), 
40 


Head, 21 

Hexapoda, 1 

Hypognathous head, 22, 
24 

Hypopharynx, 2, 22, 23 


Ileum, 16 

Imago, 11 

Insect, 1 

Intestine, anterior, 16 
Ischiopodite, 32 


Labium (labia), 22, 29 

Labrum (labra), 22 

Larva (larvae), 9, 10, 11, 
13, 14 


Legs, 8, 31 
Leg segments, 32 


Male genitalia, 37 
Malpighian tubules, 16 
Mandibles, 22 
Maxilla (maxillae), 22 
Mesenteron, 15, 16, 20 
Mesomere, 39 
Mesosome, 40 
Mesothorax, 29 
Metamorphosis, 10, 12, 
13, 14, 36 
Metathorax, 29 
Moulting, 15 
Mouthparts, 16, 21, 22, 
23, 24 
Muscles, 3, 28, 30, 32, 33, 
34 


Neck, 28 

Nerves, 26 

Notum (nota), 4, 6, 7, 29 
Nymph, 9 


Occipital foramen, 22 
Occipital sulcus, 22 
Occiput, 22 

Oesophagus, 16 
Opisthognathous head, 24 
Organ of Johnson, 27 
Ovipositor, 42 


Paramere, 39, 42 

Paranotal lobes, 4, 33 

Paraphyses, 41 

Pedicel, 27 

Penis (penes), 37, 38, 40, 
41 

Peritrophic membrane, 21 

Phallic rudiments, 41 

Phallobase, 40, 42 

Phallomeres, 41 

Phallosome, 40 

Phallotreme, 41 

Phallus, 40, 42 

Pharate period, 12, 15 

Pharynx, 2, 16, 23 

Phragmata, 34 


47 


48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Plates, coxal, 44 
postnotal, 30 
segmental, 5, 6 

Pleural process, 30 

Pleural ridge, 30 

Pleurites, 6, 7 


Pleuron (pleura), 4, 6, 7, 


30, 34 
Podites, 3 
Postnotal plate, 30 
Postnotum, 34 
Postoccipital sulcus, 23 
Postocciput, 22 
Preoral cavity, 16, 23 
Prescutum, 30 
Prephragma, 30 
Pretarsus, 32 
Proctodaeum, 15, 16, 21 
Prognathous head, 24 
Prolegs, 32 
Prostomium, 25 
Prothorax, 28, 29 
Protocephalon, 25 
Proventriculus, 16 
Pupa (pupae), 11, 12 
Puparium, 12 
Pylorus, 16 


Recapitulation, 14 
Rectum, 16, 17 
Regenerative cells, 20 
Retromorphosis, 14 


Sagittae, 41 
Scape, 27 
Sclerites, 28, 34 
Sclerotization, 4, 5, 7 
Scutellum, 30 
Scutum, 30 
Segmental plates, 5 
Segmentation, 2 
abdominal, 36 
embryonic, 25 
head, 22, 26 
leg, 32 
primary, 3 
secondary, 3 
segment areas, 4 
Segments, 3 
Skeleton, grooves, 8 
Somites, 3 
Spinasterna, 31 
Spiracle, 31 
Sternellum, 30 
Sternites, 6 
Sternum (sterna), 4, 5, 6, 


Stigmata, 31 

Stomodaeal valve, 16 

Stomodaeum, 15, 16, 21 

Striated border, 20 

Subgenal area, 22 

Subalare, 8, 34 

Sulcus (sulci), 5, 7, 8, 9, 
22 


Sutures, 2, 9 
Stylus (styli), 40 


Tarsomeres, 32 
Tarsus, 32 
Teloparamere, 40 
Tentorial bridge, 24 
Tentorium, 23, 24 
Tergites, 6, 7 
Tergum (terga), 4, 5, 6 
Thorax, 29 

Tibia, 32 

Trachea, 31 
Trochanter, 32 
Trochantin, 8 


Valves, 43 
Valvifers, 43, 44 
Valvulae, 43 
Veins, wing, 35 
Venter, 4, 6 
Ventral surface, 6 
Ventriculus, 16, 20 
Vertex, 22 
Vitellophags, 19 


Wings, 33 
Wing buds, 35 
Wing veins, 35 


Yolk cells, 18 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NUMBER 3 


Roebling Fund 


SOLAR VARIATION AND WEATHER 


A SUMMARY OF THE EVIDENCE, COMPLETELY 
ILLUSTRATED AND DOCUMENTED 


By 
Cc. G. ABBOT 


Research Associate, Smithsonian Institution 


(Pusiication 4545) 


i> \O - CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
. OCTOBER 18, 1963 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NUMBER 3 


Roebling Fund 


SOLAR VARIATION AND WEATHER 


A SUMMARY OF THE EVIDENCE, COMPLETELY 
ILLUSTRATED AND DOCUMENTED 


By 
Cc. G. ABBOT 


Research Associate, Smithsonian Institution 


eeeeOsocc, 


(Pusrication 4545) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
OCTOBER 18, 1963 


PORT CITY PRESS, INC. 
BALTIMORE, MD., U. S. A. 


Roebling Fund 


SOLAR VARIATION AND WEATHER 
By C. G. ABBOT 


Research Associate, Smithsonian Institution 


From 1920 to 1955, with the aid of John A. Roebling, the Smith- 
sonian Astrophysical Observatory under my direction, and later 
under that of L. B. Aldrich, made “solar-constant” observations from 
mountain tops in cloudless deserts in Africa, Asia, South America, 
and the United States. Although all the results were highly accurate, 
they were especially so from 1924 to 1944, for it was not till 1924 
that the “short method” was fully perfected, and after 1944 the 
transparency of the atmosphere was less perfect than before. 

From 1935 to the present I have sought to correlate the solar- 
constant measures with weather phenomena. I have published in 
Smithsonian Miscellaneous Collections + more than a score of papers 
on this subject. These papers and the volumes of the Annals of the 
Astrophysical Observatory, as well as several papers in outside jour- 
nals, are referred to in the Appendix. They give in detail the evi- 
dence I shall rely upon in what follows. 

I have been led to conclude firmly that variations of the sun’s emis- 
sion of radiation are associated intimately with weather changes. 
Since the death of H. H. Clayton I know of no professional meteorolo- 
gists in the world, with the exception of Dr. Irving P. Krick, who 
have acknowledged support of my main conclusion. They all, indeed, 
credit us with highly accurate solar measurements, but in the absence 
as yet of connecting theory they distrust my proofs that solar varia- 
tion has any considerable influence on ground weather. 

Being now past 91 years of age, and firmly convinced that the 
sacrificing years of residence of my colleagues on high desert moun- 
tains have given to astrophysics and meteorology a long series of 
measurements of great practical importance, I feel compelled in jus- 

1In the Appendix I give full references to all sources I refer to here. Nearly 
all are from Smithsonian Miscellaneous Collections. For brevity in the text 


I shall cite the Smithsonian publication number as “P.” so and so. 


SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 146, NO. 3 


2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


tice to them to write a summary of the whole research. I hope to 
make it so thoroughly supported by varied evidences as to convince 
the professional scientists that it can no longer be ignored and allowed 
to sink into oblivion. But it is quite impossible for me to give half the 
evidence which saturates my mind with the certainty that the family 
of regular harmonics of 273 months, in solar radiation and terrestrial 
weather, is a controlling geophysical fact. 


1. THE “SOLAR CONSTANT” 


Pouillet invented his pyrheliometer, and about 1876, after measur- 
ing the heat of sun rays at different solar altitudes, he estimated that 
the instrument would have indicated 1.76 calories per square centi- 
meter per minute if exposed outside the atmosphere at the earth’s 
mean solar distance. Langley strongly argued that since the atmos- 
phere transmits different wavelengths unequally, spectroscopic meas- 
ures are necessary additionally to pyrheliometry to estimate the solar 
constant. He invented the bolometer for this purpose and used it at 
Allegheny Observatory and on and near Mount Whitney. Errone- 
ous theory caused him to prefer 3.0 calories as the solar constant. 
K. Angstr6ém, from solar measures on the Island of Teneriffe, attrib- 
uting excessive influence to atmospheric carbon dioxide, preferred a 
value of 4.0 calories. 

In volume 2 of Annals, A.P.O.,? is demonstrated the true theory 
for the spectrobolometric determination of the solar constant. An 
improved pyrheliometer similar to Pouillet’s is described. Measure- 
ments at Washington, D. C., during the years covered by volume 2 
indicated an average solar constant of 2.20 calories. A hint of solar 
variation appeared to be indicated by results of 1903 and 1904. By 
invitation of Director George E. Hale, we made measurements of the 
solar constant on Mount Wilson, Calif., in 1905 and 1906. From 
1908 to 1920 the Smithsonian sent expeditions to Mount Wilson. 
A long-focus vertical telescope was installed in addition to solar- 
constant apparatus. Every day that solar constants were observed, 
the distribution of brightness over the diameter of the sun’s disk 
was observed by allowing the 8-inch solar image from the telescope 
to drift without a clock over the slit of the spectrobolometer, in rays 
at various wavelengths. (See fig. 52, p. 62.) 

In volume 3, Annals, A.P.O., pages 21-29, a full description of 
solar-constant measurement is given. The silver-disk pyrheliometer 


2 We thus abbreviate Smithsonian Astrophysical Observatory. 


NOs 3 SOLAR VARIATION AND WEATHER—ABBOT 3 


is described on pages 47-52. More than 100 of these instruments have 
been constructed, standardized, and sold at cost by the Smithsonian 
Institution to observers in all parts of the world. For their standardi- 
zation in absolute units, I devised the water-flow and water-stir abso- 
lute black-body pyrheliometers (see Annals, A.P.O., vol. 3, pp. 52- 
69). With certain improvements, the water-flow double-barreled 
electrical-compensation pyrheliometer has been used for standardizing 
pyrheliometers hundreds of times. It is now recognized as the world’s 
standard for measurements of solar radiation. The double-barreled 
water-flow design was suggested by V. M. Shulgin of Russia about 
1927 and was immediately adopted by us. 

About 1913, with F. E. Fowle and L. B. Aldrich, I did the original 
standardizations. We used thermometers certified in Paris and elec- 
trical instruments certified at the U. S. Bureau of Standards. Our 
solar measures from that time to this have always been expressed “on 
the scale of 1913.” During the 40 years following, whenever im- 
provements brought alterations we always made many checks and 
comparisons to keep the solar constant values still “on the scale of 
1913.” Observed solar-constant values have ranged irregularly from 
1.900 to 1.960 calories and even higher. Their mean value “on the 
scale of 1913” is 1.944. We now recognize that the single-barrel 
standard pyrheliometer of 1913 in our hands gave values about 2 per- 
cent too high. This was cured by the new instrument used since 1930. 
Various other important changes in solar-constant work have been 
made. These include restricting the sky exposure, making larger cor- 
rections for wavelengths beyond the violet and far in the infrared 
not observed daily, evaluating ozone absorption, determining personal 
equation, introduction of “the short method,” and other changes. The 
effects of all these we have applied retroactively to all the solar- 
constant determinations from 1920 to 1955. (See Annals, A.P.O., 
vols. 6 and 7.) Every published value was scrutinized extensively by 
L. B. Aldrich, Mrs. A. M. Bond, and W. H. Hoover, and generally 
by all three as a committee. So far as we have been able to bring it 
about, the solar-constant tables in volumes 6 and 7 of Annals A.P.O., 
and also published in my papers P. 4088 and P. 4213, form a homo- 
geneous series, all “on the scale of 1913.” 

Johnson, of the Naval Research Laboratory, using data from 
rockets, and with critical studies and use of our work, has published 
the solar-constant value 2.00+0.04 calories.* I doubt if any de- 


3 Johnson, F. S. On the solar constant. Journ. Meteorology, vol. 11, No. 6, 
1954. 


4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


termination depending basically on observing from mountain tops 
can claim with certainty to be within 1 percent of the absolute scale 
of heat. But as will be shown below, a series such as ours, where 
every effort was made to retain a constant scale over many years, 
can be depended on to preserve its relative homogeneity to 1/6 of 
1 percent in daily values, even though 1 or 2 percent away from the 
true absolute scale throughout. 

Volleys of criticisms of our solar-constant determinations were 
published between 1910 and 1914 by numerous authors. These we 
answered by several papers, but as they still continued we published 
(P. 2361) the extensive paper “New Evidence on the Intensity of 
Solar Radiation Outside the Atmosphere.” This has three distinct 
parts : 

(1) On September 20 and 21, 1914, two of the clearest and most 
uniform days ever experienced on Mount Wilson, we observed for 
the solar constant continuously from sunrise to 10 o'clock. This 
yielded for both days, by Langley’s spectrobolometric method, solar- 
constant values computed as between air masses 1.3 and 4.0; 4.0 and 
12.0; 1.3 and 20.0. All these six solar-constant measures (Langley’s 
method) fell between 1.90 and 1.95, which shows both the excellence 
of the sky conditions and the accuracy of the observing. 

(2) At Dr. A. K. Angstrém’s suggestion I designed, and our 
instrumentmaker Andrew Kramer constructed, five copies of an auto- 
matic combined pyrheliometer and barometer. These were flown 
by balloons from Omaha by L. B. Aldrich, with the cooperation of 
Dr. William R. Blair and his assistants from the U. S. Weather 
Bureau, on July 11, 1914. One instrument was recovered uninjured 
in Iowa. It was calibrated both before and after flight under the same 
conditions of temperature and barometric pressure that obtained dur- 
ing flight. It rose to 24,000 meters, where 24/25 of the atmosphere 
lay below. It yielded a value of 1.87 calories, a value that lies within 
the limits of solar variation, as observed in those times at Mount 
Wilson, and as expressed on the Smithsonian “scale of 1913.” 

(3) Here I quote the concluding paragraphs of our paper: 


It seems to us that, with the complete accord now reached between solar 
constant values obtained by the spectro-bolometric method of Langley, applied 
nearly 1,000 times in 12 years, at four stations ranging from sea level to 
4.420 meters, and from the Pacific Ocean to the Sahara Desert; with air- 
masses ranging from 1.1 to 20; with atmospheric, humidity ranging from 
0.6 to 22.6 millimeters of precipitable water; with temperatures ranging from 
0° to 30° C.; with sky transparency ranging from the glorious dark blue above 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 5 


KNNANS 
Wines 


L4/ 


Fic. 1—Water-flow standard pyrheliometer of 1913. 


Mt. Whitney to the murky whiteness of the volcanic ash filling the sky above 
Bassour in 1912, it was superfluous to require additional evidence. 

But new proofs are now shown in figure 10 [fig. 7, p. 10, of present paper]. 
This gives the results of an independent method of solar constant investigation. 
In this method the observer, starting from sea level, measures the solar radia- 
tion at highest sun under the most favorable circumstances, and advances 
from one level to another, until he stands on the highest practicable mountain 
peak. Thence he ascends in a balloon to the highest level at which a man may 
live. Finally he commits his instrument to a free balloon, and launches it to 
record automatically the solar radiation as high as balloons may rise, and where 
the atmospheric pressure is reduced to the twenty-fifth part of its sea level 
value. All these observations have been made. They verify the former con- 


& 


Ceti ie te 


ee 


jim 

ays) 
it aS Ci 
| | HE 
RZ iS 


Fic. 2.—Double water-flow electrical compensation pyrheliometer. 


6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


woe es Se 


Millimeters 


d 
Lot ie 


sat 


CVT Deaf) 


sib 


iS 
co OLE «Ta s 


Fic. 3.—Angstrém-Smithsonian pyrheliometer. 


clusion; for they indicate a value outside the atmosphere well within the pre- 
viously ascertained limits of solar variation. 


The solar-constant method of Langley, which we used exclusively 
until 1920, is fundamental and sound. But it requires several hours 
of observation through unchanged transparency while the sun is 
ascending or descending, so that the thickness of air its rays traverse 
alters enough to give accurate transmission coefficients for all wave- 
lengths observed. If during a morning series of measurements the 
atmosphere grows more transparent, the value obtained is too high, 
and vice versa. The opposite, of course, holds in the afternoon. We 


SOLAR VARIATION AND WEATHER—ABBOT 


NO. 


aS Saraeguenetee 2. il 

Peer 

i HEL ae 

fe et an AA 
Teeaeerae Hay q OAT A 

PCE EEE CER 

REP Ceeeeee 


SUSEUGEEEL SURES” apaaMe ety WTA 
alfa bon 


ra 


POA PAE arr 
UAL ET 
REE, COAL ay eavaurwae LT 


UA ee TL 
fits al ar ae @4suaeay meas ARAL UA 
ope AZ ae CEL faway 

72 


Hl 
eee HA 
Geico ptt 


Liha ULL 
WANE 


Fic. 4.—Logarithmic atmospheric transmission. Wavelengths 0.47 to 2.42 u. 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


*IojaworfsysAd UOOT[eq SuIpsovay—'¢ “O1T 


“ }*seH omg @ 


Op Cpe en nt 


4) *eWdLIwhsnaD oO 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 9 


eooeecaer eee el El or (| Sa al aa 
aS Th CRs ee ER 
a SS] a Se Gee ae Ae 
Ea aa GS SS) eS eS EE Eee oe 
a Pe ET 
_———————— Sas ee a 
———— a2?) a Sy 
} --——--—_——}—-| Sere eS 
Tae ee) FS] Ea eee 7 eS -—— 
— __————— 
| | -=4 Pe ee (SS — [aids ae Coal 
(rea + es) ESS SSE Se (Sen Ga ee Dee Sey 
wee ae oe ed Ee Wy CE 
SSeS ee Be] eee eee a eee |.) ee eee, | ee eee 
== = SSS ee ee 
——— ee Puspyres Oem 
Beare ee ae Ee Se Se bana Eee 
a SSE 
aa ee ee aa ee ||| ee a eee 
Geer Se) ae ie eS (||) eee ee 
ee Ee Ee ee, 
Ban Ss en ae ee ee |: ed ee 
fee a a SSS a oe a eee 1): ae ee 
a Cee ee || CN RAR 
(Eee wees 14 a ae 2, 
+} — hr. ee eee ee 9, i 
Ky Ay hittin — ——— rm i f ea 
yf AAT LT ed ee Gee) 
ing AUN ha a | aa Se eee (| ||| rH H mt sin itn tt 
Tt _ 
ai —* —t— —— Ae 
He 
es =a eee ae ee le Sa ee eae || pee eee 
SS SSS Se ie (2S SEE ee ee (ee ae hee 
(i ae al ee LS Se Beg | a ee hae PESTS eee re RES 
ae er oe ES Gages Siw Ge ES ee eee | Ee | ee ar 
ae Ee ERS Se a er ae rer eel Pee Sas 
= 
SSS (SSeS Sg SS eS 
a os = ee Ee eT 
cee enapeaIe COMCLNICIGNS ) foacer cere 
eS Se Se ee Ee eS See ee ee ee eee 
aaa ees (See a SS Se ae ee (pee See) 
et ee ee a UE ER 
Sai ae Grae nS ee a oe ee (ae ie el ees ST ae al 
ee a ee ES EET SS 
es TE SS a I ee 
Sea Fees eee) LS = | ae SS ee a Se ee) 
(eat ae ee ee) a | aS EE S| Pe ee) 
a ae ee pf 
ee er ee 2 a 
ae on Pen ee ee 
213-04 1086-——_ te 18 20 


Fic. 6.—Record pyrheliometry to 15 miles altitude. 


wished to devise a method whereby several values of the solar 
constant could be obtained per day, by intervals of observing too short 
for hurtful changes of transparency. 


2. THE SHORT METHOD 


Alfred F. Moore, observing at Calama, Chile, showed me in 1920 
a long series of observations with our sky-radiation instrument, the 
pyranometer (fig. 9, p. 13), on the brightness of a limited zone of sky 
surrounding the sun. When the transparency of the atmosphere is low, 
the sky gets brighter, and vice versa. Comparing Moore’s pyranometry 
with simultaneous determinations of atmospheric transparency at 
40 wavelengths, made by Langley’s method, I was able to draw 
families of curves throughout the spectrum of the sun, giving trans- 
mission coefficients suited to all states of sky brightness at Calama. 
(fig:) 8,'p: (11): 

This is the basis of the “short method” of solar-constant observing. 
It requires only about 10 minutes of observing by spectrobolometer, 
pyranometer, and pyrheliometer. We became accustomed to making 


sf) SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Balloon. 


1 1914. 


a 


f Vertical Sun at Mean Distance. 


2 


GCal. em. min. 


. Radiation o 


— 


Ye 
Barometer. 


Fic. 7—Maximum pyrheliometry, sea level to 15 miles altitude. 


three or even five determinations of the solar constant per day and 
could utilize days with cumulus clouds intermittently—days quite unfit 
for Langley’s method. The short method is, indeed, empirical and 
must be set up separately for each observing station by observing a 
year or more simultaneously with Langley’s method to standardize it. 
We continually improved the “short method” till 1926, but after that 
we used it exclusively except for occasional checks by Langley’s 
method. 


3. ACCURACY OF “SHORT METHOD” SOLAR CONSTANTS 


In volume 6, page 163, Annals, A.P.O., are compared the solar 
constants observed on 616 identical days at Mount St. Katherine in 
Egypt and Mount Montezuma in Chile. Winter at one station cor- 
responds with summer in the other. The difference between daily 
results ranged from 0 to 0.028 calorie. The weighted mean differ- 


PTS LI tl 
aoe 


‘ae! 


LAL. 
PT Ae 


BRETT BERET | EREET | EERE 
PABRERE!BRRP. 


2a Ge? dees HERZ Ae: 
Pay | | TT AT T el 


DS 2G EA aoa Ado Ae 2 RS APs 


BP SRP ARR ae eee ee 


BURBS SABES? se ee Ae 


ETT ERRRCER CC ERER 
eee AS A See Pe At Te 7 a ee 


SES). RV GR Bey eenee ee 


<a 
La 
sci 
=e 
ey 
EC 
A 
aR 
fala 
=— 
ae 
‘Gnm 
=e 
ei 
cilia 
ne 
HS 
ime 
=e 
rh 
17 
on 
SS 
an 


arma 
ale Vakesfiatl 
BT] BERBII 
ERE eee 
FERETLI EREBREI LS 


Shoe) Aen eee ae 
HEH A A 


I} IP , i 
TA at AE 


| 
Bock BaP ABET LED ae 
SASP Gee, ane? 2a aeeP 


Hehe bt fet AL ALY 
pA 


7 Ae 


Or2345678562 Oorz23¢5 6789 


Fic. 8—Atmospheric transmission by short method. 


If 


I2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


vn 


i 
aur 
a 


Ff Wii A | 
Re Th 
At] LES ea 

te |B LDGLSS ie aa ale lee 


Fic. 8a.—Solar constant measures repeat after 273 months. 


Fn are ees Se 
ee 


ae 
= 
ad 
ne 
bE 5 
TA) 
UY 
MW 
A 


= 
Hoe 


ence for 616 days is +0.0076 calorie. In 1953, Mrs. L. S. Hill made 
several similar comparisons. Mount Montezuma, Chile, is compared 
with Table Mountain, Calif., and with Tyrone, N. Mex. Table Moun- 
tain is compared with Tyrone. Her data cover, respectively, 891, 283, 
and 202 identical days. 

From this abundant evidence, with no indication of appreciable 
seasonal influence, I confidently take the average probable error of 
one day’s observation at one station by the “short method” to be 
1/6 of 1 percent of the solar constant. 

In volumes 6 and 7, Annals, A.P.O., are published about 6,000 daily 
values of the solar constant obtained by the short method, often by 
two or three stations observing upon the same day. Fully half of 
them are “very good” and at least as accurate as those referred to in 
Table 1. The mean value for one month would rest on at least 10 of 
them, and frequently 20 to 30. The month will be the unit I shall 
use in the discussion to follow. Its probable error can justly be 
regarded as 1/V10x1/6=1/20 of 1 percent of the solar constant. 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 13 


TABLE 1.—Accuracy of solar observations 


Mean difference Number of 


Station Pair Calories — 10,000 comparisons 
NMonteztina——dablemacr ances se hele scid evsle site sieve weiss 76.8 891 
MIGIMEZIIiNa—= YTONE “Lice sc «cles oe a Socks ew Poe be duis 79.6 283 
Montezuma—Mount St. Katherine.................. 76.0 616 
ME TUOUE tre ast cc cece cere cack tr cetens casas 77.9 202 


In tabulations which we make of forms and amplitudes of harmonic 
solar variations, these range from 4 to 91 months in periods. Results 
on the solar-constant periods rest on mean values of tabulations of 
between 10 and 100 months, their numbers increasing as their period 
decreases. So the individual points on a periodic curve computed 
from monthly values will have probable errors ranging from 
1/20+ V10 down to 1/20+ 100 of 1 percent of the solar constant. 


ES — 
Fig. 3. 


Centimeters. 


4 = a) —, 
n : 


Fig. 3. 


Fic. 9.—Smithsonian pyranometer. 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


14 


r s4dv2Ig 


4 A sels ia AN 2) “IN 49 


NOVARTIS Se 
PEAT N TL abbehal pate ital (be 


2/3 


SaLE Gert een Ta 
Arh [baba el SC Tt pas 
Gs 
at ch tort 
Pre eae az 


VCO a 
Nop aA SE EFAS Cs 
Se?  GRRMRSGSee A Oe 
Ci ON eo TS 
A LO) 
ss © OS aa a a Wan 
Se eT 


SMEs 


Spaten, 


~ / 
vy a 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT T5 


The smoothed curves representing the periods, as will appear from 
figure 10, are nearly sine curves. Their regular shapes, of course, 
make them stronger evidences of real variation than their individual 
points. But these points, as just said, have probable errors ranging 
from 1/60 to 1/200 of 1 percent of the solar constant. 


4. PERIODICITIES IN SOLAR VARIATION 


Over 25 years ago I noticed periodic repetitions of dentlike depres- 
sions in graphs of the solar-constant record. These periods appeared 
to be harmonics of 273 months. Notably appeared 273/7, 273/21, 
273/39. In my paper “Periodic Solar Variations” of June 1955 
(P. 4213) I amplified what I had said earlier (P. 4088) in May 1952. 
For convenient tabulation I gave in the two papers together the 
departures from 1.90 calories of all mean monthly solar constants 
from August 1920 to December 1952. These departures are ex- 
pressed as percentages of the solar constant. Thus the computer uses 
small positive numbers exclusively, and his results are expressed as 
percentages of the solar constant, taken as 1.944 calories per square 
centimeter per minute. 

P. 3902 graphs the march of the solar constant from 1920 through 
1951 and shows that it is closely represented by the summation of 23 
periods, all exact harmonics of 273 months. To be sure, in 1922 and 
1923 there is a depression far exceeding any observed since. Per- 
haps it represents one member of another family of solar changes of 
long periods. Unfortunately, since 1955 no measures of the solar 
constant have been made so far as I know. In P. 4462, of 1961, I 
suggest that the solar constant might be observed daily with higher 
accuracy than ours from a satellite. That might reveal solar changes 
of great importance in future years. Figure 8a, p. 12, shows the solar 
constant repeating after 273 months. 

Figure 10 graphs 26 harmonic periods in solar variation. They are 
cleared of overriding subharmonics. Small type numbers give the 
length numbers of the 26 harmonics as fractions of 273 months. All 
the curves have approximately sine form. Table 3 of P. 4213 tabu- 
lates the amplitudes of the periods and of their submultiples which 
were removed for the sake of clearness. Altogether 31 harmonic 
periods are given in this table with amplitudes ranging in solar radia- 
tion from 0.18 down to 0.02 of 1 percent. Small as they are, all these 
amplitudes are far above the probable errors derived in Section 3. 

To show the importance of clearing away the overriding submulti- 
ple periods, figure 11 is a graph of the period of 39.0 months recently 


16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


SE ee 
Pee ne eR ot a ci 
s rel 


i. 

real —] 

Z NB 
AG 

ia 

x 

Bi 


— 


ae 


ote ale Lae Ed: ae 
poi aes aes 


ae 
a 
nN 
N 
eS 
ne 


SeRCezeR aes 


Zoe Ne eee 


ris) de Ne aNeIS Pea [eee 


BRR See sex 
Rae Seer 
ESD tate 


came ae cemtadel Ge 


Fic. 11.—Thirty-nine-month period in solar variation cleared of shorter 
harmonics. 


ee , 
Ei Saabs Spee aS 


2 a 


computed, using al] the solar constant measures from the years 1923, 
January, through 1943, June. Periods 273/14, 273/21, 273/28, 
273/35, 273/91 (and 273/24 doubtful) are shown successively re- 
moved by computation and are exhibited as graphs B’, C’, D’, F'’, F’, 
leaving curves B, C, D, E, F. This leaves 273/7 almost fully cleared 
of irregularities. It is a very pretty sine curve with amplitude 0.28 
percent of the solar constant. 


5. ANOTHER TYPE OF PERIODIC VARIATION DUE TO 
SOLAR ROTATION 


The late Dr. H. Arctowski published (Proc. Nat. Acad. Sci., 
vol. 26, pp. 406-411, 1940) graphs (here fig. 12) which show vari- 
able differences in brightness over the sun’s rotating surface asso- 
ciated with corresponding variation of the solar constant. Figure 13 
shows solar-constant results, March to April 1920. A huge sunspot 
group passed centrally over the sun’s disk March 20-24 (see fig. 13). 
It produced a large depression of the sun’s radiation. L. B. Aldrich 
gave extensive evidence (Smithsonian Report for 1952) proving the 
increase of solar radiation with increased sunspot activity (see fig. 14). 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 7 


Fic. 12.—Relationship of calcium flocculi and solar constant as demonstrated by 
Arctowski 


Fic. 13.—Huge sunspot central group of March 1920 depresses sun’s radiation. 


18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


n 
s 
ed 
oO 
i 
g 
» 
o 
& 
co) 
o 
he 
a 
ad 
° 
7) 


(The numbgr near each point 
indicates the number of monthly 
means includad in each group.) 


7 
Sunspot numbers 


Fic. 14.—Solar radiation increased by rising sunspot activity as shown by Aldrich. 


When very great sunspot groups are exactly central on the solar 
disk, however, solar radiation decreases (see fig. 13). Again, from 
P. 2499, I reproduce in figure 15 results of correlation coefficients 
computed over an interval of 40 days. These show that in 1915 the 
solar rotation, with its 27-day period, brought up changes of correla- 
tion, between the days of solar-constant measures and the days fol- 
lowing, which ran through a range of 50 percent in correlation. They 
passed from +20 through 0 to —30 percent in correlation coefficients. 

Though the event is not of periodic nature, I mention here P. 3916 
and P. 3940 which show, respectively, the sharp fall in the solar 
constant when West Indian hurricanes or great magnetic storms are 
reported (see figs. 16 and 17). 


6. SOLAR VARIATION AND WEATHER 


Having explored above evidence on the intensity of solar radiation 
and on the variability of it, I now take up the subject of the influence 
of solar variation on terrestrial weather. I will first quote from my 
paper “Solar Variation a Leading Weather Element” (P. 4135, Aug. 
1953): “On January 28, 1953, the American Meteorological Society 
devoted the day to consideration of the influence of solar variation on 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 19 


ie 
Se a ae 
ER a ac 
I htalet {Adal 
oso ia lola | 


A aM er a ROTATION 27 DAYS 


a 26 3! 
Fic. fie s rotation in 27 days shown in 1915 solar-constant correlation. 


+20 


S 


SE en + 
is) 


-|0 


4l 


au : 

Bette. 

= 
2 
aa 


002 
Fic. 16.—Solar constant and West Indian hurricanes. 


20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Fic. 17.—Solar constant and severe magnetic storms. 


weather. An early speaker said he acknowledged the results of long 
continued studies of total solar variation, which had been made, as 
probably sound. But the variations found appeared to be of the order 
of 1 percent, or much less. No reasonable theory could show that 
these might have important weather influences. He distrusted statis- 
tical conclusions unless grounded on sound theory. Statistics might 
show that it is dangerous to go to bed, for a great majority of de- 
cedents die in bed.” The panel appeared to agree unanimously with 
him, for nobody supported the view that solar variation is an im- 
portant weather element. 

In Section 4, above, I show that a family of periods, exact har- 
monics of 273 months, exist in solar variation. In Section 3 I show 
that the solar constant measurements are abundantly accurate enough 
to support this fact. In figures 10 and 11 I show in many members of 
this harmonic family, how the shorter periods overlie and confuse the 
graphs of the longer ones, and how the shorter ones are computed 
and removed, leaving the longer ones as approximately sine forms. 

I now show in figure 18 exactly similar treatment of the 39-month 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 21 


period in the precipitation at Vienna, Austria. Jonathan Wexler 
computed electronically for us the 39-month period in Vienna pre- 
cipitation in terms of average percentage departures from the normal 
for all months between 1910 and 1950. I select the months when 
Wolf sunspot numbers exceeded 20. In our adopted nomenclature 
this group as presented from 1910 to 1950 is Category 2, Division 2. 
This means that the Wolf sunspot number was above 20 and the 
time interval was the second half of the years spanning 1870 to 1950. 

When one scans line A, figure 18, it naturally divides into halves. 
The average half, line B’, is computed and subtracted, leaving line B. 
From this the average third, line C’, is removed, leaving line C. From 
this, successive removals are D’, one-fifth, leaving line D; then F’, 
one-eleventh, leaving line E; and finally F’, one-thirteenth, leaving 
line F. The line F shows a nearly smooth sine curve. The small 
indentations may all be traced to accidental irregularities of pre- 
cipitation in the original observations. The amplitude of sine curve F 
is 17 percent of normal precipitation, which is 17/0.28, or about 60 


i Bo ee ea ae a a a 
ms De RT ele | 

COS ae 
aS 4 


| DY 
|S 


aeHifae 


ae ee ae 


> 


m baa | 


9 


= | 
Pale <a 


rh 


Eve 


a7aNncarans s 
POAC 
Sma a 


oa 


MET GNE lia Sp 
fide 
uc Se Rveany 


fe 
Y 


 < 
ae 
pate ea | eae 


| 
| | 
iS 


5 7 a WW 3 17 19 21 ed BS) Ge7 noc ua 33 a5 ams 


Fic. 18.—Thirty-nine-month period in Vienna precipitation, cleared of 
shorter harmonics. 


22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


times as great as the amplitude of sine curve line F (fig. 11), is in 
percentage of the solar constant. 

I will now give in figures 19 and 20 two more examples from our 
study of the period 273/9, or 30-1/3 months, in the precipitation of 
Buenos Aires, Argentina. Tables 2 and 3 of Figure 19 are from a 
direct photograph of Jonathan Wexler’s electronic computation of 
Category 1, Divisions 1 and 2, for the period of 30-1/3 months.* 
Taking first Division 1, Mrs. Hill’s graph (fig. 20) from Wexler’s 
means discloses an obvious overriding period of 273/18. Removing 
this as usual, the curve remaining discloses 273/27. After this I 
removed 273/45, then 273/63. This left a nearly smooth sine curve, 
273/9, of about 30 percent amplitude, with minimum at the fifth, and 
maximum at the twentieth month. 


* As shown on figure 19 we arbitrarily cut down precipitation when very high 
to 200 percent normal. 


PRECIPITATION 
BUENOS AIRES, ARG. FRACTION TABLE 30 AND 1/3 MO 
DIVISION 1 CATEGORY OLVISION 2 CATEGORY 1 
TABLE 2 TABLE 3 
1876 1886 1889 1899 1901 1909 1911 1922 1932 1942 


JUL SEP MAR APR NOV MEANS JUN DEC JAN MAR APR MEANS 


88 69 175 104 164 93 66 70 99 
95 90 120 = - aoe 110 158 ~=108 80 56 102 
120. ere 97-127), Qt 


156 102 138 81 122 
103 138 127 ES }e% 81 116 
95 147 185 146 9L 133 
87 145 172 124 93 124 


115 74 8 
83 79 = se alt? Ly 66 90 75 97 106 87 
110 E10 2 Pena) Ce ie, ere 
81 nae: 94 172 78 8 7 
65 ape tae nee ABM Ory), 87.10/56. RI an 
109 84 2 gil? ler Ws, ree 
104 105 76 a 7 100; <9) | S81 0G 85 
gia oy gal é a5)! 86 
114. 404) TA oe 4 er  po7 | “die 0 
142. 95 wa l22 leo 4 gor oF, Loz! “12. Erwiiee 
C2) Bee lc eee es SON ST) lie | RE) “a2 ese 


102 95 

ma hiss ue ef ioe a 110 dss 113 ieith 150 121 
ot fica on ae ie Be 123 144 NTs: 75 110 LES 
aa re haa ce res 92 123 98 110 34 101 
ae ia : oe 97 139 BY 124 70 104 
HA ; 43 49 94 113 101 161 83 98 ie) 103 
ve Wee athe oe oa ce 145 156 39 122 93 ILL 
e ae ies as ag The 165 184 33 118 94 119 
65 94 : * 


Fic. 19.—Electronic tabulation of precipitation at Buenos Aires, 304-month period. 


WEATHER—ABBOT 


SOLAR VARIATION AND 


CAT. |, DIV.2 aan 


LLC 


OO SRP Seeing eee ene eee eee Ss 


HNANAUANUE 


, 304-month period, cleared of shorter harmonics. 


‘cn mao 


Aires 


24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Turning to Category 1, Division 2, I first remove 273/18, then 
273/45. After removing these overriding submultiple harmonics, 
273/18, 273/45, there is left only the nearly smooth sine curve 
273/9. The amplitude of this final curve for Category 1, Division 2, 
is 34 percent of normal rainfall at Buenos Aires. This percentage is 
160 times the percentage of the normal solar-constant radiation given 
as 0.21 percent in P. 4213 (here fig. 10) for the period 30-1/3 
months. 

In further emphasis on the existence of subharmonics preventing 
direct recognition of the longer periods, while disclosing shorter ones 
otherwise hidden, I refer to eight telling cases: P. 4103, p. 4; P. 4352, 
pp: GS, Tike PR. 4390 4p 012. 


7. LONG-RANGE FORECASTING OF WEATHER 


As yet theory fails to explain why the family of harmonic periods 
related to 273 months, which exists in the sun’s variation of radia- 
tion, should exist also with identical periods in precipitation and 
temperature on earth. Still more surprising, the weather amplitudes 
have frequently a hundred times the percentage apparent importance 
of the identical periods present in solar-constant measures. But a 
huge body of measurements accumulated over 40 years, including 
detailed study of weather in 54 cities, proves that it is so. The periods 
in weather, to be sure, are hidden from casual view because though 
invariable in solar radiation (P. 3893, p. 22, figs. 11), their phases 
vary with place, time of the year, prevalence of sunspots, and growth 
of population. All these things alter atmospheric transparency and 
thereby shift tumes of terrestrial response to solar causes. But when 
long-time weather records are grouped with attention to eliminating 
as far as possible all these disturbing influences, and are cleared of 
overriding shorter harmonic periods, the identical periods observed 
in solar-constant measures stand out clearly in weather records all 
over the world as sine curves. 

In view of all these facts, may it not be possible to forecast rain- 
fall and temperature by adding together the amplitudes of these 
weather periods? This would require that the forms and amplitudes 
which the periods presented in past years, and the disturbances of 
their phases due to the causes mentioned above, would all remain 
sufficiently constant. 

We have made at 54 stations tests of such long-range weather fore- 
casting. On the whole the results have been very encouraging. Ex- 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 25 


amples will show both excellent successes and disappointing partial 
failures. 

About 10 years ago, working quite alone from the beginning with 
“World Weather Records,’ and computing myself every succeeding 
step, I began the study of the precipitation at St. Louis, Mo. This 
research occupied me for over three years and resulted encouragingly 
beyond expectation (see P. 4211). It was early discovered that the 
normal monthly values, customarily published with long monthly 
weather records, must be recalculated in two parts, one for low, the 
other for high sunspot numbers (P. 4090). Sunspot activity makes 
a very important difference, both in the magnitudes and in the yearly 
run of monthly means. This obtains both for precipitation and tem- 
perature. After inspecting the curves of Wolf sunspot numbers, 
it seemed best to set the boundary between low and high sunspots 
at 20 Wolf numbers. We have thus far studied 54 stations, using 
“World Weather Records” (now published up to 1950). These rec- 
ords that we have used have been electronically recomputed by 
Jonathan Wexler to give normal monthly values for sunspots 220 
Wolf numbers. The observed monthly values are all reduced to per- 
centages of these new sets of normals. 

In order to eliminate as far as practicable the shifting of phases 
attending time of the year, periods less than 15-1/6 months are 
separated into three groups. These are: January-April; May- 
August ; September-December. There are two series of these groups, 
one for sunspots less than 20 Wolf numbers, and the other for sun- 
spots above 20 Wolf numbers. As we see, exaggerated conditions as 
“smog” in Los Angeles County, Calif., the multitudes of automobiles, 
airplanes, factories, and forest fires which have attended the increase 
of population have greatly altered atmospheric transparency and 
thereby have shifted phases. We imperfectly allow for this by dividing 
the whole interval of records, say 1870-1950, into halves. 

So, up to the period 15-1/6 months, consideration of the (3 parts 
of the year) x (2 ranges of Wolf numbers) X (2 halves of the record 
interval) yields 12 groups of records for each period. Dropping the 
3 divisions of the year, from 18-1/5 to 45-1/2 months there are 
4 groups for each period. For the last 3 periods, 54-3/5, 68-1/4, and 
91 months, we drop the halving of the time interval, and retain only 
separation of 220 Wolf numbers. Altogether we separate the rec- 
ord into (12x15)+(4x9)+(3x2)=222 groups. This separation 
would restrict many groups to too few members to deserve confidence. 
So we make the assumption that, though different individually in 


26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


phases, the groups of the same Wolf number will be approximately 
of nearly the same form of distribution, and amplitude of intensity, 
and may be combined. But if combined, the individual phases must 
not be lost sight of. So when six columns of means are averaged to 
yield one general mean, the columns must be shifted bodily upward 
or downward to be as nearly as possible in the same phase. We use 
symbols, ok, Tm, Jn, to indicate shifts. Then the general mean column 
of the six numbers must be shifted back to the original phase rela- 
tions of its individual members when tabulated for the summation to 
make up a forecast. 

Beyond 15-1/6 months period, practically every period when 
plotted betrays confusion, for shorter harmonic periods override the 
period sought. This requires what is by far the most arduous com- 
putation of all. After the electronically prepared tables are received 
from Mr. Wexler they must be treated as was seen in figure 18 to 
clear the overriding shorter harmonics away. It is sometimes difficult 
to decide which submultiples are present until after one or two futile 
trials. Such repeated trials with periods 54 to 91 months in length 
are very tedious. 

The combination of 6 member columns into a general mean, as we 
do for periods less than 15-1/6 months, will best be understood by 
a numerical example. The letters a, b, c denote, respectively, data 
of January-April; May-August; September-December. Subscripts 
1 and 2 with them mean first and second halves of the records. 
As expected, these columns are not in the same phase. The signs, 
ok, fT, and |, show how much the columns must be moved up or 
down bodily to be brought into the best posture for uniform phases. 
When the mean percentage departures from normal in the final 
column of table 5 are used in the summation for prediction, the 
columns marked “ok” are to be replaced by the general mean column 
without shifting. The general mean values are to be Jowered 2 months 


TaBLE 4.—Berlin. Period, 7.0 months 
We-ler’s table. Means 


Cat. 2, Div. 1 Cat. 2, Div. 2 
a ok bi Ma C1 ok aa As ba v3 ca ok 
94 96 93 114 94 100 
97 100 96 105 97 100 
99 99 96 103 105 101 
99 95 91 100 109 97 
99 99 94 105 109 97 
99 95 101 103 115 89 


95 91 95 109 112 98 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 27 


TasLe 5. Berlin. Period, 7.0 months. 
Rearranged table with symbols unchanged. 


Gat 2aDived Cat. 2, Div. 2 
General 

Sums mean 
ar ok bi 4s c1~—) ok aol 43 bs Ws cs ok = +6 
99 99 93 109 109 100 +9 +1 
99 95 96 114 115 100 +19 +3 
99 99 96 105 112 101 +12 +2 
98 95 91 103 94 97 —22 —4 
94 91 94 100 97 97 —27 —4 
97 96 01 105 105 99 +3 +0 
99 100 95 103 109 98 +4 +1 


at bi, raised 3 months at b., and lowered 3 months at a2, so as to be 
in proper phases in the summation. As for the modification of 
results in periods from 18-1/5 to 91 months, examples of the re- 
moval of overriding submultiples are given in figures 11, 18, and 20. 


8. EXAMPLES OF FORECASTING 


Figure 21 is from a direct photograph of my work when comput- 
ing alone, beginning from “World Weather Records,” 1854-1939, and 
finally forecasting St. Louis precipitation, 1875-1879. (See P. 4211.) 
All the records, 1854-1939, of precipitation at St. Louis were used 
to fix every individual year. Every year computed, therefore, is 
equally as really a forecast as if it were after 1939 and stems from 
zero year (the median) 1895. The forecast gives a correlation co- 
efficient of +80 percent for the 5 years 1875 to 1879 between fore- 
cast and observation. Figure 22a, which includes Figure 21, is also 
a direct photograph from my comparison of forecast and event from 
1860 to 1887. Throughout this interval there is an average correla- 
tion coefficient of 70+10 percent. Figure 22b gives direct photo- 
graphs of results I obtained from forecasts of precipitation and 
temperature during the interval 1934 to 1939 for St. Louis and 
Peoria. I employed all records for the period 1854-1939 to fix the 
forms and amplitudes. In all these examples just given the depar- 
tures in forecasts and events, in short all the computation, is done 
with 5-month consecutive means. 

The reader will see that since all records, 1854-1939, were used 
impartially to fix every number forecasted or backcasted in figures 
21 and 22 and beyond, the number of years forecasted forward or 
backward is to be determined as starting at zero for the year 
(1939 + 1854) /2=1896. Hence, the forecast of St. Louis precipita- 


28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


-60 “80 -20 ° 410440 #60 5 
87s 40 60 $0 180 120 140 160 Lf 
Jen 41% 623-2073 14 460-11 42) 4B <3! 605-4) -52 95-93-15 +78 tHE -I -80 442470 -205 76 

413 -33 -33-°36-34% -77 -52 4182413 -86 -/0|-17-8 -6 1-103 -38 469496 -12 -44 +45 465 -324 73 

-30 a 26-18 -26 -58 -/01421 +3 746 -76 -56 -30 +50 +{50-|7 -30 +HF+E5-T4 86 
Apr +4 <2 -{3 452-18 0 Sd 4IF 464 +7 t7D FIESTA - (5 489 135-21 - 5 450 452 4388 126 
tH £14 220 629-9 $44 - 6-19 $45 431 F2Q 495 4B -ST 49 San493 4515-25 45 453460 +522 150 
1% 26-32 43243 +6) 4/9 Hl - 15 426425| 445 419 -39 tl 45497 4100-28 430 +55 4674496. 144 
Jur 413 414 -26 -18 +/6 4/08 +9/ -16 -/4 +30424%~-H +6 -22478 £Z04+ho 43-21 $49 4574574 642-146 
+30 taG 443 238 58 47 4110-29 47 HI! $41] -29 435 44 488 W747 475-96 452% 358454 4649 II 

-3 -33 +49 -36 -9 ~30 $9L-1F ~ 20-3] 424+ F +26 +)2 tIZASHF79 466-32 437 4604524705 [yl 
Oct 415 —4 446 -18 +3 -33~-17-/6 +2—S6 - 2-53 4/9 ee FS7 430450 454-32 +52 462 94E +263 120 
$30 -L +43 452HO 4H4 419957 HIF ~ FY -UOl-I2435Y 4353+ Met 44-9) 44S 463 tH4+452-IF0 

7% tin rea 42748 +60 491 #49451 +7 =-SA-6L44L HG FIG 4927 PRAY -30 43R F4444Z4674 163 
Jorn 1% — We IF F3ZHIY $108 FIN HIF 469 431 FO} 41 $105 +64 +33 +O pbeteB24 +50 463 4794683 173 
: = He +17 -28 ¢21 $624364307 1409 
PB 426 -3% -38 -26 -30- 17 ~ Yi -I5 4307/0) + Leh 49-25 413 4664324)25:130 

Apr 739-33 -26-36-16 39-1) -)8 ~48 465 +e [750 -BY —] -2) HO 460430 -75HNT 
4H +14 -f3 -18 484 -77 $2.-29 -32 428 Let 95 - 5O BG "80 -/6 -/9 +10 +54 42% 432 (28 

+4 26 -20 452-4 -58-(04 - [5 = 14 — 2°42 3 -) 42. hades -)3_ -§3 ~-22-)5 +10 +50 425-IF¥% j21 
Sub #12 -33 -32429-34 0-52-16 -H7 P38 $48 -FO+L24+5S -ZIH -30-(0 4/0 +57 +Z2-576 SH 
4/3 -H -26 432 .264108 - 6457 -20-3! +24] -52 +3 414-2266 -41-6 415 4524204726 [52 
=J0 -% 443-18 -18 +7 $15 HHO +2 -BO 44) 129 Steer TI 39-4] -50-2 420 46 +19 -174 13% 
Oct 8 -4 449-38 —F -304)) 92) 413 -SOM ap eHF ET ALS -24 -SU-36 -87 -S3 +L 422 4364/5 - BY 96 
#)3 -2 43-36 +3 ~ 53 sit gu = -2]-53 449 ~29-99-33 -93 -6L +7 4/0 439 +/0-/20 82 
4-30 t14 449-18 +16 353 a2) F CF t3l -4G-92-41 45-93 -40 -73 ~-EF +I -1d 32046-4272 4Y 
an +4 426 ~/3 452-14, °F B17 Hl 445 426 -SO- 62-86-61 “75-46 -G4 -694)7 -J0 +12 +7 ~ Seg TO 
412.733 720 427-2) ONLI 18 - [> 430 ~S6)-4 1-142 -16 93-44 424 -73 421 US +71 44-639 SZ 

$/U 733-52 432-26 sayy - ST RA LH 465 + b0y- 17-40 -F0-103 - 62 430 -77 423 -IZL4+L O- GI7 OF 

Apr 43°33 -26 -16 -1F +54 -i5 +2 P29 +7 -52 -37 -5% -S1 +12 -78 +56 -BO -6 -3 - 45199 
330 = ~/3 -38 -9 4/08 -32 "add sot! OS +2224 -§2-54 -TFH4S +X -|7T -5 -215 [08 
4H 2-20-36 +3 +7 +15 457 +51 -35 Py 46 4H 49 -§0 242-71 46 -30 -27 -8§ HB 10S 

Jub +H 414 432-53 +16 -30 49) 470 469 -([ 424]+ ae 421-50 -30°67 +47 -S 7-31-16 £152 IN 
HILLS -26452 436-53 +) 42[ 445-3) 425¢50 423 sgl BS -AS -1L0-66347 +5 39 -IF 4320133 

$19 -H 443429 -2 -77 4324/8 -IS-26 +I19b63 iT CES3 HR H38 + 6-50-47 4390 -45-1% +507 LIT 

Oct -30 -L +9 43243 -SB -17 +1T -4¢ -59 THIS 47L -30 +54 ~YOUG 49 946-22 4167 [27 
=F +i 443 -/F fl@ OO -I1 ~l4 -33 47 +44 UG 442 YSCWS9 -IF 42€ -304 44 +52 -H7-25 +239 134 

fa7g +13 *26 464 -3 984 HAF -F2-4[ -14 931 - 2-1 +1052 £43.-3 478-2441 457 49 -27+492 127 
Jan -39 33 -13 -36 -l +57 -52-18 -H7 +26 ~46)- TIS 415° +69 -13497 752 -I4-35 EHS 125 
tH -33-20 -)8 -34-77 -& -19-20 430 -50 #30780 OF 24S -I6 -M 2 1 132 

PID 24-32 452-266-586 tIS-+1F 4D OS -56 U7 48Y F20421 F398 757 -30 4100 147 

Apt 41%. -2%-26 429-18 6 -I1 -16 +/3 +23 -6O R4654%3 pSt!7 430-62 -32 4 159 137 
eID +14 -13 432793 t4H4¢-S2457 +51 O- 10 804+99 +9G4II FU ~ES~35 +445 146 

$30 26°19 -/8 +84 +6) -JOL +40 469 ~ 11 41%] -4) $49 -3 t4esleoths +1049 +197b6E -37 2 424 139 

Jub +4 rly -32% - 36-10 4+10%-S2 42; 44S 23) +42] -17 tH Lhe SY +92497 +9543 410-79 -424277 (23 

TH 426-26 “36-39 +7 ~6 -1Y -1F ~ T6429. P=zS6 29 $55 480479 HF -3 410 -Tl 4S a3Y¥ 10 

FIL 033 443 -1G/-26 -30 HIT -4l UY ~$4e4Z5] HOO -H2 -27 +14 463480 +1/0F-F +/0 -71 -¥7 ~f36 10 

Ov’ £13 -4 449 452718 -53 49) -[% -33 $7 +2 PR 90-45-37 446 4HY F9FS-IZ 410 -70 -SOLI104 107 
30 e243 629 HY 4-29-14 43) TYLOS -52 -61-54 430425 +76 -12 445 -GE-S1 486 96 

1I8T% - 2 <4 AH] £32 +3 444 4I9L-IF -N7 44HG +50 - 8-76-99 +7 42!) +66-/7 +20 -56-S2 ey BE 
Jan 4/2 -2-]3 -|@ +16 $6! -t7 “18 -20 +% AS +6 ~ 80-93 -/3 +4 454 -21_ +22 -59 -53-154 73 
¥13 F)4-24 -38 = 14 2108 <1) 457 422. =I) 435 -5F-76 -20-GO 44-25 410 -S6 SY -164 75% 

= 30 426-92 -36-34 ce re 4 Oo] -29 41% -39 -93 -30-8¥ 434-18 -/4 -$a-SS ~483 52. 


Ape 4 33-26 - (5 269036 $B 4/94 ~22-103 -35-FO 425-31 -TO -4¥G~S4 -Gos 5? 
tH (4-13 452.718 - “5346 oY -£6435 -83 417 -9LAHS -YO-57-S500 SE 
$12426-20 429 -F -17-*6 92.935 #1211 ~ 40°76 FF -32-13% -IF-52-475 6? 

Jul #13 233-32 2I2 +3 -s3 +e -62$¢28 42549 -HS-S6 -] -32 -80 ~-30-S7-USS ST 
-30°4 -26 </@416 @ } +H #/7 433 421 -47 -91 -10 -31 -¥F aa ee ey 
$13 -2 443-35 POY HOY HI -2Z3 4 -TTN SY 442 478 -50 87 22-30 -30 ~17 ~SE 10% 

Oct “30 -Y449 -36 -F +61 a PL pur $6¢ geF-SF1 -93 -30-24 -F -12-SF +16 LO 
SH -2 443-18 43 t1@ -17 -16 -47 424 : Ps Ten Se -98 HR TF -S ST HUIS IE 
WS TIG47 TSANG 47 41 459-33 430 44 5 P44, 497-50 -44 -30 -25 490 +2 SO 4d If 


Fic. 21.—Forecast of St. Louis precipitation, 1875 to 1879, compared to observed. 
Forecast dotted. 


29 


SOLAR VARIATION AND WEATHER—ABBOT 


3 


NO. 


"6L61 OF PEST ‘Snoliea sjseoat0F f /QQT-NOQT “Uor}epdro0I1d stno’yT 1G JO ysed0IOJ—'q ‘eZZ “DIY 


peo 
; i j Tae 
Lo th ca AEST ARM NG wg 


s24nt 


° 
a 


BEE 


Hey ibn NW Ie et 
Hg sent 
NPEEEEEREER AEG 


| 


kK 


IVWVON 40% 
SIMivwe7Q 


' , . 
PY (ant sun Mma AVN Im 
y ; 1 } 
| ; U 
‘ } 
H Yelehz G9 NOVI TENO sia co lava mw 
$AHLNAG] o= == PSH LN AS 
ANIA “NGA 
NOIL LN DGey SI ino} 16 WAIT 


30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


= 
°C 
of 
i 


BTS 


SSDS 
SR 


ae 
SSA A 
wanes 


ce 
< 
E 


; << 
me 


j 5 9 13 17 él a5 


Fic. 23.—Precipitation, Helena, Mont., 39-month period, cleared of 
shorter harmonics. 


tion for the year 1956 in P. 4211 is a “60-year forecast,” as claimed 
by the title of P. 4211. 

I will conclude these illustrative examples by graphs (see figs. 28- 
37)® taken from P. 4390 and P. 4471 on precipitation and tempera- 
ture forecasted or backcasted from data smoothed by 3-month con- 
secutive means using all the records from about 1870 to 1956. It 
will be seen that forecasts and events have about equal amplitudes. 
They evidently exhibit the same principal features. Principal and even 
minor features in prediction and event prevailingly coincide on the 
same months. But sometimes there are displacements of 1, 2, 3, or 


5 Figure 33 was prepared from the 1,032 months of records used in P. 4390, 
“A Long-range Forecast of United States Precipitation.” These records cov- 
ered the years 1870-1956 and centered on the year 1913. All 1,032 months of 
these records had equal weight in the forecasts. The observations quoted in 
figure 33 were not available till late 1960. Figure 33 was used as a slide 
at my National Academy paper of April 1961. ‘Whatever success it has 
is for being a verification of forecasts of precipitation for 14 cities from a 
zero date of 1913, 46 years previous to 1959. 


SOLAR VARIATION AND WEATHER—ABBOT 


3 


BAP ARACARICHUAICALERERL AGL 
nD SHEA 


Ha ani Stn: SUT TE 
ana YaLas afeue ei =H 
| Arter aoe Sapp She 


Haan bin SUNRESTINEBESE err 
<GnES 


PD CORES ee 
CSP LRC RECT SECT REET ETT ec =a 
AEsMASHAAS! ae PREESI TT prorat 

TEEPE re ieee peter 
ECE Het TES Preeti 


2. 2. 2” oO Oo oO 
3°os So sara 2 oho eS 


_—— — eee 


. 454-month period, cleared 


Ariz 
armonics 


ridge, 
h 


1 Bri 
shorter 


of 


24.—Precipitation, Natura 


32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


vam 
Eee 


Fre. 25.—Precipitation, St. Louis, 684-month period, cleared of shorter harmonics. 
Three months shift of A. A and C observed before and after 1900 combined in B 
for further work. 


NO. 


Ylo 
Mo 


3 SOLAR VARIATION AND WEATHER—ABBOT 


PHASE AbJuST MENT. THE 63, Mo. PERIOD. 


CaTeGory 2 ASSEMBLY. 


at bex cls @,o% box 


105 95 108 100 
102 95 406 93 
00 9% 95 92 
lo6 92 104 97 
loFS 103 98 106 
103. 4 108 lo5 


~< 


Fic. 26.—Phase adjustment for 6-column combination. Eastport, Maine, 
precipitation, 6%45-month period. 


- 


i ¢ 
~-._ Curve LT” 
~ : eS 


Giek Meat A 


95 100 +t {20 

88 97-2 108 

80 92 -7 pe 

8&& 97-1! 

91 {0 +2 103 
"99 105 +6 u0 


Ig 
Net RANGE 160s OF NonmacTova | 


| ie Oe 
RANGE 13 of. of NORMAL ToTAL CaT2 | 


Catecony lAssemBLy. 
on @r0x $42 Cor Mran O 


at; bs 


100 jos 
{00 aI 
106 .g90 
{02 87 
10% 91 
103 102 


(]7 lo7 
16 G5 
100 95 


68 9), 


73 
oO 
-3 


-7 


93).93)=4 


$7 10) 
6) 586 
“96 


+3 


33 


34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Oo 


o 


ay 


PERCENT, NORMAL PRECIPITATION AMPLITUDE 


Fic. 27.—Exact length of the 273-month master period. 


even more months, evidently for the same features. If these shifts 
could be predicted and corrected in advance, very high correlation 
coefficients would exist between predictions and events. As it is, 
correlation coefficients run from +50 to +70 percent at the dif- 
ferent cities for precipitation, and from +30 to +50 percent for 
temperature reckoned from absolute zero. 


9. ACCURACY OF PREDICTIONS 


The average percentage difference between forecasts and events 
runs from 15 to 30 percent for monthly precipitation values at dif- 
ferent cities. It runs from 1.5° to 2.7° F. monthly for temperature 


35 


SOLAR VARIATION AND WEATHER—ABBOT 


3 


NO. 


‘PpoArosqo YUM poredusod ‘gc6] 03 OS6T “YSeAA ‘oueyodS pue “D'q ‘UO uTYseM ‘UOT}eVIdIDOOI1d Jo JseI9I10,J—9Z “OI 


rere cera ! 
com avAN Ree NT EN 
ave 


my ne 
Saas 


Phin ot tt i ee 


TQaABaSaO” i 


SNWSIW PNINNNQ HLN 
WaON Wows Ssaudvdag 1 


MISCELLANEOUS COLLECTIONS VOL. 146 


SMITHSONIAN 


36 


8S61 OF 
OS6I ‘PaAtasqo 0} paseduiod ‘Z{6] 0} PO6T PUe ‘QC6T 0} OST “UUaT ‘a]]IAYseN ‘UONepIdID9I1d Jo }sed910 J—§z “DIY 


656! sé! S61 “9561 SS61 ¥561 €S61 ZS6l- iSé6t Osét 


Pee ese ayes eee 
Sea eviseleel \ieldlelblneneiGtiseesgesae 
lel a as ft ae ese ele es ele i cae ee 


£161 7161 1161 o16t 6061 806! 406 906) S061 rO6I 


Eine 
aaa el 
| 


a= 
en 
ieee 


Ui 


—I 


7 
‘ 
‘ 


! 


Percent Normat Precipivation 
+ 


== = 
fie Ee Fe bm as SS el oe de ol al ol 


& 
ia bene Oalecelemee 


i Sa 
sd 4954 


1952 aaa, 


oe So eas 
a 


Fic. 30.—Forecasts of precipitation: Cincinnati, Ohio, below, Denver, Colo., above, 1950 to 1958, 
compared to observed. 


forecasts. The shifts mentioned above heighten these average differ- 
ences above what they would otherwise be.. What most recommends 
the 10-year forecasts of precipitation is that they are about equally as 
accurate for times of wide departure from the normal as they are 
for times close to normal precipitation, as is shown for 14 cities in 
P. 4471, page 6(here fig. 33). 
10. MISCELLANEOUS MATTER RELATED TO SOLAR VARIATION 
Ionospheric observations—From measure of h’F, published by 
the Bureau of Standards, Mrs. Hill computed daily averages of 
h’F, from 1944 to 1957 (see P. 4338), for the hours 11, 12, 13. 


An earlier study had fixed the average monthly value for these 
[Cont. p. 45 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


38 


— ‘paarasqo 0} posedwioa ‘SS61 97 OS6T “Jue 


) ‘oyusurerses ‘uorjeydioa1d Jo ysed21I04—" TE “DIT 


A 
i | 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 39 


Fic. 32.—Map, forecast of precipitation, 32 stations, 1963, in 4-month means. Note: 
P, 4390 gives maps like this covering the years 1963 to 1967. 


PERCENT NORMAL —— 


40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


EG ee 
ea i as PE ad amr asl 

OS 7 ae ee 
ee ee a ee ee ie ee 
a Sa a ES BE 
Da A a 2 Sa a | 
i i aim a 
Sh es aeole eran ans on 


Hae eee 
NECMOB EN Coe. 
CECE Nes 


in 


Fic. 33.—Verification of monthly forecasts of precipitation, 1959, 1960, for 
14 stations. 


AI 


SOLAR VARIATION AND WEATHER—ABBOT 


3 


NO. 


“paarosqo 0} pasreduiod ‘Q96T 0} OS6I “O'C ‘UO SuIyse AA ‘onjesrodu9} Jo }seI91I0 J—'fPE “OLY 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


‘PaAJoSqo 0} pasedwi0d‘QO6T 0} OS6T “AGEN ‘eyeUCO ‘ainjes19dura} Jo yse09I0qJ—ce “DIY 


; bs 3 2 ig 


cSt s a 


sf 


capes 


LA WARP PAZ 


oss 


SOLAR VARIATION AND WEATHER—ABBOT 43 


NO. 3 


“‘PpaAtasqo 0} paredurod ‘O96T 0} OS6T “YeIN ‘AUD aye yes ‘oinzessduso} Jo ysev0I0.J—o¢ “OT 


09 €S ef “s 9S 


uC es 


WALA Avan 
Sas (LE eee Gi eee Cth 

2s Ee ee ee ee Le | BS 
Bg fe 205 cSh ide Noss eed ees eee 
eS Sob Galt ES eee? | ee es 


Ss +s es 2 1g 056 


ove 
ote 


44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Fic. 37.—Map, forecast of temperature, 10 stations, 1963, 4-month means. Note: 
P. 4471 gives maps like this for the years 1963 to 1967. 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 45 


hours at 315. Monthly means of Mrs. Hill’s values were subtracted 
from 315 yielding the results given in table 6. (See fig. 39, p. 48.) 

From these figures the average probable error of the monthly mean 
departures is +15 over the years 1944-1957. Combining the de- 
partures with observed mean monthly values of h’F:2, the observa- 
tions are cleared of the yearly march and become adapted to show 
if they indicate relationship to the family of harmonics in solar 
variation. The result of this comparison is graphed in figure 39 
and appears in table 7. 

Within the limit set by the probable error, the graph shows sine 
forms for 7 of the 9 solar periods included. The other 2 are too 
short and irregular to judge. 

Washington daily precipitation—The sun’s rotation period is 
approximately 27 days and is almost an exact submultiple of 2 years. 
The daily rainfall at Washington was tabulated from 1924 to 1941, 
18 years, in 9 tables of 27 columns each. The 9 means of the 27 
columns were then tabulated. This disclosed a small departure 
from exactly 27 days. The exact period of the solar rotation for 
the purpose in view was found to be 27.0074 days. P. 3765, figure 1, 
here reproduced as figure 40, shows how the rainfall of Washington 
was distributed over the period of the sun’s rotation during the 
years 1924-1941. (See p. 49.) 

The curve for years of average rainfall, No. 3, indicates that 
days 1, 2, .3,.4;.5,.12, 17,22, 24,.25, 26,.and 27, of the sun’s' rota- 
tion cycle (based on January 1, 1924) are apt to be more rainy 
than other days. From 1941 to 1954 I published yearly a pamphlet 
predicting 175 days of the next following year when precipitation 
was more likely to occur than on the other 190 (or 191) days. The 
experiment was successful beyond expectation. Indeed in 1948 no 
less than 14 brides wrote or phoned inquiring if their wedding day, 
to come in a month or six weeks, would be fair. I predicted 13 fair 
days and one probably rainy. As it turned out 14 were fair, but one 
was overcast without rain. The most outstanding success concerned 
the hostess in a restaurant. I told her that her day would probably 
be fair, but there would probably be rain the day before and the 
day after. That was exactly what happened! 


TABLE 6.—Monthly mean departures of h'F2 from 315. 


Jan. Feb. Mar. Apr. May June 
—69+9 —27+14 —27+24 +29+30 +88+26 +89+25 
July Aug Sept. Oct. Nov. Dec. 


493419 4-73-35 O11. Soo) F105 730 


Years 


46 


Tasie 7.-—Periodicitics in monthly values of departures in h'Fas 


Monthly 
periods 


2-23/24 


4-1/3 


5-1/18 


6-1/15 


10-1/9 


11-3/8 


15-1/6 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


2.1 
—8.5 
2:1 
—0.5 
pet A | 
+4.2 
pol 47 
—5.1 
E22 
—3.4 
+2.5 
—3.1 
s2.0 


March of departures and probable errors 


—2.5 +0.2 


22.1 +21 se2.1 


—72 —41 —1.7 
se2i shal se2i1 


+79 478 +0.5 
sE2.2 sE2:2 a 4 


+83 460 +7.1 
225 wees 223 
—17 -53 —73 
EZ Seat WEEE S 


pH | 


+1.6 
a0 


+0.1 
2.2 


+5.1 
2.3 
—8.7 
+23 


—0.6 
+1.8 
—3.3 
+2.0 
+4.6 
21 


a4 
+21 


—10.0 
E22 


=1.4 
+23 
£47 
+2.3 


VOL. 146 


—3.2 
2.0 
—1.4 
| 


+8.5 
a2) 


—49 
ae22 


—2.5 
+23 
—0.3 
cea 


—4.4 
+21 


472 
+21 
43 
+22 


+0.2 
2.3 


The following table 8 shows the results from 1934 to 1954, I 
give the ratios of yearly rainfall which came on 175 preferred days 


to that which came all other days. The expected ratio is 1.42. 


In the year 1952 a peak rainfall 6 times the average of all other days 
of the cycle appeared on the eleventh day. In 1953 the peak had 
fallen to 3 times but was still extraordinary. I have wondered if 
the surprise was caused by atomic bombing. After 1954 I was too 
immersed in long-range forecasts, solar boilers, and my second mar- 
riage to continue with Washington daily rainfall. 


Various periods. — 
1. The 6.6485-day period. 


TaBLe 8.—Precipitation of Washington. 


Ratio 175 preferred to 190 (191) other days 


.. 1934-1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 


ations 


La5 


148 1.06 1.45 1.10 1.28 1.56 149 1.34 


a 


Mean 
Ratio 
1.45 


Fic. 38.—Phase adjustments for 6-column combinations, Washington D.C., 
temperature, 13%0-month period. 


Many have noticed that if a rain comes Saturday it may recur on 
several succeeding Saturdays and after that on Fridays. We have 
tabulated daily temperature in New York and Washington to in- 
vestigate this observation. We discovered a well-marked period 
(fig. 41) of 6.6456 days (later improved to 6.6485 days) and also the 
half of it in the weather of New York and Washington. (See also 
P. 2499, P. 3893, P. 3990, and P. 4015.) The period was found to be 
within 1 percent or less of 1/1250 of 273 months as computed em- 
ploying the exact period of the earth’s revolution about the sun. 
The shorter period is exactly 273/2500 months. 

2. Human pulse periods. 

My friend Dr. Frances P. Marshall has a 3-year daily record of 
her pulse taken before rising each morning. She permits me to 
publish my analysis of it (see P. 4265, pp. 15-17). A very smooth 


48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Br deal 
LCL 


JUNI EEE TE 


ALLE ee 


idiot was) 
wed 
yo 
ma pee 
(aed PSY 
Bees lated 
vs eS 
ik Bae: 
Re 
rake 
Faby 
ER a 
ee ee 
oe WE 
inane: 
ae ia 
bas Se 
ete 
aii 
ee 
ae 
gi I 
ce 
rae 


See 
= 


Fic. 39.—Periods in ionospheric h’ Fz, harmonics of 273 months. 


sine curve of 7.0 months, shown in figure 42, was found after clear- 
ing the original 7 months observations of harmonics of 1/2, 1/3, 1/6, 
1/7, 1/11 of 212 days. Its amplitude is 2 pulses on 67. All the 
periods named are of course exact submultiples of 273 months (p. 51). 

3. Droughts at long periods. 

By invitation I published in the year 1938 an article on solar varia- 
tion and weather in Zvlastni Otisk, the organ of the physicists at 
Prague in Czechoslovakia. In that article, so far as I can discover 
now, appeared my first prediction of very long-range solar periods. 
I quote: “Records have been kept of the levels of the Great Lakes 
of North America regularly since 1860. In addition, partial records 
exist which fairly indicate the levels of some of the lakes since 
1837. From these sources I have prepared figure 6 [here fig. 43, p. 52] 
which indicates the water-gauge values for Lake Huron for a cen- 


NO. 


SOLAR VARIATION AND WEATHER—ABBOT 


N S 


49 


Fic. 40.—Period of 27.0074 days in Washington precipitation. See P. 3765. 


=. 


20° 


ae 


1 
Aarne 
st 


— 
a Sa ae 
_— ce 


O 
sake 


= 
ee __ es 


WASHINGTON 
Fic. 41a.—Period of 6.6485 days in Washington, D.C., weather. 


BIR 

LU 
iL UAT | 
Aa at 
nA 
2 


eS NOK 30 OS JO 
NEW sone 


Fic. 41b.—Period of 6.6485 days in New York City weather. 
50 


es 


15° 


-5° 


NOi"3 SOLAR VARIATION AND WEATHER—ABBOT 5I 


ean RANKR 

PLEA rh Wd 
EER AVILES 

ACTER ls WRN Mond 7) 


ae A A A 
Pe Rr eae ee 
SUMWATAV AABN RICAN 
Va VN LMA eh TT TNT TN 
VALI Vii AY RAUL 
COE COUR 
mers wl cP le Ua al 


oR SSL Ste 


eS a ee | ra 
Pa CA a Ad AL 
epee ary ea Ty 
tt ONE 
Fa a De 
Le lL ol a AB SS el a 


iE SEE 


Fic. 42.—Periods in human pulse rate, harmonics of 273 months. 


ae 


tury. The data are arranged in cycles of 46 years. It will be mani- 
fest that each 46-year cycle shows near its beginning a great depres- 
sion of about 10 years length from crest to trough, and a smaller 
depression of about the same length following 23 years later. The 
great depressions following years 1838, 1885, and 1929 were each 
associated with disastrous droughts in the Northwestern States and 
adjacent regions of Canada. There is much reason to expect a 
recurrence of such a drought beginning about 1975.” 

The two principal periods of 91 and 45-1/2 years, 4 and 2 times 
273 months, depress Lake Huron about 5 feet. They last before 
complete restoration of level about 10 years. A drought of less 
magnitude in the supply area of Lake Huron has the single 273- 
month period. I predicted about 1938 its recurrence in the decade 
of 1950-1960. It proved very severe in Southwest United States 
and is plainly forecasted by my study of St. Louis precipitation “60- 
year Forecast” with a forecast of its end in 1957. (See P. 4211, 


52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


oa an (ae a Sa is es 
SS oe 
Soeg aw Seta GE tribe 
Re -=ee vane 


Sissasceece 


Fic. 43.—Periods in level of Lake Huron. 


fig. 6). The two great droughts will probably begin about 1975 and 
2020, respectively. 


11, SHORT ACCIDENTAL TRENDS IN SOLAR VARIATION 


Measurements of the solar constant of radiation reveal trends of 
rising and of falling radiation several times each month lasting from 
3 to 10 days each (see fig. 45). We noticed as early as 1925 that 
the temperature at Washington seemed to respond to these acci- 
dental solar changes. This led us to make a very extensive investi- 
gation of this matter, in which we found (P. 3771, P. 3893, P. 4462) 
that not only the measures of the solar constant of radiation but 
also measures of the areas of solar flocculi and measures of Fe in the 
ionosphere precede up-and-down temperature trends. These trends, 
observed either in solar constant, calcium flocculi, or Fe changes, 
produce opposite effects, like the right and left hands, on tempera- 
tures the world over. (See figs. 45, 46, and 47.) The temperature 
changes are not small. I cannot better bring out this phenomenon 
than by quoting from pages 219-221 of my paper of 1939 (Quarterly 
Journal Royal Meteorological Society, vol. 65, No. 280, April 1939). 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 53 


Fic. 44.—Drought prediction from forecasts of precipitation, St. Louis, Mo., and 
Peoria, IIl. 


Referring to Fig. 1 of my just cited paper, here reproduced [here figure 46], 
the reader will find there two curves for each month of the year showing 
departures from normal temperature at Washington, D.C. In each month the 
curves show a well-marked opposition like the right and left hands. The sep- 
arations of the curves in the months January, February, March, April, May, 
June, August, October, November, and December range from 14° to 24° F.,, 
and evidently constitute major departures from normal temperatures. Similar 
results, showing in almost all cases opposition like the right and left hands, 
but differing widely in actual march of the pairs of curves, are shown for 
St. Louis, Helena, and Potsdam, in other illustrations in the cited publications. 
The curves are computed for all these cities starting from identical dates, 
320 in number, scattered over 12 years. Some 10 to 20 cases combined are in each 
curve shown. The data of temperature departures in each case cover 16 days 
following the starting date selected. 

How were these 320 dates selected? They are chosen as dates when solar 
variations commenced. As shown in Fig. 1 and Table I of “The dependence 
of terrestrial temperatures on the variations of the sun’s radiation,” they 
comprise all the dates during 12 years when good consecutive solar constant 
observations, made mostly at Montezuma, Chile, began to indicate rising or 
falling sequences of the sun’s output of radiation. The range of these sequences 
was small, rarely exceeding 1 percent, and their mean range is only about 0.7 per 
cent of the solar constant. Owing to the interference caused by changes in 
atmospheric transparency, superposed on the inevitable accidental errors of 
measurement, it is highly probable that some of these 320 cases are spurious. 
If the spurious cases could be eliminated the average temperature departures 
would doubtless be increased above their already large magnitudes. 


54 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


19 6 0c aes SSARDRSDORR ORGAO E eS eS 
Ee ees aR Sees PAOReReAs 
GESREE ROS 


[959 
| a0 
be PERE Eee are re 
Zz SSRSeeeeeSRSRn8 
1.950 
= 
n 
z 
ro) avenneen cco 
5) AREER Ae ee oS 
Seana datatatanaeeceans 
1.445 sseeestesaticeits H 
a il COC eee 
PCCP RSE eat 
SSSS000:~E 50 00S 0e5R88 
ai CLIRISIN Q HEHEHE 
fa 2 
aan PERS else tal petal 


EEeees 
iqas fete sneer fas tantatent fat?st 
SEPTEMBER 1934 


Fic. 45.—Solar constant, rising and falling trends. 


Our critics, however, maintain that they have demonstrated by correlation 
methods, and by references to my writings, that the supposed variation of the 
sun is largely due to defects in our methods of observing, and that the con- 
sequences deduced from such supposed variations are illusory. 

If so, we must assume that these interesting curves, which show such extraor- 
dinary inversions of temperature departures, would as likely as not result from 
a haphazard choice of any 320 dates, quite as well as from the selection of 320 
dates which were chosen because they were observed to be the dates of com- 
mencing solar changes. This is of course absurd. 

Several of my friends have urged me to omit the just-preceding part of my 
defense. They consider that meteorologists are so firmly fixed in their dis- 
belief in the meteorological importance of day-to-day solar variations that no 
meteorological evidence whatever can persuade them to reconsider the matter. 
But though I may be singular in my opinion, I regard the present argument 
as unanswerable. Though it may be futile for the present, owing to this pre- 
vailing attitude, I shall proceed to place on record still more support of my 
argument. 

First, as I showed in my paper, “The dependence of terrestrial temperatures 
on the variations of the sun’s radiation,” above cited, not only do several cities 
show large opposing trends of temperature following rising and falling sequences 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 55 


Van ls 
Napa 
Oe ie oh Pa ys a! 
Atala a 


H 


Mola 
SNARE 
Se at 


7 


Bh 
Paar ai 


fia 

ancien 
TR 

iS G@iN=. 


Fic. 46.—Temperature, Washington, D.C. Opposed effects follow rise and fall 
of solar constant and, 2 days later, of calcium flocculi observations, January 
to December. 


a3: 
iw, 
ie 


56 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


a 3 
Ms ; 

‘ & / 16): =/ |-2. 
eid Z ¢ -2 Lo fe s}-3 = 
rr) = 10 q YES eT 
17898 S oh mi 2 ps 3 2. 

2 o - cm = = - = : 
fq4l% —2 <4 Zz eA es = oy ; 
x g ig Z PSS E EES | 
ax Z fi\3 55 i? | 
445.00 1& (otitis | 
pee TN rleed | 
(4417 3310 - I igs G: 
Lh oF IONS Hy | 
Ler |so =] > y Gus 
“ j 20 r3/ ~ Mier) 
1s \/23 307 997; 16d | 


3 Blhens 16 IM LEI | 
17584 SS a oes 
ae MIP CSR 
cd Pa ae, ead ot | 
1939 % SND Dy tone eal lies 
3s =i Sus 27, : 
“ fee i 2 ee Pp: 

aug Z ; min D 
"944 1 0 -jo-= ss: 
: 4 ‘-Y-l2 -F 10 pan 

1s d = 5 - 2 om A: 

MaCS q 7% Yo | 3-25 oo; ia 
fk: ¢ Be A(z 22 le Ne Ane 
943 9 /o 17 No.3 = ASG 
ache 10 Ge Als e 716 dB: 
ae if 51-3 Lo |e Bea 5: 3h? 
is 1 57 laf 35/29) Go ~221-22.- Gi, 76. 
say fo 41 a|0s 30 -$2 -S 2H -1F 4: 
(35) 2-9 176 3b 366. = i106 2F2| 
1 hee ean 2 BO eS 
10 4 sat Nive H . 
s : - 3 | 
S |B 7 e NTIe-NIYNY shan : 
e . : 


Fic. 47.—Temperature, Washington, D.C. Opposed effects follow rise and fall 
of solar-constant and ionospheric observations. 


of observed solar variation, but 46 cases of unusually great observed solar 
changes were followed on the average by 1.95 times as large temperature changes 
in the same phases as the mean of 150 cases of all amplitudes. Again, the 
average trends of temperature following solar changes, as observed in the 
years 1924 to 1930, were nearly identical in phase, magnitude and form with 
those observed in the years 1931 to 1935. 

But now I offer a new evidence which I think is even more convincing. 
If, in reality, the observed variations of the sun were real, and influenced 
temperatures greatly for 16 days after their incidence, there still seems no 
reason to think there should have been any unusual temperature effects imme- 
diately before their occurrence. I have therefore computed for each of the 
320 dates the march of temperature departures from normal for 16 days 
preceding the dates in question. I have then computed correlation coefficients 
for Washington as between the average marches of temperature attending 
rising and falling solar sequences, both after and before the beginnings of the 
sequences of solar change. 

To fix ideas, I recall that in each division of this test there are 24 lines 
comprising 17 values each, two lines for each month of the year, selected from 
the 12 years, 1924 to 1935. These pairs of 24 lines of the divisions are 
separated into two types, one type containing 17 values for days following, 


NO. 3 SOLAR VARIATION AND WEATHER—ABBOT 57 


and 17 for days preceding the beginning of sequences of observed rising 
solar radiation. The other type comprises 17 values for days following and 
17 values for days preceding the beginning of sequences of observed falling 
solar radiation. Two correlation coefficients are to be computed, one including 
the 204 values of the two contrasted types following the supposed critical 
dates, the other for the 204 values of the two types preceding them. 

In order to avoid diluting the correlations by including extraneous influences 
due to previous conditions, each line was first reduced to the level of zero 
temperature departure, by adding to all 17 values in that line a constant quan- 
tity such as to make the average temperature departure for that line zero. 

Having thus arranged the values, correlation coefficients were computed 
between the two types for the two divisions. They resulted as follows: 


After appearance of solar change, r= — 54.3+4.9 per cent, which is sig- 
nificant. 

Before appearance of solar change, r—11.1+6.0 per cent, which is mean- 
ingless. 


The inference is obvious that the 320 dates, above described, were dates of 
real significance, since no other consideration was used in selecting them, and 
it is difficult to avoid the conclusion that they were dates when real solar 
changes began. 


In what follows I shall show that the ionospheric Fe and the 
areas of calcium flocculi are both as effective as are solar constants 
for this observation. (See figs. 45, 46, 47.) 

Graphs from other publications will show plainly how important 
this matter is. My latest paper on the subject, P. 4462 (see figs. 48, 
49), shows how a modern satellite could obtain first-rate solar con- 
stant values daily, and the results radioed to earth would give means 
to predict temperature changes all over the world for 16 days in ad- 
vance. Furthermore, if such satellite continued indefinitely to circle 
the earth we would have means to discover great changes in the solar 
radiation like that of 1922 and 1923 if they occur. This might lead 
in course of a century or more to discovery of long-period variations 
of solar radiation of high importance. 


12. DRIFT OBSERVATIONS 


As stated in the introduction, and illustrated in plate 1, our Mount 
Wilson observing station was equipped with a tower telescope. The 
solar image, about 8 inches in diameter, fell upon the slit of the 
spectrobolometer in such a way that when the telescope clock was 
stopped the solar image drifted centrally over the slit. The spectro- 
bolometer was set for any desired wavelength, and the intensity of 
that wavelength along the sun’s central diameter was recorded as the 
solar image drifted. In 1908 we began to make these drift records 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


58 


dua; uo%zD 


pa ro eee eee oo 
Weel onli | bere Lh Peers) Cerri 
2 See ORS ee 
Fete ess reise NPE Meeses Lh tas i 

ARC ee ere eer eer 
CACC REESE 
SUDN SEES SIRES 0 hs IR EN EOP Is wwe cereale =a 
Mme eaigee er ie) ei a ee 
—Jh¢ dears aa Oe Sree ee RES Ce Be Se 


IS6l - 76) 


4 Sate a : on | ie = 
BRE oo ee tS sesoi™ Brn < 
ie ae SERS 7 i i Poe % 
4 eb ey has 4 


| 

| 
rs 
ve 


mn ei). 
em | 
mec 


Fic. 49.—Washington temperature controlled 16 days by sun’s variation. The sun changes at S. Washington temperature, —5 to 16 days. 
Full curves, solar radiation rises, dotted curves, falls. Zero departures at O. See P. 4462. 


NOW 3 SOLAR VARIATION AND WEATHER—ABBOT 59 


every time bolographs of the solar spectrum were made for measur- 
ing the solar constant of radiation. This continued until 1920, when 
we removed from regular observation on Mount Wilson. 

Our purpose was to determine if changes in the solar constant are 
accompanied by correlated changes in the U-shaped drift curves. 
Dr. Langley had hoped that they would be and that drift observa- 
tions would serve as an easy means to measure solar variation, 

The entire series of comparisons between solar-constant variation 
and solar-contrast variation is studied in chapter 7 of A.P.O, Annals, 
volume 4, pages 217-258. The discussion of this long series of 
careful measures gave conflicting results, hard to understand. Some- 
times it indicated increased solar constant with increase of contrast 
in brightness between center and limb of the sun. Indeed, for the 
results of 1913, the correlation coefficient was +0.601+0.067, and 
a change of 1 percent increase of the solar constant was accompanied 
by +17 in the arbitrary solar-constant number. But at some times 
even the sign of the correlation coefficient changed from plus to 
minus. So the hope that Dr. Langley had held before his death in 
1906 proved illusory. Solar-contrast observations did not yield an 
easy way to measure the variation of solar radiation. 

In May 1952, however, P. 4088 threw new light on this diffi- 
culty. We then knew of the family of harmonic periods in solar 
variation, Synthetic solar constant values computed from these 
periodic terms marched in close accord with observed values from 
1920 to 1951 (P. 3902). So good was this agreement that I com- 
puted the probable march of the solar constant from 1900 to 1920, 
the years before good determinations had been possible. For, as I 
have said, the Langley solar-constant method, though sound and 
fundamental, must always give values too high or too low if the 
transparency of the atmosphere changes during the several hours 
required to measure it. The synthetic solar-constant values (see 
fig. 50) were based on “the short method” which has no such draw- 
back, and besides gives several values of the solar constant on each 
day of observation, thus providing mean values. (See pp. 61, 62). 

Figure 50 shows that before 1920 there is no visible correlation 
between the observed and the synthetic Mount Wilson solar constant 
values. But figure 51, in its graphic comparison of synthetic solar 
constant with observed solar-contrast values, shows a fairly high 
degree of correlation. Increased contrast goes with increased solar 
constants. So if the A.P.O. was still in short-method operation as 


60 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


formerly, and in a good location, and with a tower telescope, perhaps 
Dr. Langley’s hope might be at least partly realized. 


13. FINAL EVIDENCE 


Notwithstanding the evidences contained in the references cited 
below some meteorologists may still be reluctant to accept forecasts 
many years in advance. In the absence of conclusive theoretical 
demonstration that the small percentage changes in solar radiation 
can cause changes of identical periods of many times larger per- 
centage in weather, and that these are hidden by phase changes from 
direct disclosure, they may still withhold belief. Therefore I present 
an additional observation which is so striking that some have con- 
sidered it conclusive. 

If it is true that the 273-month family of regular harmonic periods 
exists in weather, with such amplitudes that by their summation a 
controlling influence is exerted, then it follows that the weather 
should tend strongly to repeat its features at intervals of 22 years 
9 months. I showed such a tendency in the precipitation of Peoria, 
Ill., in 1934 by figure 33 of P. 3339, reproduced as figure 1 of 
P. 4095, 1952. But now I will present a much more telling evidence 
from the records of precipitation at Nashville, Tenn. 

Taking from our files the computations on Nashville prepared for 
P. 4390 in 1958, I lengthened my forecast for Nashville through 
1970. Considering only the 6 years 1965 through 1970, I looked 
back 22 years and 9 months to the interval April 1942 to March 
1948, 6 years. 

Figure 53 gives a graphical comparison of my forecast, from 1965 
through 1970, with the observed precipitation at Nashville from 
April 1942 through March 1948. The values plotted are, as stated 
in P. 4390, smoothed by 3-month consecutive means and are depar- 
tures from the normals given in table 9, P. 4390. I have computed 
the correlation coefficient for the 6 years between the two curves of 
figure 53, and also the correlation between the two curves of figure 2, 
page 3, of P. 4390, for the 6 years 1950 through 1955, all from 
Nashville precipitation. The two correlation coefficients are, respec- 
tively, +0.469+0.061, and +0.737+0.024. 

So the correlation coefficient between the direct forecast and the 
event, 1950 through 1955, is 30 times its probable error, and the 
correlation coefficient between the forecast, 1965 through 1970, and 
the observed precipitation at Nashville, April 1942 through March 
1948 (22 years 9 months previous) is 7.6 times its probable error. 


61 


SOLAR VARIATION AND WEATHER—ABBOT 


3 


NO. 


2 


o7eeetsssneeun55555 


*PIQAISGO UOS[IAA JUNOP, YIM UOI}LIIIION ON “6N6T OF OZ6T ‘P2}seoyoeq JUR}SUOD IeIOS—‘_g ‘DIZ 


BARRE See oS rei 
CAE ANZ eet NAN 
SAR eT am RRR 


Ov6'! 


96 


62 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Fic. 51—Solar constant (backcasted) versus solar contrast, 1913 to 1920. 


BricuTNess DistRIBUTION Aone Sun's DIAMETER 
For DIFFERENT CoLoRS 


InrRA-RED InrRA-RED - R BLUE-GREEN ULtRA-VIOLET 
Az 1.53% Az -986u : an.5OSn A372 


Fic. 52.—Solar contrast, center to solar limb, observed, 5 wavelengths. 


63 


SOLAR VARIATION AND WEATHER—ABBOT 


‘snojAaid syjyuoul ¢72 ‘8b61 4 Zr6l 


‘ 


PaaArasqo SNSJ9A “Y/6T 0} SOBT ‘SBI2I0 


:UOTEPdIIIId IT[IAYSeN—'SS “OY 


64 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Considering probable changes of an accidental character that may 
have happened in the long interval of 22 years 9 months, I do not 
think a fair-minded critic should object because the latter coefficient 
is smaller than the former. For, after all, the correlation coefficient 
for the interval 1942-1948 is almost 50 percent and nearly 8 times 
its probable error. 


CONCLUDING REMARK 


In the present paper I have summarized 13 aspects of the depend- 
ence of weather on solar variation. But without unduly expanding 
the paper, I could not give half of the evidence that supports these 
positions. My best course is to give in the Appendix references to 
many papers where, by tables, illustrations, and text, the evidence 
is amplified for those who may be interested. I would like to call 
attention particularly to Publications 4088, 4090, 4103, 4135, 4211, 
4213, 4222, 4352, 4390, 4462, and 4471, where much that I have not 
crowded into this paper will be found. I have suggested there one 
theoretical hint as to the nature of the connection between solar 
variation and weather (P. 4211, pp. 10-11). Also the atmospheric 
conditions that prevent discovery of the family of regular periods 
in weather by cursory tabulations are more fully explained. I am 
still hopeful that meteorologists in America will at length see that a 
useful measure of long-range forecasting, even to a half century 
in advance, can be attained by using the records of the past with due 
attention to atmospheric conditions. 


SOLAR VARIATION AND WEATHER—ABBOT 


NO. 3 


*Sp1OI91 [9AQ] SoyYe] Jwoasd) WOIJ pojoipaid s}ysno1p 
{AZo]OIOIjJOU [eII}S9I110} pue soiskyd Ivjos url satporpotsad VWI0G 


‘uol}eIpesl IejOS 
JO a[v9S PsJepuLys s}t pue JoJOWOeYyIAd MOY-Jeyem poadosdwy 


*}SB1JUOD Iej[Os pue jUL}sUOD IPOS 

‘azey pue SdAINO }1IG 

‘YSIP Telos Ssyj}susjaaeM ¢ ‘saAINI iIG 
‘ySIp Ivjos {As Jo SsaujYyStIq ‘ JojoWOUeIAT 


‘OI6L PUL SI6T UI UOHJaII09 Aq UMOYS syUOUT ae. 
pue OSe1 pue sAep Le Spollod ‘uoTpelieA Tejos pue uolyeyoI Telos 


£lc 
"sartojed ¢6'] JO JuR}sUOD 


JPJOS SIATS ‘UOOTTeG Jsoysty 0} [PAI] vas ‘sanyeA ysoySTY JO VAIND 
‘ainssoid ‘wiu ¢Q°Z +e 
SILIO]VI /Y"] JO anjeA spsoda4 [] A[N{ UMOY JazoutorjaysAd uooTeEg 
‘96° 93 O6'T IV ‘poyzow AajsueT Aq poureyqo sonjea 9 [IV 
uOOU 0} asiJuns WOJJ [Z pue QZ ya peArasqo jJuL}sUOD APIO 


‘UOTINAII}SIP JoJOUIeIP IvjOS ‘YJSuaToAeA PUL UOT}eIPeY 
*sJojoWOT[ayIAd Piepue}s I1}s-I3}eM Pue MOP-19}8\\ 
“JoyoWIOT]OYIAG YSIP-IVATIS 

‘poyjour Aa] Sue’y ‘Juejsuod svjos 10F ainpaso1d ajaydus0oy 


‘yuR}JSUOD IE[OS IOF poyjour AdjsueyT jo Asooy TF, 
q9alqus 


SS sss NN 


(3X9} 94} JO suONIes peioqunu dy} 0} 


LXAL FHL NI SLNAWALVLIS DNILYOddNS SNOILVDILSAANI TWNIOIWO AHL OL SHINAUIATA 
XIGCNdddV 


reeseeesesccessconseid “TIIAX ASHO 'AselAZ — BEST 
TeeeeseseT “ON “28 “JOA “T]OD “OSI, URIUOSuNUS = EST BIE 
dae ie 2A) I ‘dd 
Te eee eee eeiede nese cl()GZ, ‘d 
ca erbe Oy Do cp srs 2261 
pe ees aie EH 2 Te) ‘dd “p "JOA PY @ Ia ACA ‘ULYy 
Q ‘ON ‘99 “JOA “]]OD “OSI, URTUOSYNMS  BI6I 66h 
re ah tit eeeeeeeeeeserees og ST6r §8=—-_:« 197 
"Tb ON {CQ "JOA “]JOD “OSTA, URIUOSHIUS 
St NS Sesto | “d 
Ny 9 OS Ea OAS ‘dd 
ca aieoesasyeeyg ida E16L 
OE DIE OI Sst 4 ‘dd fo ‘JOA “O'a V ‘uUUYy 
SS eEuONEGs 429) SYR SRA TES Tey ‘dd 4 "JOA "Oy uuYy 3061 
aouasa fay avd ‘ON 
14%d 


Jajai spesowmu advypjoq ay]) 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


aqeras 

aie skep Aueul Jo suvour AJUO ‘MOT 00} IO YSIY 00} SanjeA 
Aep [enprarpul soars sXemye yUR}SUOD Ie[OS JOF poyjaut Ao[sueT 

‘San[eA jSe1}U0D 

IEJOS YPM UOHP]AII0D poos A]IeZ Jnq sanjea UOs]IAA JWHOPy 
YPM UOTL[IIIOD OU STRIARY “P2}seIYIe YZHT-SO6T WeISUOD IOS 
“Leol Pue pz6T ‘Ydess potied yyuoul-¢7Z 


‘POYLIOA 6P6T “99q Ul 91n} 
-eladuio} JO UINWIUIW YIOX MON pue UO}SUIYSeAA JO SUOT}IIPIIg 
‘poriod aejos Aep-g¢p9'9 ay} Jo soUNYUr 
ainjetaduia} 0} Ajyenbd puodsal YIoX MaN pue uUosuTYysSeAA 


‘0Z6I ‘OZ ‘1eW ‘dno18 yods 
-UNS }¥9IS JO SPIO {SuUIJO}S DjIUSeUI SuIpus}je UOT}ELIeA IeOS 


“SOUBITIINY ULIPUT JSAAA Surpus}ye uorzersea 1ejoS 


‘uns [eijUsd J¥sU T[NIOY UMID][VD Jo seaie 
Ul pue 2, IJeydsouor Ul OsTe 4INdD0 Spuet} yssoyT, “Jo}e 
-a1aY} SYIOM Z JOY oinjessduis} Jo saydieu aytsoddo sonpoid 
yoryM ‘uMop pue dn spuss} Jepnsaim sey uorerpes suns Os[Ty 


ra! 
I 


‘uoelieA Iejos Aq pounusojapoid J9yyeI AA 


“‘syyuOU ¢/7 X = sep 9¢¢9'9 ‘potsad jsz0Ys suns 


‘uoTyepdise1d UoSuryse A, Ul polled Aep-77 VW 


“AdeINIIV IOF Patedusod sanjea Uor}ye}S 
“SJUDWIOINSeIU JULSUOI-IeJOS 
4afqus 


I 8°62. ‘d 


ZL «ee AG ‘d 
bP °°°9 SY ‘OL ‘ON “ZIT 104 “T10D “OstJ UerUOSYZUS 


TT CON “TIT JOA “T]OD “OST, UETUOSYTUS 
Trees 8 9OT CONT TTT JOA “]]OD ‘OSI, WeTUOSYy}TWIS 
OL sid pur pay Osjuase peu, 


‘9 'ON ‘OTT “104 “TOD “Ost WerUOsYyTUIS 
OL TON “OLT TOA “TOD “ost URTMOSYTUS 


"ON “LOT ‘Toa “TOD “OSH URTUOSHATUS 


Q 
= 


Q TTS ON “POT “TOA “1109 “OSE, UeLUOSYATUIS 
OT EON “POT ‘TOA “TOD “OSH weItOsYIUIS 


fe er oats LISS 30 ‘d 
poe sno vite sicia esis Ss iT OG ‘dd ‘9 ‘JOA “Ory. ‘uuYy 


aIuasa fay 


panuijuoI—XIdNaddy 


cSél 


OS6T 
6r61 


8r6l 
8r6l 


£61 


bret 
br6l 


crol 
awq 


880¥ 


STOP 


066¢ 


Or6e 
9T6E 


£68E 


IZLE 
SOLE 


‘ON 
19%d 


SOLAR VARIATION AND WEATHER—ABBOT 


NO. 


‘joqqy ‘Ssulyseo910 
JayyeIM Isuv1-Zuo] JO poyjoul sjoqqy Ul pasn sinpsdo1d poyreyq L ‘Z ON ‘Z ‘JOA “Buy pue ‘19g ABi9uy Aejosg ‘usinof 
‘joqqy ‘uoTjelivA Iejos pue JoYyyeEoM L ‘LON ‘[ ‘JOA “Sug pue ‘19g Asiouy IejoS ‘usnof 

*IQAOOPT 
pue ‘yoipry ‘yoqqy ‘eu1a3g Aq JuR}sUOD AeJOs dy} UO siodeg S ‘"'¢ ‘ON ‘¢ ‘JOA ‘soIskudomsy ‘rJUOD UeTUOSY}IUIS 


SHONYYGATA AUVINAWAITdd NS 


‘9 ‘d ‘Jewiou 94} Wor} sainjzredap 


[[ews pue a3iv] JO} O6Ep “q Ul uorjezIdi991d Jo sysedaI10} JO 6 
Adeind9V JO suosTIedWIOD 99G "JsedIIIOJ JIN}JeIodui9} adue1-Suo0T L CCS CON “Ep “JOA “][OD ‘OST, WeruosyyUIS 
‘SUOTJVAIISGO d}]JO}eS WOT sjseo9I10J JoyyeoM Aep-gT |) ae aie “*Z ‘ON ‘Ep “JOA “][OD “ISI, UETUOSY}WIS 
‘uoryeyIdi901d “S*A) JO 4se¥ddI0} 9SueI-SUOT 8 °° '''°'6 ON ‘6ET ‘JOA “T[OD ‘osTPL Weruosy UWS 
“syyuoU O/2 
potsed Ja}seul JO yySus] JOeKO {SUljsedaIOy JoyJeom oSueI-SuoT Loo E CON ‘QET “OA “TOD “OSI, URTUOSy WS 
“Byep Isaydsouor Ul seiToIpolag OL °°" OT “ON “GET “TOA “T]0D “OSTT UeTUOSY}TUIS 
*syyuoul 
6€ 


e/g = SHOW Z “syJUOW E/z Jo souOWIeY Te “IT/T ‘Z/T 


‘9/1 ‘€/1 ‘$ Sporssdqns yyIM Yuad1ed ¢'9 — syyuou (C°7Z = sep 
Z1Z jo poriod es[nd ueuIn yy *syjuoUl ¢1Z 0} poyejas spollag OL Pic Ce MOM ICS ce Cory CoN rece rT aG irrtns Larmor. “erie io. . LI-SI ‘dd 
‘LON ‘pET “104 “][0D “OSHA, Weruosy}TUISG 


"SS6I-S68T ‘Aro}yeAIasqG JesIshydo1jsy 34} Jo suoljei1sdo Zurpeay Toc ECON ‘TET ‘JOA “T[oD ‘ost, WeruosyyUIS 
*SUOTJELICA Jejos ul SOT}IOIPOLIag TM ce LO ae rcpt arate g ae a eee ¢1-Z1 ‘dd 
‘bp “ON 8ZI [OA “JOD ‘OSH, WETUOSYIUS 
‘SJSBIZIOF JOy}eIM Iea‘-9O BCS CON ‘QZT JOA “TOD ‘dSTT UeIUOSyIUS 
‘S}SISO[OIODJIUI [VUOISSoJOId JO Surjsojyy Q "PON ‘ZZ ‘JOA “][OD ‘oS, WeruosyyTUS 
“Aueqiy 38 uorjeydioeid pue uorjeea rejosg BCS CON “TZ “JOA “TOD ‘OSI, WeIUOSYITIWIS 
‘Spi0daI J9Y}CIM UL S[TBUIIOU Y}IM SddUIII}.19JUT 9 "CTL ON ‘ZTT ‘TOA “T[OD “OST, WeruOsyyUIS 
q9alqQus auasafay 


8S6l 
LS6T 


£S61 


1961 
196T 
096T 


6S6T 
8S61 


9S61 
SS6T 


SS6l 
SS6I 
eS6T 
es6l 


cSél 
240q 


IZbb 
cobb 
06¢P 


cstv 
Seer 


S9¢b 
cccv 


cep 
ler 
selp 
fOr 


060¢ 
‘ON 
19%d 


SMITHSONIAN MISCELLANEOUS COLLECTIONS WOE. aleKey, (NOS Sy PES al 


Smithsonian Observatory, Mount Wilson, Calif., with solar-constant outfit and 


tower telescope. 


VOL. 146, NO. 3, PL. 2 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


“UIYJIM YIOM JULJSUOI-IVJOS IO a}[Oposy} pue yeysojI07, 
‘JoyomouRIAd pue ssojyowoyoyssd YIM JaAtOsqOG “Pye ‘ureyunoyp yayqey, ‘AIOPCATASGQ) UePTUOSYIWIS 


SMITHSONIAN MISCELLANEOUS COLLECTIONS VOLES 146; NOS; (PEs 


Aldrich observing with double water-flow electric compensation pyrheliometer. 


VOL. 146, NO. 3, PL. 4 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


‘JOJOWIOTOYIAd YSIP-JOATIS YPM SulAtosqo UeU OT 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NO. 4 


EVOLUTIONARY TRENDS IN THE 
AVIAN GENUS CLAMATOR — 


By 
HERBERT FRIEDMANN 


Director 
Los Angeles County Museum 


Noe: 
at <a 
wINGTON 


(Pusiication 4532) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
OCTOBER 30, 1964 


Hh t 
7 ¢ by A 

Oy hora os 
ria? ah iy : 


Pesenl 7 


rn 


Jn Mh 
i wid) 
Dea 
ws f 


il iy i) 
nA! 
vit i 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NO. 4 


EVOLUTIONARY TRENDS IN THE 
AVIAN GENUS CLAMATOR 


By 
HERBERT FRIEDMANN 


Director 
Los Angeles County Museum 


/} <V\s Ak 
METH Cots 
INGTON Se 


(Pusiication 4532) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
OCTOBER 30, 1964 


PORT CITY PRESS, INC. 
BALTIMORE, MD., U. S. A. 


CONTENTS 


Page 

PICO WCU RTICHES otc e cote veto ny sicrs. bisie'e's s Rielederahe-ale'y, auiecls Saelema ele mileieie 1 
TM ERO UCLIOM ck etry sretetet ne: ere clans oscls hevale cushetciatoie cusleietecveuvalsyers eae avolelalvsvele level eneie 2 
PE MlOmeMetiG. ERelAMOMSHIPIS: © é:</cr<)s «1 ieis aia. e/aie:0) lata g a selainle'e Was aie ei Uieavenie she's 6 
FPeaturesiof Brood, Parasitisny in: Clamatoes ..\. 4c class sot ve ssccies eels aisiele tes 11 
ost Selectionvand hts! Evolutionmsy. «ac sisal sieves stay steleler «5 sla slater nieyaiers 14 
Tterisiley Gin ParaSHisil: 260 oe Sake Swnle ew teal dtarsis sion a aeiekbne Oates 37 
Pugaare DEENA SIG’ cine ale Pe aici sialaia stearate’ > Spe oes leita eRe alain cimaiels atombace 50 
leu bakionigt Persindia tt: Nkec st. Mes aie siete aut Mes Gas ee tads 57 
Host-Parasite (Nestling, Relationship ns iicigis coisa cance dis ddan els 58 
Bledeline Beeding by: Adult) Clamatorsiy. fo. iis. sociales scceels 61 
Pimmage Variations and: Them Significance. 20.0220. os nade deakles 62 
Es AtaitEM TEMUIESEALY Sub te enest te wl echarssoro Ml whe iw: toca are ne Weave chat a cieeannty Le Jae 70 
iieratory Belaavieie iss. yes lst ctteats toa dois ac onie Halhlta Wiee aloe aelele wid 75 
Rite vate CONCIISIOTIS « o wiic-< ul seed «kbiesieie ew bu deiae Coane laa oes 92 
Anpendixc—Anditional ‘Host! Data .oeies te ese bole LW eae seed 96 
A. Data on the Hosts of Clamator glandarius............20000000- 96 

B. Data on Additional Hosts of Clamator jacobinus................ 103 

C. Data on Additional Hosts of Clamator levaillantit.............. 105 


TRADE; oe 18 Re Rone AER A he a We a RE Oa BN ae PA 106 


; } 
i At ; 
ry 
s) 
hi 
7 Ay } 
r i 
H 
I 
; 
¥ 4 } 
A 
¥ ¥ 
ee etn? 
» “ ‘ 
‘¥ 
nd ’ 
5 ie 
i : eee 
‘ ‘ ye Abe avy tf : 
A wi nk ely Tier ne Ye haiks 
Sa eA PRY On, a 
: vies , pee y 
ee ee es ee i . re 
4 ree 44 ‘ ’ oD 4 
uJ ie i4 ; : 
} Gy PEL OVER RRA OY TE eb Ae VAD Cee me 
Eas tak ray vi Rit AP ae f ‘ a ; re mn 
DO eS Td RAY Wise » Rank? ; 
f ' A 7 i i ] 
: ee oe ee oe PET ea Ow ee Do To iy 
NiAWwe Gi Oy th Q Wie : ‘ 
: ite. Cpe VK tae ne H 
Wee a ee sf 
Seve eh eee eee) eee @ pews 
} a ot ug re i 
Ae ee asi weg amt ¢ 
P, yrs gt wl 
Peee bes ene ee Pak den 
Ay eh po de pee i) 
’ Harp! bole CUA ee cea ey ee ce Re tee bar tale e eee le Be 
1 en ie Pf ae A : ; ihe: 
Ry » ee Ue Cd 
i 


i | 1 " : > ay , , H i j . ; ' PT * 
; 8 ee ee el ee . AG aL ¢ j i, bd . . 
Oe ie a) mee ze ke es ee ies! nm’ i! wh : =e Mi gh aH vai se 
te ae oy Pa a Shite Be dlalsi Sule a f, i i , 1 2 (+ Ae ee 
' eee h We Se 107, Oh 
y 5 : it \ i 4 \ f (de 7 "4 
ER Rie PS woe ee ae * : pj. Uaheae , 
5 e, f WA bugs oP et 
‘Al te 1 al 
' Ah diet! 9) Boe Pins 


oe ti) 
Ua en Cy 


. 
7 Shy ic ly a ey y Nive 

aa a a Ye ae ere 0 ee ee ek es 

, i. i Ree yt Ve ee aT f 


an 


EVOLUTIGNARY TRENDS) IN) THE 
AVIAN GENUS CLAMATOR 


By HERBERT FRIEDMANN 


Director, Los Angeles County Museum 


ACKNOWLEDGMENTS 


To APPRAISE as accurately as possible the involved situations that 
exist in the species of Clamator, it was necessary to examine care- 
fully large segments of the preserved material. Museum study skins 
were inspected to evaluate the nature, frequency, and distribution 
of the plumage phases, and the kinds and degrees of variation within 
these phases for possible suggestive clues as to their nature. The 
changes of plumages in all the included species were reviewed for 
possible phylogenetic hints they might reveal. And the eggs of the 
cuckoos and of their hosts were examined to determine the extent of 
adaptive similarity, or the lack of it, thus avoiding undue influence by 
earlier published opinions, some of which, as suspected, turned out 
to be casual and rather uncritical estimates, or were based on geo- 
graphic segments of the total picture. 

A research grant from the Frank M. Chapman Memorial Fund 
of the American Museum of Natural History enabled me to spend 
3 productive weeks at the British Museum (Natural History) in Lon- 
don, where by far the largest assemblage of the pertinent material 
is stored, and also to spend a few days at the United States National 
Museum in Washington. To the custodians of these bird collections 
I express my thanks for their help. 

By loan of actual specimens and by correspondence from coopera- 
tive curators of their respective collections, I have been able to 
tabulate the data on material of special interest in the museums 
of Bloemfontein, Bulawayo, Cape Town, Dundo, Durban, East 
London, Khartoum, King William’s Town, Lourenco Marques, 
Nairobi, Pietermaritzburg, Port Elizabeth, Pretoria, and Salisbury, 
in Africa; of Bombay in Asia; of Basle, Berlin, Copenhagen, Genoa, 
Madrid, Milan, Paris, Stockholm, Turin, and Vienna, in Europe ; and 


SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 146, NO. 4 


2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


of Ann Arbor, Baton Rouge, Berkeley, Cambridge, Chicago, Los 
Angeles, New Haven, New York, Philadelphia, Pittsburgh, and 
Princeton, in the United States. To the officials of all these museums, 
who have kindly sent me data or specimens, I hereby tender my thanks. 
The following individuals also have helped with observational notes 
and egg-specimen data: Salim A. Ali, M. Courtenay-Latimer, G. 
Duve, R. Kreuger, D. W. Lamm, R. Liversidge, G. R. Mountfort, 
J. Ottow, C. R. S. Pitman, C. G. Sibley, C. J. Skead, G. Symons, 
V.G. L. van Someren, J. G. Williams, and J. M. Winterbottom. 

My personal field experience with Clamator, limited to the three 
species that occur in Africa, continued to play a contributing role in 
the present as in earlier studies. Acknowledgments, therefore, for 
the support that made these field studies possible are due again to 
the National Research Council, American Philosophical Society, the 
Guggenheim Foundation, and the Smithsonian Institution. 


INTRODUCTION 


The genus Clamator is one of a number of genera of cuckoos which 
are parasitic in their mode of reproduction. The vicissitudes of its 
biological history which are reviewed in this paper are of interest in 
clarifying some concepts involved in the overall problems of evolution 
of habits in the puzzling family Cuculidae. 

Brood parasitism in the cuckoos is not a “‘single-line’’ development, 
as it is in the cowbirds, the honeyguides, or the ducks, but comprises 
many genera, some of which have evolved specialized features, such 
as the evicting habit in the nestling stage. Other genera have de- 
veloped elaborate egg morphism with related host-specific gentes, 
some have an extremely restricted range of host species, others have 
a broader choice of fosterers, while still others show none of these 
refinements in their mode of reproduction. In the weaverbirds it is 
known that the parasitic habit has arisen independently in two sections 
of the family (Friedmann, 1960). Whether it arose equally inde- 
pendently in different groups of genera of parasitic cuckoos is still 
uncertain, but it has developed in various ways in the 18 genera that 
are parasites in their mode of reproduction. 

These genera, each with its own special features and its own 
peculiar problems, are of much interest to the student of adaptation 
and evolution. C. D. Darlington (1953, pp. 441-443) wrote of the 
European cuckoo, Cuculus canorus, that it was “. . . uniquely in- 
structive in its relations with the environment. Exposed from hatch- 
ing to an alien environment for innumerable generations, the be- 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 3 


haviour of the cuckoo, the instincts of the cuckoo, are determined by 
its heredity. Its migration and all its malpractices follow the pattern 
of a parent it has never seen. 

“Further, of all animals, the female cuckoo most resembles man in 
exercising a choice which fixes her offsprings’ environment . . . the 
cuckoo species is divided into mating groups which specialize in 
laying eggs in the nests of birds whose eggs theirs most closely 
resemble. The cuckoo species, like any large human community, thus 
has a spurious plasticity which derives from its variability. This 
variability, again like that of a human community, is preserved by 
natural selection, that is, by the adaptive value of a whole range of 
genetic types. The cuckoo is thus the most significant of all birds for 
the theory of heredity and environment .. .” 

Without detracting from Darlington’s estimate of the inherent 
philosophical importance of the biology of Cuculus canorus, it may be 
suggested that the present account of a group of that bird’s less 
completely specialized relatives may even enhance it by presenting 
informative perspectives and tangential views into our total concept 
of brood parasitism in this family. There is a real need for this, 
since, in spite of the known differences in the mode of parasitism in 
the various genera of parasitic cuckoos, the literature of the subject 
is devoted largely to that one species, which, it so happens, is the 
most highly evolved and specialized of all the members of its family 
and possesses many features not present in other parasitic cuckoos. 
This has resulted in an overly accented, rather one-sided emphasis in 
the usual presentations and discussions of the subject. It is hoped 
that the present study may help to correct this and to offset some 
of the literature on cuckoo parasitism. 

At the same time, the situation present in the four species of 
Clamator is, in itself, well worthy of study as a survey of the evolu- 
tionary history of a compact and relatively isolated genus of the 
family, only distantly related to Cuculus. The genus Clamator gen- 
erally is considered fairly primitive; however, its included species 
reveal much adaptive evolution and the effects of diverse and not 
altogether harmonious trends. Not only is it a primitive group of 
highly specialized species but also one that reveals to a greater de- 
gree than most that evolution may proceed at different rates in 
different characters and in different species even within a small genus, 
and that some of these trends may even be abandoned after a state 
of high perfection had been achieved. In these respects this study 
differs from most. 


4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


The majority of studies of evolution within limited groups of 
animals emphasize single characters or single aspects, such as external 
morphology, relatively minor changes in size, shape, or proportions, 
or adaptation of one part or one structure to changing habits. Also, 
the stress has been placed on characters that seem to have gone the 
whole way from a generalized to a highly specialized condition. This 
emphasis is understandable—it is most convincing to be able to re- 
construct historically the path or paths followed by piecing together 
carefully and critically all the available data. This procedure, how- 
ever, has tended to conceal, or at least to detract attention from, the 
fact that many organisms have evolved only “part way,’ and still 
have managed to survive and to succeed. This is, of course, generally 
implied or assumed in the stage elements of all more complete de- 
velopments, but it is well to underscore it where, as in Clamator, 
some of the species have stopped at “part-way”’ stages. 

The four species of the genus Clamator form a compact group 
that has been considered by Jourdain and Baker and other writers on 
parasitic cuckoos as one in which adaptive evolution in egg similarity 
to those of its usual hosts has gone as far as in any group of brood 
parasites. Yet, two of the four species have geographic segments 
(populations or races) that either never arrived at, or else appear to 
have “ignored” or to have “repudiated,” the results of the adaptive 
evolution of their respective stocks, and this situation has been arrived 
at in very different ways in the two. 

Thus, in the case of the jacobin cuckoo, Clamator jacobinus, we 
have a species which, throughout its extensive Asiatic and part of its 
African range, is parasitic chiefly on babbling thrushes, most of which 
lay bluish eggs. In Asia and in northeastern Africa the eggs of 
the jacobin are always similarly bluish or blue-green in color, but 
in most areas south of the Sahara the resident jacobins, using some 
of the same type of hosts, but more frequently, bulbuls and shrikes, 
lay only pure white eggs, which contrast strikingly in appearance 
with those of their victims. 

Turning to the great-spotted cuckoo, Clamator glandarius, we 
find that this species lays but one type of egg throughout its range. 
In the Iberian peninsula and adjacent parts of northwest Africa, it is 
almost exclusively parasitic on magpies, with the eggs of which its 
own show extreme similarity. So great, indeed, is the resemblance, 
that it has been cited frequently as an example of “perfected” adaptive 
evolution, and some not uncritical collectors have had the experience 
of collecting sets of eggs containing both species without realizing 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 5 


this until later. However, in Egypt and in most of Africa south of 
the Sahara where there are no magpies, this cuckoo parasitizes crows 
of several species, and also it lays frequently in the dark, hole nests 
of starlings. Its eggs show little resemblance to those of these hosts. 

If it could be demonstrated that the egg type of Clamator glandarius 
had evolved as an adaptation toward the use of magpies as hosts (it 
is the only species of Clamator laying speckled eggs, which egg type 
in cuckoos generally is considered an “advancement” from the more 
primitive unmarked eggs), then it would follow that the geographic 
spread of this cuckoo to areas where there are no magpies would 
appear to be a matter involving something akin to a “repudiation” 
of the specialization it had achieved earlier through natural selection 
with the magpie as the effective agent. This, if established, would 
open a rare opportunity to study the biology of a highly adapted 
species in a new environment where this adaptive excellence no longer 
is a special advantage, but where it is apparently no critical en- 
cumbrance with new and nonadapted hosts. 

In attempting to trace the course of the evolution of a group of 
organisms, or of a habit and its correlated morphological characters, 
it is a common experience to find that the trend generally is toward a 
more and more perfected stage of adaptation, eventually reaching a 
degree of perfection beyond which it cannot, or at least does not, 
go. From the general to the specialized, from the “good enough to 
survive” to the obviously advantageously adapted, seems to be the 
history of case after case. What is unusual is to find a highly adapted 
evolutionary product apparently departing from the particular set of 
conditions which its past history appears to have been concerned in 
meeting more effectively, and carrying with it in its secondary path the 
primary adaptations no longer needed or especially advantageous to it. 

On the other hand, if it should seem more likely that the great- 
spotted cuckoo developed its speckled-egg type south of the Sahara, 
and subsequently spread to Mediterranean areas, where its egg 
happened to “fit” so well with those of a new host, this would have 
to be considered as a most unusual instance of preadaptation. It 
should be kept in mind, however, that the known facts concerning 
host egg similarity, or mimicry, in parasitic cuckoos generally cannot 
be explained satisfactorily on the basis of any assumed preadapta- 
tions, but, on the contrary, indicate the degree to which real adapta- 
tions in egg coloration have been evolved. 

The situation in the jacobin cuckoo is just the opposite. The seem- 
ingly similar success of the white-egg laying population with that 


6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


of the blue-egg layers raises the problem of the efficiency as selective 
agents of the relatively uniformly blue-egg laying hosts. 

Inasmuch as host adaptation is an important part of the biology 
of brood parasitism, the picture in Clamator commends itself for 
careful study and interpretation. This is attempted in the present 
report. Still other biological problems, similarly apparently arrested 
in their development as “‘part-way”’ stages in the species of Clamator, 
concern development of plumage morphisms and of migratory be- 
havior. Thus, in two of the four species we find geographically de- 
limited melanistic plumage phases, more restricted in range in one 
than in the other, but in neither has the black morph replaced, or 
achieved reproductive isolation from, the pale, or normal, morph. 
Also, all four species are migratory in parts of their total respective 
ranges and not in other parts. The extent of migratory movement 
within a single species varies from none at all to thousands of 
miles. Geographic segments of each, not necessarily even subspe- 
cifically distinct, differ markedly from other conspecific segments in 
this important trait. These are also discussed with all available 
evidence in the following pages. 

The four species of crested cuckoos comprising the genus Clamator 
form a group of birds that still reveal much—that in other groups 
has been concealed—in their continued progress toward greater adap- 
tive excellence. 


PHYLOGENETIC RELATIONSHIPS 


The genus Clamator contains four species of crested cuckoos of 
Africa, Asia, and parts of Mediterranean Europe (fig. 1)—jacobinus, 
levaillantii, coromandus, and glandarius. It forms a natural, easily 
recognized group, characterized by a well-developed occipital crest of 
elongated feathers, by a transilient mode of remigial molt, and by 
the nares in the form of linear ovals. It agrees with the subfamily 
Cuculinae in being parasitic in its breeding, but lacks the evicting 
behavior pattern in its young. It agrees with the Cuculinae in most 
other characters, but varies from that group in the direction of the 
Phaenicophaeinae in having only 13 cervical vertebrae (14 in the 
Cuculinae and in the other subfamilies of cuckoos), and in having 
the muscle formula “ABXYAm” (Berger, 1960). No one has 
proposed merging it with any other genera, and practically all of its 
recent investigators (Stuart Baker, Berger, Friedmann, Jourdain, 
Peters, the Stresemanns, etc.) have generally agreed that it is a 
primitive group in its particular subfamily, the Cuculinae. This is 


NO. 4 


AVIAN GENUS CLAMATOR—FRIEDMANN 


Fic. 1—Geographic range of Clamator. 


8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


the subfamily that not only contains most of the genera of parasitic 
cuckoos (all but the two Neotropical Tapera and Dromococcys), 
but also contains no cuckoos that are not parasitic, as do the other 
five subfamilies. In their recent study of the molting patterns of the 
cuckoos, the Stresemanns (1961) have pointed out also the primitive 
nature of Clamator’s transilient mode of remigial molt, and have 
mentioned the absence of evicting behavior in its young as another 
evidence of primitiveness. Peters (1940) seems to have been so 
convinced of the primitive nature of the genus Clamator that he 
actually placed it at the very beginning of his list of all the members 
of the family. 

While I also conclude that the crested cuckoos are to be looked 
upon as among the primitive, oldest sections of the subfamily 
Cuculinae, I doubt that this subfamily may justifiably be placed at 
the base of the whole family. Inasmuch as nest-building, incubation 
of eggs, and care of young are features of reproductive activity in 
practically all groups of birds, it seems likely that the most primitive 
cuckoos were nonparasitic as well. From this it follows that a sub- 
family made up wholly of brood parasites could not be the most 
primitive section of a family that contains many self-breeding genera 
and species. 

The age of the genus Clamator is, of course, unknown, but some 
suggestive evidence points to its being not later than Pliocene in 
origin. This is an inference based on the fact that although the genus 
occurs over a wide area in Africa and in southern Asia, it is absent 
in the Malagasy Republic (formerly Madagascar). In his study of the 
history of the African terrestrial fauna, Lonnberg (1929) concluded 
that Pliocene faunal transfers between southern Asia and Africa gen- 
erally are absent from the Malagasy Republic regardless of the extent 
of their range in either of the two continents. The fact that Clamator 
does not occur in Malagasy suggests that the spread of the genus prob- 
ably took place during, or subsequent to, the Pliocene, at which time 
Malagasy became completely isolated as an oceanic island. As a result 
of the present study, it appears that the southern African population of 
C. jacobinus is the oldest, most primitive of existing Clamator stocks, 
and it seems that its species spread throughout much of Africa and 
thence to Asia. The fact that in southern Asia this stock gave rise 
to a more involved evolutionary development than in Africa, and 
eventually produced so different a bird as C. coromandus, which, in 
turn, seems a stage on the phylogenetic road that culminated in C. 
glandarius, suggests a very considerable antiquity for the genus in 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 9 


Asia. The need for sufficient time makes a Pliocene spread more 
probable than a post-Pliocene infiltration, presuming, of course, a 
Pliocene or pre-Pliocene origin of the group stock in southeastern 
Africa. 

The fossil record of the Cuculidae gives no pertinent data. The 
family is known from the Oligocene of France (Dynamopterus 
velox), from the middle Miocene (Necrornis, only questionably a 
cuckoo), and from numerous Pleistocene remains too recent to be 
of any use in reconstructing the history of the family (Coua, Cuculus, 
Geococcyx, Coccyzus, Crotophaga, Tapera, and Pyrrhococcyx), but 
even there nothing close to Clamator. 

It is not feasible to say, or even to guess, from what stock Clamator 
may have evolved, as there are no living cuckoos that seem likely 
ancestral forms. Yet I cannot put down the vague thought that some- 
thing like the Phaenicophaeinae in Asia, or like Ceuthmochares in 
Africa today, may be closer to—less subsequently specialized and 
hence less deviated from the original—primordial stock of the family, 
and to this extent may be looked upon as existing representations of 
the ancestral group that gave rise to Clamator. 

Recent studies by Berger (1960, p. 82), especially his myological 
dissections, coupled with his familiarity with what had been written 
of the breeding habits, parasitic or otherwise, of the genera of 
cuckoos, led him to write as follows “. . . It would appear that one 
must discount either myological data or breeding behavior in deciding 
the relationships among the cuckoos . . . Thus, if we are to place 
any value on morphological characters, we must assume either that 
parasitism has developed independently as many as four times in 
this one family (which seems highly unlikely) or that the parasitic 
habit (or tendency for it) developed in the primitive cuckoos (all 
ABXYAm) .. .” Similarly, Darlington (1957, p. 273) concluded 
from a study of the geographic distribution of the cuckoos as a 
whole, that the family is probably ancient and had a “. . . complex, 
undecipherable history.” 

Although there is fairly good agreement among students that 
Clamator is a primitive genus, there is no such concurrence as to 
what other living genera it is closest in its phylogenetic relations. The 
Stresemanns (1961, p. 328) concluded that Clamator was only 
distantly related to the other Cuculinae. Many years ago, Sharpe 
(1872, p. 68) suggested it was somewhat similar to Eudynamis, 
but this is not substantiated by Berger’s anatomical findings (1960). 
He noted that there were two basic types of cuculine muscle formulae, 


Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


and that Clamator was of one type, along with such genera as Cuculus, 
Chrysococcyx, Surniculus, Tapera, and others, while Eudynamis 
was of the other, along with Scythrops and Dromococcyx. It must 
be admitted that it is not wholly clear how significant this myological 
character may be, as each group contains genera that seem only 
distantly interrelated. Thus, in the former aggregate, Tapera is very 
different from the rest, and in the latter group all three genera are 
widely separated. 

Berger’s anatomical studies, together with earlier work by Beddard, 
Forbes, Ftrbringer, and others, give us our best evidence of relation- 
ships within the family. All the characters, osteological, myological, 
and even ecdysial, have one thing in common—they are all of suf- 
ficiently nonfunctional nature as to make them seem relatively re- 
moved from the effects of selection. Hence they may be looked upon 
as phylogenetically conservative, and, to that degree, they are reliable 
indices of relationship. Breeding habits, parasitic or otherwise, are 
more amenable to change. In fact, one of the safest deductions that 
may be made from a study of brood parasitism is that in all the 
groups in which it occurs it is a secondary situation that arose in 
stocks that were originally self-breeding. 

Inasmuch as all the members of the Cuculinae are parasitic, it 
would seem that brood parasitism had already become established in 
their common, remote, ancestral stock before they became dif- 
ferentiated into the genera as we know them today. This differentia- 
tion has resulted in a wide variety of end products, some 16 genera 
with 46 species according to Peters’ list (1940), which suggests a 
long period for its operation. This, in turn, indicates a great antiquity 
of brood parasitism in the group, an antiquity that the history of 
Clamator suggests must date from pre-Pliocene or not later than 
Pliocene time. 

In studying the genus Clamator we are fortunate in that consid- 
erable information is available on the life histories of each of its four 
species. The entire group has been considered by Baker and by 
Jourdain, two of the principal students of cuckoos’ eggs, as one in 
which adaptive evolution in egg similarity to those of its usual hosts 
has progressed as far and as successfully as in any genus of cuckoos. 
Clamator is, therefore, a primitive group of highly evolved species, 
a biological situation that is not infrequent despite its seemingly 
paradoxical nature. As Baker (1942, p. 3) put it, “. . . perfection 
or completeness in adaptation or evolution must depend upon time 

. . and therefore the most perfectly evolved egg need not and does 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN II 


not belong to the most perfectly advanced Cuckoo. The more primitive 
forms of parasitic Cuckoo, such as members of the genus Clamator, 
containing the Great Spotted Cuckoo, have probably had an infinitely 
longer existence in their present form and condition than such beauti- 
fully perfected forms as our Common Cuckoo, and we should there- 
fore expect primitive Cuckoos to have acquired a more perfect 
adaptation in their eggs than those Cuckoos more highly developed.” 

The phylogenetic relationships of the species of Clamator, as sug- 
gested by all the data brought together in this paper, ethological, 
morphological (chiefly plumage coloration and eggshell pattern), and 
distributional, as shown in the diagram (fig. 2), reveal that jacobinus 


A. glandarius 


1, jacobinus-7” 


. 
. 
. 
. 
s 


“s2, levaillantii 
Fic. 2.—Relationships within the genus Clamator. 


is the most primitive member, and that from it two lines of descent 
bifurcated. One, rather short one, led to levaillantii; the other longer 
one led to coromandus and from this to the “climax” species, glan- 
darius. The geographic movements undergone by Clamator during its 
differentiation and dispersal are shown in figure 3. 


FEATURES OF BROOD PARASITISM IN CLAMATOR 


The genus Clamator evolved from an earlier stock that was already 
parasitic, as is indicated by the fact that all of its species are parasitic. 
It is understandable, therefore, that a comparative survey of their 
habits affords no clues as to the origin of this mode of reproduction, 
although it does reveal much of the course of the development 
parasitism underwent in this particular genus. 

Compared with a highly specialized group, such as the species of 
Cuculus, Clamator is relatively simpler and shows none of the de- 
velopment of infraspecific gentes, each with its elaborate, adaptive egg 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


12 


jo [estodsiq “¢ 


weer pen IVvoyeerees 'r eee g 2 vie pwn J FV [evevuctp Fo UY Vv Dip Meu AY =e } 


“yunyjwaa] *) Jo jesiedsiq *Z ‘eIpuy 0} ‘pue BIIFY JO SOUL OF LoIIFy yseayynos woIy suuqosvl *y yo jessadsiq *] 
"40jDWUD]) JO Tesiadsip AreuoNjoAy—e¢ “o1Ly 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 13 


morphism, and none of the instinctive eviction of nestmates by the 
newly hatched young, so characteristic of some species of Cuculus. 
Also, compared with the latter genus, all four of the species of 
Clamator are far more prone to deposit multiple eggs in individual 
host nests, either multiple eggs by the same cuckoo or multiple use 
of the same nest by several cuckoos. These differences suggest that 
the refinements shown in Cuculus are evolutionary advances over 
the basic cuculine stock and that Clamator is nearer to the original, 
common ancestor in these matters. This lack of conformity or control 
in egg deposition led Mountfort (1958, pp. 54-56) to conclude that 
“, . the parasitic behaviour of the Great Spotted Cuckoo is in many 
respects more complicated than that of our familiar Cuckoo, in which 
the single nestling merely evicts its foster-brothers from the nest .. .” 
It seems truer to say that the “uncomplicated” behavior of Cuculus 
canorus is a result of much adaptive evolution, whereas the “com- 
plicated” picture in Clamator glandarius still retains much of the 
simpler, less-developed features of the basic primitive parasitic cuckoo 
that we may postulate as the remote source of both genera. In my 
introductory statement I mention that the entire literature and think- 
ing about cuckoos is overly dominated by Cuculus canorus. If specific 
evidence were needed to demonstrate this, Mountfort’s statement 
about Clamator glandarius would be a case in point. 

All the species of Clamator have the habit, common to so many 
parasitic cuckoos, of removing one or more of the host eggs from 
the nests of their victims either before or after laying in them. 
Occasionally this does not take place, and in some nests some of the 
host eggs are dented (equivalent, in survival terms, to destroyed) 
by the beak or the claws of the adult parasite. There are ample 
observational data on this in three of the species—jacobinus, coro- 
mandus, and glandarius. The lack of such records for levaillantti is 
not significant. There is no need to repeat here these observations as 
they are already on record in my earlier (1949a, 1956) accounts, 
and in that of Stuart Baker (1942). Particular mention may be 
made, however, of Mountfort’s observations (Mountfort, 1958, pp. 
54-56; Mountfort and Ferguson-Lees, 1961, pp. 98-99), one of the 
few detailed recent contributions on this habit in C. glandarius. They 
marked eggs with indelible ink so as to be able to identify them 
individually on consecutive days; they found that when the hen 
cuckoos laid they removed one and sometimes two of the magpie’s 
eggs. In no instance did they remove eggs laid by other great-spotted 
cuckoos, which raises the question as to whether they could recognize 


I4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


the small differences in the eggs. Mountfort and Ferguson-Lees 
found that as many as three cuckoos laid in one nest. 

The various topics of interest in the brood parasitism of Clamator 
are discussed in detail below (pages 14 to 62). 


HOST SELECTION AND ITS EVOLUTION 


A study of the four species of Clamator yields considerable data 
relevant to the evolutionary changes that formed their present host 
preferences. Not only are the hosts of each fairly definitely restricted 
in kind, but two of the four parasites show unmistakable signs of 
changes in their selection of favored fosterers. To this extent they 
afford glimpses of the past development of their host orientation, a 
basic part of their parasitic mode of reproduction. The two that 
show these signs of evolutionary change are C. jacobinus and C. 
glandarius, the most primitive and the most advanced members of the 
genus. In both species, the change took place together with extensive 
geographic expansion of their ranges. Furthermore, on the basis of 
the total survey it is possible to sense the course of host selection in 
the other two Clamators as well. 


Clamator jacobinus (figs. 4, 5, pp. 15, 16) 


The presumed ancestral home of the pied cuckoo, C. jacobinus, 
is in southeastern Africa, the area now inhabited by the race serratus. 
In this region, ranging from Cape Province, Natal, Transvaal, 
Orange Free State, parts of South-West Africa, and Bechuanaland, 
to Southern and Northern Rhodesia and Nyasaland, the cuckoo has 
been found to lay its eggs in the nests of 22 species, but 13 of these 
have been recorded as hosts but once, and 2 others but twice. The 
only species definitely known to be frequent and regular fosterers are 
four species of bulbuls of the genus Pycnonotus (nigricans, barbatus, 
capensis, and importunus) and two shrikes, Lanius collaris and 
Telophorus zeylonus. These 6 hosts account for 101 of the total 123 
cases of parasitism by serratus known to me. 

When the pied cuckoo began its geographic expansion, giving rise 
to the race pica in equatorial and northeastern Africa and in India, 
the population inhabiting these new areas turned from bulbuls and 
shrikes to babblers as their chief fosterers. The race pica has been 
found to parasitize some 36 species of birds, half of which have been 
so recorded but a single time. No fewer than 26 of the known hosts 
are babblers, and all the hosts for which there are 5 or more records 
are species of this group. More than four-fifths of all instances of 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 


a 
SJ Q 
/ i) 
s a i aw 
4 ic 
Us ry ne 
y Suni 


. 


‘. 
sess, ' 
J 
oe 
Day 
=) 
cre | a 
1? 
¢ 
é 
? 
Falta POT tare 
4 
<a eee ee 57S, 


. 
Le 


Fic. 4.—African range of Clamator jacobinus. 
egg record. B, Blue egg record. O, Egg of unrecorded color. 


W, White 


16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


OHOO000O0 
POOOO000NC 
OO000000000 
OOOOO00000000 
SOO0000000000009 
* POOOOO000000000090 
OBWSO00000000000000000) 
leleyelelelelelelelelelelelelelejejejeso' 
OOOOO00000000CO00000 
(ofolelelelefelelelelelelejelelejlesey” 
O0000000000009000 


lolelololelelelelelelejelejelelolr ADOOCOOOY 
OOO0000000000000 Q0000009 
O00000000000004 JOOOOOCOG? 
lelelelelelelelelelerelelese} | QO0000G 
lefelelelolelelelelelelelelo Poope 
OOOO0NOCO00O0HO¥g OG 
DOOOONONOCNCOOOS [ele} 

pel elelejeleloleleleleleleleleye) JODOY 


OGOOOOD0ONGOOODG000000SYG 
felelolelelelelelelelesey= vee 


Od90000000 

O000000000 
Seietelersselele’e) 
fe] 


® 
(©) 
< ov 
ou OD 
- ¢€ 
m2 
el = 
— — Ss 
Deen s 
———— bass oe 1 2s 
= 
SS——————SSSSS=aBaBananan=_ ed 2S 
————SSSSSS=s oS a — nn 
SanmnaSSSSSS55>>5—«¢~0*——_ = Bs = Es | 
——SSSSSSSS=== = eos 
————— is ete 
re oo oO DS 2 
——— otp = = & 
+———4 oun 5 i= ro) Oo 
== Mime 
es Go OLD 
uO OG 
SSS =a coc 
=e == 000 
=> gos 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 17 


parasitism by pica known to me refer to 7 species of babblers (94 out 
of a total of 106 nests) of the genera Garrulax (moniligerus, 
pectoralis, delessertt, and erythrocephalus) and Turdoides (caudatus, 
striatus, and affinis). 

It may be stressed that inasmuch as several species of bulbuls, 
shrikes, and babblers occur as breeding birds in considerable num- 
bers throughout the ranges of both races of the pied cuckoo, the 
difference in host selection is not something imposed on the parasites. 
In Kenya the race pica has been noted by van Someren (in litt.) to 
lay fairly frequently in the nests of a bulbul, Pycnonotus barbatus, 
(3 records), but in India there is but a single instance of a bulbul 
nest used by the parasite. Inasmuch as Kenya was an early area of 
invasion in the course of the northward spread of C. jacobinus, this 
tendency there to use bulbuls as fosterers may have been established 
very early prior to the general shift to babblers. As discussed more 
fully elsewhere (pp. 51-53) the egg of pica is greenish blue whereas 
that of serratus is pure white. 

In India, where the pied cuckoo has been studied extensively, Baker 
(1942, p. 82) had no records of its eggs in central or southern India 
from any nests other than those of babblers, though a few such had 
been reported by others. It so happens that this cuckoo, after becom- 
ing well established in India, began to expand its range northward 
into the foothills of the Himalayas. Baker noted that “when we 
come to the hills . . . we find the . . . Pied Crested Cuckoo placing 
their eggs in a great range of birds’ nests, though in most cases these 
are nests of the Larger Laughing-Thrushes, nearly all laying blue 
eggs with which the eggs of the Cuckoos do not contrast. The normal 
fosterers here are undoubtedly the Necklaced and Black-gorgeted 
Laughing-Thrushes in the Eastern Himalayas and the Striated Laugh- 
ing-Thrush in the Western. Of eggs laid in these nests I have 49, 
while I have 42 deposited in the nests of twelve other Laughing- 
Thrushes . . . With the exception of the species mentioned .. . I do 
not believe any of the others could be considered normal fosterers, 
while even these three can only be considered normal because they 
have been selected as such by birds breeding outside their own normal 
Phin: area) vss” 

In Africa, when C. jacobinus stock gave rise to what has evolved 
into C. levaillantii, that new group, even more than the more northern 
jacobinus (now pica), became attached to babblers in its brood para- 
sitism. In these birds, C. levaillantii found an abundant supply of 
fosterers and in their use found an escape from competition with its 


18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


long-established ancestral relative. To this day C. levaillanti, as we 
shall see, is parasitic chiefly on babblers. In India, when the C. 
jacobinus stock gave rise, apparently in the northern part of its range 
in the foothills of the Himalayas, to what has become C. coromandus, 
the trend shown in the relatively recent use of laughing-thrushes by 
northern jacobinus became well established in the new group. We 
find today that laughing-thrushes, chiefly of the genus Garrulax are 
the mainstay of C. coromandus. 

The tabular list of the known fosterers for each of the three races 
of the pied cuckoo shows the remarkable degree to which the hosts of 
the African serratus and the Indian pica and jacobinus differ. The 
picture is not as clear in the case of the African population of pica, 
as data on it are still rather sparse, and involve only five species of 
hosts—a bulbul, Pycnonotus barbatus, three babblers of the genus 
Turdoides (fulvus, rubiginosus, and leucopygia), and a shrike, Telo- 
phorus zeylonus. It may be expected that more extensive observations 
will add further Timaliine fosterers to this list, as there is no reason 
to suppose that African pica differ from Asiatic ones in their type of 
host choice. 


HOSTS OF CLAMATOR JACOBINUS 


Host serratus jacobinus pica 
Stigmatopelia senegalensis aequatorialis.............. 3g 
Centropus#erilliiwahlbenga.eeke sas setae een Be 
Haleyon\albiventris ‘albiventris\,44os25 ace eee eee x 
Golins, striatest minis a 1j<<c.q stunt de sccerutioe seam x 
Colitis sinters -amdiCus:. 4 o.0ais,s «peice ss syste map eee esa x 
DicKunus adsimulisnadsimiliSs «seer cele cementite x 
Campephagalphoenicea Havas. cemscc a seecisec ae oe x 
Pycnonotusicarer bengalensistecn eee. em cetaaere- ee A x 
Pycnonottis: nigricans: 42.6 a2. '/5 Su eee ek. cele x 
Py cnonotusmeapensisn ister stu s or acice a eke rere ei x 
Pycnonotus; barbatus) lavards 9... oa. eee yee doe sous ss Es 
PYCnOnOtUS batbatis davies. 6 « <ciiscblaeem mre kD ote. hae Sate a 
Py cnonotussbarbatus tenebuonss sree reee secs SZ 
Pycnonotus importunus importunus.................. X 
Pycnonotus importunus noomei................eee+- x 
Phyllastrephus terrestris terrestris..........seeeeees x ie oye 
Aegithinastiniia lime sac eee eee ee sone cae eile 553 5K 
‘Turdoidesicaudatus icandatas. 2700 2k. Week. As eee Se 
Turdoides \earle?edrlei..cs.2edels.enviee aeueoer eee ¢ oat eh < 
Turdoides, longirostris, (6 v4 doce seit aides suspense = ce cikegs Se 
undoidesermarcolms 2.1.42 cle cietaa lt aig evel Ss cae seas Lee cisge De 
Turdoides ‘subfutus’ subrutast. 7.0" oe seeks oc eeh ne ae x 


mMurdoidesrstriatus sindiamuss eect aero eee rene phe aes i x 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 


HOSTS OF CLAMATOR JACOBINUS (continued) 


Host 
Murdoides striatus somervillet 2%. 9.00. . Loca ecu cle 
@prdoides striatus: malabaricus. 22. iets Joe. ions uk 
Aen eShstrintus. Stratis ted ocho is Sik dla. ois la Baia: = reeak 
PRUEOOIIES StEIA MIS, EIST CSCEMS (20s Gia.ay ohnie ays o1a8 id osha dai crate 
Turdoides fulvus acaciae..............0.. aa ee ere eae 
Turdoides rubiginosus rubiginosus................... 
Gurdoides: jardinermatalensis®. - 2... 64 5205. 2. 
Apgdoides \afiinis.aflinis. sis¢d0 oosie Se Beg Lae 
Turdoides leucopygia: omoensis.). .. 5... .... sais. seer «5 
bas AnCeolatus: \WOOGL ic.aicicie os lass spa vceieteie ne oes old 
Garrulax moniligerus moniligerus.................. . 
Goarrulax; pectoralis:pectoralis!)). sv ues!: seed saa ule Ye 
Garrulax. ‘pectoralis + meridionalis’. <,...:).. .6.<bacces ss 
Gaeriiax striatus striatus. odo. csc fete. Glens cee chee 
Garrilasc striatussbrahmaputtaa.scacaccee cas ue eee - 
Garrulax leucolophus leucolophus................... 
Gatenlaxdelessertiowlanisinc: 323..3dfsac + asee' s ENA 
GarGitlax TInCOlisipee sass chessiacls coc. teas ca. cow es 
Garralax merulinus’ tuerilinus:).. 6.5... 52 6 es 6. vee 
Garrolax caetilatus ‘subcacralatus..< oc. ccc e sess sce 
Garrulax cachinnans cachinnans..... Ban eee GORE Cold 
Garrulax lineatus lineatus....... Sen Shares sa, aioe ba CE 
Garrulax austeni austeni........ BROS ee SRA te 
Gatrnlaxsqtiamiatas: oc../se dee <tee Saalede seals cane 
Garrulax erythrocephalus chrysopterus.............. 
Garrulax’ phoentcets) bakeri. .i 5.215535 eciee os aida oats 
Actinodura egertoni khasiana...... neces 
Alcippe nipalensis nipalensis................ S ehais fares 
Parisoma subcaeruleum Sia eth 5 oa shite edanas 
@opsychus: saularis> sanlarisci shh. Oke obs. ewe lade 
Ginclidiunmleicusumies te ees ee tet: oe aioe dats 
BSHCIELS MUACHIAIIS PUELATUS .,0.0).!0, 0:6 o.c.cme-e cat acteraaner 
WAGHCOlA FURVENEDIS. 65. a6 osc ace 00 eae 6 ole so adie one's cds 
Myophoneus caeruleus temminckii................... 
Zoothera citrina citrina........ Ly SOR. MPSS 
PRC INGeSUCHS, Si iace A MUM. Wis seo ley stad’: PADaeS 
Terpsiphone viridis perenieillata. eS each Seoeve boos eee 
Sphenoeacus afer transvaalensis...............-..2:- 
Motaciliay capensis, CapensiSac mci cm asc ese sce cst ee 
Moataetlin asainp: viduals... So0teieis. Soca esc eeeees ca 
Laniarius ferrugineus natalensis..............00000.. 
Telophorus zeylonus zeylonus...............--eee00 
Telophorus -zeylonus: phanus..,.ss...)5 csc sees sess 
PEPIS SCHACH, STICOIO GE o.515,0 a.chcic es Sn ee ee ols so once : 
Lanius schach nigriceps....... Sdn ts wince a en Peg Meg 
PEATUS COUALIS! COUALIS Ac. seo aek cee eee cent eaadels 


serratus 


nM A 


a 


mm 


jacobinus 


19 


pica 


Am nA mM va 


nw mM A 


A 


ra 


MMM Mw OM MoM MMM 


A 


20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


One important point, not revealed by the tabulation, is the relative 
frequency with which the different hosts are selected. Out of the 
total 123 African records, 59 are of species of the genus Pycnonotus, 
and 14 more are of other bulbuls, making a total of 73, or almost 60 
percent of the total, that refer to this one family of hosts; 23 are 
of one species of Lanius, and 10 more are of other shrikes, a total 
for this family of 33, or more than 25 percent of the total; only 5, 
or not quite 5 percent, are of babblers. On the other hand, out of 
106 Asiatic records, only 1 is of a Pycnonotus, but over 85 percent 
are of babblers, chiefly of the genera Turdoides and Garrulax. 

The figures given above for the frequency of parasitism on bulbuls 
and shrikes in Africa are actually below the truth, as they are based 
solely on the total of individual instances reported. They make no 
allowance for the undocumented, general statements of experienced 
collectors, such as Plowes (1944, p. 93), who wrote that practically 
every bulbul nest examined was found to contain one or more eggs 
of the jacobin cuckoo. Also, generally a lower percentage of cases of 
parasitism on frequent hosts gets into the literature because of their 
repetitive nature, whereas practically all cases of infrequent ones are 
apt to find their way into print eventually. 

As may be seen from the list of fosterers, in the great majority of 
cases the pied crested cuckoo lays its eggs in open, cup-shaped nests 
built in trees or bushes. The one record of its using a kingfisher as 
a host (Schénwetter, 1928, p. 130) and the two involving the rock 
sparrow, Petronia (de Klerk, 1942, p. 58), are the only instances of 
its parasitizing hole-nesting species. Another unusual type of nest 
choice is the lone case of a coucal, Centropus grillii wahlbergi, as a 
host. This bird builds a roofed-over, or domed, nest of fine twigs and 
grasses, on the ground. Other frequently terrestrial-located nests 
known to be used occasionally are those of two species of wagtails, 
Motacilla capensis capensis and Motacilla aguimp vidua, and of the 
grass-bird, Sphenoeacus afer transvaalensis. 

In the present state of our knowledge of Clamator jacobinus the 
only obvious difference in its overall “fitness” to all the aspects of 
its existence in India and northeastern Africa on the one hand, and 
in southern Africa, on the other, is the much lesser degree of adaptive 
similarity of its eggs to those of its common hosts in the latter area. 
Strangely enough, in the areas where there is adaptive similarity it 
appears to have value to the parasite, but in the areas where it is 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 21 


nonexistent its absence seems quite unimportant. Baker’s data (1942, 
p. 83) on this cuckoo in India reveal a strong correlation between 
percentage of host acceptance of its eggs and the degree of egg 
resemblance involved, and, conversely, between the incidence of re- 
jection when the cuckoos’ eggs are deposited in nests of nonadaptive 
hosts and the degree of difference in the eggs of the two. Thus, of 
106 parasitized nests of “normal” (7.e., egg-adapted) fosterers, only 
1 was deserted (less than 1 percent) ; of 48 parasitized nests of ‘“‘un- 
usual” fosterers, 3, or 6.25 percent were deserted; of 8 parasitized 
nests of “abnormal” fosterers, 5, or 62.5 percent, were deserted. 
Similar figures were found (Baker, 1942, p. 85) for the red-winged 
crested cuckoo, Clamator coromandus: of 111 “normal”  fos- 
terers’ nests, 1, or 0.9 percent was deserted; of 58 “unusual” 
fosterers’ nests, 4, or 6.9 percent were deserted; of 12 “abnormal’’ 
fosterers’ nests, 6, or 50 percent were deserted. 

In the case of the jacobin cuckoo in southern Africa no such 
correlation has been found. In fact, the most frequently imposed 
upon hosts in South Africa are two bulbuls, Pycnonotus nigricans 
and P. barbatus, whose eggs are salmon to pinkish white, blotched 
and blurred with reddish brown and grayish lavender, very different 
from the pure white eggs of the local race of the jacobin cuckoo. 
Yet these bulbuls accept and incubate these dissimilar eggs. Perhaps 
the next commonest host in that area is the fiscal shrike, Lanius 
collaris, whose eggs also differ from those of the parasite about as 
much as do those of the bulbuls, being grayish green rather than 
pink, but equally speckled and blotched with brown and lilac. The 
fact that this shrike accepts the strange eggs is even more surprising, 
as it is an aggressive bird that has been known to attack and to drive 
off the cuckoos when they come too close to its nest. Yet, in spite 
of this, once the eggs are deposited in the nest, the seemingly alert, 
pugnacious host appears to be indifferent to their appearance. 


Clamator levaillantii (fig. 6, p. 22) 


The stripe-breasted cuckoo is the least known member of the genus, 
but while the total number of observed instances of its parasitism is 
less than that of the others, it is sufficient to show a marked preference 
for babblers as fosterers. The list includes 10 species, 6 of which are 
babblers, and which, together, account for more than three-fourths of 
all the records. In fact, one species, Turdoides jardinei, alone, with 


VOL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


22 


Decrees 


pipette 


« 
Say S 
i) 
© 
2 
» 
» 
, 
$ 
« 
> 


Fic. 6.—Range of Clamator levaillantit. 


polymorphism. 


f 


10n O 


B, Breeding record. Solid black area denotes reg 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 23 


some 23 records, account for half of the total. The known hosts are 
as follows: 


Phoeniculus purpureus zuluensis Turdoides jardinei jardinei 
Colius striatus striatus Turdoides jardinei natalensis 
Pycnonotus barbatus minor Turdoides jardinei kirki 
Pycnonotus barbatus layardi Turdoides jardinei emini 
Phyllanthus atripennis bohndorff Turdoides jardinei tanganjicae 
Turdoides plebeja gularis Turdoides reinwardii reinwardii 
Turdoides plebeja cinereus Turdoides leucopygia hartlaubii 
Turdoides plebeja plebeja Turdoides bicolor 

Turdoides plebeja platycircus Cossypha caffra caffra 


Of these 10 species, the first is not more than an accidental choice, 
and is based on a questionable identification of the parasitic egg 
(Roberts, 1939a, pp. 10-13). In addition to these it may be added 
that Bradfield (1931, pp. 7-9) suggested that in Damaraland the 
Burchell starling, Lamprotornis australis, was also parasitized, but 
he had no evidence other than that he had noted these starlings 
“mobbing” a stripe-breasted cuckoo. 


Clamator coromandus (fig. 5, p. 16) 


The red-winged cuckoo is parasitic chiefly on babblers, and, within 
this group, primarily on the larger laughing-thrushes of the genus 
Garrulax, some 13 species of which have been found to be victimized. 
Baker (1942, pp. 196-197) listed 265 eggs of the red-winged cuckoo 
in his collection, taken from nests of 21 species (25 species and 
subspecies) of hosts. Of these 265, all but 24 were found in nests 
of Garrulax, and no fewer than 109 from nests of a single species, 
the necklaced laughing-thrush, G. moniligera, and 37 from nests of 
the black-gorgeted laughing-thrush, G. pectoralis. 

Our knowledge of this cuckoo’s fosterers is still largely based on 
collections and observations from the Indian and Burmese portions 
of its range. In due time many hosts from other areas will be added 
to the list. The following list of known victims is based on that of 
Baker, with a few additions from other sources. 


Dicrurus adsimilis macrocercus Garrulax delesserti gularis 
Pomatorhinus erythrogenys mcclellandi Garrulax cineraceus cineraceus 
Turdoides gularis Garrulax rufogularis assamensis 
Garrulax moniligerus moniligerus Garrulax caerulatus subcaerulatus 
Garrulax moniligerus fuscatus Garpnlase wubenile 


Garrulax pectoralis pectoralis : : 
; est ; Garrulax merulinus merulinus 
Garrulax pectoralis meridionalis 


Garrulax striatus striatus Garrulax squamatus 
Garrulax striatus brahmaputra Garrulax erythrocephalus chrysopterus 
Garrulax leucolophus leucolophus Garrulax phoeniceus bakeri 


Garrulax leucolophus belangeri Actinodura egertoni khasiana 


24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Copsychus saularis saularis Zoothera citrina citrina 
Enicurus schistaceus Turdus protomelas 
Myiophoneus caeruleus temminckii Lanius schach tricolor 


As we have already noted, when the Indian population of pied 
cuckoos (pica) began extending their range northward into the foot- 
hills of the Himalayas and found themselves removed from the 
habitat of the babblers of the genus Turdoides that had served them 
as fosterers in the plains, they began using the larger species of 
Garrulax. This change was probably already incipient in the pica 
stock, as it had become very pronounced and fixed in the earlier 
evolutionary offshoot from pica that resulted in C. coromandus. 
There must certainly have been a considerable time span involved in 
the evolution of the red-winged from the pied cuckoo, whereas the 
northward spread of the latter seems to have been fairly recent. 


Clamator glandarius (fig. 7, p. 25) 


The great-spotted cuckoo is the most advanced of the four species 
of Clamator, and is closer to C. coromandus than to either of the 
others. From the circumstantial evidence of the current situation in 
the genus, it is justified to conclude that glandarius was an evolution- 
ary development from the stock at present represented by coromandus. 
Hence, it seems probable that it originated somewhere near the north- 
eastern portion of the range of that species. Moving eastward, the 
primordial glandarius came into contact with magpies, a group of 
sizable, suitable, potential fosterers until then unaffected by any 
parasitic cuckoo, and to them it became adapted with marked success. 
Advancing farther eastward glandarius met with magpies in southern 
Iran, Iraq, Lebanon, etc., in areas of warmer climate than the 
Himalayan foothills and slopes of northern Assam, Bhutan, and 
Sikkim, where its ancestors may have first encountered their magpie 
hosts. In fact, the presence of the latter birds may well have expedited 
the eastward shift of early glandarius. Being essentially a warm 
climate form, glandarius left its original locus and eventually became a 
circum-Mediterranean species, still largely in areas of sympatry 
with the magpie, although becoming allopatric with it in eastern 
Egypt, where it used crows as hosts instead. At that stage of its 
history glandarius was largely contained within the range of its magpie 
host, and its great spread to sub-Saharan Africa, completely away 
from this fosterer, came much later. 

It is conceivable, though, in the nature of things not demonstrable, 
that possible competition from the corvine parasitism of the koel, 


AVIAN GENUS CLAMATOR—FRIEDMANN 25 


NO. 


a” 


“ 


et eel et 


\Wureetesseo? 


[i— 


g range of Clamator glandarius. 


Fic. 7—Known breedin 


Area of sympatry with Pica lies north of heavy line. 


26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Eudynamis scolopacea, in northern India, may have influenced the 
eastward emigration of early glandarius to areas free from such 
difficulties. 

The close resemblance between the eggs of the great-spotted cuckoo 
and of the magpie is evidence of a long continued and very specialized 
host-parasite relationship. As Baker (1942, p. 85) wrote, the parasite 
lays, “. . . one type of egg and one type only which is so exactly like 
in colour, shape, and superficial appearance of texture to that of 
the Magpie that identification is generally extremely difficult .. .” 
It may be stressed at this time that the egg, speckled with dusky, is a 
highly “advanced” egg type for a cuckoo, the basic, primitive egg type 
in the family being unmarked white, and it has arrived long ago at 
a stage far beyond the development evinced in any other species of 
Clamator. That it is “fixed” and invariable, and that it now persists 
unchanged in the vast stretches of sub-Saharan Africa, where it does 
not match the eggs of the hosts used there, is evidence for the age 
and the finality of this ‘end product’ of adaptive evolution. 

If we were to assume, without documentation, as Voous (1960, p. 
154) has done, that C. glandarius originated in sub-Saharan Africa 
and hence, that it there evolved its egg type in the absence of any 
known host whose eggs it resembled, and then later invaded Mediter- 
ranean Africa and the Iberian Peninsula, where its eggs “fitted” 
so well with those of the magpie, we would have a most remarkable 
example of extreme preadaptation. It would be so remarkable that it 
would be difficult to accept it without extremely disturbing doubts and 
skepticism. 

If, on the other hand, it be accepted, as here postulated, that 
Clamaior glandarius, having arrived at a perfected stage of adaptive 
evolution with regard to the degree of similarity of its eggs to those 
of the magpie, its chief, and almost its only, host in Asia Minor, in 
the Iberian Peninsula, and in northwestern Africa, then expanded its 
range southward into areas where this adaptive excellence no longer 
had its former value, we would have a case of what may be called 
“repudiative evolution.” Part of the species acted as though the 
matter of egg resemblance no longer mattered, and in its new home 
used new fosterers to which it was not adapted. In a sense, this 
amounted to an escape from too specialized a form of host relation- 
ship ; one which, had it been adhered to, would have markedly limited 
the parasite geographically, for the cuckoo is a bird of warm climes, 
whereas the magpie’s range extends far to the north where the 
parasite would not be able to follow it, and the two are sympatric 
only in a limited area. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 27 


The importance of settling the question as to whether glandarius 
was originally sub-Saharan in range and later spread to the Mediter- 
ranean areas, or vice-versa, warrants a little further discussion. Voous 
(1960, p. 154), the only proponent of an African origin for the 
species, considered the hypothesis for such an origin “. . . supported 
by the occurrence of at least three other species of the genus Clamator 
in Africa...” This is incorrect as there are only two, jacobinus and 
levaillantu, while in Asia there are also two, jacobinus and coroman- 
dus. The last named is the species to which glandarius is most clearly 
related and appears to be the stock from which it arose. During a 
visit to Los Angeles in 1962, Stresemann, who has been studying very 
carefully the distributional history of the birds of Europe, agreed 
with me in considering the sub-Saharan range of glandarius as a 
recent expansion from an older circum-Mediterranean one. 

In this connection we may recall, in Dobzhansky’s (1940, pp. 312- 
321) words, that “. . . each species, genus and probably each geo- 
graphical race is an adaptive complex which fits into an ecological 
niche somewhat distinct from those occupied by other species, genera, 
and races. The adaptive value of such a complex is determined not 
by a single or a few genes, but is a property of the genotype as a 
whole. Furthermore, the adaptive complex is attuned to its environ- 
ment only so long as its historically evolved pattern remains, within 
limits, intact . . .” Clamator glandarius is highly adapted to the 
magpie, but yet part of its population has been able to abandon this 
evolved situation and to become attached to as different a host relation- 
ship as that with Spreo bicolor in South Africa. 

One cannot help but wonder if this exodus of part of the Mediter- 
ranean glandarius may have been influenced, if not caused, by intra- 
specific competition in a too populous stock of the species, after its 
adaptive evolution had seemingly expedited its existence. Haldane 
(1932, p. 119) pointed out that there is a fallacy in the concept that 
“’. . natural selection will always make an organism fitter in its 
struggle with the environment. This is clearly true when we consider 
the members of a rare and scattered species. It is only engaged in 
competing with other species, and in defending itself against in- 
organic nature. But as soon as a species becomes fairly dense matters 
are entirely different. Its members inevitably begin to compete with 
one another .. .” 

Inasmuch as the other three species of Clamator parasitize almost 
exclusively birds that build open, “saucer-shaped” nests, it may be 
assumed that a similar host choice is, or originally was, basic in 
glandarius as well, and that the use of hole-nesting starlings is a 


28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


relatively recent development. The surprising thing is not only that 
it was able to make this change, but that it had previously gone so 
far in the road of adaptive specialization to a host with which its area 
of sympatry was so limited. 

The magpie genus Pica occurs throughout Europe, including the 
Mediterranean islands, and Asia north of the tropics (2.e., north of 
the Arabian Peninsula, Baluchistan, Pakistan, India, Burma, Assam, 
and the Malayan countries), east to the western part of North 
America, and south from Gibraltar to northwestern Africa (Morocco, 
Tunisia, Algeria). In all the vast extent of this primarily Holarctic 
range, it is sympatric with its “highly adapted” brood parasite, 
Clamator glandarius, only in the Iberian Peninsula, adjacent portions 
of northwestern Africa (Morocco, Tunisia, Algeria), parts of south- 
eastern Europe, Cyprus, and the Near East as far as Iran. This 
area of sympatry is thus a somewhat peripheral part of the range, 
both of Pica and of Clamator glandarius (whose geographically 
most extensive range is African south of the Sahara all the way to 
the Cape, and in eastern Egypt). An instance of the degree of 
sympatry of C. glandarius and Pica in southwestern Europe is the 
absence of both from the Balearic Islands although both occur in the 
Iberian Peninsula and in Morocco. 

If the egg coloration of Clamator glandarius evolved to match that 
of Pica, this must have taken place in this limited area where the two 
occur together. The bulk of informed opinion regards the close 
egg resemblance as something arrived at by adaptive evolution, and 
not as a fortuitous coming together of a parasite and a host whose 
eggshells were similar in color, pattern, and size. The latter inter- 
pretation would assume an improbable and unlikely happening, al- 
though it cannot be ruled out as a possible explanation. The fact that 
throughout its range, the great-spotted cuckoo lays only this one 
type of egg suggests that its original range was just those areas 
where its egg type was adapted to a prevalent host. This further 
suggests that the Pica-allopatric portions of its present range in 
Egypt and in Africa south of the Sahara must have been a more 
recent extension of its distribution. 

Amadon (1947) ascribed a marked change of bill form and of 
feeding habits in a Hawaiian honeycreeper, genus Hemignathus, 
to a sudden ecological shift of its ancestral population. Mayr (1959, 
pp. 177-178) considered that such a shift into an entirely new 
ecological niche may well have been the type of occasion attendant 
upon the emergence of many major evolutionary novelties. When 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 29 


Clamator glandarius extended its breeding range into sub-Saharan 
Africa, the ecological shift was one that involved a marked change 
in host choice, even a virtual repudiation of a previously highly 
evolved egg adaptation, but, as it did not involve any apparent, drastic 
alteration in the daily life or feeding habits of the adult cuckoo, no 
comparable evolutionary change transpired. 

It follows from this that, whereas in the case of self-breeding birds 
the entire biology of the species is a closely coordinated unit (almost 
what in current commercial jargon is referred to as a “package 
deal”) on which selection may operate, in the case of brood parasites 
there is cleavage resulting in two fairly separate parts. The evolu- 
tionary climate ambient to the egg and nestling stages is that of the 
host species and has relatively minor connections with, and repercus- 
sions upon, the selective factors surrounding the life of the adult 
parasite. This may have helped make it possible for Clamator 
glandarius to invade vast new areas and to remain unchanged. Con- 
currently, it must be assumed that the new, non-egg-adapted hosts, 
suddently parasitized by the newcomer, had no previous need to 
evolve any particular acuity of discrimination and thus were rela- 
tively easily susceptible to parasitism. 

As far as casual observations go (and these are all that have been 
recorded in the literature), the great-spotted cuckoo seems equally 
successful in the various portions of its range. It might be expected 
that the wide discrepancy in the degree of host adaptation it shows 
in tropical and southern Africa on the one hand, and in the Mediter- 
ranean area on the other, would be reflected in its local numerical 
status, but the available evidence does not point to any such effect. 
It must be admitted, however, that the data are still very superficial 
and imperfect. If anything, the fact that unusually large numbers of 
its eggs are often found in single nests of its corvine hosts in the 
areas where the cuckoo is nonadapted might even suggest a relatively 
greater abundance of the parasite in proportion to the available host 
population there. 

The known hosts of the great-spotted cuckoo, as listed here, are 
primarily birds of two families, the Corvidae (crows, jays, magpies, 
and piapiacs) and the Sturnidae (starlings). The other three in- 
cluded species are a kestrel, which was probably an “unintended” host 
choice as the bird was using an old magpie nest, and two South 
African ground-tunnel nesters, the hoopoe and the ground wood- 
pecker, which may have been “acceptable” to the parasite because of 


30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


their general similarity to the nesting tunnels of the pied starling, 
Spreo bicolor, a favored and frequent host there. 

The predominant role played by members of the Corvidae as 
fosterers of this cuckoo is indicated by the fact that of a total of 172 
nests parasitized, 141 were of various corvids; 89 nests belonged to 
crows of 6 species (Corvus corax, corone, ruficollis, albus, rhipidurus, 
and capensis) ; 1 was of a raven, Corvultur albicollis, 48 were of 2 
species of magpies (Pica pica and Cyanopica cyanus), 1 was of a 
piapiac (Ptilostomus afer), and 2 were of a species of jay (Garrulus 
glandarius). Of the remaining 31 parasitized nests, 28 belonged to 7 
species of starlings. Of these 14 were of the pied starling, Spreo 
bicolor; 5 of the red-winged starling, Onychognathus morio; 4 of the 
glossy starling, Lamprotornis nitens; while of the other species single 
instances only have been reported so far. 


Falco tinnunculus tinnunculus Pica pica bactriana 
(in an old Pica nest) Pica pica galliae 
Upupa epops africana Pica pica melanotos 
Geocolaptes olivaceus- Pica pica mauritanica 
Corvus corone sardonius Garrulus glandarius krynicki 


Corvus corone corone 

Corvus ruficollis edithae 
Corvus corax corax 

Corvus albus 

Corvus capensis capensis 
Corvus capensis kordofanicus 
Corvus rhipidurus 


Ptilostomus afer 

Acridotheres tristis tristis 
Onychognathus morio morio 

Spreo bicolor 

Spreo albicapillus 

Lamprotornis nitens phoenicopterus 


Corvultur albicollis Lamprotornis caudatus 
Cyanopica cyanus cooki Lamprotornis chalybeus cyaniventris 
Pica pica pica Lamprotornis chalybeus sycobius 


As we have seen in the case of Clamator jacobinus, in the present 
species also, it is the sub-Saharan segment of its total membership 
that is the less well adapted in its egg coloration. However, in both 
these species, the available observational evidence gives no grounds 
for assuming that the sub-Saharan birds are less “successful” than 
their more completely and more perfectly adapted northern segments, 
insofar as “success” may be implied from ability to survive in num- 
bers over a vast area. 

We cannot, however, deduce from this that adaptation has lost its 
value and significance in one geographic portion of the total distribu- 
tional range of this one genus of birds, while remaining advantageous 
elsewhere in the same genus, as well as in most of the rest of the 
animal kingdom. Certainly the case for the natural selective value 
of adaptation generally is so strong, so well-nigh invariable, that we 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 31 


cannot easily accept its apparent unimportance here. At least part 
of the answer to this puzzle lies in the fact that because these two 
cuckoos were studied earlier in India and in the Mediterranean lands 
than in sub-Saharan Africa we have come to accept the adaptive 
excellence reported for them from those areas as an essential and 
necessary aspect of their natural economy. But now we know that, 
advantageous as this may be, it is not essential, and that the two 
species can and do survive without it. Actually, this is implied even 
in the course of the evolution of the climax adaptation in the areas 
where it has transpired, as countless less completely adapted gen- 
erations had to survive to provide the material out of which was 
achieved the greater perfection, which in time supplanted the less 
adapted birds. 

We have, then, a superficially similar situation in southern Africa 
in both the great-spotted and the jacobin cuckoos, but one which 
appears, on more careful study, to be due to opposite evolutionary 
trends. In the jacobin it seems probable that the southern popula- 
tion, serratus, is the original, primitive segment of the species that 
has remained as it was while giving rise to the more advanced pica 
and jacobinus, an evolution involving primarily the change from un- 
pigmented to pigmented eggshell. On the other hand, the fact that 
in sub-Saharan Africa glandarius is not only bereft of the adaptive 
advantage its egg evolution had given it in Mediterranean lands, but 
further that in its southern range there is a striking difference in 
the numerical relationship of parasite-host eggs in parasitized nests 
in the two areas causes the southern population of this species to 
seem relatively so inept that it may only be explained on the basis 
of the recency of its invasion into that area. 

What has happened with C. glandarius is paralleled by a similar, 
though less extensive, move in the jacobin cuckoo. Although 
the geographic spread of C. jacobinus from Africa to India is 
something that happened relatively early in its evolutionary his- 
tory, the species has expanded its range in India more recently by 
advancing higher into the hills. Thus, Baker (1942, p. 83) con- 
sidered it possible that its present breeding in the hills up to 6,000 
feet and even higher in Assam and in the central Himalayas was a 
“modern extension of its breeding habitat. In the Plains... its normal 
fosterer, or group of fosterers is so completely established that excep- 
tions are very, very few. In the lower hills the Cuckoo adheres 
closely to the Necklaced and the Striated Laughing-Thrushes, but 
above the normal elevation of the breeding areas of these birds or 


32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


where these birds are not found, or are rare, it launches out into 
the use of all kinds of nests which bear some resemblance to those 
they usually cuckold...” 

The timespan involved in the southward spread of Clamator 
glandarius to sub-Saharan Africa need not have been great. The case 
may well have been similar to the recent rapid, almost “explosive,” 
spread of the cattle egret, Bubulcus ibis. In both instances the advanc- 
ing birds filled vacant ecological niches. The cattle egret had no com- 
petition from other herons because it was a dry land bird and lived 
largely on insects, not an aquatic feeder on fishes, tadpoles, etc. The 
great-spotted cuckoo was parasitic on corvids, a group until then un- 
molested by any parasitic birds in Africa. The spectacular spread of 
the collared turtle dove, Streptopelia decaocto, in Europe during the 
past 50 years is a parallel example. 

It seems that the relatively recent, but very extensive, geographical 
expansion of C. glandarius originally was motivated by the bird rather 
than by its environment. This statement may require a little elabora- 
tion to make its meaning clear. Evolutionary changes are often the 
result of a double process of selection; selection by the environment 
of the most advantageous, best adapted structural, functional, or 
behavioral organization in the organism, and also selection by the 
animal of the most comfortable, the most nearly optimal environ- 
ment. The capacity for making a choice among available environments 
is inherent in all animals that are able to move about freely. In effect, 
this results in a process of sorting out the members of a species 
environmentally instead of selectively eliminating the less fit in the 
original ecological situation. 

Implied in the phrase “sorting out” is what appears to have been 
behind the great move to sub-Saharan Africa. The part of the original 
circum-Mediterranean population of C. glandarius that was relatively 
less completely “fit” was the part that moved on to new territory— 
in this case, to equatorial and southern Africa. That it was less 
delicately, or less nicely, adapted to its original hosts than was the 
part that stayed in the Mediterranean area is still evidenced by its 
lack of adjustment in its egg deposition to the size of the total re- 
sulting clutches in the nests of its victims. This significant difference 
in the two geographic segments of C. glandarius is discussed in detail 
in our account of the intensity of parasitism (see pp. 38-47), but a little 
additional comment seems called for here. 

While it is obviously impossible to state precisely what factor, or 
factors, motivated the dispersal of part of the glandarius population 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 33 


from its Mediterranean homelands, it seems likely that it was growing 
population pressure, such as we have seen recently in the case of the 
cattle egret, mentioned above. The latter bird had increased greatly 
in numbers in Africa prior to its sudden geographic advance. Coin- 
cident with a situation of overpopulation, it may be remembered that 
the nature and the intensity of natural selection varies with different 
degrees of abundance of a species. When a species is numerically 
uncommon the selection pressure it experiences is exerted chiefly by 
the environment, whereas when it is more abundant the selection is 
often between members of its own species. It was selection of the 
latter kind that seems to have been involved in the emigration of the 
less adapted members of the glandarius population. 

The lack of any fine control in the intensity of parasitism, as 
evinced by multiple-egg deposition and the resulting uncorrelated egg 
complements in parasitized nests, in sub-Saharan glandarius is more 
than a matter of an as-yet-unachieved adaptation. It is also an 
indication that the cuckoo is a recent arrival and is increasing in 
numbers, because at a time when the size of the population of a 
species is growing, selection is usually relatively weak, and such 
excesses as extreme multiple parasitism would be tolerated, whereas 
in a stable, “climax” situation this would be less apt to succeed. 

Conversely, selection is apt to be stronger when the population 
of a species is decreasing. This must have been the case in the 
Mediterranean glandarius when part of the species emigrated south- 
ward, thereby reducing the intraspecific competition and permitting 
a more active environmental selection. This may actually have con- 
tributed to the development of an even better controlled host-parasite 
relationship there. As Carter (1954, p. 255) has stated, “. . . the 
population that survives the decrease of numbers will be a selected, 
and not a random, sample of that at the preceding maximum. Only 
the better adapted are likely to survive ... It follows from this 
that adaptive evolution will be accelerated at the time of decrease . . .” 

As the great-spotted cuckoo extended its range into sub-Saharan 
Africa, where there were no magpies, it undoubtedly used at first 
the nests of various species of crows for its egg laying, just as it had 
already done in eastern Egypt and the Near East. However, while 
it continued to use the arboreal nests of corvine hosts throughout 
its new domain it also extended its host choice to include such very 
different types of nest structures as those of an earth-tunneling 
starling, Spreo bicolor. It is known that in some animals specific 
types of nest structure may act as isolating mechanisms, preventing 


34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


mismating. So marked a change as that between the arboreal, bulky 
nests of masses of twigs and sticks of the magpies and crows, and 
the terrestrial nesting tunnels of the pied starling might seem more 
than sufficient to have functioned in its impact on the behavioral 
patterns of the cuckoos in much the same manner as an isolating 
mechanism. However, the differences involved, real as they are to 
human eyes, did not appear to affect the parasite. 

In this connection I may say that I have tried to find a place 
where both the pied starling and one or more species of crows were 
present in numbers as breeding birds and where the great-spotted 
cuckoo also bred, but have not been able to do so. Such a locality 
might give an observer the opportunity to study the host choice of 
the parasite where both types of hosts were equally available. 

The matter of host nest selection appears to affect the life and 
activities of the cuckoos only during the brief moments of actual 
ovulation by the hens. It may be remembered that mating or copula- 
tion by the cuckoos does not take place in or at the nests of any of 
the hosts, and that the cock cuckoos do not necessarily even know 
which nests receive the eggs they may have fertilized. 

While the difference between the two extreme types of egg de- 
positories used—the open, dish-shaped, arboreal, stick nest of a crow 
and the long earth-tunnel of a pied starling—are great, the change 
probably was not as abrupt as it might seem. To begin with, the host 
to which the great-spotted cuckoo’s evolution has made it most ade- 
quately adapted is the magpie, a bird which customarily makes large 
nests of small branches, twigs, and sticks, roofed over, with an 
entrance on one side, and usually constructed in large thorny bushes 
or on the upper branches of tall trees. From this it was not a great 
change for the parasite to use nests of the crow in eastern Egypt 
and the Near East, the chief difference being that the nests of the 
latter were open, not roofed over, but were constructed of similar 
materials and in generally similar types of situations. From one 
species of crow to another (from C. corone in Egypt, Iraq, etc., to 
C. albus and others in sub-Saharan Africa) involved no vital change 
for the parasite, but the change from these to hole-nesting starlings 
seems quite marked. However, even this was neither abrupt nor 
as drastic as one might assume. In former British Somaliland (now 
a part of the Somali Republic), a somewhat intermediate stage has 
been reported by Archer (1961, in Archer and Godman, pp. 649-659). 
He found several eggs of the great-spotted cuckoo in nests of the 
white-capped starling, Spreo albicapillus, a species that builds bulky, 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 35 


domed nests of twigs and coarse grasses high up in trees—nests quite 
similar in their main features to those of the magpies, the parasites’ 
primary fosterers. From laying eggs in these domed, internally 
dusky, if not dark, somewhat tunnellike egg chambers of Spreo 
albicollis it was not a great step to using the darker nests of true 
tree-hole nesters, of other species of starlings, such as Lamprocolius 
nitens and L. caudatus and Acridotheres tristis. More of a change 
was involved in the shift from these to terrestrial burrowing hosts 
such as Spreo bicolor, but even here there may have been a transition 
stage, as this starling is said to nest in a variety of sites such as are 
used by Sturnus vulgaris as well as in its more usual earth burrow. 
Priest (1948, p. 118) indicated that this variety of nesting sites in- 
cludes crevices on walls, under the eaves of houses, in trees, as well 
as breeding in tunnels in soft river banks or cuttings, or in mine 
shafts. I am informed by Dr. Winterbottom that the nest record 
files of the Percy Fitzpatrick Institute of African Ornithology extend 
this list of sites to include haystacks and even a crevice of a concrete 
platform in the sea (obviously near shore). 

Once the cuckoo had become used to the pied starling as a fosterer, 
it could be expected to be attracted to it regardless of just where the 
nest was built. From terrestrial burrow nest-sites of this starling 
it was no great change to utilizing other similar nests, such as that 
of the ground woodpecker, Geocolaptes olivaceus. The pied starling 
is the most frequently used host in eastern South Africa today. 

A partial parallel to what transpired in Clamator glandarius, as 
outlined above, has also been reported occasionally for the Indian koel, 
Eudynamis scolopacea, a cuckoo parasitic also very largely, in fact 
almost solely, on crows. Baker (1942, p. 197) listed two starlings, 
Acridotheres tristis and Graculipica nigricollis, among its known 
hosts, the former one or two times, the latter more often. However, 
as far as known, the koel has not adapted itself to terrestrial-nesting 
hosts. Baker listed 209 eggs of the koel in his collection. Of these, 
16 were laid in nests of the black-necked mynah, Graculipica nigricol- 
lis, 2 were with Acridotheres tristis, 6 in nests of 2 species of magpies, 
and the other 185 in nests of 2 species of crows. 

Even the European cuckoo, Cuculus canorus, has been known to 
lay occasionally in the underground nests of the wheatear, Oenanthe 
oenanthe. Furthermore, and more directly pertinent, it may be re- 
called that one egg of Clamator jacobinus has been reported from a 
ground-tunnel nest of a kingfisher, Halcyon a. albiventris, and 
another from a tree-hole nest of a sparrow, Petronia superciliaris. 


36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Also, Clamator levaillantit has been known to lay in the hole nest of a 
kakelaar, Phoeniculus purpureus. The tendency to utilize such nesting 
sites has certainly not developed in jacobinus to the extent it has 
in glandarius, but these instances show that it is not outside the range 
of possibilities even in the former. 

One further thought emanating from a consideration of this prob- 
lem of host selection may be added here. Recent studies on many 
species of self-breeding birds, particularly in North America and 
Europe, have shown that nest-site selection is fairly rigid and fixed 
in its major elements. Slight vegetational differences often are 
critical to various species in the precise location of their nest sites. 
So widespread is this tendency that it is only proper to apply it to a 
review of the situation in brood parasites as well. In these birds 
nest-site selection would be altered to host selection based on the 
types of nest-sites used by the latter, and would be expressed in terms 
of host specificity as far as the parasites are concerned. On the whole, 
avian brood parasites fall into three main categories in this respect. 
Some exhibit little or no such specificity; others are specific in their 
host choice as individuals only (“individual-host specificity,” as found 
in the European cuckoo, Cuculus canorus) ; and in still other parasites 
the entire species is specific on one or a small group of related hosts 
(“species-host specificity’’). 

Inasmuch as nest-site selection does reflect trenchant and remark- 
ably uniform criteria in each species of self-breeding birds (potential 
hosts), and inasmuch as there is no reason to assume that parasitic 
species, especially early in their evolutionary history, were necessarily 
different from self-breeding birds in their response to familiar and 
uniform environmental details, it may be that species-host specificity, 
as opposed to individual-host specificity, was the original situation in 
brood parasites and that a broader range of host choice developed 
from it later. This would imply that the broad spectrum of hosts 
subsequently arrived at may have evolved even as a negation of the 
original selection pressure that operated in an earlier atmosphere of 
species-host specificity. From this concept it would follow that rigid, 
if not actually “obligate,” parasite-host specialization is a basic rather 
than an ultimate condition. This is quite the reverse of the often 
assumed pattern of species-host specificity arising from individual- 
host specificity. 

In this connection it may be pointed out that the one species of 
brood parasite whose descent from a self-breeding form is most 
obvious and clear, the screaming cowbird, Molothrus rufoaxillaris, 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 37 


started with, and has not deviated from, a fixation upon a single- 
host species, its close and antecedent relative, the bay-winged cow- 
bird, M. badius. Furthermore, the parasitic Viduinae are still in the 
original stage of limited range of hosts, their Estrildine relatives, the 
waxbills. It also seems not improbable that the great-spotted cuckoo, 
Clamator glandarius, early became involved with, and went through 
its eggshell evolution with a single host, the magpie, Pica pica; the 
same is true of the stripe-breasted cuckoo, Clamator levaillantu, with 
its chief host, the babbler, Turdoides jardinei, and of the koel, 
Eudynamis scolopacea, with its use of crows. The evolution of 
host egg similarity obviously is facilitated by species-host specificity. 


INTENSITY OF PARASITISM 


By intensity of parasitism two quite separate things are implied. 
The percentage of the total nests of frequently used hosts that are 
parasitized gives one aspect of the parasite-host situation. The fre- 
quency with which individual parasitized nests are found to contain 
more than one egg of the cuckoo adds still another element of the 
total picture. On the whole, increase in frequency of multiple-egg 
parasitism on the same pair of hosts is something that is super- 
imposed on the basic situation. While in some areas where the host 
nests are very numerous the incidence of multiple parasitism appears 
to be lower than in places where there are relatively few hosts for 
the number of cuckoos, this cannot yet be demonstrated convincingly, 
as in no area have the data been sufficiently extensive and intensive 
to give a precise survey of the numerical status of the hosts and of 
the parasite, or of the percent of nests of the favorite fosterers that 
are parasitized. Perhaps the nearest approximation to the kind of 
information needed is that afforded by Mountfort (1958, p. 54). In 
his fieldwork in Spain, he examined 7 nests of the magpie on one 
afternoon and found that 5 of them were parasitized by the great- 
spotted cuckoo, an incidence of simultaneous parasitism of 71.4 per- 
cent in a circumscribed area. Mountfort did not list the numbers of 
eggs or of young of either the host or the parasite in each of these 
nests, so his observations tell us something of the percent of magpie 
nests parasitized but not how intensively they had been affected. 
However, a compilation of all the cases he mentioned shows that of 
eight parasitized nests, none were found to have only a single cuckoo 
egg apiece. His figures are quite different from those given below 
for 28 other parasitized magpie nests, all also from Spain. Mountfort 
found some 50 occupied magpie nests in that country in 1956, but 


38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


did not specify if the 8 parasitized ones that he discussed were the 
only ones so affected. 

The lack of sufficiently comprehensive or detailed quantitative ob- 
servational evidence makes it necessary, at least for the present, to 
rely on the available data on multiple parasitism of individual nests 
as our primary method of evaluating the host-parasite relations. Of 
all the species of Clamator the one that is most revealing of evolution- 
ary change in this important regard is the great-spotted cuckoo, C. 
glandarius, and we may, therefore, begin with it. The data on this 
species are as follows. 


Clamator glandarius 


Of a total of 172 parasitized nests, containing 407 eggs of the 
cuckoo, 82 had a single one each, 41 had 2, 13 had 3, 13 had 4, 10 
had 5, 5 had 6, 1 had 7, 4 had 8, 2 had 10, and 1 had 13. In other 
words, some 47 percent of the nests contained single eggs of the 
parasite. However, considering that the first cuckoo egg laid in a 
nest was a “single” one at that time, and counting only the subsequent 
eggs as “multiples,” the total is 172 singles and 235 multiples. In 
other words, multiple eggs were almost 50 percent more frequent 
than singles. While this is a general condition, it does not give a 
representative picture of the situation as it really is in any one 
geographic fraction of the total range of the parasite. 

To make the data more comparable, we may eliminate for the 
moment all cases involving hosts other than species of crows. Six 
species of the genus Corvus (corone, corax, albus, capensis, ruficollis, 
and rhipidurus) are parasitized, and together account for more than 
half of all the records (fig. 8). Out of 43 parasitized crow nests 
found in Spain, Asia Minor, and Egypt (the bulk of the records are 
from Egypt) containing a total of 54 cuckoo eggs, we find that 33 
nests had 1 each, 9 had 2, and 1 had 3. In other words, over 75 
percent of the nests contained single eggs of the cuckoo, and, all in 
all, multiple eggs of the parasite were less than 25 percent as frequent 
as were singles. 

Out of 35 parasitized crow nests from south of the Sahara—from 
former Italian Somaliland (now a part of the Somali Republic) to 
Nigeria and south to South Africa—only 5 had single cuckoo eggs, 
7 had 2, 3 had 3, 6 had 4, 7 had 5, 2 had 6, 2 had 8, 2 had 10, and 
1 had 13; a total of 148 cuckoo eggs for the 35 nests as compared 
with 54 eggs from 43 nests in the Mediterranean lands. In tropical 
and southern Africa less than 20 percent of the nests had single eggs 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 39 


of the cuckoo, and, all in all, multiple eggs of the latter were three 
times more frequent than single ones. 

In the Iberian Peninsula it is definitely established that crows are 
only infrequently parasitized and that magpies are the principal hosts. 
In Egypt, and in Africa south of the Sahara there are no magpies, 
and the various species of Corvus are regularly victimized. Judging 


12} 
| 


13] O 


Number of Clemator eggs per nest 


8 10) 


1 re) x 


Tes Ge SO! 12.014 16) 6 t8) 20 22) 24 265628 30632345 36 


Number of nests 


Fic. 8.—Frequency of multiple Clamator eggs in nests of Corvus. 
O, In sub-Saharan Africa. X, In Mediterranean area. 


from all the records available it appears that multiple eggs of the 
great-spotted cuckoo in Mediterranean lands are found more fre- 
quently in the nests of magpies than in those of crows. Thus, in 
Spain, Lilford (1866, p. 184) found a magpie nest with eight eggs 
of the cuckoo and five of the host, while Saunders (1869, p. 401) 
reported others with four and six cuckoo eggs in them, although he 
noted that other magpie nests had only one or two of the parasitic 
eggs apiece. As mentioned above, Mountfort (1958, pp. 54-56) 
found eight parasitized magpie nests in Spain. The number of eggs 
or young of the great-spotted cuckoo in these were as follows: One 
nest had six; two had five ; two had three; and two had two. 

In reply to my inquiry, J. D. Macdonald very kindly sent me the 
data on 28 parasitized sets of the magpie, all taken in Spain, and now 


40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


in the British Museum. In these the combinations of eggs of the 
parasite and of the host were as follows: 1 cuckoo egg in each of 10 
nests with from 1 to 6 eggs of the magpie; 2 cuckoo eggs in each of 
12 nests with from 2 to 8 of the magpie; 3 cuckoo eggs in each of 
3 nests with from 1 to 3 of the magpie ; and 4 cuckoo eggs in each of 
3 nests with from 0 to 6 of the magpie. 

Data are at hand on nine parasitized nests of the Spanish blue- 
winged magpie (Cyanopica cyanus cooki) from the Iberian Penin- 
sula, ex Rey (1872, p. 143) and others, plus three sets in the British 
Museum; all of these had only single eggs of the cuckoo, with from 
one to five of the host. It is not possible, however, to say whether 
there is a significant difference in the intensity with which the 
two species of magpies are parasitized, although the available data 
appear to suggest that there may be. In both sets of data (Pica pica 
and Cyanopica cyanus), few of the sets approach the maximum size 
recorded for complete, unparasitized clutches—up to eight or nine 
eggs of either of the magpies. It can only be conjectured if this may 
have been due to elimination of host eggs by the parasite. 

However, in the total count of instances of all host species, the 
recorded numbers of eggs of the fosterers in the individual nests do 
not consistently follow any variation directly proportional to the 
number of parasitic eggs found with them. We may recall that in 
his discussion of the great-spotted cuckoo as a parasite on species of 
magpies and of crows, birds larger than itself, Lack (1947, p. 323) 
reasoned that it might “. . . be anticipated that the host could raise 
more young than a single Cuckoo, and in fact, the young Clamator 
does not eject the members of the host brood, which are raised with it. 
However, the argument of this paper is that the full clutch of the 
Corvid host is determined by the average maximum number of young 
which the parents can successfully raise, hence even one additional 
nestling should upset the balance. It is therefore interesting that, 
according to Baker (1942) and Jourdain (in Witherby, et al., 1938- 
41) the parent Clamator removes one egg of the host species. Jourdain 
states further that a Clamator sometimes lays more than one egg in 
the same nest, in which case it is thought to remove one host egg 
for each egg of its own.” 

Jourdain (1925, p. 657) did expressly state that the female great- 
spotted cuckoo usually removes an egg of the fosterer when laying 
one of her own, but in a later paper (1936, p. 739) he further wrote 
that “. . . in some cases the eggs of the Magpies are removed by the 
Cuckoos, for on one occasion I met with a Magpie’s nest containing 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 4I 


four eggs of the Cuckoo but none of her own. Occasionally a 
Magpie manages to keep her clutch apparently intact, though unable 
to prevent the Cuckoos from depositing an egg or two. Thus, one 
bird was flushed from a nest with ten eggs, eight of her own and a 
couple of Cuckoos’ eggs ...” It seems from this that Jourdain’s 
evidence was not unvarying, but still pointed to regular egg removal. 
Mountfort (1958, pp. 54-56) has added more evidence in support of 
this habit. He marked with indelible ink all the eggs in a number of 
parasitized magpie nests. “. .. The notes made subsequently at these 
nests . . . proved clearly that not only as many as three different hen 
cuckoos were laying in one Magpie’s nest but that, as more eggs were 
laid, so the number of the host’s eggs diminished. Moreover, on at 
least two occasions the addition of one Great Spotted Cuckoo’s egg 
coincided with the disappearance of two Magpie’s eggs...” 

The picture revealed by the present data may be viewed graphically 
in figures 9, 10, and 11, in each of which is shown the distribution 
of the actual records, the number of instances of each particular com- 
bination of egg numbers being indicated in the graphs. By contrast, 
to emphasize the scattering, uncorrelated nature of this distribution, 
the dotted line represents the theoretical arrangements we should 
expect ideally from Lack’s postulated clutch-size relationship. 

All cases falling to the left, or below, the dotted diagonal line can 
only be interpreted as in agreement with the Lack relationship, as 
they represent clutches of eggs either collected or observed, but in 
all cases it is not only possible, but even probable, that had they been 
watched for subsequent days they would have had more eggs of 
either the host or of the parasite, or both, and would thus have moved 
closer to the line. The cases above and to the right of the dotted 
line are instances of disagreement with the postulated relationship, 
and it is their frequency with relation to those that concur with the 
diagonal line, and the degree by which they exceed this relationship 
that indicates the lack of adjustment between the parasite and its 
hosts. 

It may be noted that the great-spotted cuckoo is relatively well 
adjusted in the intensity of its parasitism to magpies (fig. 12), very 
well adjusted to crows in Mediterranean lands, especially in eastern 
Egypt, and quite obviously little or not at all adjusted to crows and 
starlings (fig. 13) in sub-Saharan Africa. Because of the historical 
accident by which this cuckoo came to be studied in Spain earlier 
and more extensively than elsewhere, we have come to think of it as 
primarily a parasite on magpies and, because of that, we are apt to 


42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


think of its use of other hosts almost as relatively unusual. This 
concept is, as we now know, erroneous. As a matter of fact, the 
parasitism of this species on the magpie is remarkable in two very 


Number of Clamator glandarius e398 per nest 


Number of Corvus eggs per nest 


Fic. 9.—Distribution of Clamator glandarius eggs in nests of Corvus, and the 
number of instances of each particular combination of egg numbers. 


dissimilar ways. For one thing, the adaptive resemblance in egg 
coloration between them is so close as to imply a lengthy evolutionary 
relationship between the two birds. On the other hand, at least in 
terms of their respective present distribution, the two are sympatric 
only in a very small, peripheral portion of the range of each. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 43 


A question that may arise from a perusal of this situation, and that 
merits some discussion is the following. It may be asked whether the 
striking difference in the degree of correlation between the egg num- 
bers of the great-spotted cuckoo and of its hosts in the parasitized 
nests of the latter on the two sides of the Sahara might be explained 
by assuming that the sub-Saharan cuckoos do not have the habit of 


Number of Clamator glandarius eggs per nest 


Number of Corvus eggs per nest 


Fic. 10.—Distribution of Clamator glandarius eggs in Corvus nests in the 
Mediterranean area, and the number of instances of each particular combination 
of egg numbers. 


removing one or more of the hosts’ eggs when laying in the nest, 
as their Mediterranean counterparts are known to do. Against this 
explanation we may note that the egg-removing habit is also known 
in C. jacobinus, the most primitive member of the genus, and in C. 
coromandus, while lack of evidence on this habit in C. levaillantu 
cannot be looked upon as implying its absence. In other words, it 
appears to be a basic part of Clamator behavior, and it would be 
surprising if one population of the most advanced member of the 
genus no longer showed it. Furthermore, in numerous parasitized 


44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


nests of Spreo bicolor in South Africa, many of the hosts’ eggs were 
found to be pecked prior to the hatching of the parasites, and hence 
this damage, comparable to egg removal, could only be attributed to 


Number of Clamator glandsrius eggs per nest 


Number of Corvus eggs per nest 


Fic. 11.—Distribution of Clamator glandarius eggs in Corvus nests in sub- 
Saharan Africa, and the number of instances of each particular combination of 
egg numbers. 


the cuckoo. In the case of the long, narrow tunnels of the Spreo 
nesting sites it might be difficult for the parasite to remove the eggs, 
and pecking them may be a “substitute” behavior. To this extent it 
is evidence of the basic egg-removing habit in sub-Saharan glandarius. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 45 


Furthermore, out of 13 of the parasitized nests of the Cape rook, 
Corvus capensis, in southern Africa, described later in this paper 
(p. 99), 6 instances, containing a single cuckoo egg apiece, held from 


Number of Clamator eggs per nest 


Number of Pica eggs per nest 


Fic. 12.—Intensity of Clamator parasitism to Pica indicated by number of 
instances of each particular combination of egg numbers. 


1 to 4 eggs of the host, while in 7 nests with 2 cuckoo eggs apiece, 
there were from 1 to 3 eggs of the rook. This suggests at least a 
certain frequency of host egg removal, although it is true that other 
more intensively parasitized nests of the same species of host in 
southern Africa did not bear this out. 


46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Recent knowledge gives us no reason for assuming any density 
dependent genetic factor that may operate in such a way as to control 
and to maintain a proper “spread” of multiple parasitism with 
reference to the resultant combined clutch size of the parasite and the 


Number of Clamator glandarius eggs per nest 


1 2 
Ta 
6: a 
a San 


Number of Spreo bicolor eggs per nest 


Fic. 13.—Intensity of Clamator glandarius parasitism to Spreo bicolor indicated 
by number of instances of each particular combination of egg numbers. 


host. Indeed, it is difficult to imagine just what such a factor or 
factors might be. The available evidence, incomplete as it is, suggests 
that it is not chiefly a matter of differential development of the habit 
of host egg removal by the parasite that is responsible for the striking 
difference we have found in the circum-Mediterranean and in the sub- 
Saharan populations of Clamator glandarius. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 47 


The high incidence of multiple eggs in Clamator glandarius, par- 
ticularly in sub-Saharan Africa, raises the question as to the situation 
in the other members of the genus. A summary of all the available 
data gives the following figures, none of which comes up to those 
we have just considered. 


Clamator coromandus 


Baker’s (1942, p. 152) data on 171 nests containing 225 eggs of 
this cuckoo reveal 171 were singles when laid, and 54, or 31.5 percent 
were multiples. Of the parasitized nests 139, or 81.3 percent had 1 
cuckoo egg apiece, and 32, or 18.7 percent had been parasitized more 
than once. 


Clamator jacobinus 


Data on 220 nests containing 290 eggs of this species show 220 
were singles when laid, and 70, or 24 percent, were multiples. Inas- 
much as this is a species which has extended its range (from Africa 
to Asia) and is, in this regard, somewhat comparable to C. glan- 
darius, we may treat the Asiatic separately from the African data. 

Asia: 106 nests containing 142 eggs, of which 106 were singles 
when laid, and 36, or 25 percent, were multiples. Of the 106 nests 
parasitized, 84, or 78.2 percent, had but a single cuckoo egg each; 22, 
or 21.8 percent, had been parasitized more than once. 

Africa: 114 parasitized nests were found containing 152 eggs, of 
which 114 were singles when laid, and 38, or 25 percent, were mul- 
tiples. Of the 114 nests, 98, or 86 percent had a single cuckoo egg 
each, and 8, or 14 percent had been parasitized more than once. 

We may point out that the absence of any significant difference 
in the ratio of multiple-egg deposition in the two great segments of 
this species is decidedly different from the comparable picture in the 
great-spotted cuckoo. It adds one more support to the contention 
that the great geographic spread of glandarius and of jacobinus 
were, in an evolutionary sense, quite opposite of each other. In the 
former species it was a very recent (late) advance after a high evolu- 
tionary development had been achieved ; in the latter species it was an 
early spread prior to the evolution of egg adaptation. In glandarius 
the geographic “advancers” still reveal a lesser degree of efficiency 
in their host relationship than do their “stay-at-home” ancestors; in 
jacobinus this is not true. Of the total 220 nests parasitized by the 
latter species, throughout its range, 182 contained 1 egg each of the 
cuckoo, 21 had 2, 9 had 3, 5 had 4, 2 had 6, and 1 had 7. 


48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Clamator levaillantii 


Data on 23 nests containing 28 eggs of the stripe-breasted cuckoo 
reveal that 23 were singles when laid, and 5, or a little under 20 
percent, were multiples. Of the 23 parasitized nests, 20, or 87 percent, 
had but a single cuckoo egg each, while 3, or 13 percent, had been 
parasitized more than once (2 had 2, and 1 had 4 cuckoo eggs). 


Other Genera 


A still higher incidence of multiple eggs is known to characterize 
the brood parasitism of another genus of cuckoos in southeastern 
Asia, the koel, Eudynamis scolopacea. This large, sexually dimorphic 
cuckoo is parasitic chiefly on crows, to the eggs of which its own 
show much resemblance. In its host selection it is, thus, comparable 
to Clamator glandarius. Baker (1942, p. 153) gives the following 
data on the koel, culled mainly from observations in India. 

Out of 223 koel eggs laid in 93 parasitized crow nests, 93 were 
singles when laid, and 130, or a little over 58 percent were multiples, 
Of the 93 nests, 36, or 38.7 percent, held a single koel egg each; 57, or 
61.3 percent, contained multiple eggs. The greatest number of 
parasitic eggs in any one nest was 16, but in 75 percent of the nests 
1, or not more than 2, koel eggs were present; 8 nests had 3 koel 
eggs each, 7 had 4, 4 had 5, 1 each held 6, 7, 9, 11, and 16 koel eggs 
respectively. That Baker’s data are not atypical is shown by many 
published observations on this cuckoo by others. Thus, to take only 
a single such note, Hopwood (1912, pp. 1211-1212) found koels to 
be unusually abundant at Arakan, Burma, and to be “wasteful” of 
their eggs. In one crow nest, apparently forsaken by its builder, he 
found seven koel eggs, which appeared to have been laid by at least 
three different individuals, and none of the crow. 

In the European cuckoo, Cuculus canorus, by contrast, we find 
the birds almost always lays but one egg in a nest and it is relatively 
seldom that more than one hen uses the same nest. Thus, Baker 
listed 3,711 eggs of several races of this cuckoo in his collection, and 
from his various statements it is possible to estimate that in only 86 
out of 3,617 parasitized nests were there more than a single cuckoo 
egg. To put it a different way, out of 3,711 eggs of Cuculus canorus 
3,617 were singles when laid, and 94, or about 2.5 percent, were 
multiples. 

While in Cuculus canorus, with its wide range of egg variability, it 
is possible to distinguish multiple eggs from the same hen, from eggs 
of multiple hens, this usually is not readily feasible in some species 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 49 


of Clamator. Baker (1942, p. 152) has claimed otherwise, and he 
had behind him many years of experience, even if not of the most 
critical sort, when he wrote that “. . . several eggs of all the species 
of the genus Clamator ... may be found in the same nest, obviously 
the production of two or more Cuckoos .. . For instance, in the set 
with six Pied Crested Cuckoos’ eggs it is easy to see that they 
must have been laid, two each, by three different cuckoos...” Yet, 
elsewhere in his work he stressed the fact that each species of 
Clamator lays a single, invariable egg type. The only variations that 
might be expected would be quite minor, and in many cases these 
would hardly suffice to distinguish the eggs of individual parents. 

In the case of C. glandarius the greater variability of eggshell pat- 
tern makes it possible to distinguish between multiple eggs of the 
same hen and eggs of different individual cuckoos. Here there are 
acceptable records of more than one parasite laying in the same nest. 
Mountfort and Ferguson-Lees (1961, pp. 98-99) found as many as 
three cuckoos laying in single nests of magpies in Spain. What 
is true of glandarius, and, it seems, of coromandus, may or may not 
be true of jacobinus and levaillanti. There are not yet the necessary, 
careful observations to prove or to disprove this in these two species. 

Because of this it is not profitable at this time to attempt to 
particularize our discussion of multiple parasitism in Clamator below 
the species level. In instances of maximal numbers of eggs in single 
nests it is highly probable that multiple hens were involved, but in 
cases of two, or even three, eggs in a nest, there is no certain way of 
telling. 

Brood parasitism is a more precarious mode of reproduction than 
is self-breeding, as it involves all the risks normally attendant upon 
the nests it utilizes plus the elements of desertion of the nests or 
destruction of the eggs by the hosts. It is therefore plausible that any 
improvement, or any increased discrimination, in the matter of egg 
deposition would provide a basis on which natural selection would 
operate and, conversely, a basis from which the effects of such selec- 
tion might be inferred. By and large, the chances of success for the 
parasite tend to decrease when more than a single egg is laid in the 
same nest. Many hosts may stand for a single imposition but not for 
repetitive ones without deserting the nest; others simply could not 
hatch and rear more than one or two of the parasitic young. It 
follows, therefore, that an original tendency to lay multiple eggs in 
the same nest would eventually be modified by natural selection, and 
that the relative frequency of such multiple eggs would tend to de- 
crease in areas where selective pressure was in operation. 


50 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


The situation we have just outlined in Clamator glandarius, es- 
pecially in Africa south of the Sahara, and in Eudynamis scolopacea 
in southern Asia, appears, at first glance, to go counter to this idea. 
However, one important factor that enables their excessively multiple 
egg deposition to continue as a reasonably well-functioning habit is 
that both of them use large hosts, mostly birds as large as, or even 
larger than, themselves, that are capable of incubating successfully 
more of the eggs and of rearing more of the young of the cuckoos 
than is the case with other species of cuckoos, regularly parasitic on 
birds smaller than themselves. In the latter situation one cuckoo 
egg is often close to the limit of the hatching potential of the host, and 
in such cases multiple-egg deposition would merely bring about the 
loss of the nest and its contents. In other words, the fact that in the 
case of the great-spotted cuckoo and the koel the parasite-host size 
ratio tends to favor the host has the effect of lessening, if not eliminat- 
ing, any selective pressure against multiple parasitism. This has made 
it possible for sub-Saharan glandarius to become established over a 
vast area. It seems, however, that as the frequency with which it 
selects smaller birds, such as starlings instead of crows, as fosterers 
increases, it will again be committing itself to the selective pressure 
to which it is temporarily fairly immune. 


EGG MORPHISM 


The great development of egg morphism in Cuculus canorus, with 
its connotations of adaptive evolution, could only have come about 
from a basic wide range of original variations in eggshell coloration. 
However, this is a highly specialized species, far removed from the 
situation in the crested cuckoos of the genus Clamator. In fact, the 
very simplicity of the whole matter of eggshell coloration in Clamator 
permits some suggestive glimpses into the early stages of a process 
that has not gone far in this genus, but that has not only advanced, 
but, in doing so, has obscured its past history, in Cuculus. 

The basic, primitive, unspecialized type of eggshell in the cuckoos, 
as a family, is unpigmented, unmarked white. In this character the 
cuckoos agree with the doves, the parrots, the owls, the touracos, and 
also with the bulk of the scansorial and picarian families. The eggs 
of the relatively unchanged, nonparasitic cuckoos are either plain 
white or tinged with plain, unmarked, pale bluish, and this may be 
looked upon as the original, basic type in all cuckoos. However, 
numerous kinds of cuckoos lay eggs that are pigmented in a plain, 
overall tone, and, in the most highly specialized species, we find some 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN SI 


that lay eggs that are patterned, speckled, or blotched as well. The 
situation in the species of Clamator is discussed below in detail for 
each of the four forms, but in general it may be said that each species 
has but a single egg type, except for C. jacobinus, where we find two 
types, but which are geographically distinct, so that even there, in any 
one area there is but a single type. In C. levaillantw there is some, 
as yet not wholly satisfactory, evidence for incipient egg morphism ; 
in C. coromandus and in C. glandarius, only one type apiece occurs. 


Clamator jacobinus 


So evident does it seem that uniform white is the primitive egg- 
shell type that I am influenced by it in considering the southeast 
African (serratus) population of C. jacobinus, with its pure white 
eggs, as the old, primitive portion of that species, and its other two 
races, pica and jacobinus, with their pale greenish-blue eggs, as more 
advanced. There is no other character that lends itself to judging 
their relative phylogenetic positions. In all the great number of eggs 
of this cuckoo taken in Ethiopia, the Somali Republic, Uganda, most 
of Kenya, and in India, not a single one has been found that was not 
plain, greenish blue. One white egg, taken from the oviduct of a 
collected female, was reported from Doinyo Narok, in southern 
Kenya, by Jackson (1938, pp. 495-496). One other white-shelled 
oviduct egg obtained near Timbuctu in Mali (formerly part of French 
Equatorial Africa) by Paludan (1936, p. 292). In an earlier study 
(1949a, p. 20) I suggested that the white color of this particular egg 
might have been due to the fact that it was “unfinished,” an oviduct 
egg not quite ready to be laid, and that it might have been about 
to receive some bluish pigment. This has been countered by the re- 
sults of Harrison’s recent study (1963, pp. 154-155), which show that 
the pigment is distributed throughout the thickness of the whole 
shell in bluish eggs of this cuckoo. The example from Timbuctu, 
therefore, must be looked upon as a definitely white egg. The white 
Doinyo Narok example appears to have been even closer to laying 
time when collected. These two are puzzling records that cannot be 
“explained away” easily. The Doinyo Narok one is separated from 
the nearest (more southern) record of a pure white jacobinus egg 
by over 600 miles, the closest ones being from Nyasaland! It may be 
mentioned that the species has not yet been found to breed in 
Tanganyika or in the northern half of Mozambique, so there are no 
records of blue eggs between Doinyo Narok and Nyasaland either. 
The Timbuctu specimen is even more remote from known white eggs 


52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


of the species. Blue eggs are known from near Nairobi (Ngong), not 
too far to the north of Doinyo Narok. Clancey (1960, p. 29) has 
identified a breeding male specimen of the cuckoo from Lake Magadi, 
quite close to Doinyo Narok, as of the subspecies pica, a race whose 
eggs are bluish. 

In South Africa, South-West Africa, Bechuanaland, the Rhodesias, 
southern Mozambique, and Nyasaland, the known eggs (and there 
are many) are pure white, except for one blue oviduct egg from 
Bulaya, 8°33’S., 30°07’E., near Lake Mweru, in northeastern North- 
ern Rhodesia. It is possible that the local race of the cuckoo there 
is pica and not serratus, as it cannot be said, on present evidence, 
that pica may not be the breeding form of Ruanda-Urundi (now 
the Republic of Ruanda and the Kingdom of Burundi) and the 
eastern part of the Republic of the Congo near Lake Tanganyika, 
and that its range may extend south to Bulaya. On the map of 
the breeding ranges of the races of this cuckoo the egg color has been 
indicated, B for blue, W for white. 

Several recent authors have considered pica inseparable from ser- 
ratus, but while close I prefer to keep them distinct, as did Clancey 
(1960). As far as our immediate problem is concerned the only 
difference is that if they were merged we would have two egg types 
in one race, although geographically separate from each other, 
whereas in our present arrangement, each race has one egg type, but 
still the species has two. Inasmuch as C. jacobinus is the only species 
of the genus that has developed two distinct, and constant, egg types, 
it may be pointed out that the more “advanced” of the two, the 
pigmented, or greenish-blue, type is apparently the type developed in 
the stock that gave rise to C. levaillantii and to C. coromandus, both 
of which lay similar, unmarked bluish or bluish-green eggs. 

The origin of two egg types, or, more precisely, the advent of the 
pigmented one in a species originally laying only unpigmented white 
eggs, is a problem for the solution of which no real clues exist, al- 
though we have already noted that nonparasitic cuckoos lay eggs that 
are either white or bluish white, and that there may be a tendency in 
the basic, primordial cuculine stock as a whole to produce some blue 
eggshell coloring. However, a parallel case has recently been described 
in a totally unrelated, primitive parasitic cuckoo, the Neotropical 
Tapera naevia. This bird, parasitic primarily on furnariids and den- 
drocolaptids, all of which lay white eggs, was known to lay white 
eggs as well. Haverschmidt (1961, pp. 353-359) has found, in 
Surinam, that this cuckoo lays two types of eggs, both unmarked, 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 53 


one white, the other bluish green. Of 13 eggs taken near Paramaribo, 
6 were white, 6 were bluish green, and 1 was white with a bluish 
tinge. In other words, in Tapera naevia we have a comparable trend, 
but without geographic separation of the two types as in Clamator 
jacobinus. Other examples, among nonparasitic birds, of two such 
egg types sympatrically, are Diplootocus moussieri and Phoenicurus 
ochrurus gibraltariensis, (Etchecopar, 1942). In this connection it 
may be noted that Etchecopar (1946, p. 160) suggested that the 
blue egg color might be a result of the greater humidity in the 
more tropical areas. He cited no evidence in support of this notion, 
and, indeed, it is not clear that the more equatorial birds do experi- 
ence more humidity than do their more austral relatives. It is not 
clear just what he had in mind when he referred to the “. . . grande 
facilité la coquille a se tacher sous l’influence d l’humidite, il nait alors 
de grandes macules bleu fonce tres particulieres a ces oeufs. .. .” 
The blue eggs are not blue spotted; they are uniform in their blue 
coloration. 


Clamator levaillantii 


The eggs of the stripe-breasted cuckoo are uniform, glossy, pale 
bluish to greenish blue, somewhat pitted, and average 26 x 20.4 mm. 
Most of the recorded eggs are of this type, but there is evidence of 
some egg morphism in this species. Pale pink eggs, finely and faintly 
speckled with slightly darker pink, attributed with strong presumptive 
evidence to this cuckoo, have been taken in two nests at Kafanchan, 
Northern Nigeria, by Serle (1939, p. 689) and at Balgowan, Natal, 
by Bell-Marley (Friedmann, 1949a, pp. 44-45). Searle was of the 
opinion that the pink eggs were an adaptation to the similarly pinkish 
eggs of the local fosterer in Nigeria, Turdoides plebeja; no such seem- 
ing adaptation was involved in the Natal records, where the host (two 
cases) was the Cape robin chat, whose eggs were not pinkish, but 
pale greenish blue flecked with brown. It is true that in some in- 
stances the eggs of this bird are almost covered with light salmon- 
pink flecks. 

To these two egg types we may add that it has been suggested in 
print that this cuckoo may occasionally lay pure white eggs. The 
evidence, if it may so be termed, is far from conclusive, but the 
case may be given here. Milstein (1954, pp. 4-5) observed two 
Clamator levaillanti showing much interest in a yellow-vented bulbul’s 
nest. They repeatedly came toward the nest and each time were 
attacked by the bulbuls. The cuckoos, fluttering wildly, never at- 


54. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


tempted to fight back, even when one of the bulbuls yanked out a 
tuft of whitish breast feathers from one of the intruders. Milstein 
watched this repeated series of attacks for over an hour and a half. 
Several hours later he returned and examined the nest which he 
found contained two eggs of the bulbul and four larger, pure white 
eggs of a cuckoo. One of the white eggs was on the rim of the nest, 
almost falling out; this one Milstein was inclined to assume had been 
laid by the stripe-breasted cuckoo during his absence. It was very 
slightly pinkish, which he interpreted as indicating extreme freshness. 
The white eggs measured 25.2 to 27 by 20.7 to 22.2 mm., agreeing with 
known eggs of both levaillanti and jacobimus in size, and with 
jacobinus eggs in color. If they were laid by Jevaillantiu, they add a 
third egg type to the known range of its egg colors. 


Clamator coromandus 


The red-winged crested cuckoo lays but a single type of egg, uni- 
form greenish bluish, slightly glossy, and averaging about 26.9 x 22.8 
mm. According to Baker, who has studied it more extensively than 
anyone else, its eggs show good adaptive resemblance to those of its 
usual hosts, laughing-thrushes of the genus Garrulax. 


Clamator glandarius 


The great-spotted cuckoo also lays but one egg type, pale greenish 
white to pale greenish gray, abundantly spotted and flecked with 
fairly evenly distributed tiny dots and larger markings of various 
shades of yellowish brown, reddish brown, umber, grayish brown, 
and slate gray, many of the blotches with a pale lilac tone or under- 
marking. In many eggs the blotches tend to be more numerous to- 
ward the large pole and in some they almost fuse to form a ring there, 
but in others there is no such local concentration. In size they vary 
from 29.1 to 35.2 by 22.6 to 26.5 mm. 

This is definitely an “advanced” egg pattern, more developed than 
the uniform ones of the other three species of Clamator. In a sense 
the pinkish eggs of levaillantu with their faint speckling of darker 
pink may be looked upon as foreshadowing the development that took 
place in glandarius. Makatsch (1955 pp. 218-220) has discussed the 
evolution of spotted from uniform egg coloration in cuckoos, and 
concluded, as I do, that these patterned eggs represent the climax 
stage, and not, as Baker and von Boxberger did, the early stage. 

The eggs of the great-spotted cuckoo are smaller than those of its 
corvine hosts, although as large as, or larger than, those of its sturnid 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 55 


fosterers. In the other species of Clamator their eggs are generally 
larger than those of their victims. 

The egg pattern of the great-spotted cuckoo is so closely adapted 
to that of the magpie, its primary host in Spain, that it has often 
been mentioned as an example of perfected adaptive evolution. As 
we have had occasion to discuss elsewhere in this paper, there can be 
little doubt that it developed with the magpie as the chief host, and 
that the range of the cuckoo has subsequently been extended to areas 
outside the range of this fosterer. 

As is so frequently the case, adaptations that seem, to the in- 
vestigator, obviously functional, and, hence, readily understandable, 
suddenly seem to be unimportant and unnecessary when the organism 
possessing them moves into a different situation. In Portugal, and 
also in parts of Spain, the great-spotted cuckoo parasitizes the blue- 
winged magpie, Cyanopica cyanus cook, and, in Egypt and in sub- 
Saharan Africa it uses even more divergent hosts. Etchecopar (1946, 
p. 165) admitted the striking similarity in the eggs of this parasite 
and those of Pica pica, but was moved to state that when the host 
was Cyanopica it was difficult to see any special resemblance (“. . . 
ou il est difficile de voir la moindre trace d’adaptation . . .”). 

Recently Tomlinson (1962, p. 260) has stated that he found the 
great-spotted cuckoo parasitized the black crow (Corvus capensis) 
and the pied crow (Corvus albus) in South Africa, and that its eggs 
varied in color to match those of the host, pinkish in the case of 
capensis, greenish in albus! Fraser (1962, p. 343) and Calder (1962, 
p. 344) rightly questioned the identification of the pink “cuckoo” egg 
noted by Tomlinson. In the light of all we know at present there is 
no reason for thinking that the great-spotted cuckoo lays more than 
one type of egg. However, as I discussed in my earlier account 
(1949a, pp. 44-45) and in the description of egg morphism in Clama- 
tor levaillantii in the present paper, there are on record some four 
instances of pinkish eggs attributed with some presumptive evidence 
to the stripe-breasted cuckoo. In addition to these, Priest (1934, vol. 
2, pp. 238, 245-247) reported a speckled, pinkish egg, supposedly of 
a cuckoo, in a nest of a pied crow (which lays greenish eggs that 
contrasted very strongly with it). He suggested that it might be 
either the stripe-breasted or the great-spotted cuckoo, and noted that 
its size, 29 x 23 mm., favored the latter identification. In discussing 
this record I suggested that it might have been a “runt” egg of the 
black crow ; this suggestion would be even more appropriate in Tom- 
linson’s record, as there the nest and the egg would both be identified 
to the same species. 


56 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


The problem of egg morphism, is, as we have seen, not a prominent 
one in the genus Clamator. Still, by virtue of what it reveals in 
Cuculus, where it is well developed, it raises one further point that 
is worth discussing here. In the European Cuculus canorus we have 
a species with a wide range of eggshell coloration, and we have rea- 
sonably good evidence for the existence within the species of numbers 
of different gentes, each specific on a definite host species. The 
existence of two or more gentes sympatrically increases the efficiency 
of each in exploiting host egg mimicry, and allows a greater popula- 
tion of cuckoos to exist in a limited area. Regardless of the reality 
of these gentes, and no doubts as to their existence are here implied, 
it is true that the only way in which we may be made aware of them 
is by the fact that each individual hen cuckoo lays a single egg type, 
and is specific in its host choice, while the species lays a wide range of 
eggs and uses many species of hosts. From this it follows that if 
Cuculus canorus laid but a single type of egg it could still have gentes, 
but we would be unable to sense their existence and would have no 
reason even to conceive that there might be any. In the crested 
cuckoos of the genus Clamator we have seen that each species has 
but a single egg type, except for incipient variation in levaillantu and 
geographic variation in jacobinus. Consequently, no suggestion of 
gentes has ever been raised in studies of this genus, and, indeed 
there would seem to be nothing on which natural selection might 
have favored the development of such infraspecific categories. Still, 
we cannot rule out the possibility that the hens of each of the four 
species may be individually host specific (as in glandarius in the 
Iberian peninsula, where all the hens are essentially specific on the 
same host, the magpie). If this should prove to be the case, we would 
have, in effect, undistinguishable but yet actual gentes in the species 
of Clamator. 

In this connection we may recall Southern’s (1954, p. 223) con- 
clusion about gentes in Cuculus canorus to the effect that those gentes 
which are highly adapted in egg mimicry probably thereby sacrifice 
a certain degree of what plasticity their ancestors may have had, and 
with it the ability to turn successfully to new and very different hosts. 
In effect, Clamator glandarius in the Iberian Peninsula and in ad- 
jacent parts of western Mediterranean Africa is comparable to a 
single highly specialized gens in Cuculus canorus. Yet it has been 
able to utilize remarkably dissimilar hosts in sub-Saharan Africa. 

Whereas in Cuculus there is a definite trend for small egg size, 
relative to the size and weight of the adult bird, a trend which has 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 57 


enabled the members of that genus to parasitize birds much smaller 
than themselves, no such reduction in relative egg size is found in 
Clamator. For that matter, diminution of egg size is found in all 
species of Cuculus, but not in other genera of parasitic cuckoos. 
Within the species Clamator jacobinus we do find a slight geographic 
reduction in egg size, but nothing comparable to the situation in 
Cuculus. The greenish-blue eggs of C. 7. pica average slightly smaller 
than do the white ones of C. 7. serratus, but the difference, while 
significant and, in an evolutionary sense, suggestive, is not trenchant 
as there is extensive overlap in the sizes of the two groups. Thus, 
eggs of southern African serratus vary from 24.1 to 28 by 20.8 to 
23 mm., with an average of 25.5 by 22 mm.; while those of pica 
from Ethiopia vary from 22 to 25 by 20 to 22 mm., with an average 
of 23.5 by 20 mm., and eggs of pica from India range from 21.9 
to 28 by 17.6 to 21.4 mm., with an average of 24.3 by 19.4 mm. The 
eggs of south Indian, nominate jacobinus are slightly smaller still, in 
keeping with the lesser size of the birds of that race. 

The development of brood parasitism and the varying features it 
exhibits in different genera of cuckoos make it clear that each genus 
needs to be studied independently before we may attempt to gen- 
eralize. In Clamator the evolutionary history of the egg size and 
coloration differs from that in Cuculus; it reveals no marked reduc- 
tion in size and while it has achieved remarkable adaptive similarity 
to those of its hosts in color it has done this without developing 
any extensive egg morphism within any of its species. 


INCUBATION PERIOD 


Rapid development of the embroyos, or shortening of the incuba- 
tion period of the eggs, is generally considered as advantageous to a 
parasitic bird, as it may result in the parasite hatching before its nest- 
mates and thereby gaining a “start”? on them. This would seem par- 
ticularly pertinent to parasites that do not attempt to evict their nest- 
mates but grow up with them. If this concept were infallible we 
might expect to find a slight, but significant, change in this direction 
from the most primitive species of Clamaior, the pied cuckoo, C. 
jacobinus, to the most advanced, the great-spotted cuckoo, C. glan- 
darius. The few facts available, are, surprisingly enough, contrary to 
this postulated condition. The incubation period for jacobinus, as 
worked out carefully by Liversidge (1961, p. 624) in four instances, 
was between 11 days + 14 hours and 12 days + 12 hours, while in 
glandarius, Mountfort (1958, pp. 54-56) found it to be 14 days. As 
yet, no data are available on the other two species. 


58 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


It should be kept in mind that even with its longer incubation of 14 
days, glandarius averaged 3 to 4 days less in its incubation period than 
the magpie hosts it used in Spain, where Mountfort studied it. This 
would be true for its other corvine hosts as well. It may be that 
the greater size of glandarius, as compared with jacobinus is reflected 
in its longer incubation period, but this is by no means established. 
The incubation periods of the various hosts—shrikes, bulbuls, and 
babblers—that are used by jacobimus are shorter than those of the 
magpies and crows used by glandarius. It may be that the change 
in host choice in the latter offset any advantage that more rapid 
embryonic development might otherwise have given it. 


HOST-PARASITE NESTLING RELATIONSHIPS 


The development of brood parasitism in Clamator has not included 
the development of eviction by the newly hatched young. 

In some parasitic cuckoos, notably those of the genus Cuculus, the 
newly hatched bird, while still featherless and with still unopened 
eyes, evicts from the nest in which it finds itself other nestlings and 
eggs. This it does by pushing against them and slowly burrowing 
under them until it gets them on its back, when it climbs slowly to 
the rim of the nest, where with a final and violent, muscular effort it 
heaves them out of the nest. Thereupon, it falls back into the nest, 
where it rests momentarily before tackling the next nestmate. This 
evicting behavior usually lasts only until the fourth day of life, after 
which the nestling cuckoo tolerates anything that may be in the nest 
with it. 

This highly peculiar, and obviously instinctive, behavior is one of 
the features associated with brood parasitism that has not been de- 
veloped by the species of Clamator. In C. glandarius and C. coro- 
mandus we have ample numbers of observations to be able to state 
that usually eviction by the newly hatched cuckoo does not take place. 
In glandarius, the elimination of the host young that often happens is 
due to their being either starved or smothered by their parasitic nest- 
mates, and their dead bodies removed by their own parents. Thus, 
Mountfort (1958, pp. 54-56) wrote that in only one magpie nest in 
Spain did he find young of the host and of the great-spotted cuckoo 
together, and this was only for a very brief period, as the nestling 
magpie was gone 2 days later. It had hatched 3 days after the eggs 
of the parasite, and the emerging nestling was never able to overcome 
this disadvantage. Mountfort concluded that the shorter incubation 
period of the cuckoo (shorter by 3 days) doomed the young magpie, 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 59 


and that only such young magpies as hatch from eggs laid well before 
those of the parasite can have any chance of surviving. 

Similarly in South Africa, Miss M. Courtenay-Latimer (in lit.) 
watched a nest of a hoopoe, Upupa e. africana, that originally con- 
tained four eggs of the host and one of the parasite. The cuckoo egg 
hatched on the same day as the first host egg; the other three hoopoe 
eggs hatched on the following 2 days. The young hoopoes disappeared 
3 days later, but their eviction or removal was not observed. The fact 
that they did not disappear until 3 days after hatching argues against 
eviction by the young cuckoo, and makes it appear likely that they 
perished in the matter of food competition with it. 

Another case in point is one reported by Meyer (1959, p. 85). 
Near Que Que, Southern Rhodesia, he found a nest of a glossy 
starling, Lamprotornis chalybeus, containing a young great-spotted 
cuckoo several days old (the quills just appearing on the wings and 
tail), a young starling, fully feathered and estimated to be 10 to 14 
days old, a dead young starling, estimated to have been dead for 
from 2 to 3 days, and trampled into the bottom of the nest, and one 
cracked, unhatched starling egg. Three days later the young cuckoo 
and the young starling were still in the nest; on the following day 
only the parasite was there and was seen being fed by the foster- 
parents; there was no trace of the missing young starling. Six days 
later the cuckoo, still in the nest, was fully feathered; the following 
day the nest was empty, but two days later the starlings were seen 
feeding the fledged parasite. Here we have another example showing 
the absence of eviction by the young cuckoo. In this particular in- 
stance it would appear that the starling must have hatched some days 
before the cuckoo. 

In C. levaillantii we still lack such observations but there is no 
reason for thinking that the picture there is any different. Actually 
there are unpublished data of N. R. Hyslop (editorially referred to 
in Bokmakierie, vol. II, 1959, p. 19) that are said to confirm the 
absence of evicting behavior in this species. In the case of the pied 
crested cuckoo, C. jacobinus, alone, has anyone even suggested that 
eviction may take place and even here there is no conclusive evidence 
for it. In the few instances where this has been suggested it was not 
possible to establish that the ejection was deliberate or even that it 
was done by the nestling cuckoo. 

Skead (1962, pp. 72-73) observed a nest of the forktailed drongo, 
Dicrurus adsimilis, containing a young jacobin cuckoo and three 
drongo eggs. Two days later one of the eggs was found on the 


60 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


ground below the nest; the following day another egg was found 
there; still 4 days later the drongo chick, which had hatched in the 
meantime, was lying below the nest. The finding of these eggs and 
of the nestling drongo under the nest suggested (but only suggested) 
that the eviction was done by the young parasite. Skead was careful to 
point out that this inference required proof. 

The estimated age of the young cuckoo was three days when the 
first of the host’s eggs was found below the nest, and 8 days when 
the drongo chick was so recorded. If the eviction was done by the 
young cuckoo, this would imply a much longer duration of the 
evicting instinct than occurs in Cuculus, a genus in which the habit 
is well established. 

It is impossible to state that the young cuckoo was not responsible, 
but there are other cases known where eviction definitely did not 
take place. Skead himself (1951, pp. 172-173) described a case in 
which a nestling jacobin cuckoo tolerated eggs and young in the 
same nest for up to 4 days, and another nest in which another young 
parasite made no attempt to evict the eggs for 4 days during which 
the nests were under observation. It not infrequently happens in a 
crowded nest that activity by one of the nestlings may sometimes 
result in the accidental pushing of one of the eggs or young out of 
the nest. Also, in parasitized nests, the young parasite often is larger 
and grows relatively faster than its nestmates and by successful com- 
petition with them for the food brought by the adults may starve them 
to death. In such cases the dead young are removed, not by the young 
parasite, but by their own parents as a matter of nest sanitation. 

Furthermore, there are observations of still other nests in which 
the host young and the young jacobin cuckoo grew up together to the 
fledgling stage, and were seen together even after leaving the nest 
(Godfrey, 1939, p. 3; Bates, 1938, p. 125). These are clearly cases 
in which no eviction by the young parasite took place. In the two 
instances described by Skead, if any eviction by the young cuckoo 
might have been involved, it did not occur for some days after hatch- 
ing, which is not the case in Cuculus. On present evidence it is doubt- 
ful that young Clamators have the habit of methodically and deliber- 
ately ousting their nestmates during the early stages of their nestling 
life. 

An unusual type of host-parasite nestling relations was observed in 
India by MacDonald (1960, pp. 174-175). He watched a nest of a 
jungle babbler, Turdoides striatus somervillei, that contained a nest- 
ling jacobin cuckoo and a young babbler. The young parasite was 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 61 


more advanced in its development than its nestmate and it was found 
to leave the nest and forage and then return to it to be fed by the 
foster-parents, who were more or less bound to the nest by the pres- 
ence in it of their own less advanced chick. This was noted re- 
peatedly, and suggested a degree of resourcefulness quite unexpected 
in a bird at the nestling-fledgling stage. It also was another instance 
of mutual survival, or, in other words, of the absence of evicting 
behavior by the young cuckoo. 


FLEDGLING FEEDING BY ADULT CLAMATORS 


The feeding of well-grown, fledged, young crested cuckoos by 
adults of their respective species has been reported for two of the 
four species of the genus. In no case has convincing, corroborative 
evidence been placed on record, but inasmuch as such behavior has an 
evolutionary interest as atavisms it is necessary to mention them 
here. The data are as follows: 

In India, Gill (1925, p. 283) claimed that he had often watched 
adults of the jacobin cuckoo, C. jacobinus feeding fully fledged 
young of their own kind, and that koels, Eudynamis scolopacea, do 
this even more often and regularly. If this is correct the observations 
have not been reported subsequently by other field students, and it is 
possible that Gill mistook an adult female for a fully grown young 
merely because he saw it being fed by another jacobin cuckoo. It is 
known that this cuckoo does indulge in courtship feeding and it may 
be this was what Gill really saw. Thus, in South Africa, Godfrey 
(1939, p. 26) watched a melanistic and a pale morph of the jacobin 
cuckoo feeding on caterpillars on the ground. The pale bird was seen 
to pick up a caterpillar, pass it a few times back and forth along its 
beak, and then to approach the black-phase bird with this in its bill. 
It mounted the latter, gave it the caterpillar, and then mated with it. 

Similarly, many years earlier, in northeastern Africa, von Heuglin 
(1869-1873) wrote of the great-spotted cuckoo, C. glandarius, that 
he thought it occasionally took care of young of its own kind. It is not 
possible to decide from his wording exactly what actions he witnessed, 
but it may have been more a matter of premigrational flocking, as no 
evidence of actual feeding of the young by the adults has been noted 
since anywhere in its range. That this may be the real condition from 
which von Heuglin gathered his impression is suggested by an ob- 
servation of Ivy’s (1901, p. 22), who, in eastern Cape Province, 
found a pair of adults with five young birds late in February. He 
considered that “. . . the old birds collected their broods previously 


62 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


to migrating...” The most that may be said in the case of this 
cuckoo is that fledgling feeding is yet to be proved. 

The mere act of courtship feeding, as shown in the jacobin 
cuckoo, is in itself an atavistic behavior, and if it should eventually be 
found to be coupled with even occasional feeding of fully fledged 
young (other than by mistaking them for adult females) this 
would further strengthen the suggestion that Clamator is a fairly 
primitive genus of parasitic cuckoos. 


PLUMAGE VARIATIONS AND THEIR SIGNIFICANCE 


Before we discuss the polymorphisms which have been well es- 
tablished in two of the four species of Clamator, jacobinus and levail- 
lanti, it is necessary to review the extent and the nature of the 
variations found in the “normal” plumages as well as in their 
melanistic phases. 

Although not pertinent to the immediate problem of polymorphism, 
the plumage variations of the climax species C. glandarius also may 
be described and discussed in this section of the paper, as they have 
pertinent evolutionary implications as well. C. coromandus calls for 
no comment here. 


Clamator levaillantii 


The “normal,” i.e., the white-breasted, plumage has the entire un- 
derparts from chin to vent white, with black streaks on the chin, 
throat, breast, and upper abdomen, these streaks narrowing to shaft 
lines on the feathers of the sides, flanks, and undertail coverts, the 
streaks heaviest on the throat and upper breast; all the rectrices with 
broad white tips crossing both webs, and with a white patch on the 
outer eight primaries, this patch not visible from above in the closed 
wing because of the overlapping secondaries ; underwing coverts white 
with very variable amounts of black markings. The entire upper 
surface of the head and body is solid black. 

Here we find variations in the mental and pectoral stripes from 
specimens in which these dark marks are narrower than the white 
interspaces (the lateral portions of the feathers are here involved) 
to others in which the dark marks are broadened to the degree that 
they become practically coalesced to form almost solid black areas on 
the chin and upper throat. Although no geographic races of the 
stripe-breasted cuckoo are recognizable, much.attention has been paid 
by authors to the degree of the variation, especially in the heaviness, 
the length and width, and the darkness of the blackish stripes on 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 63 


the throat and breast of the adult birds. Gyldenstolpe (1921, pp. 246- 
247) considered some of this variability to be a matter of age, the 
stripes being narrower in younger birds than in older ones. He also 
thought the stripes were broadest and darkest in the males. Examina- 
tion of large numbers of specimens, especially in London, convinced 
me that no correlation with age or sex may be maintained. Chapin 
(in litt., 1961) wrote that while these stripes were variable through- 
out Africa, they seemed to average heaviest in specimens from the 
northeastern part of the continent—Ethiopia and Somalia. In five 
specimens from there in the American Museum of Natural History 
the throat was so broadly streaked as to be almost completely black. 
Examples of this extreme type came from Giamo, Bissidimo, Godja- 
Mariam, and Maraco, in Ethiopia, and from Warsangli-Mush Hated, 
5,000 feet, in the Somali Republic. Another equally dark bird came 
from much farther to the south, from Machame, Tanganyika; in it 
the chin was fairly solid black, the throat less solidly so. In the 
British Museum I have seen specimens just as heavily marked with 
black from Mount Lotuke, in the Didinga Mountains of Sudan; also 
from Usambara Mountains, Pangani River, Tanganyika, and even 
one from as far to the southwest as Damaraland. All of these ex- 
amples were adult males. Furthermore, in all the areas involved, 
other examples were much less heavily striped than those mentioned 
here. 

One of the palest birds seen was from Tembura, in the Bahr-el- 
Ghazal Province of Sudan. It had only narrow black streaks on the 
chin and throat, these disappearing on the upper breast, in marked 
contrast to the Mount Lotuke dark extreme which not only had the 
black streaks almost coalesced on the chin and throat, but had these 
markings continuing very broadly over the entire breast, tapering 
caudally, but with narrow black shaft streaks on the entire abdomen, 
sides, flanks, and thighs. These abdominal shaft streaks were even 
more pronounced in the Usambara birds. Perhaps the extreme 
variant in this character of all the birds seen was a female from 
near Mombasa, collected together with five in the melanistic “albo- 
notatus’ plumage phase. In it the entire underparts were heavily 
streaked with black, from chin to vent. One from Kyambu, near 
Nairobi, Kenya, described by van Someren (1922, p. 51), was said to 
have the black stripes reaching the abdomen as well. 

Turning now to the opposite extreme, a male from Gunnal, in 
Portuguese Guinea, from the other side of the continent, was almost 
as lightly marked as the Tembura bird. Recently, in a report on Gabon 


64 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


specimens, Rand, Friedmann, and Traylor (1959, p. 271) noted the 
great variation in the underwing coverts and axillars; from almost 
wholly dull white to largely blackish. The Tembura specimen in 
London had no blackish at all under the wings. Years ago, a specimen 
from Danger River, Gabon, was used as the basis for the description 
of a very pallid “race” caroli. This specimen was studied in 1962 
together with the Tembura one, as well as with a very extensive series 
of others. It was found to be paler, less streaked with black on the 
chin and throat, but the difference between it and the Tembura bird 
was not great enough to suggest that it might represent a distinct 
race. The type of caroli had the terminal white spots on the rectrices 
larger than in the Tembura specimen. The extreme pallor of caroli 
suggests, if anything, that just as in coastal Kenya Jevaillantii may 
produce completely melanistic phases (albonotatus), so elsewhere 
it may almost approach jacobinus in its lack of dark ventral mark- 
ings. Certainly the geographic distribution of color extremes— 
darkest birds from Ethiopia, Somali Republic, Kenya, Sudan, Tan- 
ganyika, and Damaraland in South-West Africa, and lightest ones 
from Sudan, Gabon, Portuguese Guinea, and Rhodesia—indicate that 
they are haphazard in their occurrence, and hence not significant 
taxonomically. In none of these areas are the birds uniform in their 
variational trends. 

Another variable character of the “normal” phase of levaillanti 
is the length of the feathers forming the crest. Here again, examina- 
tion of long series from all parts of the range, tends to rule out the 
supposed significance of any local extremes. At one time in these 
investigations it seemed that birds from Ethiopia tended to have on 
the average longer crests than birds from elsewhere, but measure- 
ments failed to corroborate this. 

The melanistic phase, originally described as a separate species 
under the name C. albonotatus, has been found (with one exception) 
only in the narrow coastal belt of Kenya, south to Usambara Hills, in 
Tanganyika, and north to southern Jubaland in the Somali Republic. 
I have been able to examine 21 of the 26 recorded specimens known 
to me. Since relatively few investigators have examined this plumage, 
and none with as ample material, the following notes on its variations 
are here recorded. 

In general this phase may be described as being black all over, 
except for terminal white spots on the outer rectrices (this varies 
from the two outermost pairs to the four outermost pairs in different 
individuals), and a white patch on the inner webs of the eight outer 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 65 


primaries. However, in four of the specimens there was some whitish, 
in the form of edges on the feathers of the throat, breast, abdomen, 
and undertail coverts. The amount of such whitish areas varied in- 
dividually ; in one unsexed example from Takaungu the feathers of 
the abdomen and the undertail coverts were broadly edged with gray- 
ish white; in another from the same place these pale edgings were 
very narrow. In this connection, it may be recalled that in our discus- 
sion of the variations in the normal phase, we described (supra) 
one example, collected together with melanistic birds, in which the 
entire underparts were white, heavily streaked with black from chin 
to vent. This specimen might equally well be described as a melanistic 
polymorph in which all the ventral feathers had white edgings. 

The white spots at the tips of the outer tail feathers not only vary 
in the number of rectrices on which they occur, but also in the size 
of the individual spots; in some cases they are restricted to the outer 
web, in others they extend across both webs of these feathers. The 
presence of these tail spots is the only constant difference, aside from 
the total size of the bird, and its corresponding wing and tail dimen- 
sions, between this phase of Jlevaillantit and the corresponding 
melanistic morph of C. jacobinus serratus. Very occasionally a speci- 
men of levaillantii may lack these white tail spots, as in one taken 
near Lake Chahafi, Kibwezi, southwestern Uganda, reported by Pit- 
man (1931). He implied that there was another similar one from 
former French West Africa in the British Museum, but I failed to 
find it when I examined the series there in 1962. 

A number of the specimens of this black phase studied were in 
various stages of molt. They revealed that the juvenal plumage (or, 
at least, subadult plumage) is uniformly dull fuscous brown on the 
entire upper side of the body and head. In some examples the entire 
underparts, as well, were of this color, but in others the abdomen 
and sides were paler, more of a dirty brownish white. Even these 
young specimens had the white wing speculum as in the adults. 


Clamator jacobinus 


Unlike C. levaillantii, this species varies geographically in its plum- 
age characters, and has been divided into three recognizable races. 
Typical jacobinus, a small race (wings 135.5 to 150; tail 146 to 172 
mm.), with white throat and breast, the feathers of the lower throat 
and breast with the faint, dusky shaft streaks either practically want- 
ing or pale and very narrow, hairlike lines, occurs in southern India 
and Ceylon, and is partly migratory as some of its members winter in 


66 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Africa. Another race, pica, similar in color characters, but larger 
in size (wings 149 to 164 mm.; tail 173 to 197 mm.), has two discon- 
nected breeding ranges; one part breeds in Baluchistan, and in 
northern India from the Kashmir Himalayas west to West Pakistan, 
and Nepal, south to United Provinces, and Kutch, and winters 
largely in Africa, south of the Sahara; another large segment of the 
subspecies breeds in sub-Saharan Africa from Senegal east to Ethio- 
pia and the Somali Republic, southward to Angola, northern portions 
of Northern Rhodesia, and southern Kenya. The third race serratus, 
similar in size to pica, is dimorphic in the southeastern portion of its 
range. Its pale morph is like pica but with the shaft streaks of the 
throat and breast feathers darker and often heavier, and the entire 
pectoral area and the sides of the body tinged lightly, or more 
heavily, with grayish. The dark morph is black except for the white 
wing patch. This is the race of Africa south of the Zambezi River, 
where it occurs only during the breeding season, October to March, 
wintering in tropical Africa along with pica and jacobinus. In the 
western part of its breeding range, especially in South-West Africa, 
it approaches in coloration the race pica; but the birds breeding there 
agree better on the average with pale serratus, and, furthermore, 
they lay white eggs like serratus (pica and jacobinus lay blue-green 
eggs). There is reason to believe that serratus may breed north to 
southern Kenya, but no form of the jacobin cuckoo has yet been 
found to breed in Tanganyika. 

The chief evolutionary interest in the facts discussed in this section 
is that the range of variation in C. Jevaillantu, from the heavily striped 
pattern to the almost unstriped anterior underparts, as in the type 
example of “carol” and in the Tembura example, comes very close 
to bridging the gap in this character between this species and C. 
jacobinus (fig. 14). We have also seen that the range of variation 
in the pale morph of the latter varies from birds with the chin, throat, 
and breast entirely white, devoid of any streaks or marks, to others 
with well-developed, but narrow, dusky shaft lines on the feathers 
of these parts. The lack of greater difference between the most defi- 
nitely lined jacobinus and the least striped Jevaillantii almost es- 
tablishes a variational continuum. There is a more sizable break be- 
tween the dimensional characters of the two, but here again, the gap 
between the smallest Jevaillantii and the largest jacobinus is not very 
great in proportion to their size. Also, the juvenal plumages of the 
two species, and the melanistic phases are extremely similar, 
save in dimensions. The eggshell color of C. jacobinus in India and 
in northeastern Africa is similar to that of levaillantii. It is true that 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 67 


in the more southern population of jacobinus (serratus) we find a 
markedly different egg type, but the difference here is just as great 
between sections of this species as between it and Jevaillantu. There 
can be no reasonable doubt as to the close relationship of the two 
species. 

Inasmuch as it is thought that the southern serratus is nearer to 
ancestral C. jacobinus than are the other races of the species, it would 
follow that the evolutionary trends in plumage pattern involved a 


3 4 


Fic. 14.—Variation in pectoral markings. 


Clamator levaillantti: 1, Darkest; 3, palest. 
Clamator jacobinus: 2, Darkest; 4, palest. 


loss of the pectoral shaft stripes giving rise to C. 7. pica and C. 7. 
jacobinus on the one hand, and a great intensification of the same 
character giving rise to C. levaillantii. From the serratus-like pri- 
mordial stock both developments arose and diverged. 

It may be remarked that it seems (to human eyes) that some 
plumage characters of no great biological significance are tenaciously 
retained during evolutionary changes while others of no more obvious 
utility are altered. An example of the former is the white wing patch 
found in both the pale and the melanistic morphs of both jacobinus 
and Jevaillantii, although this may have a recognition and a releasing 
function in both species. 


Clamator glandarius 


The one feature of greatest evolutionary significance, or, to put it 
in a different way, the one phylogenetic clue of greatest interest in 
the plumages of this, the climax species of the genus, is the fact that 


68 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


its juvenal plumage has the remiges extensively reddish, suggesting a 
relationship with C. coromandus. Aside from this, there is only one 
point worth mentioning in any detail, the possible geographic variation 
in the adult C. glandarius. 

Clancey (1951, p. 141) separated the population breeding in Africa 
south of the Sahara from the nominate, Mediterranean basin birds, 
and gave them the name choragium. The characters of this southern 
race were smaller size and warmer buff on the throat and breast with 
less pronounced dark shaft streaks on the feathers. Judging by 
Clancey’s account, Gilliard examined for him the material, totaling 
101 specimens, in the American Museum of Natural History and 
apparently agreed in considering the two populations as distinguish- 
able. In the course of my studies I have examined nearly 200 addi- 
tional examples, including 177 in the British Museum, and I found 
that there was very extensive overlapping in all these characters. The 
buff tone of the throat and breast in fully adult birds may average 
very slightly warmer in choragium, but great care must be taken to 
compare birds of the same age, as the young of both populations are 
warm tawny buff on the throat and breast and adults of both largely 
lack this color and are not more than pale buffy cream with a slight 
ashy tinge. I could not find any constant difference in the development 
of dusky shafts on the feathers of this area in the two groups of 
birds. Furthermore, the size characters showed more overlapping than 
the figures given in Clancey’s paper, and at best were not more than a 
slight average difference, hardly enough to warrant nomenclatorial 
recognition. 

Using only breeding season examples, to eliminate possible migrants 
of the other population, I found that in males the wing length varied 
from 190 to 223 in Mediterranean birds, from 185 to 218 in southern 
ones : The tail length in these males varied from 186 to 226 in Mediter- 
ranean examples, from 181 to 219 mm. in southern ones. The females 
showed similar overlapping. Finally, as a test, I found myself unable 
to relegate the majority of specimens to subspecies without looking 
at the localities on their labels. I therefore do not accept choragium 
as a valid race, at least not as a race of utility in taxonomic work. 

It may be mentioned that I examined these birds with the expecta- 
tion that choragium would be corroborated, as I was aware of interest- 
ing differences in the host relations and host-parasite adaptations in 
the two populations. I can only look upon the results as providing 
unexpected support for the relative recency of this cuckoo as a breed- 
ing bird in sub-Saharan Africa, a conclusion already suggested by a 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 69 


study of their relatively poorly adjusted host relations in the African 
part of their range. 

The situation present in the sub-Saharan Clamator glandarius as 
compared with that in the circum-Mediterranean segment of the 
species is reminiscent of Thorpe’s (1930) biological races in insects. 
He called attention (p. 189) to instances among species of insects and 
allied groups in which geographically isolated populations with little 
or no structural or pigmentary differences were definitely differ- 
entiated biologically or behaviorally. These he considered essentially 
the same as subspecies, but in which the racial characters are aspects of 
the living rather than of the preservable parts of the specimens. As 
I have already indicated I do not think it advisable to give separate 
taxonomic or nomenclatorial rank to the two sections of Clamator 
glandarius, as the size and color characters ascribed to choragium are 
too slight and the overlap too great to make it a “usable” subspecies, 
although choragium does reveal a trend toward differentiation, as yet 
not well developed, and there is in its life history a biological difference 
in its host relations and in its range of host selections. The very fact 
that its behavioral character results in relatively poor coordination 
in its host relations, coupled with the independent fact that its struc- 
tural modifications are still only faintly developed, suggests that 
“choragium” is a new, possibly as yet only an incipient, race. 

One further item in the plumage cycle of C. glandarius deserves 
mention. Its juvenal plumage is blackish on the top of the head and 
nape, not gray as in the adult. Because of this, Jourdain (1925, p. 
661) suggested that this might have been produced through adaptive 
evolution to achieve some degree of resemblance to the plumage of 
the nestlings of its Palaearctic corvine hosts. Jourdain stressed that 
the only conspicuous parts of the young bird while in the nest are 
the crown and nape, and he accordingly discounted the pale tawny 
chin and throat coloration. However, it may be recalled that the 
critical moments for the nestling are when the foster-parent comes 
with food. At such times the young cuckoo, as well as host young, 
raise their heads and open their mouths widely and clamor for food, 
and at such moments the throat would be no less visible than the 
crown and nape. I cannot help but consider Jourdain’s suggestion 
as an “armchair speculation,” and it certainly does not apply to the 
sturnid hosts the parasite uses in Africa. On the other hand, Cott 
(1940, p. 422) was convinced enough to write that the nestling of 
this cuckoo has a plumage “. . . whose crown has been influenced by 
natural selection, but whose throat has been neglected—so that while 


7O SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


the latter resembles that of its parents, the former imitates that of 
its nest-mates .. .” 


POLYMORPHISM 


Two of the four species of Clamator have produced well-established 
melanic morphs or plumage phases in restricted portions of their 
respective ranges. These two are C. jacobinus and C. levaillantu, and 
it so happens that the black phases of the two are extremely similar. 
Their geographic ranges are quite dissimilar in extent, however; that 
of jacobinus occupies a large area of southeastern Africa, while the 
corresponding one of Jevaillantii is restricted to a narrow coastal 
strip of northeastern Tanganyika and of Kenya. 

Both of these morphs appear to be, in every way, good examples 
of polymorphism in the sense defined by Ford (1945, p. 73). In his 
definition a polymorphic species is one in which there are two or more 
distinct phases or forms simultaneously in the same habitat, even in 
the same deme and the same local population, and in which these 
forms occur in sufficient numbers that even the least common of 
them is too numerous to be accounted for by a continuous series of 
recurrent, identical mutations. To this basic concept may be added, 
as was pointed out by Carter (1954, p. 259) the further thought that 
the characters of these polymorphs must be such that they do not 
blend on crossing ; in other words, they must be controlled by single 
genes, or at least by small groups of genes that are closely linked in 
their mode of inheritance. Otherwise, the normal interbreeding that 
goes on in any local population of a species would tend to transform 
these polymorphs into a broad but continuous spectrum of variation. 
In the case of C. jacobinus we have ample field evidence that crossing 
between the two color morphs takes place frequently and yet no 
intermediate plumages are known. For C. levaillantii we still lack 
field observations of similar crossing between the morphs as little 
work has been done in the restricted area of its polymorphism.2 In 


1 As described in our discussion of migratory behavior (p. 86) Lamm noted 
seeing two levaillantii, one in the normal and one in the black phase, at Vila 
Luisa, southern Mozambique. There can be no reasonable doubt as to his identi- 
fication of the pale morph, and if the two were really a “pair” this would be a 
case suggesting that the situation between morphs in this species is the same as 
in jacobinus. The fact that the locality is so far south of the known range of 
the melanistic morph of levaillantii makes this sight record somewhat uncertain. 
Furthermore, Lamm’s notes are not conclusive as to whether the two birds were 
really a pair. He merely saw a black-plumaged bird near the normal levaillantii; 
he observed no sign of mutual interest between them, although he wrote, 
“probably a pair.” 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 71 


this connection it may be recalled that Stresemann (1947, pp. 518- 
519) suggested that the “. . . allele-producing mechanism has besides 
some physiological effect increasing viability in this special environ- 
ment (but not elsewhere) which results in its being favoured by 
selection .. .” In support of this suggestion he pointed out that a 
shrike, Laniarius ferrugineus sublacteus, also has a melanistic morph 
(L. nigerrimus) in the same limited area of coastal Kenya and the 
lower portion of the valley of the Tana River. 

While the restricted geographic coincidence of melanistic poly- 
morphism in two widely dissimilar and unrelated birds as a cuckoo 
and a shrike may be suggestive, it remains that in neither species do 
we have as yet any observational data as to the relative abundance 
of the two plumages, to say nothing of the frequency of crossing 
between their phases. In the absence of such information we can only 
interpret the situation in C. levaillantu as probably similar to what 
we know in the related C. jacobinus in southeastern Africa, and in 
that species it is difficult to see that either morph has any selective 
advantage over the other. In connection with Stresemann’s sugges- 
tion, it may be recalled that C. jacobinus also occurs in coastal Kenya 
and the Tana Valley, and has produced no melanistic morphs there 
although it has done so far to the south. If there were something 
in the ecological situation of coastal Kenya that might favor such 
melanisms we might expect it to have produced some visible mani- 
festation in C. jacobinus as well. That local ecology is not directly 
important in the establishment of polymorphism is further indicated 
by the fact that in the extensive area of southeastern Africa where 
jacobinus has two phases, Jevaillanti occurs in a single, “normal” 
or “pale” phase. In other words the two species are sympatric in the 
two areas where one and not the other is polymorphic. 

The melanistic morph of C. jacobinus occurs as a breeding form 
throughout Natal, the eastern Cape Province, the eastern half, or 
more, of the Transvaal, the Orange Free State, and to Bechuanaland 
(Mahalapye), Southern Rhodesia (Bulawayo), Northern Rhodesia 
(Livingstone), and Nyasaland (Misanje). It is decidedly rare in 
the Rhodesias and Nyasaland, and finds its greatest abundance in 
Natal, the eastern Cape Province, and eastern Transvaal. It also 
occurs in southern Mozambique, but I know of but one actual speci- 
men record from there, and it had no exact locality other than 
“Mozambique” on its label. 

The corresponding black phase of C. levaillantii is known, as far 
as I have been able to learn, from some 26 specimens in the museums 
of the world. Of these, no fewer than 18 were collected within 50 


72 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


miles of Mombasa (Mombasa, Kilifi, Mazeras, Takaungu, Rabai 
Forest, Sokoke Forest); 1 came from Malindi; 3 from the lower 
portion of the Tana Valley (Kosi, Kau, and near Lamu); one 
example was recorded from southern Jubaland (Jebeir), the most 
northern locality from which the black morph has been reported; 
the type specimen (of “albonotatus,’ under which name this morph 
of levaillantu was first described as a new form) was collected in the 
Usambara Hills, northeastern Tanganyika, and another was taken not 
too far away, on the Pagani River. The one remaining specimen, now 
in the collection of the Academy of Natural Sciences, Philadelphia, 
was taken at an altitude of 10,000 feet on Mt. Kenya, in February 
1919! This locality is far removed geographically and ecologically 
from all the others. The specimen was originally in the Blayney 
Percival Collection, and since Percival is generally considered to have 
been a careful and reliable labeler of his birds, there is no valid reason 
for questioning this record. 

It does, however, point out an interesting fact, namely that the 
tendency to produce melanistic morphs is not absolutely restricted to 
the area where these have become well established. While it cannot 
be proved that this specimen from the high slopes of Mt. Kenya was 
not a migrant from the coastal lowlands, this is extremely improbable. 
It would be a strange migration indeed for a bird of the hot coastal 
belt to migrate to an altitude of 10,000 feet on Mt. Kenya. Further- 
more, we have no other evidence of the melanic morphs spreading out 
from their restricted habitat, such as we might expect if they were 
regularly migratory. Yet, it must be admitted that Percival (in 
Bannerman, 1910, p. 704) wrote that ‘‘albonotatus’ seemed to visit 
the coastal belt of Kenya for about 6 weeks only in the year, which 
suggests seasonal movement. On the other hand, he collected two ex- 
amples there in March, only a few weeks different in season from 
his Mt. Kenya bird. It seems, from all these considerations, more 
likely that the latter was a case of an individual melanism cropping 
up as an isolated occurrence. It may be mentioned that there is 
evidence of similar, sporadic, widely spaced cases of melanic poly- 
morphism in C. jacobinus as well. 

In the latter species occasional black-phase birds, indistinguishable 
from southeast African melanistic serratus, have been taken at Port 
Gentil, Gabon (November 3), south of Lake Tchad (in July), at 
Kulme, Darfur, Sudan (July 11), Kordofan, Sudan (no date) and 
at Sagon River, Ethiopia (June 4). Reichenow (1902, p. 78) listed 
the Kordofan record in the synonymy of the pale-vented jacobinus, not 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 73 


of serratus. However, Strickland (1850, p. 219) was aware of the 
difference, and commented that to his knowledge this “. . . Cape bird 
has never before, I believe, been obtained to the north of the equator 
. . .” Certainly the November bird from Gabon, in very fresh plum- 
age, cannot have been a migrant from southeastern Africa, where 
at that time of the year serratus is breeding. Furthermore, the Kulme 
and Lake Tchad birds, taken in July, and the June specimen from 
Sagon River, are all in the same stage of molt as Natal birds are in 
February. This indicates that whether they were resident in the areas 
of capture, or whether they wandered there from elsewhere, they may 
not have come from southeastern Africa, where the molting season 
differs by a third to a fourth of a year from theirs. 

The ranges of the melanistic phases of C. jacobinus and C. levail- 
lantii are not readily expressed in terms of vegetational areas. Thus, 
if we take Keay’s 1959 Oxford “Vegetation Map of Africa south of 
the Tropic of Cancer,” we find that the black-plumaged C. 7. serratus 
overlaps in its breeding range, the “Relatively Dry Woodlands and 
Savannas” (characterized by savannas of tall grass with Acacias as 
well as other trees) and the “Temperate and Subtropical Grassland” 
(pure grassland above 3,500 feet). The melanistic phase of C. 
levaillantit appears to be contained within, but is not coextensive 
with, the “Coastal Forest Mosaic” area. 

To clarify the recorded data, it may be stated at this point that 
the old report of a black-phase serratus from Denkera, Fantee, Ghana, 
listed by a number of authors in their compilations, is based on an 
error. The actual specimen involved, examined by me in London in 
1962, is not a Clamator at all, but a black cuckoo, Cuculus cafer. 
Similarly, the supposed record of melanistic serratus from Lamu, 
Kenya, cited by several writers on east African birds, is actually 
based on an example of the black phase of C. levaillantii, to which it is 
properly referred in the present study. 

To return to Ford’s illuminating appraisal of the whole question of 
polymorphism, it appears that the situation in the two species of 
Clamator fits the definition of what he terms neutral polymorphism. 
It may be explained that Ford distinguishes three types—transient, 
neutral, and balanced polymorphism. The first is, as its name sug- 
gests, a polymorphism in the process of spreading through a popula- 
tion, but once it becomes fixed and ceases to spread it is no longer 
to be termed transient, but becomes either neutral or balanced. When 
a variant phase, or morph, has a selective advantage only as long as 
it does not dominate numerically the total population in which it 


74 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


occurs, and loses this advantage when it becomes more widely 
prevalent, it is said to be a balanced polymorphism. However, when 
such a form appears to be without obvious selective value or adaptive 
advantage, it is termed a neutral polymorphism. It is not clear that 
the distinction between neutral and balanced polymorphism is bio- 
logically factual; it might be better to say that the neutrality of 
certain balanced polymorphisms is merely an observational inference 
rather than an established genetic fact. It is in this restricted sense 
that the situation in Clamator may be described as neutral. In the 
light of present evidence, at least as far as Clamator jacobinus is 
concerned, the “normal” morph and the black phase seem equally 
well adapted to their common environment. The birds interbreed 
freely and act as though they recognize no differences between them, 
and furthermore the black form is not noticeably spreading geo- 
graphically or becoming increasingly numerous where it occurs to- 
gether with the white-breasted form (and it is not known to occur 
anywhere as the sole morph). In some localities it is apparently as 
common as, and in a few spots even more numerous than, the pale 
morph, but there is no evidence to suggest that the ratio has changed 
appreciably in the past half century or more of observations. It must 
be admitted that this absence of evidence is not nearly as good a 
support as would presence of negative evidence have been. The 
“evidence,” if it may so be termed, is chiefly the memory and rec- 
ollections of observers of long residence, unsupported by critical 
records and notes. 

In the case of C. levaillantii we can only assume that the situation 
is also one of neutral polymorphism, as we do not have the direct 
evidence available in C. jacobinus. It seems, however, a safe assump- 
tion. 

Polymorphism is an expression of gene frequency, and neutral 
polymorphism implies a fairly stable frequency picture. Inasmuch 
as the occurrence of melanistic morphs of both levaillantit and of 
jacobinus away from their geographically restricted areas of de- 
veloped neutral polymorphism is so sporadic and infrequent, it follows 
that these two wide-ranging species, each with geographically con- 
tinuous, uninterrupted, nonfragmented distribution patterns, have 
local populations whose gene pools seem to be fixed and seem to be 
kept unavailable to adjacent populations of their own kind.? There 


2JIn the case of C. jacobinus this statement is intended to cover only the 
African part of its range; its extensive Asian population is of course effectively 
cut off from the larger African one. No black-phase birds of this cuckoo have 
ever been noted in Asia (from which area I have examined at least 200 examples, 
as well as read and checked the many observations and records in the literature). 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 75 


may possibly be some unknown and, as yet unsuspected ecological 
factor in the area of polymorphism of each of these species that has 
made possible a local decrease in selective pressure and thereby 
enabled two morphs of each to develop in a state of passive neutrality. 
This may be something akin to what Stresemann vaguely suggested, 
although transposing the factor from the “allele producing mecha- 
nism” to the environment. There is no evidence even tangential to 
this concept that may be cited, and the thought is merely inserted for 
its suggestive value. Considering that both species are highly migra- 
tory, at least in their southern sections (exactly the section where 
jacobinus is polymorphic), it is difficult to account for this apparent ge- 
netic isolation in the light of present knowledge. It seems that in coastal 
Kenya, Percival’s statement (cit. supra) notwithstanding, levaillantii, 
with its local black phase, is relatively nonmigratory, but no such as- 
sumption can be maintained for either in southeastern Africa. The 
absence of migrant, or of “wintering” examples of black-phase C. 
jacobinus serratus from equatorial Africa during the southern winter, 
when it is known that both phases are absent from their relatively 
well-observed austral breeding range and have presumably gone north, 
is a real puzzle. This is discussed more fully in our account of 
migratory behavior (see pp. 84-85). In the present connection it may 
be hypothecated that, wherever they may “winter,” all the individuals 
of southeastern serratus return to their home area for breeding, and 
thus remain unmixed with their adjacent conspecific populations. 

Aside from the well-developed melanistic morph in the adults of 
jacobinus and levaillanti, the latter species also has a rufescent 
juvenal morph, reminiscent of the hepatic phase of the young in 
Cuculus canorus. I know of only two examples of this rufescent 
phase, both from the northeastern portion of the Republic of the 
Congo (former Belgian Congo). One such bird, a young male, taken 
at Poko, in the Uelle district on July 31, now in the British Museum, 
is bright cinnamon rufous above and below, only the remiges and 
retrices being darker, less reddish, as in the “normal” juvenal. The 
other one came from near Beni, in the Ituri district. Another, possibly 
partial, rufescent bird may be one mentioned by Granvik (1934, p. 
24) as having the undertail coverts pale rufous, although the rest 
of the plumage was probably “normal” as it evoked no comment from 
the describer. 


MIGRATORY BEHAVIOR 


The evolutionary picture of migratory behavior in the genus pre- 
sents some peculiar features. All four of the included species are 


76 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


strictly migratory in some parts of their respective ranges and not 
in other parts. (In the case of C. coromandus, the available evidence 
is not conclusive about an area of permanent residence, and, hence, 
of a population of nonmigratory birds.) The extent of migratory 
movement varies from none at all to thousands of miles. All three 
of the species breeding in South Africa (glandarius, jacobinus, and 
levaillantw) are absent from that area during the southern winter ; 
all three have resident populations in tropical Africa. The first named 
of these is present in its Mediterranean breeding grounds in the 
Iberian Peninsula and northwestern Africa only during the northern 
summer, and then migrates to equatorial parts of Africa, apparently 
chiefly in the eastern half of the continent. The migration of this 
section of the species is thus exactly the opposite, both in direction 
and in time of the year, from that of the South African glandarius. 
In other words, we find in these cuckoos that geographic segments, 
not necessarily even subspecifically distinct, differ markedly from 
other conspecific segments in their migratory behavior. It need hardly 
be added that in the majority of birds that have spatially distinct 
breeding and nonbreeding quarters, migration is an important, well- 
formulated and patterned, presumably evolved and inherited, part of 
their annual life cycle. Yet in the crested cuckoos of the genus 
Clamator, this migratory behavior is manifested only in sections of 
each of the included species. 

In this geographic fragmentation of migratory behavior within the 
members of each species, we have something that may be likened to 
partial migration, with the difference that here the “partial” element 
is geographic, not individual. Partial migration is a term used chiefly 
for species in which some individuals are regularly migratory while 
others, breeding in the same area, are nonmigratory, resident birds. 
Inasmuch as there is no evidence to prove that the migratory South 
African populations are ecologically or geographically cut off from 
their nearest resident counterparts in tropical Africa, we cannot 
postulate an interference effectively isolating them into discrete, non- 
intercommunicating colonies or gene pools. Even allowing for a 
reduced frequency of such intercommunication, we may come back 
to something akin to partial migration (in an overall species view) as 
a valid way to express their migratory tendencies. 

Partial migration of the more usual sort has been studied in the 
North American song sparrow by Nice (1937), and in a variety of 
European passerine species by Lack (1943-44). Their findings are 
of interest here. Nice (1937) found that in the song sparrow, 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 77 


Melospiza melodia, in the Columbus, Ohio, area, migratory and non- 
migratory behavior was not correlated with age, and apparently was 
not a matter of inheritance. Thus, nine resident fathers had seven 
resident and two migratory sons, and nine migratory fathers had 
seven resident and four migratory sons. Among 61 males, it was 
found that 24 remained consistently resident, 31 were consistently 
migratory, and 6 changed from one to the other of these behavior 
groups. Among 43 females, 5 were consistently resident, 37 always 
migrated, and 1 changed from resident to migrant. Some years later 
Lack (1943-44) reported on a study of partial migration in a number 
of species of European birds, and showed that in all cases the females 
and the young of the year showed a noticeably greater tendency to 
migrate south in the autumn than did the adult males. 

Baker (1942, p. 4) was aware of the partial nature of the migra- 
tion of cuckoos other than Clamator. He went so far as to conclude 
that “. .. most genera and, indeed, most species of migratory 
Cuckoos include a race which is more or less sedentary. For instance, 
the Common Cuckoo, the most migratory form of Cuckoo, has a race, 
the Khasia Hills Cuckoo, which can hardly be called migratory at all. 
It breeds in the eastern sub-Himalayas and spreads into the plains of 
Burma and India in winter, while some individuals remain all the year 
round in their summer quarters. If ... migration has in many 
cases been forced upon birds because of the insufficiency of food 
supply during the breeding season it may well be, . . . that cuckoos 
were originally tropical or sub-tropical oriental birds and their extreme 
limits, East and West, are those to which they have extended under 
this pressure .. .” 

In his recent (1962) survey of bird migration, Dorst noted that a 
considerable number of species of birds are composed of sedentary, 
migratory, and partially migratory populations, which could be looked 
upon as “physiological races” or sections of the total population of 
each species. From this he drew the logical conclusion that migration 
cannot be regarded as a specific character as it really belongs within 
the framework of populations within the species. 

In some species of other birds migration is a characteristic of one 
race or subspecies and not of another. A case that may be mentioned 
is the Oregon junco, a North American finch studied experimentally 
by Wolfson (1942). This bird has a migratory and a purely resident 
race in northern California. To test their migratory tendencies 
Wolfson experimentally subjected groups of individuals of each kind 
to increasing numbers of hours of light, either natural or artificial, 


78 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


but only the individuals of the regularly migratory race responded by 
becoming restless. This indicated that not only was a predisposition 
for migratory behavior necessary but that it could be manipulated. 
What can be manipulated experimentally by the investigator may 
also be effected out of laboratory conditions by “natural” causes. 

At this point it seems useful to note the results of some recent 
studies because they correct a concept of migration based too largely 
on what has been recorded in north-temperate areas of the world. 
The pattern of migration there, with its easily accepted geographic 
inferences and correlations, is usually expressed in terms of Pleisto- 
cene climatic fluctuations. However, Moreau (1951) has shown that 
bird migration is probably as old as bird flight and that what happened 
during the Pleistocene in Europe and North America merely deter- 
mined the geographic details of the migrations of individual species; 
but not the migratory behavior itself. It is true that most of the 
major, “best organized” migrations of considerable geographic magni- 
tude seem to have reflections of Pleistocene events, but we realize that 
migration may have begun anywhere, anytime, with different groups 
of birds (Drost, 1950, p. 231). Pleistocene glaciation was not its 
cause. Moreau (1951) and Mayr and Meise (1930) indicated that 
migration may have originated in any localities where seasonal food 
scarcity may have caused some birds to move away seasonally and 
thus have a better chance of survival. Ostensibly, it would seem that 
this would be acted upon by natural selection, and in this way migra- 
tory behavior would become established, with or without any influence 
of Pleistocene glaciation, and, in some cases, probably was much 
earlier than Pleistocene in origin. In defense of this argument Moreau 
(1951, p. 247) cited cases of migration entirely within warm areas, 
and mentioned among them “. . . the Indian population of the cuckoo, 
Clamator jacobinus, which travels all the way to East Africa after 
breeding (je. 

Cuckoos, as a group, are birds with a great tendency or predisposi- 
tion toward migration. Many years ago, W. L. Sclater (1906) cal- 
culated that of the 814 species of birds then known to occur in South 
Africa, 731 were resident, and only 21 were to be considered as 
African migrants (as distinguished from European and Asiatic winter 
visitors), and of these 21 no fewer than 9 were cuckoos. Many years 
prior to Sclater, Emin Pasha, prior to 1888, (published by Schwein- 
furth, et al., 1888, p. 392) noted the seasonal wanderings of a number 
of purely African savannah birds in “Equatoria” (the southern part 
of the present Sudan and the adjacent area of the Republic of the 
Congo), among which he mentioned Clamator. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 79 


It is among the cuckoos, purely insectivorous in their feeding, 
that we find some of the most remarkable of geographic migrants. 
We may take, as an example, the long-tailed cuckoo, Urodynamis 
taitensis, that makes an unusually long, and largely nonstop, over- 
seas journey from its New Zealand breeding grounds to the islands of 
Polynesia, some of which are as much as 4,000 miles away. (Bogert, 
1937). Another notable example is the little bronze cuckoo, Chryso- 
coccyx lucidus, also a New Zealand bird, that goes more than 2,000 
miles across the South Pacific to the Solomon Islands (Fell, 1947). 
The common cuckoo of Europe makes a similarly impressive journey 
from its northern breeding area to tropical and even to southern 
Africa. In fact, numerous writers have made particular mention of 
the fact that the young of the year of this species make this spectacular 
trip by themselves with no possible aid from, or accompaniment by, 
the adults of their own kind, with whom they have had no experi- 
ence. 

The concepts of migratory tendencies, even if they are not more 
than a periodic psychobiological restlessness, originally not rigidly 
correlated with, or controlled by, heredity, and of migration apart 
from the rigid seasonal climatic fluctuations of Pleistocene glaciation- 
induced patterns, make it possible to look upon the Clamator situation 
as less enigmatic and less paradoxical than it first seemed to be. Con- 
sidering the pronounced migratory tendencies of its relatives in the 
subfamily Cuculinae, it would be surprising if the species of Clamator 
were not also somewhat migratory. The extent to which this behavior 
is developed differs in the four species of the genus. To explore these 
differences further, we may now turn to the situation in each species, 
as far as the present, still incomplete, data will permit coordinated 
presentation. 


Clamator jacobinus 


It is definitely known that the population (subspecies serratus) 
that breeds in Natal, Transvaal, Cape Province, and Southern Rho- 
desia, is absent from those areas from late March to October (south- 
ern “‘winter’”’), and that individuals of the pale morph of this race have 
been collected during these months in Nyasaland, in the open grass- 
lands of the southern and eastern parts of the Republic of the Congo, 
former Belgian Congo, (Aru in the Upper Uele, Mahagi Port in the 
Ituri, and near the base of Ruwenzori) and in Uganda (Mohokya, 
Fajao, and Kebusi in May, Butiaba in November). In Darfur, Lynes 
(1925, p. 354) found serratus (recorded by him as jacobinus, but 
corrected by Jackson and Sclater, 1938, p. 497) in June and August. 


80 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


In Sudan the pied crested cuckoo generally is reported by Cave and 
MacDonald (1955, p. 174) to be a fairly common nonbreeding visitor 
from March to October in the southern part of the country, from 
June to late August in Darfur. It may be noted, however, that many 
years ago Emin collected an egg of this cuckoo at Lado (Hartlaub, 
1881, p. 114) while Butler (in Sclater and Mackworth-Praed, 1919 
p. 642) stated that it breeds near Khartoum, where he found newly 
fledged young on October 5 (Butler, 1908, p. 245). 

In British Somaliland (now part of the Somali Republic) Archer 
and Godman (1961, pp. 663-668) reported it as present in May and 
June, the main breeding season of many potential passerine hosts. 
There are as yet no definite breeding records from that area but the 
pied cuckoo may well prove to breed there. 

In Ethiopia, serratus has been collected as early as April 7 to 8, at 
Gato River, near Gardula (Friedmann, 1930, pp. 268-272) together 
with examples of the race pica, and at Sagon River, June 4. In 
Eritrea, K. D. Smith (1957, p. 309) classed it as a migrant, present 
from June to September, possibly breeding there in summer. 

In Angola this cuckoo (subspecies pica) is found only from October 
to May, from Huila and eastern Mocamedes to Cuanza Norte to 
Luanda and along the coast from Benguela to Cuanza; it migrates 
north of the equatorial forest in “winter.” For these summary data 
I am indebted to M. A. Traylor of the Chicago Natural History 
Museum, who further informs me that there are breeding records 
from Chibia in February and from Huila in December, and that 
some of the Huila specimens show an approach to the pied phase of 
serratus. 

In Tanganyika, Moreau (1937b, pp. 22-23) noted that while no 
race of C. jacobinus had yet been found to breed in that country, 
pied morphs of serratus were known to appear there as migrants, as 
well as the paler, white-breasted pica, some of the latter race presum- 
ably coming from India. He further noted that examples of this 
cuckoo from extreme western portions of Tanganyika may belong to a 
“|... population different from that occurring in the rest of the 
territory; the date of the influx accords with a possibility that they 
might be birds coming south from spending their off-season in Darfur 
and the Sudan.” Thus, at Kigoma, in late October, Packenham 
found pica became abundant, and he found that a female collected as a 
specimen record had an enlarged ovary, ostensibly a bird on its way 
to its breeding grounds. 

In another paper, Moreau (1937a, pp. 5-7) reported that white- 
breasted birds (C. 7. pica) had been collected in northern, central and 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 81 


southern Tanganyika between December and April. All his own 


northern Tanganyika records were of “. . . silent non-breeding birds 
in worn plumage or very slow and irregular moult like the February 
birds of extreme southern Tanganyika . . . This influx of non-breed- 


ing birds into northern Tanganyika fits in strikingly with Whistler’s 
hypothesis that the Indian population migrates to Africa after breed- 
ing in the northern summer ; and clearly the birds, also non-breeding, 
that are so common in Darfur June to September must have quite a 
different origin . . .” 

In Kenya and Uganda, Jackson (in Jackson and Sclater, 1938, pp. 
495-496) found pica to be a local migrant, rarely if ever remaining 
long in one locality, arriving in November and leaving in April and 
May. He noted these birds, apparently traveling north, from March 
20 to April 16 at Nimule, Uganda, and moving south in November at 
Lake Albert. In the Nyando Valley he found them common early 
in May and scarce at the end of that month. “The same influx and 
departure after a few weeks’ sojourn takes place in the coast and 
bush-veld regions of Kenya Colony . . . It is particularly plentiful 
in the Taru wilderness in November and December, and again in 
April . . .” However, there is now definite evidence that pica breeds 
in Kenya (Ngong) and in Uganda, so here it appears that there are 
resident birds, migrants from elsewhere in Africa, and migrants from 
India, making the resulting situation difficult to interpret with cer- 
tainty in many specific instances. The intra-African migrants appear 
to be of both pica and serratus stocks. Similarly, there is some evi- 
dence that both pica and jacobinus wander to Africa from India. 

It is unfortunately true that, so far, we have no direct proof, of 
marked individual birds, demonstrating the migration of pied cuckoos 
from India to Africa, but there are inferential considerations that 
strongly point in this direction. Long ago Whistler (1928) compiled 
an account of the postulated migration in the hope that it might 
stimulate observers in India to fill in the gaps in the information he 
was able to bring together. He showed that the bird (pica) is ex- 
tremely numerous in northern India during the rainy season, when 
it breeds there, and that it is definitely absent from there the rest of 
the year. He expressed his attitude by stating that if the birds do not 
leave India and go to Africa “. . . we cannot say at present where 
so great a mass of individuals can winter unrecorded; it can only be 
in southern or southeastern India or in Ceylon . . . Legge’s evidence 
appears to have ruled out Ceylon. As to southern and southeastern 
India, we have no definite evidence either for or against the sup- 
position...” 


82 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


In a later paper (1931, p. 193) he pointed out that the absence of 
examples of north Indian birds (pica) in the extensive series collected 
in southern India (jacobinus) at all times of the year “. . . virtually 
settles that our northern migrants go to Africa . . .”—a conclusion 
which has been accepted and implemented, without definite proof, by 
many others since Whistler’s paper. Smythies (1953, p. 326) found 
that the jacobin cuckoo “. . . seems to leave Burma altogether in the 
winter, possibly migrating to Africa.” In his recent compendium on 
Indian ornithology, Ripley (1961, p. 175) stated that pica (serratus 
of his book) reaches, on its winter migration, Gujarat, Bombay, 
Andhra, and northwestern Madras. “. . . The main wintering range 
appears to be to the west, south of the Sahara in Africa. Rainy 
season wanderings of this form and the next (jacobinus) prevent 
exact definition of the breeding zones in central India.” Meinertz- 
hagen (1954, p. 308) reported that in Arabia, a presumably logical 
area through which migrants between India and Africa might be 
expected to pass, the species was known as a migrant in the south- 
western part of that peninsula, where specimens were obtained near 
Aden on March 31 and April 22, in the Amiri district in May, at 
Hadda near Mecca on April 2. He noted that a pair was obtained in 
Asir on June 26, which “may denote breeding.”’ If these were breed- 
ing birds, and not delayed migrants, they constitute the only evidence 
for the pied crested cuckoo in Arabia other than on migration. On 
the basis of extensive personal experience with both Asiatic and 
African birds, Meinertzhagen concluded that some of the birds that 
breed in northern India and Baluchistan appear to go to Africa in 
the northern winter. 

Grant and Mackworth-Praed (1948, pp. 171-172) attempted to 
study the migration of these birds on the basis of the dates of molting 
specimens in the British Museum. They started with the opinion that 
Indian specimens should be in molt from September to November, 
South African breeders, from April to June, and birds from other 
parts of Africa, from June to August. The fact that in India birds 
taken from September to November were in molt was in line with 
these dates, and from all these considerations it was thought that any 
molting examples taken in Africa during September, October, and 
November should be Indian migrants. Their examination failed to find 
any such material and they were forced to conclude that none of the 
African records could be considered definitely as migrants from 
India, and they ended with the statement that the “. . . only evidence 
we still have of this species visiting Africa from India in the non- 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 83 


breeding season is the fact that it does leave India .. .” In 1962 I 
went over the material in the British Museum with a hope of finding 
some clues that Grant and Mackworth-Praed might have overlooked, 
and to examine recently acquired specimens that they had not seen. 
My results suggested a greater spread of months for molting of 
African birds, which clouded or obscured the whole picture to the 
degree that it was not feasible to demonstrate Asiatic origins by 
molting dates. 

Clancey (1960, pp. 27-31) has, I think, made the only convincing 
contribution to this problem. He stated that not only do north Indian 
birds (pica) migrate to Africa, but he found that so do many of the 
smaller, typical jacobinus of southern, peninsular India and Ceylon. 
The birds of this subspecies are identifiable by their smaller size 
and consistently white throats and breasts (like pica in this latter 
character), and they are known to breed only in India, Assam, Burma, 
and Ceylon, but they occur in Africa as far south as Nyasaland, 
Southern Rhodesia, and southern Mozambique. The African records 
fall between September and April, which agrees with the fact that 
the birds should be back in India for the breeding season. Unlike 
north Indian pica, the race jacobinus is only partially migratory, some 
individuals remaining throughout the year in southern India and 
Ceylon while others reach Africa where they disperse over a wide 
area. The fact that some south Indian birds do migrate to Africa 
increases the probability that similar movements occur in north Indian 
pica as well. 

The migration of C. jacobinus between India and Africa, does have 
some peculiar features. Ali (in litt.) has informed me that as far 
as he knows no other long-distance land migrant arrives in India at 
the commencement of the southwest monsoon season as this cuckoo 
apparently does. He further assured me that there is no evidence that 
any seasonal lack of insect food could operate as the reason for this 
bird to leave India after the close of the breeding season. 

As discussed elsewhere in this report (p. 51) it seems that the 
southern African population (serratus) of C. jacobinus is the oldest, 
most primitive segment of the species, and of the genus, as it exists 
today, and that after it gave rise to pica in equatorial Africa, the 
latter spread to Asia and became established there. The present 
migration of pica between northern India and Africa thus is an an- 
nual reflection of an original movement in the past history of the 
species, a situation existing (or, at least, so interpreted) in many 
other migratory birds. Ticehurst (1922, p. 531) postulated a route 


84 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


from northern India to Africa for a number of Indian breeding 
species that are absent from there in winter ; among them are Agro- 
bates g. familiaris, Caprimulgus e. unwini, Merops apiaster, Glareola 
pratincola, Cuculus canorus, and others. To these may be added 
another cuculine species, the lesser cuckoo, Cuculus poliocephalus, 
that breeds in Asia, beyond the Himalayas, and winters in numbers 
in East Africa (Moreau, 1937b, p. 42), while Ali (in litt.) informed 
me of similar migratory behavior in Indian breeding Merops super- 
ciliosus persicus, Coracias garrulus semenowt, and Muscicapa striata 
neumanni. 

Before leaving C. jacobinus, it is necessary to discuss the melanistic 
morph of the race serratus in connection with its migratory move- 
ments. This black phase is frequent in the eastern parts of the breed- 
ing range of serratus—Natal, Cape Province, Orange Free State, etc., 
and, like the pale morph, this one is absent from South Africa during 
the southern winter. These melanistic individuals are, in a sense, 
critical material, as the pale morphs could not be distinguished from 
similar birds resident in more tropical areas to which they presumably 
migrate. Yet, aside from a small number of black-plumaged birds 
(four), this phase has been conspicuously absent from collections made 
throughout Africa outside of their southeastern breeding range. The 
four black-phase birds, indistinguishable from southeast African 
serratus, that have been taken are as follows: At Port Gentil, Gabon 
(November 3), south of Lake Chad (July), at Kulme, Darfur (July 
11), and at Sagon River, Ethiopia (June 4). These pose a very 
puzzling problem that cannot be completely resolved. These have been 
discussed briefly in our account of polymorphism (see p. 71) but our 
interest in them at this point is in their implications concerning their 
geographic movements. The November Port Gentil, Gabon, speci- 
men, in very fresh plumage, can hardly have been a migrant from 
southeastern Africa, where at that time of the year serratus is breed- 
ing. The dates and the respective stages of molt and of feather wear 
of the other examples do not fit closely the seasonal chronology of 
the southern birds, and in this respect they suggest that they might be 
considered as individual (and rare) instances of melanism of the 
more northern race pica. In southeastern Africa serratus is dimorphic, 
and the melanistic phase is common, but if the four northern records 
of black-phase birds, listed above, are not serratus, or, at least, are 
not unquestionably of that subspecies, it would follow that not a single 
completely convincing example of the black morph of serratus has 
yet been collected away from its breeding range. There is no inherent 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 85 


reason why pica may not produce an occasional melanistic individual 
as serratus does in such numbers, although breeding examples of 
such have not been found as yet. If the Gabon, Lake Chad, Darfur, 
and Ethiopian black examples are looked upon as pica, where do 
the black serratus go when they leave their breeding range? If they 
are serratus, why have so few of this phase been collected during 
the southern winter while so many more of the pied phase have 
been taken? The discrepancy in numbers of winter specimens of the 
two is not at all consistent with their numerical status (almost equiva- 
lence in some localities) in southeastern Africa during the southern 
summer. Is it possible that the bulk of eastern serratus, which would 
include most of the black morphs, migrate a relatively short distance 
into Mozambique, an area where relatively little collecting has been 
done, and where Lamm (1955, p. 33) found this cuckoo (recorded 
binomially by him, but almost certainly serratus) from December 
through February? The only evidence, if it may be called that, sug- 
gesting that some of the melanistic serratus from southeastern Africa 
may wander far beyond Mozambique, even as far as southern Ethiopia, 
is that Mearns (in Friedmann, 1930, pp. 272-274) not only collected 
one bird, already mentioned, at Sagon River, on June 4, but saw four 
there, June 3 to 6, and two others at Turturo, June 15 to 17. If 
Mearns was correct in his identification of these sight records, this is 
the only instance known of a substantial, as opposed to a casual or 
individual, movement of these dark serratus. It is certainly not likely 
that the breeding pica of southern Ethiopia frequently produce melanic 
morphs in a limited area, or we would have had some other evidence 
of it by now, and, hence, if these records of Mearns are accepted 
they must be looked upon as migrant serratus. In support of this 
latter interpretation it may be noted that Mearns collected two ex- 
amples of the pied plumage phase of serratus at Gato River, near 
Gardula, southern Ethiopia, April 7 to 8, together with other examples 
of the white-breasted race pica (ibid., pp. 268-272). That pied 
morphs of serratus could reach southern Ethiopia as early as April 
7 suggests either a very rapid migration, which is not very likely, 
or that some of the southern birds must start north considerably 
before others. 

In Southern Rhodesia, where we might expect to find the black 
phase with some regularity either as a breeder or as a migrant, M. P. 
S. Irwin informs me that he has never seen one in life, and that the 
collections in Bulawayo contain a single Southern Rhodesian example, 
taken at Forest Vale, near Bulawayo, on November 20, and another 


86 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


from Livingstone, Northern Rhodesia, collected on October 19. 
These, and one other from Nyasaland, are the only black serratus out 
of a series of 64 skins from the Rhodesias and Nyasaland in the 
collections of the National Museum of Southern Rhodesia. In Nyasa- 
land, Benson (1953, p. 35) noted two reliable sight records of the 
black-phase serratus, one from Fort Johnston in March, the other at 
Monkey Bay, in November. Benson assumed that these birds were 
transients in Nyasaland, and called them migrants from the north. I 
presume this means that they were looked upon as migrants coming 
from (November) their more northern wintering grounds on their 
return to their southern breeding area, or (March) returning to the 
north for the off-season. 

To return, in our discussion, to southern Mozambique, Lamm (cit. 
supra) mentioned that in early December he saw both color phases of 
Clamator levaillantu; however, without the specimens (which were 
not collected), it is impossible to be certain that the black individual 
was really levaillantui and not serratus, for the dark morph of the 
former has not been found south of extreme northeastern Tanganyika. 
In reply to my inquiry, Lamm has informed me that this sight record 
was made at Vila Luisa on December 10, 1950. His notebooks record 
an “. . . all black cuckoo with white wing patch; near it another, 
black above, white below heavily streaked on the chest, probably a 
pains? 

It may also be mentioned that Pakenham (1948, p. 99) saw a black 
crested cuckoo in Zanzibar, April 10, which he considered as probably 
C. j. serratus. On the basis of the geographic proximity of Zanzibar 
to the known range of the black phase of Jevaillantii, Pakenham’s 
bird may have been of this species. The mere sight record, un- 
fortunately cannot be identified, and remains relatively useless. 

To summarize, the peripheral populations of the jacobin cuckoo, 
serratus, in Africa south of the Zambezi River, and pica in northern 
India, are highly migratory; typical jacobinus of southern India is 
partly migratory, and serratus and pica in much of tropical Africa 
are apparently fairly resident in some places and move about without 
obvious correlation with season, climate, rainfall, or other noticeable 
factors in other localities. In large areas of tropical Africa a breeding 
form and two or more migrant, either transient or “wintering,” 
populations often occur together. The movements of southeastern 
serratus, as evidenced by its melanistic morphs, are still unclear, but 
there is no question as to their going north during the southern winter. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 87 


Clamator levaillantii 


As in C. jacobinus, the population of the stripe-breasted cuckoo 
breeding south of the Zambezi leaves that area after the end of the 
southern summer in late March, and does not return until October. 
Elsewhere in Africa it has been noted as a local migrant, or at least 
as a fluctuating element in the avifauna, locally present one day and 
absent the next. In Nyasaland it has been recorded from early 
October to May and even to June, and has been known to breed there. 
It is assumed by Benson (1953) that it migrates to somewhere to the 
north for the rest of the year. In the Rhodesias, where it also is 
known to breed, it is also seasonal, although further data are needed, 
especially from Northern Rhodesia, to clarify the local picture. Thus, 
in that country the earliest spring date is given as November 8, a 
month later than in South Africa (!) and the latest autumn date as 
May 4. Grant and Mackworth-Praed (1952, p. 506) wrote that it 
passes through Northern Rhodesia in November and December to 
breed farther south, and concluded that “‘there is certainly a northern 
and a southern breeding bird but this is probably not the whole story.” 

The seemingly haphazard occurrence of the species in localities 
where it has been found to be present or absent without obvious sea- 
sonal correlations, was stressed by Jackson (1938, pp. 497-498) in 
both Kenya and Uganda, although the species has been recorded 
there throughout the year. In Tanganyika the picture also is still 
confusing. Moreau (1937b, p. 23) noted that the only localities in 
that country where the species had been recorded as breeding were 
Iringa, from February to March; the east side of Lake Nyasa, in 
May; at Kilosa, in April. He recorded that it had been seen at 
Kigoma and at Uvinza in November, when it was molting. He con- 
sidered it not unlikely that the nonbreeding birds in Kenya and 
northern Tanganyika may have been migrants from Ethiopian breed- 
ing grounds, while the southern Tanganyikan birds “. . . in the east 
up as far as the Central Line represent a different population breed- 
ing there and with their own movements. . .” 

In coastal Kenya and the adjacent parts of northeastern Tan- 
ganyika, Percival (in Bannerman, 1910, p. 704) concluded that 
the stripe-breasted cuckoo was present as a “visitor” for a matter of 
only about 6 weeks in the year. However, this is erroneous, as speci- 
mens of the local melanistic morph have been taken in that restricted 
area in every month of the year except July and August, and the 
present lack of records for those 2 months is not indicative of ab- 


88 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


sence. Further evidence of the nonmigratory status of this cuckoo 
in that area is afforded by the fact that during the more than half 
a century since Percival’s work many and very comprehensive collec- 
tions and observations have been made in practically all parts of Kenya 
at all times of the year, and not a single example of the black morph 
(the so-called “albonotatus”) has ever been collected outside of the 
coastal strip, except for one very surprising, but apparently ac- 
ceptably authenticated record, taken by Percival, at 10,000 feet on 
Mount Kenya. As mentioned in our discussion of polymorphism (pp. 
70-75) this last record would seem better interpreted as an unusual 
local melanism of the population of C. levaillantu resident on Mount 
Kenya, than as a migrant from the coastal lowlands. 

In the Republic of the Congo (former Belgian Congo), Chapin 
(1939, pp. 181-182) treated it as a resident bird in the Uele and in 
most other lowland savannahs, absent from forested areas, but known 
to breed in May and October (fledglings taken). Further to the north, 
in the Sudan, Cave and Macdonald (1955, p. 174) found it to be a 
common nonbreeding visitor between March and October, while in 
Darfur Lynes (1925, p. 354) concluded it was a rather infrequent 
summer visitor to the West Basin. In Mali, according to Malzy 
(1962, p. 34) the stripe-breasted cuckoo is a local migrant, common 
at the close of the rainy season, seen at Bamako from July to 
November. 

It is not clear as yet if the species leaves its Ethiopian breeding 
grounds (where it breeds from June to September) during the north- 
ern winter, but it may well do so in the highlands, thereby adding 
to the confusing population in Kenya to the south and in Sudan to 
the west. In Eritrea, K. D. Smith (1957, p. 309) called it a “pre- 
sumed resident” but had only scanty evidence. 

The movements of the species in West Africa are still uncertainly 
known. Bannerman (1933, pp. 108-110) could only say thata “... 
corresponding movement to those which take place in East Africa 
certainly occurs in West Africa, but observers being fewer we have 
less data . . .” He found from his compiled records that it appeared 
to have been met with only seldom between July and November south 
of latitude 12°. It is known to breed in Ghana (February) and in 
Nigeria (July). In the latter country, Marchant (1953, p. 45) found 
it to be an uncommon transient from December to February. By 
this he probably meant to infer that it wintered somewhere to the 
south and bred to the north, but he made no geographic guesses as 
to how far in each direction its migration extended. The species is 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 89 


present throughout the year in Gambia, but the local population is 
increased by migrants in June. 

Insofar as it is possible to summarize all these data, we may say 
that the species is clearly seasonal in Africa south of the Zambezi 
River ; occurs throughout the year in equatorial Africa, where, how- 
ever, its numbers are swelled during the southern winter months, 
and where it appears to comprise several populations, each with its 
own movements. It has not yet been ascertained to breed north of 
Ethiopia, the Republic of the Congo, Nigeria, Ghana, and Liberia, 
but probably does so. Inasmuch as it is not possible to separate, 
taxonomically, any geographic forms of this cuckoo and inasmuch as 
the known breeding records show a general, although spotty, distri- 
bution, it follows that the picture is somewhat like that in C. jacobinus, 
but wholly contained within the African continent. 

The species seems to be scarcer now than formerly in the south- 
eastern portion of its range. Thus, in the late years of the 19th cen- 
tury the Woodwards (1899) found it at the Umfolozi River, in 
Zululand, while today Clancey (im litt.) informs me that he has never 
met with it in Natal and considers it a very rare bird in the southern 
portion of its range. I also never encountered it in Natal, but only 
in the northern Transvaal (at Moorddrift, in December), where it 
was breeding. Even where it is common it is usually less numerous 
than the jacobin, although there is local variation in its numbers. In 
the Ashanti forest and the northern sections of Ghana, Lowe (1937, 
p. 635) reported it as abundant and present everywhere in the grass 
savannahs and in the open clearings in the forest. 


Clamator coromandus 


This is the one species of the genus that may have no nonmigratory 
populations or individuals, but available information is insufficient 
to establish this. The species ranges over an area where there never 
have been many resident observers and, as a result, our present data 
depend largely on specimens collected and deposited in museums. I 
have examined a large number of documented specimens of this 
cuckoo, and these, together with what has been published, yield the 
following picture. The species is known to breed in the Himalayan 
foothills from Garhwal and Nepal east to Assam at elevations of 
from about 2,000 to 8,500 feet, and in Burma at elevations of from 
1,500 to 6,000 feet; north to southeastern China (Kwangsi, Kwang- 
tung, Kiangsi, Fukien, Chekiang, and Hupeh Provinces; possibly 


go SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


also in Kweichow and Hunan) ; and southeast (rarely) to northern 
Thailand (Deignan, 1945, p. 158). 

In the nonbreeding season it wanders to Chota Nagpur, Madras, 
Mysore, and Karala in India, to Ceylon, to Thailand (where it is a 
transient in spring and autumn, never found in winter), to the entire 
length of the Malay Peninsula (except the eastern side), the Indo- 
Chinese countries, Lingga Archipelago, Sumatra, Java, Celebes, and 
Borneo, and occasionally to the Philippines. In Burma, Smythies 
(1953, pp. 326-327) considered it a local migrant, but its movements 
there have not yet been worked out in detail or with any accuracy. 
Similarly, the seasonal movements of this cuckoo in southeastern 
China are yet to be defined with precision. Thus the Caldwells (1931, 
p. 240) considered it only as a migrant in southern Kwangtung, but in 
the northern portions of that Province they found it a not uncommon 
resident. 


Clamator glandarius 


Both the northern and the southern extreme populations of this 
species are highly and regularly migratory ; the individuals breeding in 
equatorial portions of Africa are assumed (but not proved) to be non- 
migratory. In South Africa and South-West Africa, north to South- 
ern Rhodesia, Nyasaland, and southern Mozambique, the species is 
present only from September to March. In its Mediterranean breed- 
ing ground, where the seasons are reversed, the great-spotted cuckoo 
arrives at about the time the southern birds go north. Thus, Strese- 
mann (1928, p. 703) noted that this cuckoo arrives from its tropical 
African winter quarters as early as the beginning of February in 
upper Egypt and Morocco, in early April in Gibraltar, and that it 
leaves again for the south in July and early August, and, in Egypt, 
even as early as June. He pointed out that there was an obvious 
correlation between its migration dates and its host requirements. It 
had to establish itself on its breeding grounds before the prospective 
hosts began to lay. In northwestern Africa, where the hosts are 
magpies, whose early egg dates are from late March to early April, 
and in upper Egypt as soon as there are no new nests of its corvine 
hosts (the crows are all beyond this stage in June), the cuckoo begins 
to leave for equatorial Africa. This seems to imply a more hurried 
departure than is characteristic of the birds breeding south of the 
Zambezi River. Thus, in Southern Rhodesia, Smithers, et al. (1957, 
p. 67) record it as breeding from October to January, but not leaving 
for the north until April. Meinertzhagen (1930, pp. 345-347) found 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN gt 


it was absent from August to December in Egypt (not merely upper 
Egypt); he noted a marked northward passage at Wadi Halfa, 
Aswan, and Luxor in early February, when groups of from 10 to 20 
individuals were seen passing slowly down the Nile Valley. The 
same author (1954, pp. 306-307) found this cuckoo to be a regular 
but infrequent migrant in Arabia, considerable numbers going through 
in late March and April—a rather late date compared with the earlier 
passages farther west. 

In Eritrea, K. D. Smith (1957, p. 308) recorded a definite influx 
of birds in the coastal plain between December and March. The birds 
were common in summer (July and August) below 3,000 feet, and 
were absent from Eritrea in the winter. 

Cave and Macdonald (1955, p. 174) considered this cuckoo both 
a resident and a nonbreeding (wintering) visitor in Sudan, but were 
unable to say to which of these categories most of the individuals 
belong. In the Darfur Province, Lynes (1925, pp. 353-354) worked 
out the local situation in greater detail. He concluded that there were 
two distinct groups of cuckoos, one composed of individuals that bred 
farther to the south in equatorial or in southern Africa, and which 
spent their off-season farther north than Darfur, and merely passed 
through the area twice a year, and another group of Mediterranean 
breeding birds that migrated through Darfur in smaller numbers than 
the southern breeders. The southern breeders passed through Darfur 
from May until August, reaching their greatest numbers in June and 
July. Lynes found that these included adults, immature birds, and 
birds of the year, the last varying from 3 to 6 months in age, but 
not in molt, while the adult and subadult birds were molting. He 
further noted that in its middle period the passage was rapid, the 
birds arriving chiefly very early in the morning after some amount of 
nocturnal travel, and moving on during the day, lingering only to 
feed. The migration ended in late July, and no more were seen for 
3 months, except for one stray young bird about 4 months old 
collected on August 20. The Palaearctic breeding migrants passed 
through Darfur in November and December. Lynes found no 
resident breeding great-spotted cuckoos in Darfur, but it would seem 
that further observations may demonstrate that the species breeds 
there regularly, although perhaps not abundantly. It may be recalled 
that some years after Lynes did his field studies, Madden (1934 pp. 
94-95) saw a young fledgling of this cuckoo being attended by its 
foster-parents, a pair of the starling, Lamprotornis caudatus, at 
Khuwei, southern Darfur. Farther to the east, at Dembo, near the 


g2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Bahr-el-Ghazal, the cuckoo has been reported breeding as well, 
and also to the west of Darfur, in Mali, where Malzy (1962, p. 34) 
reported some migrants and some “sedentaires.” 

Madden’s notes from southern Darfur included a very marked 
northward migration through El Fasher in May and June, apparently 
of birds that had finished breeding somewhere to the south, and also 
migrants at Ngala in late June, also apparently of southern birds 
passing through to spend the off-season somewhere to the north of 
Darfur. 

Farther to the south, in the Kagera Park, in the Republic of the 
Congo, Curry-Lindahl (1961, p. 270) recorded a migratory influx 
of these cuckoos January 27 and 28. In Uganda and in Kenya, van 
Someren and others have put on record observations that add up to a 
somewhat obscured picture because of the obvious difficulty of dif- 
ferentiating in the field resident from migrant birds. There is an 
influx of nonbreeding (wintering) visitors from the north, and it is 
possible that southern breeding birds also reach those areas in the 
southern winter. Van Someren (1931, p. 24) found that he could 
distinguish migrants from resident birds, from post-mortems of 
collected specimens, the migrants usually being very fat, the local 
residents not so. Birds seen in Kenya after May were mostly resident, 
which suggests that relatively few individuals from south of the 
Zambezi River reach Kenya. Jackson (1938, pp. 493-495) was in- 
clined to doubt some of van Someren’s statements, but he overlooked 
the fact that the latter had specifically mentioned fledged juvenal 
birds in May in Kenya, which must have been locally raised. 

A similar situation also appears to occur in Tanganyika, but the 
total evidence is much scarcer. There is definite evidence of breeding, 
hence of resident birds, in December at Unyanganyi, and in March at 
Iringa. In Nyasaland the species occurs from mid-September to mid- 
March, with the greatest number of birds noted between September 
and November. 


SUMMARY AND CONCLUSIONS 


The genus Clamator originated in southeastern Africa in Pliocene 
or pre-Pliocene time, from a primordial stock that appears to have its 
least changed, current representative in the southern race of C. 
jacobinus (serratus). From its original locus it expanded its range 
over most of sub-Saharan Africa and spread to India and southeast 
Asia, and thence to the Mediterranean basin as well. In its early 
northward progression in Africa the original jacobinus stock gave 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 93 


rise to a larger, pectorally heavily striped derivative that became 
levaillantti, and, somewhat later, in southern Asia, to an equally large, 
red-winged form, the present stock of which is coromandus. This 
last-named group, in turn, gave rise to what developed into glandarius, 
which emigrated westward from northern India to the Near East, 
Egypt, and to the western portion of the Mediterranean, the Iberian 
Peninsula, Morocco, and Algeria. Much later this form suddenly 
expanded its range southward to encompass much of Africa south as 
far as Cape Province. 

The genus evolved very early from a primordial Cuculine stock 
that was already parasitic in its breeding, but that had not yet de- 
veloped the evicting behavior in the young or the tendency to host- 
adaptive variable egg morphism with the concomitant development of 
host-specific gentes. In the course of its subsequent history Clamator 
never developed either of these features as did the more specialized 
genus Cuculus. Its original eggshell coloration was plain, unmarked 
white as in C. jacobinus serratus, and in this form there is no sign 
of host selection with species reference to egg similarity. From this 
was developed a plain bluish or blue-green egg coloration, as still 
present in the two northern races of jacobinus, pica and the nominate 
subspecies, as well as in Jevaillantit and in coromandus. In these 
segments of the genus the choice of fosterers has been arrived at 
with definite correlation to general egg similarity. Finally, in the 
most advanced species, glandarius, we have a patterned, speckled or 
blotched, egg coloration superimposed on a pale greenish ground color, 
and in this case the inference to be derived from the evidence is that it 
developed together with an early fixation upon magpies as hosts. 

In the case of glandarius, with its unusually fine egg adaptation 
toward this host choice, we find evidence that this restriction, both in 
fosterer and in geographic range, became disadvantageous for the 
species as a whole, and that a large segment of its population under- 
went a great geographic emigration, in a way comparable to what in 
morphological evolution has been termed an “escape from specializa- 
tion.” 

The start of this escape from host restriction on a fosterer of very 
limited sympatry had already begun in eastern Egypt where Corvus 
was utilized in the absence of Pica. The glandarius population that 
expanded over much of sub-Saharan Africa was apparently the less 
perfectly adapted portion of its species in its old Mediterranean home- 
land, as is still evidenced by the great disparity in host-parasite egg 
ratios shown in its uncorrelated multiple parasitism in its newer 


94 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


breeding range. In the course of its rapid spread over this vast 
Pica-allopatric area, it has not altered its eggshell pattern although it 
has broadened greatly its range of acceptance of host nest types, 
enabling it to parasitize such divergent fosterers as arboreal, open- 
nesting corvids and earth-tunnel nesting sturnids. As is pointed out 
in the present paper, the fact that glandarius was parasitic largely on 
birds of greater size, capable of rearing multiple parasites as well as 
some of their own young, gave the cuckoo an immunity from selective 
pressure, but as it increases its use of smaller, sturnid hosts the 
parasite may find itself affected by this pressure, from which it is 
relatively protected as yet. 

A factor that appears to have been of considerable importance in 
the advent of the geographic spread that consummated in the emigra- 
tion to sub-Saharan Africa of the less perfectly adapted portion of 
the original circum-Mediterranean population of glandarius was the 
shift in the main stress of selective pressures when that original 
population became numerically high. Until that demographic satura- 
tion had been reached, and especially while the species was developing 
through its adaptive evolution with regard to its primary host, the 
chief focus of natural selection was between glandarius and its en- 
vironment (including in the latter, its magpie fixation). Once glan- 
darius had become successful and numerous, the primary selective 
pressure was between members of its own kind, and it is this change 
that seems to have been involved in its geographic “shedding” of those 
segments that were less able to stand the new orientation of natural 
selection. This left only the better adapted individuals in the original 
homeland, which is the reason for the difference still apparent be- 
tween them and their sub-Saharan emigres. 

Another evolutionary trend found in some other groups of brood 
parasites, a gradual shortening of the incubation period, is absent in 
Clamator. In fact, the meager data available suggest just the opposite, 
although the more advanced species of the genus, with longer in- 
cubation periods, tend to make use of hosts with still longer ones. 

To summarize, brood parasitism in Clamator has achieved a high 
degree of adaptive excellence by virtue of a restriction of host choice 
to birds of generally similar eggs (northern races of jacobinus, and 
levaillantii, coromandus, and glandarius), and only relatively recently 
has this smoothly functioning correlation been upset by a portion of 
the membership of the most advanced, most “‘perfectly” adapted 
species, glandarius. Traces of incipient tendencies toward egg mor- 
phism may be detected in levaillantii, but they have not developed 
very far. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 95 


The original Clamator stock, as represented by C. jacobinus ser- 
ratus had a tendency toward plumage polymorphism, a trend that 
could, potentially, enhance the process of subsequent differentiation 
into discrete taxonomic entities. This polymorphism remained 
localized in the ancestral home—southeastern A frica—although traces 
of the tendency still occasionally crop up elsewhere in African 
jacobinus and, strangely enough, in a very limited portion of its 
range, in levaillantii. No trace of polymorphism has been found in 
Asiatic jacobinus, in coromandus, or in glandarius. The lack of evo- 
lutionary consequences of this early polymorphism is due to the fact 
of its neutral nature. 

The variations in “normal” plumages show clearly that levaillanti 
was derived from jacobinus; the phylogenetically conservative tend- 
ency of juvenal plumage characters indicates that glandarius arose 
from a coromandus-like stock. The fact that Clamator, during its 
very long existence, has produced only 4 species, as against 12 in 
the younger Cuculus, or 12 in Chrysococcyx (including “Chalcites’), 
coupled with the evolutionarily inert nature of its polymorphic trends, 
suggests that the genus is one that has been relatively less affected 
by evolutionary change. 

Similarly, migratory behavior has remained less completely formu- 
lated and less rigid in its manifestations in many sections of the 
genus, even varying markedly in different segments of individual 
species. We have noted the entire range of behavior from absence of 
migration to local migration, to partial migration, to total and regular 
seasonal mass movements of great geographic extent. 

We must remember that, like other organisms, birds, their struc- 
tures and their habits, do not evolve; they are evolved. The creatures 
are merely the material on which evolutionary processes exert their 
influence and on which they leave their marks and it is from a study 
of these marks that we reconstruct their past history and experience. 

Clamator has existed in a less active, more “secluded,” evolutionary 
arena than some other genera of its family. Nonetheless, it has had a 
long, continuous, and successful history, and in the course of this 
great duration it has shown an early adaptation in egg coloration to 
a then new and fairly definite set of host species, and much later, in 
its climax form, a partial escape from the overly restrictive results of 
this rigid host specificity. In between these two important incidents 
in its development, it has pursued a fairly even and relatively un- 
eventful existence, although involving differentiation into four species, 
each with considerable geographic shifting of stock. This brings out 
the fact that, in studying a group of organisms, the concept of their 


96 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


evolution is not completely the same as that of their history. The 
former should, technically, be limited to the chronology, and, wherever 
possible, the explanation, of the changes that transpired in the crea- 
tures during their history, but should not be considered as the whole of 
their story, even though it may involve its most salient features. An 
organism may, and often does, survive environmental changes with- 
out undergoing change in itself, as we have seen in the case of 
Clamator glandarius. This is certainly a part of its history, but is 
hardly something that ordinarily would be considered in an account 
of its evolution. 


APPENDIX 
ADDITIONAL HOST DATA 
A. DATA ON THE HOSTS OF CLAMATOR GLANDARIUS 


So many new species have been added to my original (1949a, pp. 
10-15) host catalog and so much additional information has been 
amassed on some of the others, and so many changes in nomenclature 
have come about that it seems better to present a new catalog than to 
attempt to present only the new material with the multitude of cross- 
references needed to collate them with what was known before. 
Where the present data suggest no alteration in the earlier state- 
ments, they are given very briefly. 

Two birds, not in the subjoined catalog, have been mentioned in the 
literature as hosts, but there is no evidence to support these state- 
ments. The North African little owl, Athene noctua glaux, has been 
mentioned as a victim, based on a very indefinite statement by Canon 
Tristram (1859, p. 77), which may best be ignored. A year later 
Des Murs (1860, p. 218) wrote that the great-spotted cuckoo laid 
“without doubt” in the nests of the thrush, Turdus merula, but in 
the more than a century since then no one has reported an actual 
instance. This statement should also be ignored. 

The data on documented hosts are given below. 


Falco tinnunculus Linnaeus Kestrel 


Jourdain’s single record (1920, p. 72; Friedmann, 1949a, p. 10) 
has remained unique. The fact that the kestrels were using an old 
magpie’s nest probably was a contributing factor in attracting the 
attentions of the cuckoo, but it should be mentioned that all the eggs 
were fresh; in other words, the kestrels were already in occupancy 
when the cuckoo came there. It suggests that it is the nest itself, 
rather than the actual appearance of its owners, that is of first im- 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 97 


portance to the parasite. The record involves the nominate race of 
the kestrel. 


Upupa epops Linnaeus Hoopoe 


Two records of the South African race (africana) of this bird as 
a host of the great-spotted cuckoo have come to my attention. Miss 
M. Courtenay-Latimer (in litt.) found a hoopoe’s nest at Bailey Sta- 
tion, eastern Cape Province, on September 9, 1931, containing four 
eggs of the hoopoe and one of the cuckoo. One of the hoopoe eggs 
and the cuckoo egg hatched on September 17; two more of the host 
eggs hatched the next day and the last one on the following day. 
The young hoopoes remained in the nest for 3 days, when they were 
found dead and partly devoured by ants nearby. Their actual removal 
from the nest was not observed, so it is not possible to state whether 
they were evicted by the young parasite or had died and were re- 
moved by their parents. The cuckoo chick remained in the nest for 
4 weeks, and the foster-parents were seen feeding it after it fledged, 
until it flew strongly. 

A few years later, near Tregarthens Folly, Cape Province, Miss 
Courtenay-Latimer saw a hoopoe feeding a fledgling great-spotted 
cuckoo on November 7, 12, 15, and 20, 1934. The fact that this 
parent-young relationship was observed on numerous days shows it 
was not a casual, temporary interest of a food-laden adult in a 
clamorous fledgling from another bird’s nest, and also indicates a 
duration of postfledging feeding of at least 2 weeks. The first record, 
entailing an egg of the parasite on September 9, must be one of the 
earliest egg dates for this cuckoo in South Africa. 

The hoopoe has never been found to be parasitized in Europe, but 
Meinertzhagen (1948, p. 563) noted a single cuckoo apparently 
closely associated with a hoopoe in Ushant, Brittany, on April 16, 
1947. Meinertzhagen did not imply parasitism or any other reason 
for the observed association and did not even state if the cuckoo was 
a young bird. If not for Miss Courtenay-Latimer’s observation on 
the South African race of the hoopoe, no one would have suspected 
any host-parasite situation in Meinertzhagen’s terse report. Indeed, as 
written, it affords no basis for any such interpretation, but the ques- 
tion does arise. 


Geocolaptes olivaceus (Gmelin) Ground Woodpecker 


At Waverly Haasfontein, eastern Cape Province, on October 5, 
1952, Miss M. Courtenay-Latimer (in litt.) observed a ground wood- 


98 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


pecker feeding a fledgling great-spotted cuckoo about a week out of 
the nest. This is the only record known to me of this woodpecker 
as a host. 


Corvus corone Linnaeus Hooded Crow 


Two races of this crow have been found to be victimized by the 
great-spotted cuckoo, the nominate form in southern Spain (one 
record, now in the British Museum) and the race sardonius in Israel 
and in Egypt. In the last-named country the hooded crow is the chief, 
if not the invariable, host, and, in all, over 50 Egyptian instances 
have come to my attention. The numerical relations between host and 
parasite eggs there are shown in our graph (fig. 10, p. 43), which 
gives a picture quite different from that of other crows south of the 
Sahara. 


Corvus ruficollis Lesson Brown-necked Raven 


According to Archer (in Archer and Godman, 1961, vol. 3, pp. 
649-659) the local race (edithae) of this bird is the chief host of the 
great-spotted cuckoo in former British Somaliland (now part of the 
Somali Republic). He gave data on five instances from his fieldwork 
in that area, the actual localities being Arori Plain, Burao, Baraad, 
and Oadweina. Two other parasitized nests were collected by M. E. 
W. North, at Brava, former Italian Somaliland (now part of the 
Somali Republic), and sent to the Coryndon Museum. I am indebted 
to J. G. Williams for information about them. Belcher (1949, p. 37) 
reported another parasitized nest near Gabredarre, Ogaden, former 
Italian Somaliland (now a part of the Somali Republic). In this 
case, the nest contained a nestling of the host in addition to eggs 
of its own and of the parasite. 


Corvus corax Linnaeus Raven 


Hartert (1912, p. 956) recorded this bird (typical race) as a host 
of the great-spotted cuckoo in Spain, but the basis for this statement 
is not given; Valverde (1953, p. 294) notes it as found by Lord 
Lilford at Aranjuez. It remains a unique record. The only para- 
sitized set of eggs of any species of Corvus from Spain that I know 
of is a set of Corvus corone in the British Museum. 


Corvus albus Miiller Pied Crow 


The number of records of this crow as a victim of the great-spotted 
cuckoo has been more than doubled since my first (1949a, pp. 11-12) 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 99 


account. I have data on 18 such cases, from South Africa, Southern 
Rhodesia, Nyasaland, Tanganyika, and Nigeria. The number of 
cuckoo eggs per nest varied from 1 to 13, the majority of nests 
having from 1 to 5 of the parasitic eggs. 


Corvus capensis Lichtenstein Cape Rook 


This rook, unique among crows in that it lays pinkish, and not 
greenish, eggs, with which the eggs of the cuckoo contrast markedly 
in coloration, is known to be imposed upon in South Africa and 
Southern Rhodesia (nominate race) and in former British Somaliland 
(race kordofanicus). Of the southern race, there are 13 records in my 
files; of kordofanicus, 2 records (both ex Archer and Godman, 1961, 
vol. 3, p. 657). Five of the southern records are given in my earlier 
reports (1949a, p. 12; 1949b, p. 514). The additional eight are as 
follows: five are from Southern Rhodesia (Beatrice, Banket, Salis- 
bury, and Selukwe) and three from eastern Cape Province, South 
Africa. The number of cuckoo eggs in these sets varies from one to 
four; in six instances there was a single egg of the parasite with 
from one to four eggs of the host; in seven nests each had two eggs 
of the cuckoo with from one to three of the host; one nest had three 
cuckoo and five rook eggs, and one nest contained four eggs of the 
parasite and five of the host. 

Since the above was written, Pitman (1962, p. 23) has recorded 
that as many as nine eggs of the cuckoo have been found in a single 
nest of a Cape rook in South Africa. The exact data on this nest 
were not presented. 


Corvus rhipidurus Hartert Fan-tailed Raven 


As I mentioned in my first account (1949a, pp. 12-13), Lort Phillips 
and his party found in 1885 that in northern former Italian Somaliland 
(now a part of the Somali Republic) nearly all of the examined nests 
of this bird contained eggs of the great-spotted cuckoo, and that in 
one nest there were no fewer than eight eggs of the parasite with four 
of the raven. Archer (im Archer and Godman, 1961, p. 657) recorded 
two parasitized nests found in former British Somaliland (now a part 
of the Somali Republic), one at Sheikh and one at Ariarleh, the 
former with three eggs of the raven and two of the cuckoo, the 
latter with one of the host and four of the parasite. 

Archer found the brown-necked raven, Corvus ruficollis edithae, 
to be the most frequently used host in the Somali Republic, and in his 
discussion he appears to include Lort Phillips’s data as pertaining to 


100 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


that species and not to rhipidurus. In this he was in error. The 
Phillips party was in the field from the end of December until the 
beginning of April, so all the cases of parasitism they observed must 
have come between these dates. Archer’s dates for parasitism on 
edithae were all later, April 29 to June 9, as stated in our discussion 
of that bird. 


Corvultur albicollis (Latham) White-necked Raven 


This species was added to the known hosts of the great-spotted 
cuckoo by McLachlan and Liversidge, in their revision of Roberts’ 
“Birds of South Africa” (1957). No details were given; it was 
merely listed as a host. 


Cyanopica cyanus Pallas Azure-winged Magpie 


The subspecies cooki of this magpie is a frequent victim of the 
great-spotted cuckoo in Portugal and Spain. All in all, I have learned 
of some 11 instances of parasitism on this bird, an increase of 4 
over those listed in my earlier account (1949a, pp. 13-14). 


Pica pica (Linnaeus) Magpie 


The magpie is the primary, almost the exclusive, host of the great- 
spotted cuckoo in the limited portions of the ranges of the two 
species where they are sympatric. It is also the one host to which 
the egg coloration of the parasite is unusually finely adapted. All in 
all, counting nests with eggs, nests with young, and cases of magpies 
attendant upon fledgling cuckoos, over 80 instances of parasitism 
on this host have come to my attention. Geographically they range 
from Spain, southern France (Arles), northwestern Africa (Mo- 
rocco, Tunis, Algeria), and Cyprus, to Turkey (Ankara), and Asia 
Minor. 

Five subspecies of the magpie are involved in these records: 
melanotos in the Iberian Peninsula, pica in Asia Minor, bactriana in 
Iraq, galliae in southern France, and mauritanica in northwest Africa. 
The graph (fig. 12, p. 45) illustrating our present account of “in- 
tensity of parasitism” shows the frequency of multiple eggs of the 
cuckoo in nests of the magpie. This is based on egg records only and 
does not include cases involving nestlings or fledglings. 

The most recent study of the host-parasite relations of the magpie 
is that of Mountfort (1958, pp. 54-56), whose fieldwork was done 
in Spain. He found both birds were very common, and that in spite 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN IOI 


of heavy parasitism the magpies seemed to be thriving. He and his 
party found some 50 occupied nests in 1956. That there is some 
variation in the demographic relations of the two species is suggested 
by Wadley’s notes (1951, p. 75) made in central Anatolia, in Asiatic 
Turkey. He found that the great-spotted cuckoo was well distri- 
buted there, but the total population was quite small, although the 
magpie was numerous throughout. 

Mountfort’s experience seemed to indicate that parasitism was 
generally fatal to the magpie eggs and young, as he wrote that “.. . 
in only one nest did we ever find the young of both species together, 
and this was only a very brief period. The nest in question at one time 
contained three eggs of each species, two of those of the Magpie being 
dented. The remaining Magpie egg hatched three days after those 
of the Great-Spotted Cuckoo. Two days later the nestling Magpie 
had disappeared, presumably having been either smothered or starved. 
Herein lay the crux of the matter, for the incubation period of the 
Great-Spotted Cuckoo is only fourteen days whereas that of the 
much larger Magpie is seventeen to eighteen days. Unless therefore 
the young Magpies can hatch out from eggs laid well in advance of 
those of the parasite, they can have no hope of survival . . .” 


Garrulus glandarius (Linnaeus) Common Jay 


In my earlier account (1949a, p. 15) I mentioned that Tristram 
(1866, p. 282) considered it probable that this jay was parasitized in 
Palestine, but he listed no actual records. Yet, on this basis, several 
authors have repeatedly referred to this jay as a fosterer of the 
cuckoo. Since then, Makatsch (1955, p. 152) definitely reported two 
parasitized nests of the jay in Asia Minor, of the subspecies, G. g. 
krynicki, collected by Kriiper, one on May 9, 1882, and the other on 
May 6, 1901. Each had an egg of the great-spotted cuckoo. I am 
not aware of any other instances of parasitism on this jay. 


Ptilostomus afer (Linnaeus) Piapiac 
Recorded without data, as a victim of the great-spotted cuckoo, 

by Mackworth-Praed and Grant (1952, p. 505). 

Acridotheres tristis tristis (Linnaeus) Common Mynah 


This mynah, introduced into South Africa, has recently been 
found to be parasitized by the great-spotted cuckoo at Estcourt, 
Natal, where Godfrey Symons (1962, p. 343) observed a parasitized 
nest, containing four eggs of the mynah and one of the parasite. 


102 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Onychognathus morio (Linnaeus) Red-winged Starling 


In my first account (1949a, p. 14) I listed three records; since then 
I have learned of six more. The additional cases make it clear 
that the red-winged starling must be looked upon as a regular victim 
of the great-spotted cuckoo in eastern South Africa. Elsewhere in its 
range this starling has not yet been found to be parasitized. 


Spreo bicolor (Gmelin) Pied Starling 


The statement in my earlier account (1949a, p. 14) may be ampli- 
fied as I now have data on many more instances of parasitism on 
this starling. It is the most frequently reported fosterer in eastern 
South Africa, and is the only earth-tunnel nester regularly and fre- 
quently used by the parasite. As many as six cuckoo eggs have been 
taken from one nest of this bird. The graph (fig. 13, p. 46) illustrat- 
ing the discussion of “intensity of parasitism’’ shows the relationship 
in egg numbers of starling and cuckoo in 14 instances. 


Spreo albicapillus Blyth White-capped Starling 


This starling was added to the known hosts of the cuckoo by 
Archer (in Archer and Godman, 1961, p. 657), who found, at Sheikh, 
in former British Somaliland (now a part of the Somali Republic), 
on May 14, a nest of this bird containing two eggs of the parasite in 
addition to five of the host. 


Lamprotornis nitens (Linnaeus) Red-shouldered Glossy Starling 


Three additional cases have been forthcoming from Natal since 
the four (not three as erroneously recorded) cases mentioned in my 
earlier account (1949a, p. 14). All refer to the host race L. n. 
phoenicopterus. 


Lamprotornis caudatus (St. Miill.) Long-tailed Glossy Starling 


At El Obeid, Kordofan, Sudan, in November 1932, Madden (1934, 
p. 94) noted a pair of long-tailed starlings feeding a recently fledged 
great-spotted cuckoo. This is still the only record for this host. When 
I first commented on this case (1956, p. 378) it was the only instance 
of a hole-nesting bird being parasitized by this cuckoo north of 
South Africa. Since then, a similar case, involving another, but allied, 
species of starling, has been reported from former British Somaliland 
(now a part of the Somali Republic), and still another from Southern 
Rhodesia. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 103 


Lamprotornis chalybeus (Hemprich and Ehrenberg) 
Blue-eared Glossy Starling 


Two races of this starling have been recorded as victims of the 
great-spotied cuckoo, cyaniventris in the Somali Republic, and sy- 
cobius in Southern Rhodesia. In former British Somaliland (now a 
part of the Somali Republic), Archer (im Archer and Godman, vol. 
3, 1961, p. 657) found a nest at Sheikh, on May 26, containing one 
egg of the starling and two of the parasite. It is the only record that 
has come to my notice for the race cyanivenitris. 

The southern race, sycobius, was added to the known fosterers of 
the great-spotted cuckoo by Meyer (1959, p. 85), who found a nest 
near Que Que, Southern Rhodesia, on November 15, 1958, which 
contained a young cuckoo, still devoid of feathers, but with the quills 
just appearing on the tail and wings, a young starling, fully feathered, 
about 10 to 14 days old, a dead young starling, and a broken, un- 
hatched starling egg. Three days later the two nestlings were still 
there, but on the following day the young cuckoo was the sole oc- 
cupant and remained there for another week, when it fledged and was 
seen attended by its foster-parents. 


B. DATA ON ADDITIONAL HOSTS OF CLAMATOR JACOBINUS 


Since my first host catalog, a number of additions have been sent 
to me or have appeared in print. These records, with their pertinent 
documentation, are here reported. While these species are additions 
to the earlier catalog they are all infrequently used fosterers, as 
might be assumed from the fact that they have only recently been 
so recorded. They all come from Africa, where the chief hosts, 
bulbuls of the genus Pycnonotus and shrikes of the genus Lantus, 
have been reported in this capacity so many times since the last 
(Friedmann, 1949a) catalog as to leave no doubt as to their primary 
role in the economy of the pied crested cuckoo. 


Centropus grillii Hartlaub Black-bellied Coucal 


A coucal is an unusual host as it builds a fairly domed-over nest 
on the ground, a site not usually favored by the pied cuckoo. The 
one known record comes to me from Dr. Johan Ottow (in litt.), who 
has in his collection a parasitized set of the present coucal species, 
taken at Baviaans Krantz, near Rustenberg, Transvaal, November 
30, 1952. The set contained one egg of the host and one of the 
parasite. The record refers to the race wahlbergi of the coucal and 
serratus of the jacobin cuckoo. 


104 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Dicrurus adsimilis (Bechstein) Glossy-backed Drongo 


In my earlier account (1949a, p. 35) I knew of only one old record 
of Bowker’s, quoted by Layard (1877), which was considered ques- 
tionable. This record remained unique until 1962, when Skead (1962, 
pp. 72-73) found a parasitized nest in which the drongos successfully 
reared a jacobin cuckoo. The nominate race of the drongo, and the 
race serratus of the parasite are here involved. 


Turdoides jardinei (Smith) Arrow-marked Babbler 


This babbler, a very frequent fosterer of the stripe-breasted 
cuckoo, Clamator levaillantii, has been listed as a host of the jacobin 
as well in Northern Rhodesia, by Benson and White (1957, p. 43). 
On geographic grounds this would involve the host race T. 7. 
natalensis. 


Terpsiphone viridis (St. Miill.) Paradise Flycatcher 


Skead (1955, p. 46) found a nest of the paradise flycatcher with 
an egg of C. jacobinus at Fleet Dutch Kloof, King William’s Town, 
eastern Cape Province, December 18, 1954. This is the only instance 
known to me of the pied cuckoo laying in the nest of this species. 
This flycatcher is one of the smallest victims yet recorded. The record 
refers to the race perspicillata of the host, and the race serratus of 
the parasite. 


Sphenoeacus afer (Gmelin) Grass Bird 


In my 1956 discussion (p. 379) I mentioned that I saw a parasitized 
set of eggs of this bird, taken at Inyanga, Southern Rhodesia, by E. F. 
Allen, in the Victoria Memorial Museum, Salisbury. This is still the 
only real record, but I have been informed of one other indefinite 
one since then, also in the general region of Salisbury. The record 
refers to the race transvaalensis of the host and serratus of the 
parasite. This is one of the few birds nesting on, or close to, the 
ground that are occasionally parasitized. 


Motacilla aguimp Dumont Pied Wagtail 


The African pied wagtail was added to the known hosts of the 
pied crested cuckoo by van Someren (1956, p. 236) who found it to 
be imposed upon by no less than three species of cuckoos in the 
Ngong area, near Nairobi, Kenya; the present one, the solitary cuckoo, 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 105 


and the didric cuckoo. The lone record involving the jacobin cuckoo 
refers to the race vidua of the host, and pica of the parasite. 


Telophorus zeylonus (Linnaeus) Bakbakiri 


Additional instances of parasitism on this shrike by the jacobin 
cuckoo bring the total number of cases known to me up to seven, 
and make it clear that this bird is a fairly frequent and regular host 
choice. Of the seven records, six refer to the nominate race of the 
host, one of the grayish, western race phanus. 


C. DATA ON ADDITIONAL HOSTS OF CLAMATOR LEVAILLANTII 


The stripe-breasted cuckoo is still less often observed, and hence 
less completely known, than jacobinus or glandarius. Observations 
since my 1949 host catalog have served chiefly to emphasize the fact 
that babblers of the genus Turdoides form the main reliance of this 
cuckoo. Not only have numerous attitional instances of parasitism on 
the arrow-marked babbler, T. jardinei, come to hand, but also two 
more species of the same genus have been found to be parasitized. 
A single record of parasitism on a coly has also come to my attention, 
but this bird is at best only an irregular or a very occasional victim. 

On the whole, C. Jevaillantii, in its host choice resembles the Asiatic 
population of C. jacobinus, but, as far as present data indicate, is more 
generally restricted to species of Turdoides. 

One observation on the chief host, Turdoides jardinet, calls for 
mention here. Jubb (1952, p. 162) watched a fledgling stripe-breasted 
cuckoo with a family group of arrow-marked babblers and wrote that 
the young parasite “. . . was able to imitate the chatter so char- 
acteristic of babblers on the wing ...’ This would imply some 
vocal adaptation to a host species, such as Nicolai (1961) has sug- 
gested in some of the parasitic Viduinae. In both cases the sugges- 
tion needs further support before it may be appraised. 


Colius striatus Gmelin Speckled Coly 


One record of this coly as a fosterer of the stripe-breasted cuckoo 
has been reported. White and Winterbottom (1949) noted that an 
egg of this cuckoo was found in a coly nest at Ndola, Northern 
Rhodesia, in December, by Hudson. This coly has also been found 
to be victimized very occasionally by the jacobin cuckoo, but it is not a 
regular host to either species. The typical race of the coly is involved 
in the present record. 


106 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Turdoides plebeja (Cretzschmar) Brown Babbler 


In Gambia, Ross A. J. Walton collected two parasitized nests of 
the brown babbler (local race T. p. platycircus) at Kambo, North 
Bank Division; one on April 30, 1945, with one egg each of the 
babbler and the cuckoo, and the other on July 4, 1945, with three eggs 
of the host and one of the parasite. The Adamawa race of this 
babbler, T. ~. gularis, had been known earlier to be parasitized in 
Nigeria. Both sets of eggs were acquired for his collection by 
R. Kreuger of Helsinki, to whom I am indebted for the data. The 
set taken on April 30 has since gone to the collection of Dr. Johan 
Ottow, who informed me (in Itt.) that the locality on his set was 
given as Churchill Town, St. Mary, Gambia. 


Turdoides reinwardii (Swainson) Blackcap Babbler 


R. Kreuger of Helsinki (in litt.) informed me that he received from 
Ross A. G. Walton, a set of one egg of this babbler with one of the 
stripe-breasted cuckoo, taken at Kambo, North Bank Division, Gam- 
bia, on May 1, 1944. This is the only instance I know of this babbler 
as a host. The record refers to the nominate race. 


BIBLIOGRAPHY 


ABARONI, J. 
1932. Bemerkungen und Erganzungen zu R. Meinertzhagen’s Werk “Ni- 
coll’s Birds of Egypt.” Journ. f. Ornith., vol. 80, pp. 416-424. 
ALEXANDER, Boyp. 
1900. An ornithological expedition to the Zambezi River. Ibis, ser. 7, 
vol. 6, pp. 70-109. 
Aut, SAL™ A. 
1927. The Moghul emperors of India as naturalists and sportsmen. Pt. II. 
Journ. Bombay Nat. Hist. Soc., vol. 32, pp. 34-63. 
193la. The origin of mimicry in cuckoos’ eggs. Journ. Bombay Nat. Hist. 
Soc., vol. 34, pp. 1067-1070. 
1931b. Notes on the pied crested cuckoo (Clamator jacobinus) in the Alibag 
Taluka (Kolaba District). Journ. Bombay Nat. Hist. Soc., vol. 34, 
pp. 1071-1072. 
1934. The Hyderabad State ornithological survey. Pt. IV. Journ. Bom- 
bay Nat. Hist. Soc., vol. 37, pp. 124-142. 
1953. The birds of Travancore and Cochin. Bombay, 436 pp., 22 pls. 
1955. The birds of Gujarat. Pt. II. Journ. Bombay Nat. Hist. Soc., vol. 52, 
pp. 735-802. 
ALLEN, GLOVER MorrILL. 
1930. The birds of Liberia. Report of the Harvard African Expedition to 
the African Republic of Liberia and the Belgian Congo, vol. 2, 
pp. 636-748. 
ALLEN, S. STAFFORD. 
1863. Letter to editor. Ibis, vol. 5, pp. 363-364. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 107 


ALLOUSE, BASHIR E. 
1953. The avifauna of Iraq. Publ. No. 3, Iraq Nat. Hist. Mus., 163 pp. 
Amapbon, DEAN. 
1947. Ecology and the evolution of some Hawaiian birds. Evolution, vol. 1, 
pp. 63-68. 
1950. The Hawaiian honeycreepers (Aves, Drepaniidae). Bull. Amer. Mus. 
Nat. Hist., vol. 95, pp. 151-262. 
ANDERSSON, CARL JOHAN. 
1872. Notes on the birds of Damara Land and the adjacent countries of 
South-West Africa. London, xlviii + 394 pp., 4 pls. 
ANONYMOUS. 
1961. Miscellanea. Oolog. Rec., vol. 35, p. 31. 
ARCHER, GEOFFREY, and GopMAN, Eva M. 
1961. The birds of British Somaliland and the Gulf of Aden. Vol. 3, 
pp. 628-1042, 7 pls.; vol. 4, pp. 1045-1570, 7 pls. 
ASHTON-JOHNSON, J. F. R. 
1961. Notes on the breeding birds of Cyprus. Pt. II. Oolog. Rec., vol. 35, 
pp. 17-22. 
BAKER, EpwARD CHARLES STUART. 
1906-07. The oology of Indian parasitic cuckoos. Journ. Bombay Nat. Hist. 
Soc., vol. 17, pp. 72-83, 351-374, 678-696. 
1907. Additional cuckoo notes. Journ. Bombay Nat. Hist. Soc., vol. 17, 
pp. 876-894. 
1913. The evolution of adaptation in parasitic cuckoos’ eggs. Ibis, ser. 10, 
vol. 1, pp. 384-398. 
1921-22. Cuckoos. Some theories about the birds and their eggs. Bull. Brit. 
Ornith. Club, vol. 42, pp. 93-112. 
1923. Cuckoos’ eggs and evolution. Proc. Zool. Soc. London, No. 19, 
pp. 277-294. 


1927. The fauna of British India including Ceylon and Burma. Birds. Vol. 4, 
2d ed., London, 471 pp. 


1934. Nidification of the birds of the Indian empire. Vol. 3 (Ploceidae- 
Asionidae), London, 568 pp. 
1942. Cuckoo problems. London, 207 pp. 
Baker, H. R., and IncLis, CHARLES M. 
1930. The birds of southern India including Madras, Malabar, Travancore, 
Cochin, Coorg, and Mysore. Madras, 504 pp. 
BatpaMus, AuGcust Cart EpuarD. 
1892. Das Leben der europaischer Kuckucke. Nebst Beitragen zur Le- 
benskunde der wtbrigen parasitischen Kuckucke und Starlinge. 
224 pp. 
Bancs, OutTRAM, and LovertpcE, ARTHUR. 
1933. Reports on the scientific results of an expedition to the southwestern 
highlands of Tanganyika Territory. III. Birds. Bull. Mus. Comp. 
Zool., vol. 75, No. 3, pp. 143-221. 
BANNERMAN, DaAvip ARMITAGE. 
1910. On a collection of birds made by Mr. A. B. Percival in British East 
Africa. With field-notes by the collector. Ibis, ser. 9, vol. 4, 
pp. 676-710. 


108 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


1912. On a collection of birds made by Mr. Willoughby E. Lowe on the 
West Coast of Africa and outlying islands; with field notes by the 
collector. Ibis, ser. 9, vol. 6, pp. 219-268. 
1921. A systematic list of the birds of Sierra Leone. Ibis, ser. 11, vol. 3, 
pp. 283-302. 
1922. The birds of Southern Nigeria, Pt. 2. Rev. Zool. Africaine, vol. 10, 
fasc. 2, pp. 88-208. 
1923. Report on the birds collected during the British Museum expedition 
to the Ivory Coast (French West Africa). Ibis, ser. 11, vol. 5, 
pp. 667-748. 
1933. The birds of tropical west Africa. Vol. 3, London, 487 pp., 12 pls. 
1953. The birds of west and equatorial Africa. 2 vols., London, 1,526 pp., 
54 pls., 433 text figs. 
1955. The birds of the British Isles. Vol. 4, Edinburgh, 257 pp. 
BANNERMAN, Davin A., and BANNERMAN, W. M. 
1956. Birds of Cyprus. Edinburgh, 384 pp. 
BarTELs, MAX. 
1910. Zu Dr. C. Parrot’s “Beitrage zur Ornithologie Sumatras und der 
Insel Banka.” Journ. f. Ornith., vol. 58, pp. 484-488. 
Bates, GEorRGE LATIMER. 
1930. Handbook of the birds of West Africa. London, xxiii + 572 pp., 
2 maps. 
Bates, R. S. P. 
1938. On the parasitic habits of the pied crested cuckoo [Clamator jacobinus 
(Bodd.)]. Journ. Bombay Nat. Hist. Soc., vol. 40, p. 125. 
Bates, R. S. P., and Lowrner, E. H. N. 
1952. The breeding birds of Kashmir. Oxford University Press, 368 pp., 
5 pls. 
BEDDARD, FRANK E. 
1885. On the structural characters and classification of the cuckoos. Proc. 
Zool. Soc. London, pp. 168-187. 
BépE, PAUL. 
1926. Notes sur l’ornithologie du Maroc. Mem. Soc. Sci. Nat. Maroc, 
No. 16, pp. 25-150. 
BELCHER, CHARLES FRANCIS. 
1930. The birds of Nyasaland. London, xii + 356 pp., 1 map. 
1949. Eggs from Somalia. Oolog. Rec., vol. 23, pp. 35-41. 
1950. Notes on some eggs collected in northern Nyasaland. Oolog. Rec., 
vol. 24, pp. 2-10. 
BENSON, CONSTANTINE WALTER. 
1940. Further notes on Nyasaland birds. Pt. II. Ibis, ser. 14, vol. 4, 
pp. 387-433. 
1945. Notes on the birds of southern Abyssinia. Pt. 2. Ibis, vol. 87, pp. 488- 
509. 
1953. A check list of the birds of Nyasaland. Blantyre, vi + 118 pp., 1 map. 
BENSON, CONSTANTINE WALTER, and Pitman, C. R. S. 
1956. Some breeding records from Northern Rhodesia. Pt. II. Oolog. Rec., 
vol. 30, pp. 21-27. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 109 


Benson, CONSTANTINE WALTER, and WHITE, CHARLES MaTtHEW NEWTON. 
1957. Check-list of the birds of Northern Rhodesia. Lusaka, 166 pp., 28 pls., 
1 map. 
Bercer, ANDREW J. 
1960. Some anatomical characters of the Cuculidae and the Musophagidae. 
Wilson Bull., vol. 72, No. 1, pp. 60-104. 
Bertioz, M. JACQUES. 
1934. Etude d’une collection d’Oiseaux du Tchad (A. E. F.). Bull. Mus. 
Hist. Nat. Paris, 2d ser., vol. 6, No. 6, pp. 490-496. 
Biswas, BISwWAMOY. 
1960. The birds of Nepal. Pt. 2. Journ. Bombay Nat. Hist. Soc., vol. 57, 
pp. 516-546. 
BLANCHET, ALFRED. 
1955. Les Oiseaux de Tunisie. Memoires de la Soc. des Sci. Nat. de 
Tunisie, No. 3, 84 pp. 
BLANForD, WILLIAM THOMAS. 
1870. Observations on the geology and zoology of Abyssinia made... 
in 1867-68. London, xii + 487 pp. 13 pls. 
1876. Eastern Persia. An account of the journeys of the Persian Boundary 
Commission, 1870-71-72. Vol. 2, Zoology and Geology, 516 pp. 
28 pls. London. 
Bocert, CARDINE. 
1937. Birds collected during the Whitney South Sea Expedition, XXXIV. 
The distribution and the migration of the long-tailed cuckoo (Uro- 
dynamis taitensis Sparrman). Amer. Mus. Novitates, No. 933, 
pp. 1-12. 
Bouvet, GEorGEsS. 
1961. Faune tropicale. XVII. Oiseaux de l’Afrique tropicale (Pt. 2), 
pp. 421-798. 
BoxBeErGER, LEO von. 
1927. Zur fortpflanzung indischer Kuckucksarten. Beitr. z. Fortpflanzungs- 
biol. d. Vogel, vol. 3, pp. 24-25. 
BrapDFIELD, R. D. 
1931. The crested cuckoos of South Africa. Ostrich, vol. 2, pp. 7-9. 
Braun, FRITz. 
1908. Unsere kenntnis der Ornis der Kleinasiatisches Westktiste. Journ. 
f. Ornith., vol. 56, pp. 539-629, 
BreuM, A. E. 
1853. Zur Fortpflanzungsgeschichte des Cuculus glandarius. Journ. f. 
Ornith., vol. I, pp. 144-145. 
1861. Noch einmal das Brutgeschaft von Oxylophus glandarius. Journ. f. 
Ornith., vol. 9, pp. 392-394. 
BRELSFORD, W. VERNON. 
1942. Migration notes, Chinsali, Northern Rhodesia. Ostrich, vol. 13, 
pp. 38-41. 
BRoEKHUYSEN, G. J. 
1956. Occurrence and movement of migratory species in Rhodesia and 
southern Africa during the period 1950-1953. Pt. II. Ostrich, 
vol. 27, pp. 159-167. 


IIo SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


BroM_ey, F. C. 
1952. Short records and notes. Ostrich, vol. 23, pp. 131-132. 
BrosseEt, A. 
1956. Les oiseaux du Maroc oriental de la Mediterranée a Berguent. 
Alauda, vol. 24, pp. 161-205. 
Brown, LEsiie H. 
1948. Notes on birds of the Kabba, Ilorin and N. Benin provinces of 
Nigeria. Ibis, vol. 90, No. 4, pp. 525-538. 
BRUNEL, J. 
1958. Observations sur les oiseaux du Bas Dahomey. L’Oiseau et Rev. 
Francaise d’Ornith., vol. 28, pp. 1-38. 
BuckKNILL, JoHN A. 
1908. A description of some portion of the oological collection of South 
African birds’ eggs in the Transvaal Museum, Praetoria. Journ. 
S. Afr. Ornith. Union, vol. 4, pp. 69-102. 
Butter, ARTHUR LENNOX. 
1908. A second contribution to the ornithology of the Egyptian Soudan. 
Ibis, ser. 9, vol. 2, pp. 205-263. 
CABANIS, JEAN. 
1878. Uebersicht der V6gel Ost-Afrikas, welche von den Herren J. M. 
Hildebrandt und v. Kalckreuth gesammelt sind. Journ. f. Ornith., 
vol. 26, pp. 213-246. 
Caper, Davin R. 
1962. Colour of cuckoo eggs. African Wild Life, vol. 16, No. 4, p. 344. 
CALDWELL, Harry R., and CALDWELL, JoHN C-. 
1931. South China birds. Shanghai, 447 pp. 
CAMPBELL, Louts W. 
1948. Nest-building adaptability of the eastern red-wing. Wilson Bull., 
vol. 60, p. 244. 
CAPEK, V. 
1896. Beitrage zur Fortpflanzungsgeschichte des Kuckucks. Ornith. Jahr- 
buch, vol. 7, pp. 41-72, 102-117, 146-157, 165-183. 
CARPENTER, C. J. 
1933. Contribution a l’Etude d’ornithologie Marocaine. Les Oiseaux du 
Pays Zaian. Bull. Soc. Sci. Nat. du Maroc, vol. 13, pp. 23-68. 
Carter, G. S. 
1954. Animal evolution, a study of recent views of its causes. Rev. ed. 
London, 368 pp. 
Cave, Francis O., and Macponatp, J. D. 
1955. The birds of the Sudan. Edinburgh, 442 pp., 24 pls., 2 maps. 
CHAMBERS, W. J. 
1863. Letter to the editor. Ibis, vol. 5, pp. 474-475. 
CHAPIN, JAMES PAUL. 
1923. Ecological aspects of bird distribution in tropical Africa. Amer. 
Naturalist, vol. 57, pp. 106-125. 
1939. The birds of the Belgian Congo. Part. II. Bull. Amer. Mus. Nat. 
Hist., vol. 75, 632 pp. 
1953. The birds of the Belgian Congo. Part III. Bull. Amer. Mus. Nat. 
Hist., vol. 75A, 821 pp. 
1954. The birds of the Belgian Congo. Part IV. Bull. Amer. Mus. Nat. 
Hist., vol. 75B, 846 pp., 27 pls., 45 text figs. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN III 


CuHuss, E. C. 

1914. A descriptive list of the Millar collection of South African birds’ eggs. 

Annals Durban Mus., vol. 1, pt. 1, pp. 29-106. 
CLANCEY, Pomp A. 

1951. Notes on birds of the South African subcontinent. Annals Natal 
Mus., vol. 12, pt. 1, pp. 137-151. 

1960. Miscellaneous taxonomic notes on African birds. XV.4. The races 
of the jacobin crested cuckoo Clamator jacobinus (Boddaert) oc- 
curring in Africa south of the Zambesi River. Durban Museum 
Novitates, vol. 6, pt. 2, pp. 27-31. 

CocHRANE, J. H. 
1863. Letter to editor. Ibis, vol. 5, pp. 361-363. 
Cotr, Hucu B. 

1940. Adaptive coloration in animals. London, 508 pp., 49 pls. 

1954. Allaesthetic selection and its evolutionary aspects. Jn Evolution as a 
process. Ed. by J. S. Huxley, A. C. Hardy, and E. B. Ford. 
Pp. 47-70. 

Currie, A. J. 

1916. The birds of Lahore and the vicinity. Journ. Bombay Nat. Hist. Soc., 

vol. 24, pp. 561-577. 
Curry-LInDAHL, Kat. 

1960. Ecological studies on mammals, birds, reptiles and amphibians in the 
eastern Belgian Congo. Part II. Annales du Musée Royal du 
Congo Belge Tervaren (Belgique), ser. 8, zool., vol. 87, 170 pp., 
20 pls. 

1961. Exploration du Parc National Albert et du Pare National de la 
Kagera. II. Mission K. Curry-Lindahl (1951-1952, 1958-1959), 
fase Jolepp ee pis. 

DarLincTon, C. D. 
1953. The facts of life. London, 467 pp. 
DARLINGTON, PHIiP J., JR. 

1957. Zoogeography; the geographic distribution of animals. New York, 

675 pp. 
Davies, C. J. 
1911. Notes on the birds of the District of Matatiele, East Griqualand. 
Journ. S. Afr. Ornith. Union, vol. 7, pp. 23-48. 
De Beer, GAvIN R. 
1940. Embryos and ancestors. Oxford University Press, 108 pp. 
DEIGNAN, HERBERT GIRTON. 

1945. The birds of northern Thailand. U.S. Nat. Mus., Bull. 186. 616 pp., 

9 pls., 4 maps. 
DE KLerK, W. DE K. 

1942. Notes on nests of birds in Adelaide, Cape Province. Ostrich, vol. 13, 

pp. 57-63. 
Des Murs, O. 

1860. Traité général d’oologie ornithologique ... au point de vue de la 

classification. Paris, 640 pp. 
DoszHANSKY, THEODOSIUS. 

1940. Speciation as a stage in evolutionary divergence. Amer. Naturalist, 

vol. 74, pp. 312-321. 


TI2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Dorst, JEAN (translated by C. D. Sherman). 
1962. The migrations of birds. 476 pp. Boston. 
DovavuD, JOSEPH. 
1957. Las migrations au Togo (Afrique occidentale). Alauda, vol. 25, 
pp. 241-266. 
Drost, RUDOLF. 
1950. Study of bird migration 1938-1950. Proc. 10th Intern. Ornith. Con- 
gress, Uppsala, pp. 216-240. 
ELLENBERGER, P. M. 
1951. Notes on some birds of the Ndanga (Zaka) District of Southern 
Rhodesia. Ostrich, vol. 22, pp. 17-24. 
ERGENE, SAADET. 
1945. Tiirkiye Kuslari. Istanbul, 361 pp., 104 pls. 
ERLANGER, CARLO FREIHERR VON. 
1900. Beitrage zur Avifauna Tunisiens. Journ. f. Ornith., vol. 48, pp. 1-105. 
1905. Beitrage zur Vogelfauna Nordostafrikas, pt. III. Journ. f. Ornith., 
vol. 53, pp. 433-499, 
EtcHécopar, R. D. 
1942. Polymorphisme des oeufs d’oiseux. L’Oiseau et Rev. Francaise 
d’Ornith., n.s., vol. 12, pp. 121-132. 
1946. Considerations sur le dernier ouvrage de Stuart Baker: “Cuckoo 
Problems.” L’Oiseau et Rev. Francaise d’Ornith., ns., vol. 16, 
pp. 153-167. 
FELL, H. BARRACLOUGH. 
1947. The migration of the New Zealand bronze cuckoo, Chalcites lucidus 
lucidus (Gmelin). Trans. Royal Soc. New Zealand, vol. 76, 
pp. 504-515. 
FinscH, Ottow, and HartLaus, CARL JOHANN GUSTAVE. 
1870. Die Vogel Ost-Afrikas. In Baron Carl Claus von der Decken’s Reisen 
in Ost Afrika in 1859-61, vol. 4, viii + 897 pp., 11 pls. Leipzig. 
FiscHeEr, G. A. 
1885. Uebersicht der in Ostafrika gesammelten Vogelarten, mit Angabe der 
verschiedenen Fundorte. Journ. f. Ornith. vol. 33, pp. 113-142. 
FiscHer, G. A., and REICHENow, A. 
1878. Uebersicht der von Dr. G. A. Fischer auf Sansibar und wahrend 
einer Reise durch das Kiistenland von Mombassa bis Wito gesam- 
melten oder sicher beobachteten Vogel. Journ. f. Ornith., vol. 26, 
pp. 247-268. 
Forses, W. A. 
1885. The collected scientific papers of the late William Alexander Forbes. 
London, xv + 496 pp., 25 pls. 
Forp, E. B. 
1945. Polymorphism. Biol. Review, vol. 20, pp. 73-88. 
Frabe, F. 
1951 (=1953). Catalogo das Aves de Mocambique. Annais Junta de Inves- 
tigacoes Colonais, vol. 6, tomo 4, fasc. 4. Estudios de Zoologia, 
pp. 7-294. 
FRASER, W. 
1962. Colour of cuckoo eggs. African Wild Life, vol. 16, No. 4, p. 343. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN II3 


FRIEDMANN, HERBERT. 
1929. The cowbirds. A study in the biology of social parasitism. Spring- 
field, Ill., 421 pp., 28 pls., 12 text figs. 
1930. Birds collected by the Childs Frick expedition to Ethiopia and Kenya 
Colony. Part I, Non-Passeres. U.S. Nat. Mus., Bull. 153, 516 pp. 
1949a. The parasitic cuckoos of Africa. Washington Acad. Sci., Mono- 
graph No. 1 (1948), xii + 204 pp., 10 pls. 
1949b. Additional data on African parasitic cuckoos. Ibis, vol. 91, pp. 514- 
519. 
1956. Further data on African parasitic cuckoos. Proc. U.S. Nat. Mus., 
vol. 106, pp. 377-408. 
1960. The parasitic weaverbirds. U.S. Nat. Mus., Bull. 223, 196 pp. 
FRIEDMANN, HERBERT, and LoverIDGE, ARTHUR. 
1937. Notes on the ornithology of tropical East Africa. Bull. Mus. Comp. 
Zool., vol. 81, pp. 1-413. 
FURBRINGER, Max. 
1888. Untersuchungen zur Morphologie und Systematik der Vogel ... 2 
vols, Amsterdam. 
Gripert, ANT. 
1925. L’Oxylophe geai en Camargue. Rev. Francaise d’Ornith., vol. 9, année 
17, No. 190, pp. 59-60. 
Gu, 2... Hi. 'N: 
1925. A description of the nests and eggs of the common birds occurring in 
the plains of the United Provinces. Journ. Bombay Nat. Hist. Soc., 
vol. 30, pp, 273-284. 
Grtet, H. 
1960. Observations sur l’Avifaune du Massif de L’Ennedi (Tchad). 
L’Oiseau et la Rev. Francaise d’Ornith., vol. 30, No. 2, pp. 99-134. 
GrapKkov, N. 
1960. Systema Avium Rossicarum Ordo Cuculi ou Cuculiformes. L’Oiseau 
et Rev. Francaise d’Ornith., vol. 30, special supplement, pp. 215-217. 
GopFrEy, Roser. 
1934. Notes on cuckoos from eastern Cape Province. Oolog. Rec., vol. 14, 
pp. 67-69. 
1939. The black-crested cuckoo, Ostrich, vol. 10, pp. 21-27. 
GRANT, CLAUDE HENRY BAXTER. 
1915. On a collection of birds from British East Africa and Uganda pre- 
sented to the British Museum by Capt. G. P. Cosenss. Ibis, ser. 10, 
vol. 3, pp. 400-473. 
GRANT, CLAUDE HENRY Baxter, and MAcKwortH-PRAED, Cyril WINTHROP. 
1936. Notes: 1. On Cuculus clamosus Latham and Cuculus jacksoni Sharpe. 
2. On the status of Coccystes albonotatus Shelley and Coccystes 
caroli Norman. Bull. Brit. Ornith. Club, vol. 56, pp. 123-126. 
1948. On the movements in the non-breeding season of Clamator jacobinus 
serratus (Sparrman). Bull. Brit. Ornith. Club, vol. 68, pp. 171-172. 
GraNvIK, Huco. 
1923. Contributions to the knowledge of East African ornithology. Journ. 
f. Ornith., vol. 71, suppl., 280 pp., 11 pls. 


II4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


1934. The ornithology of northwestern Kenya Colony with special regard 
to the Suk and Turkana District. Rev. Zool. Bot. Africaines, 
vol. 25, fasc. 1, pp. 1-190, 4 pls. 

Gray, W. J. 

1945. Some notes on the nesting of certain birds in northern Nyasaland. 

Ostrich, vol. 16, pp. 49-54. 
Grote, HERMANN. 

1936. Neue Beitrage zur Kenntnis der palaearktisches Zugvogel in Afrika. 

Mitt. Zool. Mus. Berlin, vol. 22, pp. 45-85. 
GYLDENSTOLPE, NILs. 

1921. Zoological results of the Swedish expedition to central Africa 1921. 
Vertebrata 1, Birds. Kungl. Svenska Vetenskaps. Handl., ser. 3, 
vol. 1, No. 3, 326 pp., 2 pls. 

HAAGNER, ALWIN Kart. 

1905. A further contribution to the ornithology of Modderfontein, Trans- 

vaal. Journ. S. Afr. Ornith. Union, vol. 1, pp. 48-56. 
HAAGNER, ALWIN Kart, and Ivy, Ropert H. 
1907a. The birds of Albany Division, Cape Colony. Journ. S. Afr. Ornith. 
Union, vol. 3, pp. 76-116. 
1907b. Sketches of South African bird-life. Cape Town, 316 pp., 158 figs. 
HAARTMAN, Lars VON. 

1957. Adaptation in hole-nesting birds. Evolution, vol. XI, pp. 339-347. 
HALpANE, J. B. S. 

1932. The causes of evolution. London, 235 pp. 
Harpy, A. C. 

1954. Escape from specialization. In Evolution as a process. Ed. by J. S. 

Huxley, A. C. Hardy, and E. B. Ford. Pp. 122-142. 
Harpy, E. 
1940. Bird migration. Quart. Rev., London, vol. 274, pp. 219-232. 
Harrison, C. J. O. 

1963. Eggshell pigmentation in the Jacobin cuckoo, Clamator jacobinus 

(Bodd.). Bull. Brit. Ornith. Club, vol. 83, pp. 154-155. 
HArTERT, ERNST. 

1912. Die Vogel der palaearktischen Fauna, vol. 2. Berlin, pp. xxiv + 833- 
1764. 

1915. List of a small collection of birds from Hausaland, Northern Nigeria. 
Novitates Zool., vol. 22, pp. 244-266. 

1927. Die Brutzeiten der Vogel in Klein-Afrika (final part). Beitr. z. 
Fortpflanzungsbiol. d. Vogel, vol. 3, pp. 188-204. 

HARTERT, Ernst, and JouRDAIN, FRANCIS CHARLES ROBERT. 

1923. The hitherto known birds of Marocco. Novitates Zoologicae, vol. 30, 

pp. 91-152. 
HARTLAUB, GUSTAV. 

1881. Beitrag zur Ornithologie der dstlich-aquatorialen Gebiete Afrika’s. 
Nach Sendungen und Noten von Dr. Emin Bey in Lado. Abhandl. 
Naturwiss. Ver. Bremen, vol. 7, pp. 83-128. 

HAVERSCHMIDT, FRANCOIS. . 

1961. Der Kuckuck Tapera naevia und seine Wirte in Surinam. Journ. 

f. Ornith., vol. 102, pp. 353-359. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN II5 


HEINRICH, GERD. 
1958. Zur Verbreitung und Lebensweise der Vogel von Angola. Journ. 
f. Ornith., vol. 99, pp. 322-362. 
HEINROTH, OSKAR. 
1922. Die Beziehungen zwischen Vogelgewicht, Eigewicht, Gelegegewicht 
und Brutdauer. Journ. f. Ornith., vol. 70, pp. 172-285. 
Hevuciin, Martin THEODOR VON. 
1869-1873. Ornithologie Nordost-Afrika’s der Nilquellen und Kusten- 
Gebiete des Rothen Meeres und des nordlichen Somal-Landes. 
2 vols., ccexxv + 1512 pp., 51 pls. 
(Hewitt, Joun). 
1931. A guide to the vertebrate fauna of the eastern Cape Province. Pt. 1. 
Mammals and Birds. 256 pp. 
Hoescu, WALTER. 
1934. Lamprocolius nitens bispecularis als Wirtsvogel von Clamator glan- 
darius. Ornith. Monatsb., vol. 42, pp. 68-70. 
1939. Kritische Betrachtungen zum Problem des Brutparasitismus bei den 
afrikanischer Viduinen, Beitrag. zur Fortpflanzungsbiol. der Vogel, 
vol. 15, pp. 200-214. 
1955. Der Vogelwelt sudwestafrikas. Windhoek, 300 pp. 
HoescuH, WALTER, and NIETHAMMER, G, 
1940. Die Vogelwelt Deutsch-Stidwestafrikas. Journ. f. Ornith., vol. 88, 
Sonderheft, 404 pp., 1 pl., 76 figs. 
Hopwoop, Cyrit. 
1912. A list of birds from Arakan. Journ. Bombay Nat. Hist. Soc., vol. 21, 
pp. 1196-1221. 
Hug, F. 
1945. Nouveaux cas de nidification du Coucou-Geai, Clamator glandarius 
en France. Oiseau, vol. 15, pp. 89-93. 
1952. Nouvelles observations sur le Coucou Geai en France. Oiseau, vol. 22, 
pp. 303-316. 
1953. Note sur le Coucou-geai dans le Midi de la France. Oiseau, n.s., 
vol. 23, pp. 297-299. 
Hume, A. O., and Oates, E. W. 
1890. The nests and eggs of Indian birds. Vol. 2, 420 pp. London. 
HUux.ey, JuLian S. 
1927. On the relation between egg-weight and body-weight in birds. Journ. 
Linn. Soc. London, Zool., vol. 36, pp. 457-466. 
1955. Morphism and evolution. Heredity, vol. 9, pp. 1-52. 
Irpy, Lt. Cor. L. Howarp. 
1875. The ornithology of the Straits of Gibraltar. London, 236 pp. 
Ivy, R. H. 
1901. Notes on the nesting and other habits of some South African birds. 
Ibis, ser. 8, vol. 1, pp. 18-28. 
JACKSON, FREDERICK JOHN (completed by W. L. Sclater). 
1938. The birds of Kenya Colony and the Uganda Protectorate. 3 vols., 
1592 pp., 24 pls., 241 text figs. London. 
James, H. W. 
1925. Birds observed in the Somerset East District, Cape Province, Union 
of South Africa. Ibis, ser. 12, vol. 1, pp. 621-648. 


116 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Jerom_E, C. H. 
1943. Difficulties of amateur ornithologists in identifying the eggs of para- 
sitic birds. Ostrich, vol. 14, pp. 99-103. 
Jones, HUMPHREY. 
1945. Notes on some birds of the Northern Rhodesia copperbelt. Ostrich, 
vol. 16, pp. 176-183. 
JouRDAIN, FRANCIS CHARLES RoseErT. 
1910. Exhibition of eggs from Cyprus. Bull. Brit. Ornith. Club, vol. 27, 
p. 27: 
1920. Exhibition of a clutch of eggs of the kestrel with an egg of the great 
spotted cuckoo. Bull. Brit. Ornith. Club, vol. 40, p. 72. 
1925. A study on parasitism in the cuckoos. Proc. Zool. Soc. London, 
pp. 639-667. 
1936. The birds of southern Spain. Ibis, ser. 13, vol. 6, pp. 725-763. 
JourRDAIN, FRANCIS CHARLES ROBERT, and SHUEL, RONALD. 
1935. Notes on a collection of eggs and breeding habits of birds near 
Lokoja, Nigeria. ibis, ser. 13, vol. 5, pp. 623-663. 
Jovetic, R. 
1955. Apparition du Coucou-geai (Clamator glandarius [L.]) en R. P. 
De Macedoine (Yougoslavie). L’Oiseau, n.s., vol. 25, pp. 206-210. 
Juss, R. A. 
1952. Some notes on birds of Southern Rhodesia. Ostrich, vol. 23, pp. 162- 
164. 
Keay, R. W. 
1959. Oxford vegetation map of Africa south of the Tropic of Cancer. 
24 pp. of text, 1 map. 
KINNEAR, NorMAN B. 
1922. On the birds collected by Mr. A. F. R. Wollaston during the first 
Mt. Everest expedition. Ibis, ser. 11, vol. 4, pp. 495-526. 
KINNEAR, NorMAN B., and WHISTLER, HucuH. 
1930. The Vernay scientific survey of the eastern Ghats. Ornithological 
Section, Journ. Bombay Nat. Hist. Soc., vol. 34, No. 2, pp. 386-403. 
Koenic, ALEXANDER. 
1920. Die Sitzfiissler (Insessores) Aegyptens. Journ. f. Ornith., vol. 68, 
Sonderheft, 148 pp. 
1931 (n.d.). Katalog der Nido-Oologischen Sammlung im Museum Alex- 
ander Koenig. Band I. 352 pp. 
KRupeErR, TH. 
1875. Beitrag zur Ornithologie Klein-Asiens. Journ. f. Ornith., vol. 23, 
pp. 258-285. 
KUMMERLOWE, Hans, and NIETHAMMER, GUNTHER. 
1935. Beitrage zur Kenntnis der Avifauna Kleinasiens (Paphagonien- 
Galatien). Journ. f. Ornith., vol. 83, pp. 25-75. 
Kutter, Dr. 
1878. Betrachtungen iiber systematic und Oologie vom Standpunkte der 
Selectionstheorie. Journ. f. Ornith., vol. 26, pp. 300-348. 
Lack, Davin LAMBERT. 
1942. Ecological features of the bird fauna of British small islands. Journ. 
Animal Ecol., vol. 11, pp. 9-36. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 117 


1943-44. The problem of partial migration. Brit. Birds, vol. 37, pp. 122- 
130, 143-150. 
1947a. The significance of clutch size. Ibis, vol. 89, pp. 302-335. 
1947b. Darwin’s finches. Cambridge University Press, 208 pp., 9 pls., 27 figs. 
1951. Population ecology in birds. A review. Proc. 10th Int. Ornith. Con- 
gress, 1950, Uppsala, pp. 409-448. 
1954. The natural regulation of animal numbers. Oxford, 343 pp. 
LAFERRERE, MArc. 
1956. L’Oxylophe Clamator glandarius (Lin.) dans les pinédes de 1’Esterel 
et de la cote Varoise. Alauda, vol. 24, pp. 275-286. 
LAMBRECHT, KALMAN. 
1933. Handbuch der Palaeornithologie. 1022 pp., 4 pls., 209 text figs. 
Lamm, DonaLp W. 
1955. Local migratory movements in Southern Mozambique. Ostrich, 
vol. 26, pp. 32-37. 
LAYARD, EpGar LEOPOLD. 
1875-84. The birds of South Africa, new ed. London, 890 pp., 12 pls. 
1877. Note on two African cuckoos of the genus Coccystes. Proc. Zool. 
Soc. London, pp. 465-466. 
LeicH, M. G. 
1942. Notes on the jacobin cuckoo. Ostrich, vol. 13, pp. 105-107. 
LEVAILLANT, FR. 
1806. Histoire Naturelle des Oiseaux d’Afrique. Vol. 5, pp. 160. Paris. 
LEVEQUE, R. 
1957. Notes sur la distribution et l’extension du Coucou-geai en France 
mediterranéenne. Alauda, vol. 25, pp. 227-229. 
LitForp, Lorp. 
1866. Notes on the ornithology of Spain. Ibis, No. 5, vol. 2, pp. 173-187. 
LINCOLN, FREDERICK CHARLES. 
1939. The individual vs. the species in migration studies. Auk, vol. 56, 
pp. 250-254. 
LINSDALE, JEAN Myron. 
1937. The natural history of magpies. Pacific Coast Avifauna, No. 25, 
234 pp. 
LIvERSIDGE, R. 
1961. Pre-incubation development of Clamator jacobinus. Ibis, vol. 103a, 


p. 624. 
Livesey, T. R. 
1935. Cuckoo problems. Journ. Bombay Nat. Hist. Soc., vol. 38, pp. 734- 
758. 


LONNBERG, EINAR. 
1929. The development and distribution of the African fauna in connection 
with and depending upon climatic changes. Ark. Zool., vol. 21 A, 
No. 4, pp. 1-33, 5 figs. 
LovERIDGE, ARTHUR. 
1922. Notes on East African birds (chiefly nesting habits and stomach 
contents) collected 1915-1919. Proc. Zool. Soc. London, pp. 837- 
862. 


118 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Lowe, WILLoucHByY P. 
1921. The birds of Tasso and adjoining islands of the Rockelle River, Sierra 
Leone. Ibis, ser. 11, vol. 3, pp. 265-282. 
1937. Report on the Lowe-Waldron expedition to the Ashanti forests and 
Northern Territories of the Gold Coast. Ibis, ser. 14, vol. 1, 
pp. 635-662, 830-864. 
LucANUS, FREDRICH VON. 
1921. Zur Frage der Mimikry der Kuckuckseier. Journ. f. Ornith., vol. 69, 
pp. 239-258, 
Lynes, HuBERrT. 
1924-25. On the birds of north and central Darfur, with notes on the 
west-central Kordofan and North Nuba Provinces of the British 
Sudan. Ibis, ser. 11, vol. 6, pp. 399-446, 648-719; ser. 12, vol. 1, 
pp. 71-131, 344-416, 541-590, 758-797. 
1934. Birds of the Ubena-Uhehe highlands and Iringa uplands. Journ. f. 
Ornith., vol. 82, suppl., 147 pp., 15 pls. 
1938. Contribution to the ornithology of the southern Congo Basin. Lynes- 
Vincent tour of 1933-34. Rev. Zool. Bot. Africaines, vol. 31, 
pp. 1-128, pls. 1-13. 
MacDona.p, MALCOLM. 
1960. Birds in my Indian garden. London, 192 pp. 
MackwortH-PrRAED, CyrIL WINTHROP, and Grant, CLAUDE HENRY BAXTER. 
1937. Systematic notes on East African birds—Part XIV. 32. On the 
relationship of Clamator serratus (Sparrm.) Clamator jacobinus 
pica (Hempr. and Ehr.), and Clamator hypopinarus (Cab. and 
Heine). Ibis, ser. 14, vol. 1, pp. 402-406. 
1952. Birds of Eastern and North Eastern Africa, vol. 1, xxv + 836 pp., 
59 pls. London, New York. 
MappEN, J. F. 
1930. Bird migration in the Red Sea Province. Sudan Notes and Records, 
vol. 12, pp. 123-135. 
1934. Notes on the birds of southern Darfur. Sudan Notes and Records, 
vol. 17, pp. 83-101. 
MAKATSCH, WOLFGANG. 
1951. Uber den Brutparasitismus der Vogel. Urania, vol. 14, pp. 210-216. 
1955. Der Brutparasitismus in der Vogelwelt. 236 pp. 
MALprRANT, R. 
1952. Faune du Centre Africain Francais (mammiferes et oiseaux), ed. 2. 
In Encyclopedie biologique, vol. 15, 616 pp., 32 pls., 129 figs. Paris. 
MALprRANT, R., and Macratcuy, A. R. 
1949. La faune de l’Equateur Africain francais. Tome 1, Oiseaux. 460 pp., 
119 figs., 12 pls. 
Mazzy, P. 
1962. La Faune Avienne du Mali. L’Oiseau et Rev. Francaise d’Ornith., 
vol. 32, special number, pp. 1-81. 
MarcHaAnt, S. 
1953. Notes on the birds of south-eastern Nigeria. Ibis, vol. 95, pp. 38-69. 
1963. The breeding of some Iraqi birds. Ibis, vol. 105, pp. 516-557. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN II9 


Mayr, Ernst. 
1957. Evolutionary aspects of host specificity among parasites of vertebrates. 
Premier Sympos. sur la specificite parasitaire des parasites de Ver- 
tebres. Union Internat. Sci. Biol., ser. B, vol. 32, pp. 7-18. 
1959. Change of genetic environment and evolution. Jn Evolution as a 
process, ed. by J. S. Huxley, A. C. Hardy, and E. B. Ford. Pp. 157- 
180. 
Mayr, Ernst, and MEtIse, WILHELM. 
1930. Theoretisches zur Geschichte des Vogelzugs. Vogelzug, vol. 1, 
pp. 149-172. 
McLacuian, G. R., and LiversinceE, R. 
1957. Roberts’ birds of South Africa. Rev. ed. 504 pp., 56 pls. Cape Town. 
MEIKLEJOHN, M. F. M. 
1940. Notes on migratory birds from the southern shores of Lake Victoria. 
December 22, 1939, to February 28, 1940. Ostrich, vol. 11, pp. 33-40. 
MEINERTZHAGEN, RICHARD. 
1922. Notes on some birds from the Near East and from tropical East 
Africa. Ibis, ser. 11, vol. 4, pp. 1-74. 
1930. Nicoll’s birds of Egypt, vol. 1, 348 pp., 15 pls. London. 
1948. The birds of Ushant, Brittany. Ibis, vol. 90, pp. 553-567. 
1951. Some relationships between African, Oriental, and Palearctic genera 
and species, with a review of the genus Monticola. Ibis, vol. 93, 
pp. 443-459. 
1954. The birds of Arabia. 624 pp., 18 pls. Edinburgh, London. 
MEISE, WILHELM. 
1930. Zum Problem der Brutschmarotzertums. Beitr. Fortpflanz Vogel, 
vol. 6, pp. 119-131. 
1937. Zur Vogelwelt des Matengo-Hochlandes nahe dem Nordende des 
Njassasees. Mitt. Zool. Mus. Berlin, vol. 22, pp. 86-160, 4 pls. 
Meyer, H. F. 
1959. Great spotted cuckoo parasitizing a hole-nesting species in Southern 
Rhodesia. Ostrich, vol. 30, p. 85. 
MIstTEIN, PETER. 
1954. An egg problem. Bokmakerie, vol. 6, pp. 4-5. 
Mo ttTon1, EpGArpo. 
1940. Gli Uccelli dell Africa Orientale Italiana. Parta Prima. Milano, 
261 pp., 38 pls. 
Moore, H. J. and Boswett, C. 
1956. Field observations on the birds of Iraq. Pt. II. Publ. No. 10, Iraq 
Nat. Hist. Mus., pp. 111-213. 
MorEAvu, REGINALD ERNEST. 
1937-38. Amami Memoirs. The avifauna of the mountains along the Rift 
Valley in north central Tanganyika Territory (Mbulu Distr.). 
Pts. 1 & 2. Ibis, ser. 14, vol. 1, pp. 760-768; vol. 2, pp. 1-32. 
1937. Migrant birds in Tanganyika Territory. Tanganyika Notes and 
Records, No. 4, pp. 17-50. 
1938. Bird migration over the north western part of the Indian Ocean, 
the Red Sea, and the Mediterranean. Proc. Zool. Soc. London, 
vol. 108A, pp. 1-26. 


I20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


1949. Special Review. Friedmann on African cuckoos. Ibis, vol. 91, pp. 529- 
537. 
1951. The migration system in perspective. Proc. 10th Internat. Ornith. 
Congress, 1950, pp. 245-248. 
1952. The place of Africa in the Palearctic migration system. Journ. Anim. 
Ecol. vol. 21, pp. 250-271. 
1954. The main vicissitudes of the European avifauna since the Pliocene. 
Ibis, vol. 96, pp. 411-431. 
Morean, A. G. 
1956. The redwing. Journ. St. Andrews College Nat. Hist. Soc., Grahams- 
town, Cape Province, vol. 3, No. 1, p. 15. 
Morcan, RuHopes W. 
1875. On the nidification of certain south-Indian birds. Ibis, ser. 3, vol. 5, 
pp. 313-323. 
MountrFort, Guy. 
1958. Wild paradise. Boston, 240 pp., 59 pls. 
Mountrort, Guy, and Fercuson-Lees, I. J. 
1961. The birds of the Coto Dofiana. Ibis, vol. 103a, pp. 86-109. 
MULLER, AUGUST. 
1879. Zur Ornithologie der Insel Cypern. Journ. f. Ornith., vol. 27, pp. 385- 
393. 
MussELWHiteE, D. W. 
1923. Oviposition of cuckoo in domed nests. Bull. Brit. Ornith. Club, 
vol. 44, pp. 30-32. 
NeEuHRKorN, A. 
1899. Katalog der Eiersammlung. Braunschweig, 256 pp., 4 pls. 
NEUMANN, OSCAR. 
1900. Beitrage zur Vogelfauna von Ost- und Central-Africa. Pt. III. Journ. 
f. Ornith., vol. 48, pp. 185-228. 
1904. Vogel von Schoa und Siid-Athiopien. Ploceidae. Journ. f. Ornith., 
vol. 52, pp. 321-410. 
Nice, MARGARET Morse. 
1937. Studies on the life history of the song sparrow. Pt. I. Trans. 
Linnaean Soc. New York, vol. 4, pp. 1-247. 
Nicoral, J. 
1961. Die Stimmen einiger Viduinen. Journ. f. Ornith., vol. 102, pp. 213- 
214. 
NIETHAMMER, G. 
1955. Zur Systematik der Vogel des Kaoko-Veldes (Siidwestafrika). Bon- 
ner Zool. Beitrage, vol. 6, pp. 173-195. 
Nosre, HEATLEY. 
1902. Forty-four days’ nesting in Andalucia. Ibis, ser. 8, vol. 2, pp. 67-89. 
NorMAN, GEoRGE CAMERON. 
1888. Note on the geographical distribution of the crested cuckoos (Coc- 
cystes). Ibis, ser. 5, vol. 6, pp. 396-409. 
Oates, EuGENE W., and RE1p, SAVILLE G. 
1903. Catalogue of birds’ eggs in the British Museum. Vol. 3, 349 pp., 
10 pls. London. 
OcitviE-GRANT, WILLIAM RoBERT. 
1912. Catalogue of birds’ eggs in the British Museum. Vol. 5, 457 pp., 
22 pls. London. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN I21I 


OsmasrTon, B. B. 

1916. Notes on cuckoos in Maymyo. Journ. Bombay Nat. Hist. Soc., vol. 24, 
pp. 359-363. 

1927. Notes on the birds of Kashmir. Pt. II. Journ. Bombay Nat. Hist. 
Soc., vol. 32, pp. 134-153. 

PAKENHAM, R. H. W. 

1948. Field notes on the birds of Zanzibar and Pemba Islands. 5th ser. 

Ibis, vol. 90, pp. 98-102. 
PALUDAN, KNup. 

1936. Report on the birds collected during Professor O. Olufsen’s expedi- 
tion to French Sudan and Nigeria in the year 1927; with field- 
notes by the collector, Mr. Harry Madsen. Vidensk-Medd. fr. 
Dansk naturh. Foren, vol. 100, pp. 247-346. 

1938. Zur Ornis des Zagross-Gebietes, W. Iran. Journ. f. Ornith., vol. 86, 
pp. 562-638. 

Peters, JAMES LEE. 

1940. Check-list of birds of the world. Vol. 4, 291 pp. Cambridge. 
Puiiuirs, W. W. A. 

1948. Cuckoo problems in Ceylon. Spolia Zeylanica, vol. 25, pt. 2, pp. 45-60. 
Pixg, E. 

1954. The birds of Blythswood and some notes on birds of the district. 
Ostrich, vol. 25, pp. 115-129. 

Pinto, A. Rosa. 

1953. Uma lista sistematica das Aves de regiao extremo sul da Provincia de 

Mocambique. Bol. Soc. Est. Mocambique, No. 77, pp. 1-27. 
PITMAN, CHARLES Rosert S. 

1931. Variation in the plumage of Lavaillant’s cuckoo, Clamator cafer 
(Licht.). Bateleur, vol. 3, p. 50. 

1962. Comments on some breeding records in “The Birds of British Somali- 
land and the Gulf of Aden” in vol. III by Sir Geoffrey Archer and 
Eva M. Godman. Pt. 1. Oologists’ Record, vol. 36, pp. 22-27. 

1963a. Comments on some breeding records in “The Birds of British Somali- 
land and the Gulf of Aden,” vol. IV, by Sir Geoffrey Archer and 
Eva M. Goodman. Oologists’ Record, vol. 37, pp. 49-54. 

1963b. The breeding of the great spotted cuckoo, Clamator glandarius L., in 
Iraq, and a note on the Iraq eggs of the Kashmir magpie, Pica pica 
bactriana Bp. Bull. Brit. Ornith. Club, vol. 83, pp. 9-11. 

PLoweEs, DARRELL CHARLES HERBERT, 

1946. Data of birds’ eggs in my collection. Pt. 2. Ostrich, vol. 17, pp, 111- 
121. 

1947. The birds of Ladysmith, Natal. Ostrich, vol. 18, pp. 134-154. 

PLoweEs, Darrett C. H., and Cusack, E. 

1944. A brief survey of the birds of Bloemhof district. Ostrich, vol. 15, 
pp. 81-103. 

Ponce DE LEoNn, Pepro Diez, and Deetyen, Hans. 

1956. Uber die Vogelwelt der Umgebung von Madrid. Ornith. Mitteilungen, 
vol. 8, pp. 124-126. 

Poutton, E. B. 

1926. The evolution of the colours and patterns of cuckoos’ eggs and its 
relation to that of insect resemblances such as mimicry. Proc. Ent. 
Soc. London (for 1925), pp. xcii-xciv. 


122 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Priest, Cectt DAMER. 
1934. The birds of Southern Rhodesia, vol. 2, 553 pp., 10 pls., 172 figs. Lon- 
don and Beccles. 
1948. Eggs of birds breeding in Southern Africa. Glasgow, 180 pp., 20 pls. 
PRINGLE, V. L. 

1948. Egg records of cuckoos. Ostrich, vol. 19, pp. 155-156. 
Procter, F. W. 

1909. Exhibition of eggs. Bull. Brit. Ornith. Club, vol. 23, pp. 59-60. 
Ranp, AUSTIN LooMER, FRIEDMANN, HERBERT, and TrAytor, MELvIN A. 

1959. Birds from Gabon and Moyen Congo. Fieldiana, Zool., vol. 41, no. 2, 
pp. 223-411. 

Raw, W. 

1921. Field notes on the birds of Lower Egypt. With contributions by 
Col. R. Sparrow and the Rev. F. C. R. Jourdain. Ibis, ser. 3, 
vol. 3, pp. 359-387. 

REICHENOW, ANTON. 

1897. Uber die Pflegeeltern von Coccystes serratus. Ornith. Monatsb., 
vol. 5, p. 112. 

1902-03. Die Vogel Afrikas, vol. 2. 752 pp. 

Ret, S. G. 
1885. Winter notes from Morocco. Ibis, ser. 5, vol. 3, pp. 241-255. 
REISER, O. 

1913. Ueber die Erbeutung eines Haherkuckucke, Clamator glandarius (L.) 
und eine wichtige Oertlichkeit zur Beobachtung des Vogelzuges. 
Ornith. Jahrb., vol. 24, pp. 81-85. 

RENNIE, FERGUS. 
1944. Letter to editor. Ostrich, vol. 15, p. 144. 
RENSCH, BERNARD. 

1924. Zur Entstehung der Mimikry der Kuckuckseier. Journ. f. Ornith., 
vol. 72, No. 4, pp. 461-472. 

1925. Das Problem des Brutparasitismus bei Vogeln. Sitzungsber. Ges. 
Naturf. Freunde, Berlin, pp. 55-69. 

Rey, EuGENE. 

1871. Jugendkleider und Eier einiger Vogel aus Klein Asien. Journ. f. 
Ornith., vol. 19, pp. 459-462. 

1872. Zur Fortpflanzungsgeschichte unseres Kukuks. Zool. Garten, pp. 241- 
243. 

1872. Zur Ornis von Portugal. Journ. f. Ornith., vol. 20, pp. 140-155. 

1892. Altes und neues aus dem Haushalte des Kuckucks. Leipzig. 108 pp. 

RHODESIAN ORNITHOLOGICAL SOCIETY. 
1959. Report on nest record cards. Bokmakeirie, vol. 11, pp. 19-20. 
Rrprey, S. DILton. 

1961. A synopsis of the birds of India and Pakistan; together with those 
of Nepal, Sikkim, Bhutan, and Ceylon. xxxvi + 702 pp., 2 double 
maps (one colored) with overlays, The Bombay Natural History 
Society, Madras. 

Roserts, AUSTIN. 

1924. Synoptic check list of the birds of South Africa. Annals Transvaal 

Mus., vol. 10, pp. 89-195, pls. 2-5. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 123 


1932. Migration of African birds. Ostrich, vol. 3, pp. 97-109. 
1938. Reviews and records. Ostrich, vol. 9, pp. 105-119. 
1939a. Notes on the eggs of parasitic birds in South Africa. Ostrich, vol. 10, 
pp. 1-20, 100-117. 
1939b. Swifts and other birds nesting in buildings. Ostrich, vol. 10, pp. 85-99. 
1942. Review of Baker’s “Cuckoo Problems.” Ostrich, vol. 13, pp. 182-183. 
Roserts, AUSTIN (rev. by McLacu tan, G. R., and LiversincE, R.). 
1957. Birds of South Africa. Cape Town, 504 pp., 56 pls. 
Roserts, E. L. 
1954. The birds of Malta. Malta, 168 pp. 
SAUNDERS, Howarp. 
1869. Notes on the ornithology of Italy and Spain. Ibis, ser. 2, vol. 5, 
pp. 391-403. 
SCHIERMANN, G. 
1926. Beitrag zur Schadigung der Wirtsvogel durch Cuculus canorus. 
Beitr. z. Fortpflanzungsbiol. der Vogel, vol. 2, pp. 28-30. 
SCHLUTER, WILHELM : 
1859. Zweifel am Nichtbriiten des Orylophus glandarius. Journ. f. Ornith., 
vol. 7, pp. 238-239, 
ScHMALHAUSEN, I. I. 
1949. Factors of evolution. The theory of stabilizing selection. (Transl. by 
I. Dordick, ed. by T. Dobzhansky). Philadelphia, 327 pp. 
ScHONWETTER, M. 
1928. Anmerkungen zum Nehrkorn—Katalog (final part). Beitr. z. Fort- 
pflanzungsbiol. d. Vogel, vol. 4, pp. 129-136. 
ScHOUTEDEN, HENRI. 
1954. Faune du Congo Belge et du Ruanda Urundi. III. Oiseaux non 
Passereaux, Tervuren, 434 pp. 
SCHWEINFURTH, GEORGE; RATZEL, FRIEDRICH; FELKIN, Ropert W.; and Hart- 
LAUB, GUSTAV. 
1888. Emin Pasha in Central Africa, being a collection of his letters and 
journals. London, 547 pp., 2 pls., map. 
ScLATER, WILLIAM LUTLEY. 
1903. The birds of South Africa, commenced by Arthur Stark. Vol. 3, 
416 pp. London. 
1906. The migration of birds in South Africa. Journ. S. Afr. Ornith. Union, 
vol. 2, pp. 14-21. 
1924. Systema avium Aethiopicarum. Pt. 1, 304 pp. London. 
ScLaTerR, WILLIAM LuTLEY, and MAcKworTH-PRAED, CyrIL WINTHROP. 
1919, A list of the birds of the Anglo-Egyptian Sudan, based on the col- 
lections of Mr. A. L. Butler, Mr. A. Chapman, and Capt. H. Lynes, 
R. N., and Major Cuthbert Christy. Pt. III. Ibis, ser. 11, vol. 1, 
pp. 628-707. 
SERLE, WILLIAM. 
1939. Field observations on some Northern Nigerian birds. Pt. I. Ibis, ser. 
14, vol. 3, pp. 654-699. 
1940. Field observations on some Northern Nigerian birds. Pt. II. Ibis, 
ser. 14, vol. 4, pp. 1-47. 


I24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


1943. Further field observations on Northern Nigerian birds (cont.). Ibis, 
vol. 85, pp. 264-300. 

1948. The birds of Sierra Leone, pt. 2. Ostrich, vol. 19, p. 187-199. 

1957. A contribution to the ornithology of the eastern region of Nigeria. 
Pt. I. Ibis, vol. 99, pp. 371-418. 

SHARPE, RICHARD BOWDLER. 

1872. On recent collections of birds from the Fantee County, in Western 
Africa. Ibis, ser. 3, vol. 2, p. 66-74. 

1873. On the Cuculidae of the Ethiopian region. Proc. Zool. Soc. London, 
pp. 578-624. 

SHELLEY, GEORGE ERNEST. 

1881. List of birds recently collected by Dr. Kirk in Eastern Africa. Proc. 
Zool. Soc. London, pp. 561-602. 

1885. On Mr. E. Lort Phillips’s collection of birds from Somali-Land. 
Ibis, ser. 5, vol. 3, pp. 389-418. 

1891. Catalogue of birds in the British Museum. Vol. 19: Indicatoridae, 
Capitonidae, Cuculidae, and Musophagidae, pp. 209-456. 

SHuFELpDT, R. W. 

1901. The osteology of the cuckoos (Coccyges). Proc. Amer. Philos. Soc., 

vol. 40, pp. 4-51, pls. 1-2. 
S1sLEy, CHARLES G. 

1960. The electrophoretic patterns of avian egg-white proteins as taxonomic 

characters. Ibis, vol. 102, pp. 215-284. 
Simmons, R. M. 

1930. Migration of the pied crested-cuckoo (Coccystes jacobinus). Journ. 

Bombay Nat. Hist. Soc., vol. 34, pp. 252-253. 
SjJOstTEptT, Y. 

1910. Wissenschaftliche Ergebnisse der schwedischen zoologischen Expedi- 
tion nach Kilimandjaro, dem Meru und dem umgebenden Mas- 
saisteppen Deutsch-Ostafricas, 1905-06; vol. 1, Vogel, 173 pp. 

SKEAD, C.J; 

195la. Cuckoo studies on a South African farm. Ostrich, vol. 22, pp. 163-175. 

1951b. Recent records of interest. Ostrich, vol. 22, p. 197. 

1955. Jacobin crested cuckoo parasitizing paradise flycatcher’s nest. Ostrich, 
vol. 26, p. 46. 

1962. Jacobin crested cuckoo, Clamator jacobinus (Boddaert), parasitizing 
the fork-tailed drongo, Dicrurus adsimilis (Bechstein). Ostrich, 
vol. 32, pp. 72-73. 

SmitH, K. D. 

1951. On the birds of Eritrea. Ibis, vol. 93, pp. 201-233. 

1957. An annotated check list of the birds of Eritrea. Ibis, vol. 99, pp. 307- 
337. 

SMITHERS, Reay H. N., Irwin, M. P. Stuart, and Peterson, Mary L. 

1957. A check list of the birds of Southern Rhodesia. Cambridge University 
Press, 175 pp. 

SMyYTHIES, BERTRAM E. 
1953. The birds of Burma. 2d ed. London and Edinburgh. 668 pp., 30 pls. 
1960. The birds of Borneo, London and Edinburgh. 561 pp., 45 pls. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 125 


SouTHErN, H. N. 
1954. Mimicry in cuckoos’ eggs. Jn Evolution as a process, ed. by J. S. 
Huxley, A. C. Hardy, and E. B. Ford. Pp. 219-232. 
SPEncE, J. M. 
1957. Notes on the black crested cuckoo, Clamator jacobinus serratus. 
Ostrich, vol. 28, p. 126. 
STADLER, Hans. 
1954. La voix du Coucou-geai, Parc d’Aranjuez 6. VI. 29. Alauda, vol. 19, 
pp. 178-180. 
STRESEMANN, ERWIN. 
1924. Mutationsstudien VIII. Clamator serratus (Sparrman). Journ. f. 
Ornith., vol. 72, pp. 79-83. 
1926. Uebersicht tiber die . . . Mutationsstudien I-XXIV under ihre wich- 
tegsten Ergebnisse. Journ. f. Ornith., vol. 74, pp. 377-385. 
1928. Aves-Vogel. Jn Kuckenthal und Krumbach, Handbuch der Zoologie 
(1927-34), vol. 7, 899 pp., 944 figs. Berlin and Leipzig. 
1947. Laniarius nigerrimus (Rchw.): a mutation of Lawarius ferrugineus 
sublacteus (Cassin). Ibis, vol. 89, pp. 518-519. 
STRESEMANN, VEsTA, and STRESEMANN, ERWIN. 
1961. Die Handschwingen—Mauser der Kuckucke (Cuculidae). Journ. f. 
Ornith., vol. 102, pp. 317-352. 
STRICKLAND, H. E. 
1850. List of birds procured in Kordofan by Mr. J. Petherick. Proc. Zool. 
Soc. London, pt. 18, pp. 214-221. 
SWYNNERTON, CHARLES FRANCIS Massy. 
1908. Further notes on the birds of Gazaland. Ibis, ser. 9, vol. 2, pp. 391- 
443. 
1918. Rejection by birds of eggs unlike their own; with remarks on some 
of the cuckoo problems. Ibis, ser. 10, vol. 6, pp. 127-154. 
Symonps, EpMonp. 
1906. Notes on some migratory visitants to Kroonstad, Orange River 
Colony. Journ. S. Afr. Ornith. Union, vol. 2, pp. 24-26, 
Symons, GopDFREY. 
1962. Indian mynah host to cuckoo. African Wild Life, vol. 16, p. 343. 
SzreLasko, A. 
1913. Die Bedeutung der Eischalenstruktur der Vogel fiir die Systematik. 
Journ. f. Ornith., vol. 61, pp. 229-261. 
TAYLOR, J. SNEYD. 
1945. Notes on the birds of the Van Ryneveld’s Pass Irrigation Lake, 
Graaff-Reineet. Ostrich, vol. 16, pp. 44-54. 
Taytor, LIONEL E. 
1906. The birds of Irene, near Pretoria, Transvaal. Journ. S. Afr. Ornith, 
Union, vol. 2, pp. 55-83. 
THEMIDO, ANTONIO ARMANDO. 
1952. Aves de Portugal. Memorias e estudos de Museo Zoologico da Uni- 
versidade de Coimbra, No. 213, 241 pp. 
TuHopay, J. M. 
1953. Components of Fitness. Jn Symposia of the Soc. Exp. Biol. VII. 
Evolution, pp. 96-113. 


126 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


THorPE, WILLIAM Homan. 
1930. Biological races in insects and allied groups. Biol. Reviews and 
Biological Proc. Cambridge Philosophical Soc., vol. 5, pp. 177-212. 
TICEHURST, CLAUDE B. 
1922-1923. The birds of Sind., pt. 3. Ibis, ser. 11, vol. 4, pp. 526-572; 
vol. 5, pp. 1-43. 
ToMLInson, D. 
1962. Cuckoo eggs vary in colour. African Wild Life, vol. 16, p. 260. 
TownseEnpD, F. S. 
1921. Eggs of Coccystes cafer. S. Afr. Journ. Nat. Hist., vol. 3, pp. 220-222. 
TRISTRAM, H. B., in HEwitson, WILLIAM C. 
1859. Recent discoveries in European oology. Ibis, vol. 1, pp. 76-80. 
1866. On the ornithology of Palestine. II. Ibis, ser. 2, vol. 2, pp. 280-292. 
UNDERWOOD, GARTH. 
1954. Categories of adaptation. Evolution, vol. 8, pp. 365-377. 
VALVERDE, JOSE A. 
1953. Notes sur le Coucou-geai en Castille. L’Oiseau et Rev. Francaise 
d’Ornith., vol. 23, pp. 288-296. 
VAN SOMEREN, VicTtoR GURNET LOGAN. 
1916. A list of birds collected in Uganda and British East Africa with 
notes on their nesting and other habits. II, Ibis, ser. 10, vol. 4, 
pp. 373-472. 
1922. Notes on the birds of East Africa. Novitates Zoologicae, vol. 29, 
pp. 1-246, 7 pls. 
1929. Notes on the birds of Jubaland and the Northern Frontier, Kenya. 
Journ. E. Afr. and Uganda Nat. Hist. Soc., vol. 9, pp. 25-70. 
1931. Catalogue of the European and Asiatic Migrants to Kenya and 
Uganda. Journ. E. Afr. and Uganda Nat. Hist. Soc., spec. suppl. 
No. 4, 40 pp. 
1932. Birds of Kenya and Uganda, being addenda and corrigenda to my 
previous paper in “Novitates Zoologicae” XXIX, 1922. Novitates 
Zoologicae, vol. 37, pp. 252-380, 4 pls. 
1956. Days with birds, studies of the habits of some East African species. 
Fieldiana, Zool., vol. 38, 502 pp. 
VERHEYEN, RENE. 
1953. Oiseaux. In Exploration du Pare National de l’Upemba. Mission 
G. F. De Witte, fasc. 19. Brussels, Institut des Parcs Nationaux 
du Congo Belge, 687 pp., 5 pls., 45 figs. 
VINCENT, ALFRED WILLIAM. 
1946-47-49. On the breeding habits of some African birds (cont.). Ibis, 
vol. 88, pp. 48-67; vol. 89, pp. 163-204; vol. 91, pp. 111-139. 
VINCENT, JACK. 
1934. The birds of nothern Portuguese East Africa. Pts. 4-5. Ibis, ser. 13, 
vol. 4, pp. 495-527, 757-799. 
Voous, Kare. H. 
1959. The relationship of the European and Aethiopian avifaunas. Proc. 
First Pan-African Ornith. Congress, pp. 34-39. 
1960. Atlas of European birds. 284 pp. 
Wan ey, N. J. P. 
1951. Notes on the birds of central Anatolia. Ibis, vol. 93, pp. 63-89. 


NO. 4 AVIAN GENUS CLAMATOR—FRIEDMANN 127 


WEIGOLD, Huco. 
1913. Ein Monat Ornithologie in den Wtisten und Kulturoasen Nordwest- 
mesopotamiens und Innersyriens. Journ. f. Ornith., vol. 61, pp. 1-40. 
WHIsTLER, Hucu. 
1916. A note on some birds of the Gujranwala District, Punjab. Journ. 
Bombay Nat. Hist. Soc., vol. 24, pp. 689-710. 
1928. The migration of the pied crested cuckoo (Clamator jacobinus). 
Journ. Bombay Nat. Hist. Soc., vol. 33, pp. 136-145. 
1931. An open letter to the editors. Journ. Bombay Nat. Hist. Soc., vol. 35, 
pp. 189-195. 
1935. Popular handbook of Indian birds. 513 pp., 20 pls. 
WHIstTLER, HucH and Kinnear, NorMAn B. 
1934. The Vernay scientific survey of the eastern Ghats (ornith. sect.), 
pt. IX. Journ. Bombay Nat. Hist. Soc., vol. 37, pp. 515-528. 
Wuire, CHARLES MATTHEW NEwTON, and WINTERBOTTOM, JOHN MIAL. 
1949. A check list of the birds of Northern Rhodesia. Lusaka, 168 pp. 
WIckKHaM, P. F. 
1930. Notes on the birds of the Upper Burma hills. Journ. Bombay Nat. 
Hist. Soc., vol. 34, pp. 46-63. 
WINTERBOTTOM, JOHN MIAL. 
1942. A contribution to the ornithology of Barotseland. Ibis, ser. 14, vol. 6, 
pp. 337-389. 
1960. The zoo-geographical affinities of the avifauna of the western Cape 
Province. Ibis, vol. 102, pp. 383-393. 
1962. Some manuscript notes of S. F. Townsend for the period 1878-1925. 
Ostrich, vol. 33, pp. 66-71. 
WirtHeErBY, Harry Fores. 
1928. On the birds of central Spain with some notes on those of south-east 
Spain. Ibis, ser. 12, vol. 4, pp. 587-663. 
WITHERBY, Harry ForsEs; JOURDAIN, FRANCIS CHARLES ROBERT; TICEHURST, 
NorMAN F.; and Tucker, BERNARD W. 
1938-41. The handbook of British birds. London, 4 vols. 
Wotrson, ALBERT. 
1942. Regulation of spring migration in juncos. Condor, vol. 44, pp. 237- 
263. 
Woop, JoHN. 
1911. Notes from East London, Journ. S. Afr. Ornith. Union, vol. 7, p. 51. 
Woopwarp, R. B., anp Woopwarp, J. D. S. 
1899. Natal birds. Pietermaritzburg, 215 pp., 1 pl. 
Woottey, C. H. F. 
1946. Jacobin and black-crested cuckoo. Ostrich, vol. 17, p. 370. 
ZEDLITZ, OTTO GRAF VON. 
1910. Meine ornithologische Ausbeute in Nordost-Afrika. Journ. f. Ornith., 
vol. 58, pp. 731-807. 
1915-16. Das Siid-Somaliland als zoogeographisches Gebiet, III. Journ. 
f. Ornith., vol. 63, pp. 1-69; vol. 64, pp. 1-120. 


% ; Tee 
le $ , 
rr 
°. k . 
he 
; ; 5 
' 6 
~H 
4 ’ 
iA md i ; 
A 
f 
r 
; ‘ 
" i f 
amin 
| ots 7. 
Fiji ' ‘ayy OF, RS na 
" oh 
ah PT 
Sar ee 
> ‘ . 
' My x 


et oe es Cae 


ee ee 


eee ee 


Sts 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


~ VOLUME 146, NUMBER 5 


SOME BEHAVIOR PATTERNS OF 
PLATYRRHINE MONKEYS 
I. THE NIGHT MONKEY 
(AOTUS TRIVIRGATUS) 


By 
M. MOYNIHAN 


Director, Canal Zone Biological Area 
Smithsonian Institution 


(Pustrcation 4533) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
OCTOBER 23, 1964 


Sky 
Vif} 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NUMBER 5 


SOME BEHAVIOR PATTERNS OF 
PLATYRRHINE MONKEYS 
hE NIGHT MONKEY 
(AOTUS TRIVIRGATUS) 


By 
M. MOYNIHAN 


Director, Canal Zone Biological Area 
Smithsonian Institution 


ME INCRE SS 


(Pustication 4533) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
OCTOBER 23, 1964 


PORT CITY PRESS, INC. 
BALTIMORE, MD., U. S. A. 


CONTENTS 


Page 
PHEFOGHCHONM | io.0 wars op ARG Balsls sisye ¥'s)o jn eb 4s aceidaleeralle Sagat s brobderaa eit 1 
General Habitus, Locomotion, and’ Feeding. 0... .)5 502. a0. bead wees aeloace « 3 
Ordinary, Grooming and\Gleaning Activities: .: 4...0.5..% odicncies Sactene - 6 
ostiles Behaviat chest gtiods ob. lvias vee eilaes bs ste eleibed scenes tat 8 
OVEEE AttACle GRAMION. 5.5 pins 5s. oooee ae «pon #6 dwisie te SPE tae sete 9 
OvertuMlarmtor escapes Behavior oo. sa. docc das deel taioen. pe cesee 12 
DNS lip PE Chet IGOR sig eciai dcx dae alae ee wie casts oh Sea OE £5 
SAVER cuneeteht 6 sec hicink: wales dale kiebiamieyh ataeiaic oad wSeiee sire « Sehate 16 
SUGHE PGC POSURCS aie hen ase e bce x dows misls wee cee 6 Pea 18 
Other. Visiel, PatterhS.g icc ie p< aaleems sie seid Lawes om exes ane 20 
ME SoS). Gaia aks <a aba Sins csicahce oo bala Sede» Beene 22 
ResonAanp Ghats se co ioue oss da secs nlokesidtas Pocaweae dente 25 
PEARSE ere Saint. SOS eae os nine pe Ra etia S's Le dk eMee seer ma 31 
goaNee SALI eta ce eng oh ase ean Via oh aaah a ca os we Suelae ral mace onloimna STAG 32 
(Ceara eer eb hey mile Sa ha Al, as eee le a Ne ee 36 
Locomotory Notes—Gulps and Sneeze-grunts...............+-- 38 
MIR EMSSEEI tk. 80>, one Menu ee Ns, 2k SE are Is als Ti Ae Oe 2 ES 43 
SAO Ue araHAISILE ocr -taele. so Gk aI ee cc tie Moke eee oes ae wae Woe cba ee 47 
Sexual Behavior and Associated or Related Patterns................---. 48 
RR NUE Bt toy Share o Eetatubs w cralng 2g Chat apnay date ara apelin e Sra te Se 49 
LECT: SE eR ae Aap ee ae Ce OT MIROMPSSE Roh (ELSES 18 om Nnet See. Wy an SEIT 52 
Copulations, Allogrooming, and Associated Patterns............... 54 
BSA Se Seine ce arco oie San’ aa 5) Soe sual eer oudeoy tera Se En aE 59 
enauiar ore wou “Aniials. fh. oe cl Ses ccs Rodee Re ak eavontea tes 63 
Previous Descriptions of Night Monkey Behavior Patterns.............. 78 
PRRs She), aad Sisdi2 tap kv s 6 a's ale bere SOC MEES ae 6 BR Re ARE a ee ee 80 
Pea UA OUIALSAES NS, Gites) 13024". sich yas Sivas in Waahucat's at Gey teamahe oo Sais eee Hema 82 
CE Ce PAA Seiko 68 Min aus "sd valns wat GRA aT IS ea A EI 82 
ILLUSTRATIONS 
TEXT FIGURES 
1. Two typical semierect preleaping postures.................-s0000- 5 
ee Ott sHeAG-dOwin OOSEITES..« sicdoa < in. win os vinee nial daa oteciem aaladae 15 
EVRR ANT Chat OSETIEES apoyo ts shahp: i oiet ein chavo, Sin 16, ach eycheym ataatalet leLeys cegena (ats Danese iat atare 19 
mr. typical Grin Grunt, uttered by an adult... 3... 5.0522 onan ews 23 
5. A relatively short Scream, uttered by an adult.................006. 31 
Ge Two lone: Sereams; uttered by an adult... 0.50208. 6 clei ctemecies 32 
fn Wow htrll. attered: by art adult... 0 s-..</.\6 4s «ts aiala ge lamielals s aiaisibyais: 33 
8. One Moan, followed immediately by one Low Trill, uttered by an 
ALUN Earache eM NTR ARE ia ble diac rcye sro; a es eraleie Sc ates aroleretar at atepereiarekenmetenele 33 
9. One Sneeze-grunt, followed immediately by one Low Trill, uttered by 
INS AMER aa eia evra oi ere ie ohm 3,0: cd! wi grssave v1 a'Wys) are) atalarave ya. ciaramautarein cid 34 


iv 


SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 


One Moan utteredibysanradtltyytrctspverctor oferoforsreleve'=\olatole)sietaiste ol tetersterans 


+ wo) Gulps; uttered by ‘an adult's siasb. Gickis wise sss ds s\ce es werner panes 
. A single “twanging” noise heard while recording adults............ 


Three: Hoots, uttered by aniadg@lt females. ..... 5 ees Me aioe ccc me 
Positions and postures sometimes assumed by mated individuals during 
ATO SROOMIME Hass aisle oietarerctatetatetets tote ereveraieoiove) e. = cic ei aieoreloneebeter Torateloreye 
Four typical “pure” Squeaks, uttered by an immature............... 
A single very loud Squeak, uttered by an immature................. 
A “pure” Squeak, followed by a longer note more or less perfectly 
intermediate between a Squeak and a short Scream. Uttered by 
AM! AMMALUTEy 445 gs ibd ey alee low ae eae WR Ree Dalerae Thee ets tobe ake Seta ate teva reaee 
A series of four notes uttered by an immature.............0+-.000- 


. Two loud trilling patterns, intermediate between “pure” High Trills 


and Screams, but most similar to the former. Uttered by an im- 
ATMAEUET Se oie ea are eieyaieve clea alesis hist eve) oelose/eo) oie) ch eve etetttoloiiioehel cl o1 ele! ctereyeneteiens 
wo) loots, uttered vby, ani immature anale vinci ci.s eicetereteis Sotelo terete 


> Av single: “Brrrrp sound, uttered by an immature... /022. 5226-0. 


Anniant) uttering: ‘Squeaks si:'s)s <x scss s's.n ac cyale a dicts cee ele iors 


146 


SOME BEHAVIOR PATTERNS OF 
PLATYRRHINE MONKEYS 
tte NIGHT MONKEY: 
(Aotus trivirgatus) 


By M. Moyninan 
Director, Canal Zone Biological Area 
Smithsonian Institution 


Tuis is the first in a series of papers on some behavior patterns of 
New World monkeys. The main emphasis of these papers will be 
comparative. Special attention will be paid to social signal patterns 
(including hostile and sexual signals) and other patterns that differ 
significantly in the various species. It is hoped that analyses of these 
patterns may throw some light on the evolution of the group as a 
whole. 


INTRODUCTION 


The Night Monkey (also called “Owl Monkey” or “Douroucoult’”’) 
is a rather small monkey of distinctive appearance. 

Although individuals vary considerably in proportions, all adults 
have relatively small heads and more or less elongated bodies. Their 
limbs are of moderate length, the hind limbs being somewhat longer 
than the front. Their tails are long and heavily furred, and not at all 
prehensile. The back, most of the head, and the outer surfaces of the 
limbs may be gray, brown, or even reddish brown, while most of 
the underside of the body and the inner surfaces of the limbs are 
usually clear buff or caramel colored. The upper part of the tail gen- 
erally is the same color as the back, and the terminal half or two- 
thirds is biackish. The face and forehead are marked with conspicuous 
black and white stripes and patches (see the accompanying figures for 
an indication of the color pattern of Panamanian Night Monkeys). 
The distinctive appearance of the face is enhanced by the very large 
eyes. In bright light, the pupils are greatly contracted and the orange 
or hazel irides are conspicuous. In very dim light, the pupils expand 
enormously and the eyes appear to be completely black. 

Young Night Monkeys look more like marmosets in some ways. 
They have relatively larger heads and shorter bodies and tails than 


SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 146, NO. 5 


2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


adults ; and their eyes are relatively smaller, in proportion to the rest 
of the head (see figure 22). 

This account is based primarily upon studies of captive animals, 
checked (whenever possible) by observations of wild animals living 
under natural conditions. 

Fifteen Night Monkeys were kept in cages at the Smithsonian In- 
stitution laboratory on Barro Colorado Island between August 1958 
and July 1963, for periods ranging from several months to more than 
3 years. Some individuals were kept in small cages, 8’ x 4’ x 4’, inside 
a house; but the majority were in larger cages outdoors in the forest. 
The outside cages varied in size from 6’ x 6’ x 8’ to 30’ x 12’ x 8’. Gen- 
erally each cage contained a single individual, a mated pair, or a 
family group; sometimes, however, larger groups (up to five in- 
dividuals) were kept together in a large cage for a few days. All 
the Night Monkeys kept in captivity on Barro Colorado Island were 
wild-caught animals from mainland areas in Panama (probably, in 
most cases, from regions east of the Canal Zone) or were the 
offspring of such animals. 

Some other captive Night Monkeys were observed: briefly in the 
National Zoological Park, in Washington, D.C., in October 1959 
and July 1960, and in Iquitos, Peru, in December 1958. 

All observations of wild Night Monkeys were made on Barro 
Colorado Island, in both apparently mature and old second-growth 
forest (see Bennett, 1963). 

The captive animals on Barro Colorado Island were sometimes 
observed from blinds; but such concealment usually was not neces- 
sary. Blinds were not used for observations of other captive in- 
dividuals or the wild animals in the forest. Some of the captive in- 
dividuals on Barro Colorado Island were accustomed to continuous 
but dim artificial light at night. This did not appear to affect their 
behavior after the first few nights. All other individuals were observed 
without using artificial light whenever possible. Sometimes moonlight 
was sufficient to reveal at least the most conspicuous and vigorous 
movements. Even in nearly complete darkness, it was sometimes 
possible to follow the behavior of the animals by the sounds of their 
movements and vocalizations. When artificial light was necessary, 
flashlights were used (either intermittently or continuously). This 
certainly affected the behavior of the animals, but did not always 
alarm them greatly or cause them to escape immediately. 

According to Hershkovitz (1949), there is only one species of 
Night Monkey: Aotus trivirgatus; and the Panamanian population 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 3 


belongs to the subspecies griseimembra. The exact provenance of the 
animals in the National Zoological Park in Washington, D.C., was 
unknown. The animals observed in Iquitos supposedly were caught in 
nearby areas and, therefore, were presumably representatives of the 
nominate subspecies (Hershkovitz, op. cit.) or the possibly distinct 
subspecies vociferans (Cabrera, 1957). 

All references to the behavior of other platyrrhine monkeys in the 
following pages are based upon personal observation, unless specifi- 
cally stated otherwise. 


GENERAL HABITUS, LOCOMOTION, AND FEEDING 


Some characteristic activities of Night Monkeys in the wild have 
been described in a number of earlier publications. The best general 
survey is by Cabrera and Yepes (1940), who summarize the reports 
of many early naturalists and travelers in South America; and the 
best account of Night Monkeys in Panama is by Enders (1935). It 
may be useful to recapitulate some of this information, and add a 
few details about locomotory and feeding habits, before discussing 
other aspects of behavior. 

Night Monkeys differ from all other “true’’ monkeys in being 
almost completely nocturnal. Both captive and wild individuals are 
most active immediately after nightfall and just before dawn. 

Wild Night Monkeys sometimes start to become active in the 
evening before it is completely dark, and may continue to move about 
for some minutes after it has started to become light in the morning. 
Captive individuals easily learn to move about in the daytime, and 
their adjustment to continuous artificial light at night is usually 
rapid. They appear to see quite well in all but the brightest light. This 
is not surprising, as there is some evidence (see, for instance, Hill, 
1957, and Rohen, 1962) that the nocturnal habits of Night Monkeys 
are specialized rather than primitive among Platyrrhini. (The terms 
“primitive” and “specialized” are used, throughout this paper, as 
defined by Simpson, 1961.) 

Wild Night Monkeys apparently are purely arboreal. On Barro 
Colorado Island, they usually move and feed in trees at heights 
ranging from 20 to at least 100 feet above the ground, and apparently 
never come lower than 10 feet above the ground. During the daytime, 
they sleep in holes in trees. 

The fur of Night Monkeys is very thick and soft. This softness 
may be an adaptation to nocturnal and arboreal habits, possibly 
helping to reduce or muffle the sound of movements. It is probably 


4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


not coincidental that such unrelated animals as Kinkajous (Potos), 
Olingos (Bassaricyon), and Woolly Opossums (Caluromys), which 
are also nocturnal and arboreal, have fur of similar texture. 


Fic. 1—Two typical semierect preleaping postures. 


Night Monkeys are extremely active and agile. (Hill, 1956, says 
that they are fairly slow moving; but this is certainly not true of 
individuals in the wild or of healthy captive individuals in adequate 
cages.) Like monkeys of the genera Callicebus, Pithecia, and Cacajao, 
they are expert and powerful leapers. Rapid movements through the 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 5 


trees are usually accomplished by great bounds. When leaping, most 
of the propulsive power appears to come from the hind limbs, the 
forelimbs being used chiefly to absorb the shock of landing. The 
long and rather heavy tail seemingly is used as a counterweight, never 
as a prop or support. (This is another resemblance to Callicebus 
and Pithecia.) Night Monkeys run along branches less frequently 
than monkeys of the genera Cebus and Saimiri or tamarins such as 
Saguinus geoffroyi. (The generic names used in this paper follow 
Hershkovitz, 1958.) 

As noted by Hill (1957), Night Monkeys are among the most 
nearly completely quadrupedal monkeys. Not only do they usually 
use both fore and hind limbs in locomotion, but they generally grasp 
a supporting branch with one or both hands even when sitting quietly. 
Just before leaping, however, they may take their hands off the 
supporting branch and sit or stand in a semierect posture, balancing 
on the hind limbs alone (see figure 1). The hind limbs generally 
are flexed in this posture, the back may be nearly straight or curved 
to an appreciable extent, and the front limbs may hang downward or 
the hands may be slightly raised (presumably in preparation for 
grasping a branch after the leap). An individual may remain poised 
in this way for several seconds or more. 

Night Monkeys seem to eat a great variety of fruits and insects 
under natural conditions, and learn to eat artificial foods without 
difficulty in captivity. Their method of approaching and “handling” 
nonmoving food in captivity is interesting. They generally begin by 
sniffing at the food. (For other indications that they may have a good 
sense of smell, see below.) After sniffing for several seconds, an 
animal usually will seize the food with its teeth, and then sit up. Only 
then, in most cases, will it raise one or both hands to hold and possibly 
manipulate the food as it begins to chew. 

The various ways in which different species of platyrrhine monkeys 
use their hands and fingers have not been well described. Some in- 
correct statements and misinterpretations are current in some of the 
published literature. Sanderson (1957), for instance, states that 
Night Monkeys have opposable thumbs and use their hands in much 
the same way as human beings. In a later passage, he also implies 
that they usually oppose the first two fingers of the hand against the 
last three (in much the same way as monkeys of the genera Pithecia 
and Alouatta). Both statements are somewhat misleading. 

Night Monkeys often spread their fingers more or less radially 
when grasping an object. At such times, all the fingers tend to diverge 


6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


from one another to approximately the same extent. At other times, 
all five fingers of the hand may be kept pressed closely together. 
When some appreciable opposition does occur, it is perhaps most 
common for the thumb to be opposed to the other four fingers as a 
group; but this is by no means always the case. I have, in fact, seen 
Night Monkeys arrange their fingers in every possible way when 
grasping objects of different sizes and shapes. The first two fingers 
may be partly opposed to the last three ; or the first three fingers may 
be partly opposed to the last two; or the first four fingers may be 
opposed to the fifth. 

It is quite probable that this type of hand and use of fingers is 
primitive among Platyrrhini. (The actual structure of the Night 
Monkey hand is described in detail by Biegert, 1961.) 

The most important general feature of the social behavior of Night 
Monkeys is their slight degree of gregariousness. On Barro Colorado 
they are seldom found in groups larger than a single pair or a single 
family of two parents with one young. 


ORDINARY GROOMING AND CLEANING ACTIVITIES 


New World monkeys perform two types of grooming : Self-groom- 
ing and mutual or social grooming, 1.e., the grooming of one animal 
by another. The former type may be called “autogrooming” and the 
latter “Allogrooming.” (These terms have been suggested by the terms 
“autopreening” and “allopreening” applied to birds by Cullen, 1963.) 

Most monkeys devote considerable time and effort to autogroom- 
ing ; but different species tend to do this in slightly different ways. 

The autogrooming of Night Monkeys consists almost entirely of 
scratching with the hands and/or the feet. Only very rarely does a 
Night Monkey use its teeth during autogrooming to “nibble” at its own 
fur or skin. The relative frequency of different types of autogrooming 
movements performed by some adult and subadult individuals in 
captivity on Barro Colorado Island is shown in the accompanying 
tabulation. 

The actual frequency of autogrooming movements may vary widely, 
depending upon the situation. Most of this variation seems to be a 
function of the amount of dirt or ectoparasites in the fur, or of its 
disarrangement. In captivity, however, it is evident that most adult 
individuals tend to perform fewer autogrooming movements, on the 
average, than young individuals. This may be because adults tend to 
be less tame than young individuals in captivity. Night Monkeys 
seldom, or perhaps never, perform autogrooming movements when 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 7 


they are either aggressive or alarmed. In this respect, they are quite 
different from many other New World monkeys, e.g., species of the 
genera Alouatta, Ateles, and Cebus, which tend to perform particularly 
frequent and/or vigorous scratching in such circumstances. 

It is my impression that young Night Monkeys also may tend to 
perform relatively more scratching movements with their feet, and 


TABULATION OF ORDINARY GROOMING AND CLEANING 


REACTIONS 
Number1 
Autogrooming performance of bouts 
Secarchine tue face withOne Hand ..0,¢oéu.e.c «es oa aie vs ov mieiia simiera stare ore 23 
—the side of the head with the hand of the same side........... 1 
=== ENEICHESE Wit One wanderer ices tomate clercores cine slaiiniclacetelete 1 
—the side of the body with the hand of the same side........... 24 
—one hind leg with the hand of the same side.................. 19 
—one foot with the hand of the same side..................000- 5 
==fhe-senital remion: with one: hatid ./.2.)¢3. welts Sika . vases seni 4 
== tHentatlL with ONE wade meena « cas-tetcihacs wpsie, ch dfors.cispsisistuaih ih ays afetons 4 
—onelarim of hand with the other hand so... rac. ..5 20 sie orn oc 16 
=——themtatleWithe DOtin hands jnseces . calsdioc ce cine e cece t ue certs 3 
==therChinmewathronetoote ee ser os cie cs ceric dee eae eto fe 
=f NE pACE AW El OLE LOOEL ssciiss siciele nie Sido Saath Salaam ote 11 
—the side of the head with the foot of the same side............. 1 
= THER CHEStU Wit lan ONE sO OLer ta sisi oiaporsibie: teas, oteue ere Sisioud asin Silas eie Gels 1 
—under one arm with the foot of the same side................ 1 
——One atrinnwith Loot Of the same Sid@o.. 2+... cece ccucccescue 7 
—the side of the body with the foot of the same side............ 7 
Nrbblinessthestarkotethe tare. seemenesttiocclackion gic line Ooo ee 1 
Cleaning performance 
MMR cll Su ARAL arpa lop 0 29S orale tel ics ska va areata Hd CUTS Ae Peters 3 
upping the nose aiong the substrate... . o.s..0006 ee denaseedae.ctnion 2 


1 This count is based upon observations of six individuals; two immatures and four adults. 
The longest period of observation of a single individual was 1 hour. The shortest period was 
12 minutes. 


relatively fewer scratching movements with their hands, than adults 
in similar situations. There may be two reasons for this: (1) Young 
individuals find it more difficult to balance with one hand off the 
ground and (2) they can reach most parts of their (relatively shorter) 
bodies with their feet more easily than can adults. 

Unlike most other platyrrhines, Night Monkeys seldom perform 
allogrooming except in obviously sexual or partly sexual circum- 
stances (see below). This presumably is correlated with their slight 
degree of gregariousness, but it is not due simply to lack of opportu- 
nity. Mated Night Monkeys rarely groom one another even when 


8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


they are close together if they are not also performing other 
sexual patterns. 

A Night Monkey with dirt on its face sometimes removes the dirt 
by rubbing the face along a branch. These reactions are actually rare, 
but seem to be performed more frequently by Night Monkeys than 
by any other platyrrhines with which I am familiar except the tamarins 
of the genus Saguinus. 

Night Monkeys clean their hands and fingers by licking. 

The tabulation of autogrooming movements on page 7 also includes 
the cleaning movements performed by the same individuals during 
the same periods of observation. 


HOSTILE BEHAVIOR 


The term “hostile” may be applied to all behavior patterns resulting 
from a tendency to attack and/or a tendency to escape. In this sense, 
it is synonymous with the term “agonistic” as used in many other 
discussions of behavior (e.g. Scott, 1958). 

The term “tendency” will be used in a broad sense throughout 
this paper, to designate any “readiness to show a particular type of 
behavior” (Marler, 1956). 

In preliminary and essentially descriptive accounts of behavior 
based on observation rather than experiment, such as the present 
account of the Night Monkey, it is usually not necessary to distin- 
guish between a great number of qualitatively different tendencies. 
Most social reactions can be characterized as the products of such 
broad tendencies as attack, escape, gregariousness, copulation, pairing, 
etc? 


1 The tendencies involved in the production of an otherwise ambiguous be- 
havior pattern may be revealed by analysis of the other activities usually asso- 
ciated with the pattern. Conversely, the term “tendency” may provide a con- 
venient and concise way of summarizing the circumstances in which a behavior 
pattern occurs. Thus, for instance, when a signal pattern such as a ritualized 
posture or a vocalization (see below) is said to be produced by the attack 
tendency, this means that an animal performing the pattern is likely to perform 
overt attack movements during or immediately after the pattern and/or give some 
outward indication that it “wants” to attack. When one signal pattern is said 
to be produced by a stronger attack tendency than another signal pattern, this 
means that the former is more likely to be accompanied or followed by overt 
indications of attack than the latter and/or is likely to be accompanied or followed 
by stronger indications of attack than the latter. (It should be stressed, in this 
connection, that when a particular behavior pattern is said to be produced by 
one or more particular tendencies it does not mean that other tendencies may not 
be activated in an animal performing the pattern at any particular time.) 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 9 


This type of terminology has been criticized on various grounds, 
Among the most recent critics is Andrew (1963). The term “tendency” 
as used here is certainly a great oversimplification, but its convenience 
in actual practice may outweigh its disadvantages in theory. Most 
ethologists have been driven to use the same or a similar term, or the 
same concept in more or less disguised form, when describing a variety 
of different types of social behavior. 

Andrew (op. cit.) suggests that all or most signal patterns of pri- 
mates may be caused by “stimulus contrast.” This is undoubtedly true 
—if “stimulus contrast” is defined broadly enough. But platyrrhine 
signal patterns are certainly not produced by a single range of qualita- 
tively similar stimulus contrasts of differing strengths. Certain stimuli 
usually (“normally”) provoke only sexual patterns, others generally 
provoke only hostile patterns, still others usually provoke only parental 
responses, etc. Thus the causation of platyrrhine signal patterns can 
be described in terms of stimulus contrasts only if distinctions are 
made between qualitatively different types of stimulus contrasts, e.g., 
between hostile stimulus contrasts and sexual stimulus contrasts. It 
may be doubted if such a system is any more convenient or useful 
than the employment of terms such as tendency. It is certainly more 
difficult to use in descriptive passages. (And, in fact, Andrew himself 
does not use it consistently throughout his own description of primate 
signal patterns. ) 

All the vocal patterns cited in the following pages will be given 
names, such as “Moan,” “Grunt,” “Gulp,” etc. These names are used 
purely for convenience. They are not meant to describe the acoustical 
properties of the sounds in detail ; this will be done in the accompany- 
ing drawings of sound spectrograms. They are meant only to suggest 
that the sounds are somewhat similar (to human ears) to the common 
human and other everyday sounds called by the same names. 

The term “sound” itself will be used in its ordinary, everyday sense 
throughout the paper. 

All the hostile behavior patterns mentioned are those of adults, both 
males and females, unless specifically stated otherwise. 


OVERT ATTACK BEHAVIOR 


Night Monkeys are among the most aggressive New World pri- 
mates. Adult individuals of the same sex usually fight savagely when 
put together in a cage. Adults of opposite sex, mates or potential 
mates, may not engage in violent fights, but they do direct a consider- 


Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


able variety of hostile patterns toward one another in at least some 
circumstances (see below). 

This aggressiveness must be one of the factors keeping pairs and 
family groups apart in the wild. Captive adults of more gregarious 
species of the genera Saimiri, Cebus, Ateles, and Alouaita usually do 
not fight with one another as frequently or as vigorously as do Night 
Monkeys kept in similar conditions. 

It is perhaps surprising, therefore, that overt fights between pairs 
and family groups of Night Monkeys seem to be relatively rare in 
the wild, at least on Barro Colorado Island. This seems to be due to 
the fact that different pairs or family groups simply do not come into 
contact with one another often. Each pair or family group seems to 
have its own territory or home range, in which it remains all or most 
of the time. (These territories or home ranges may be largely “tradi- 
tional.” Their boundaries may be established, originally, by means of 
disputes. Once established, the boundaries may be respected simply 
because the animals remember the results of the original encounters. 
If so, the territories of Night Monkeys on Barro Colorado Island are 
maintained in the same way as those of many birds occurring in the 
same area—see Moynihan, 1962a.) 

(I have seen relatively large groups of wild Night Monkeys [includ- 
ing four or five individuals] only in the immediate vicinity of major 
food sources, i.e., trees in fruit. In such cases, one pair or family 
group may have been so strongly attracted to the food that it crossed 
over the boundary of its territory or home range. Some of the indi- 
viduals in these groups showed obvious hostility toward one another ; 
and the groups always broke up within a few minutes.) 

An aggressive Night Monkey may perform a variety of displays 
(see below) immediately before and/or after attacking ; but the overt 
attack behavior itself is usually relatively simple. 

An attacking individual usually begins by advancing, walking, slowly 
and cautiously toward its opponent. Then, when it has come close 
enough, it suddenly leaps forward and tries to bite the opponent and/ 
or hit the opponent with its hands. This hitting is reminiscent of the 
jabbing of a human boxer. Usually two or three jabs are delivered 
with great rapidity one right after the other. 

It is my impression that Night Monkeys strike with their hands 
during fights relatively more frequently than all or most other platyr- 
rhines. This may be correlated with the fact that they do not have 
very long canine teeth. 

A Night Monkey that has been attacked usually retreats as soon 
as possible; but its attacker seldom follows immediately. Even when 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN eit 


very aggressive, the attacker generally pauses a few seconds, before 
starting another slow and cautious advance, preparatory to another 
attack leap. 

The Night Monkeys kept in captivity on Barro Colorado Island 
generally were quite silent while attacking ; but once I heard an indi- 
vidual “spit,” like a cat, at the very instant of jabbing an opponent. 
Hill (1960) mentions similar sounds in similar circumstances. So 
perhaps spitting sounds are a normal part of the attack behavior of 
Night Monkeys, at least in some populations of the species. 

Sometimes two individuals will try to attack one another simultane- 
ously. This may lead to a brief but extremely violent wrestling match, 
while each animal tries to bite and/or strike the other. 

All these patterns, with the possible exception of “spitting,” would 
appear to be expressions of the attack tendency alone. At least, they 
do not include any components that are unmistakable indications of 
any other tendency. 

Other attack reactions may be less direct. It is not uncommon to 
see an aggressive animal leap over, or just to one side, of its oppo- 
nent, instead of directly toward it. An aggressive animal leaping in 
this way is less likely to bite or strike its opponent than is an animal 
that has leapt more directly. In such cases, it would appear that the 
attack has been “deflected,” probably by some counteracting tendency. 

The tendency most likely to counteract attack during most hostile 
encounters is undoubtedly escape. There is considerable evidence 
than an escape tendency generally is activated to some extent, how- 
ever slight, in all or most situations in which attack is activated. Even 
the most aggressive animals sometimes interrupt their attack behavior, 
by retreating briefly and/or performing displays which probably in- 
clude an escape component. 

Mated pairs of captive animals kept in adjacent cages may become 
engaged in vigorous disputes with one another. In such cases, it is 
obvious that usually males try to attack males while females try to 
attack females. Males also tend to attack more frequently, on the 
average, than do females. 

Captive Night Monkeys seldom “redirect” (see Bastock, Morris, 
and Moynihan, 1953) attack upon individuals other than the ones 
provoking the attack. They certainly perform redirection attacks 
much less frequently than do individuals of many other species of 
Platyrrhini in similar circumstances. This seems to be due partly to 
the fact that direct expression of their aggressiveness is seldom or 
never impeded by any positively “friendly” gregarious tendency. Even 


12 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


more important, mated Night Monkeys apparently never redirect 
attacks upon their mates or young. 

This partial inhibition seems to be one of the more distinctive fea- 
tures of the behavior of Night Monkeys. It must be advantageous, 
and it may also be directly correlated with the social structure of the 
species. Although there is no proof, it seems likely that the pair-bonds 
of Night Monkeys are comparatively strong and stable. (At least, it 
is not uncommon to find two Night Monkeys in the same area of the 
forest, behaving in more or less the same way, for months at a time. 
It is difficult to believe that the individuals involved are not the same 
throughout the whole period.) Pair-bonds seem to be weaker and/or 
less continuous in some other New World primates, such as Alouatia 
palliata and Cebus capucinus, and the males of such species do per- 
form redirection attacks upon females and young with considerable 
frequency (at least in captivity). There may have been strong selec- 
tion pressure against such behavior in Night Monkeys simply because 
it is particularly important for them to avoid doing anything that 
might interfere with the maintenance of their pair-bonds. 

The only redirection attacks observed during the present study 
were performed by one captive individual on Barro Colorado Island, 
when it was placed in a cage between two other cages that also con- 
tain Night Monkeys. When this individual became engaged in a dis- 
pute with one of its neighbors (fighting through the intervening wire 
mesh), it would occasionally interrupt the fight to make a rapid and 
absolutely unprovoked attack upon its neighbor on the opposite side. 

I have never seen Night Monkeys jump up and down in rage, or 
shake branches of trees, or break off and drop branches, like so many 
other species of both New World and Old World monkeys (see, for 
instance, Carpenter, 1934 and 1935, Ullrich, 1961, and Hinde and 
Rowell, 1962). They may lack such patterns because they are rather 
small and light in weight. 

Many other species of monkeys perform “play wrestling,” all or 
most of which seems to be a type of partly inhibited attack. This also 
seems to be almost or completely lacking in Night Monkeys, presum- 
ably as another consequence of their slight degree of gregariousness. 
Even when several young Night Monkeys are kept together in the 
same cage, they do not perform any wrestling which appears to be 
anything but ordinary, uninhibited attack. 


’ 


OVERT ALARM OR ESCAPE BEHAVIOR 


The simplest alarm reaction of Night Monkeys is a brief “freeze.” 
In the most common form of freeze, an animal remains motionless 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 13 


in whatever posture it was in when it first perceived a disturbing 
stimulus. Such freezes tend to be performed as reactions to distant 
stimuli or familiar near stimuli. They are not usually associated very 
closely with vigorous movements or displays in such circumstances. 

Sometimes an animal will close its eyes, briefly, immediately before, 
during, or immediately after pausing in a simple freeze. 

The most conspicuous escape behavior of Night Monkeys in the 
wild or in large cages is simply rapid, running or leaping, retreat. 
This tends to be performed as a reaction to some strong, near and/or 
unfamiliar stimulus, such as actual attack or other overt aggressive 
behavior by another individual of the same species or the sudden 
approach of a “potential predator” such as a strange human being. 
Active retreats frequently are preceded and/or followed by elaborate 
displays. 

In similar social circumstances, a Night Monkey in a small cage 
may either run or leap around its cage very rapidly or, alternatively, 
perform a pattern which is reminiscent of the simple freeze described 
above but somewhat more complex. In this pattern, the animal sits 
motionless with its head lowered (such lowering is not characteristic 
of ordinary freezes). The head is sometimes lowered almost to the 
level of the hands or the branch on which the animal is sitting. The 
animal may look straight down, or keep its head turned sideways in 
order to fixate the disturbing stimulus. At the same time, the limbs 
may be drawn in under the body and strongly flexed. This is obvi- 
ously a preparation for (or an “intention movement” of) leaping. As 
a result of the lowering of the head, the back is more or less curved 
(but not raised). Some typical variations of this ‘“head-down posture” 
are shown in figure 2. 

It is possible that Night Monkeys also assume head-down postures 
in the wild under completely natural conditions. I never actually saw 
such reactions in the forest on Barro Colorado Island (they would be 
extremely difficult to distinguish in trees at night) ; but some of the 
animals kept in large cages, which approximated natural conditions, 
did assume head-down postures from time to time. 

The captive individuals in large cages did not, however, assume 
head-down postures as frequently as individuals of many other spe- 
cies of platyrrhine monkeys in similar situations. And their head- 
down patterns were not as exaggerated in form as the homologous pat- 
terns of some related species. 

All these overt escape or alarm patterns are usually silent. The 
fact that Night Monkeys customarily are silent during rapid retreats 
from potential predators is another distinctive feature of the species, 


14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


as all or most other platyrrhines utter special “Warning Notes” in 
such circumstances. It is probable, however, that the Night Monkey 
does have one pattern in its vocal repertory which is at least partly 
homologous with the Warning Notes of some other species, although 
uttered in a slightly different range of circumstances and subserving 
a different function (see discussion on page 43). 

All the alarm and escape reactions of Night Monkeys would appear 
to be expressions of the escape tendency alone, in much the same way 
that the aggressive reactions described above seem to be expressions 
of the attack tendency alone. The circumstances in which the various 
patterns occur would suggest that simple freezes are produced by 
weaker motivation than either active retreat or head-down postures. 
The strength of the escape tendency may be approximately the same 
in the latter two patterns, the head-down posture being assumed only 
by individuals unable or unwilling to retreat, possibly for any one of 
several different reasons. 

The head-down postures are the most interesting of these patterns 
from a functional point of view. Both head-downs and ordinary 
freezes seem to be primarily attempts to hide, and to hide the whole 
animal, i.e., to prevent a potential opponent or predator from noticing 
any or all parts of the motionless animal. Some of the head-down 
patterns may also be advantageous in a slightly different way. When 
an animal in a head-down posture looks straight downward, its black 
and white facial pattern and its eyes usually are partly or wholly 
concealed. These facial features seem to be the characters that release 
attack most frequently, or toward which attacks are usually directed. 
Thus an individual in a head-down posture of this type may not be 
attacked by an opponent or (possibly) a predator, even though the 
rest of its head and body are perfectly visible, and have in fact been 
noticed. In such cases, the head-down may function as “appeasement,” 
reducing the attack tendency of an opponent without increasing its 
escape tendency (see Moynihan, 1955). The closing of the eyes dur- 
ing some ordinary freezes may subserve a similar function, although 
presumably less effectively. 


DISPLAY BEHAVIOR 


The most common hostile behavior patterns of Night Monkeys are 
“displays.” . 

The term “display” may be used to include all “ritualized’’ patterns, 
1.e., all vocal patterns and all movements and postures that seem to 
have become specialized in form and/or frequency to serve signal 


NO. 5 BEHAVIOR OF THE NIGHT MON KEY—MOYNIHAN I5 


Fic. 2—Four head-down postures. 


16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


functions. (This does not mean that all patterns that may function as 
signals are necessarily displays. The head-down and eye-closing pat- 
terns, for instance, may be functioning as signals when and if they 
appease, but there is no really convincing evidence that they have be- 
come specialized to serve such a function. They are not more fre- 
quent or exaggerated in form than might be expected of pure evasive 
or avoidance patterns. ) 

Following a common convention, the initial letters of apparently or 
certainly ritualized patterns will be capitalized throughout the subse- 
quent account. 


SWAYING 


The captive Night Monkeys on Barro Colorado Island sometimes 
began to sway from side to side, instead of retreating or freezing, 
when they saw a predator or potential predator such as an ocelot 
or a human being some distance away. The Swaying movements 
were usually smooth, regular, moderately rapid, and extended over 
a fairly wide arc (1.e., a Swaying animal would first lean several 
inches to one side of the vertical and then several inches to the other 
side). Sometimes an individual would sit on its haunches, grasping 
its perch with its hands, while it swayed. At other times it might 
stand up in a more or less extreme preleaping posture. In either case, 
the head might be kept facing straight forward or turned from side 
to side, the animal looking right as it swayed to the right and looking 
left as it swayed to the left. 

Most Swaying was silent; but sometimes “Gruff Grunts” and/or 
“Gulps” (see below) were uttered at the beginning and/or toward 
the end of a bout of Swaying, especially when similar or identical 
sounds were uttered before and/or after the bout itself. 

An animal performing very brief Swaying sometimes appeared to be 
doing nothing more than “‘peering”’ from side to side to get a better 
view of the disturbing stimulus; but most Swaying was much too 
prolonged and exaggerated and stereotyped in form to be serving 
this function alone. As exaggeration and stereotypy are characteristic 
of most displays in most species of animals, it seems probable that the 
Swaying of these Night Monkeys was ritualized. (The undoubtedly 
homologous Swaying of some other New World monkeys may be 
even more exaggerated in form and is undoubtedly ritualized.) 

Unfortunately, the function(s) and, to a lesser extent, the causa- 
tion of the Swaying of Night Monkeys remain partly obscure, pri- 
marily because the behavior was never observed in the wild. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 17 


One comment may be inserted here. Experience has demonstrated 
that ritualized patterns performed by captive individuals are always 
found to be part of the “natural” repertory of the species, performed 
by wild individuals under natural conditions, when it is possible to 
study the behavior of wild individuals in detail. This is certainly true 
for all the other New World monkeys that I have been able to observe 
at length both in captivity and in the wild. Captive animals may per- 
form displays in situations that are different from those in which wild 
animals perform the same displays (see also below), but they do not 
perform displays that are not also performed by wild animals in some 
situations. Thus it is safe to assume that wild Night Monkeys also 
perform Swaying, although I did not actually observe it during my 
relatively brief observations in the wild. 

The Swaying of the captive animals was frequently followed by 
overt escape. This, and the fact that it was apparently always pro- 
voked by the sight of a predator or potential predator, would indi- 
cate that it was produced by activation of the escape tendency. It 
seems likely, however, that some other tendency was also involved, 
as all or most other ritualized patterns of Night Monkeys and other 
species are produced when at least two tendencies are activated simul- 
taneously. If so, the other tendency was probably attack, as some 
Swaying was accompanied by Gruff Grunts, and these notes certainly 
include an attack component (see below). If attack was activated 
during silent Swaying, it was obviously much weaker than escape, 
but it may have been relatively slightly stronger than in the previously 
described unritualized overt escape and alarm patterns. There was 
no indication of any significantly close or regular association between 
Swaying and any other type of social (“friendly” or sexual) behavior. 

Swaying may discourage predators by letting them know that they 
have been seen. An alerted Night Monkey should have no difficulty 
in escaping from any natural predator. Night Monkeys are lighter, 
more rapid, and/or capable of leaping greater distances than any sym- 
patric species of carnivore. Most of the individual carnivores in any 
given area must be well aware of this fact. Thus a carnivore that sees 
a Swaying (and obviously alerted) Night Monkey probably will not 
bother to chase it. This, in turn, means that the monkey will not have 
to waste time by escaping. 

It seems likely that ritualized Swaying has been derived, in the 
course of evolution, from unritualized “peering.” Some of the side- 
ways components of Swaying may also have been derived from inten- 
tion movements of turning away (from an alarming stimulus). 


18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


SILENT ARCH POSTURES 


All the captive animals on Barro Colorado Island assumed dis- 
tinctive “Arch Postures” quite frequently. 

In these postures, the back was raised, usually to a considerable 
extent, and strongly arched or curved. The raising movement itself 
was always rapid, but the posture generally was maintained for an 
appreciable length of time (up to a minute in some cases). Sometimes 
an individual stood up on its hind limbs as it assumed an Arch Pos- 
ture. In this case, the front limbs were allowed to hang downward, 
usually with the elbows more or less akimbo. Generally the hands were 
held with the backs facing inward and the palms outward and the 
hind limbs were straight or only slightly bent at the knees. A typical 
Arch Posture of this type is shown in figure 3c (such postures were 
quite similar to some semierect preleaping postures in many respects, 
but could always be distinguished by the more extreme curvature of 
the back). In other Arch Postures, the performing individual did 
not stand up, but remained clasping the perch with both hands and 
feet. In such cases, all four limbs were usually nearly straight. Typi- 
cal Arch Postures of this type are shown in figures 3a.and 3b. (It 
will be noticed that the relative positions of the head and back are 
nearly the same in both Arch Postures of this type and some head- 
down postures ; but the effect is achieved by different methods in the 
two patterns, lowering the head in one case and raising the back in 
the other, and the general impression or “gestalt” of the two patterns 
is quite different. ) 

Irrespective of these minor variations in physical form, the Arch 
patterns of the captive animals could be divided into two main cate- 
gories—one silent; the other accompanied by Resonant Grunts (see 
below). 

Silent Arch Postures were assumed during all sorts of purely and 
partly hostile intraspecific encounters; but they were most common, 
and most long-sustained on the average, during the longest and most 
violent disputes between previously unacquainted individuals of the 
same sex and between “territorial” rivals in adjoining cages. They 
were often preceded and followed by a great variety of purely or 
partly hostile vocal patterns, produced by both attack and escape tend- 
encies (see below), and/or by ambivalent unritualized hostile move- 
ments. When they were closely associated with an unambiguous un- 
ritualized hostile reaction, however, it was always overt attack, not 
escape. Individuals assumed silent Arch Postures with appreciable 
frequency both before and after delivering overt attacks. 


19 


3.—Arch Postures. 


BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 
Fic 


Upper left and right, two typical Arch Postures with hands grasping the 


perch. Bottom, an Arch Posture with hands raised off the perch. 


NO. 5 


20 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


These facts would suggest that all or most silent Arch Postures are 
essentially purely hostile per se, and usually are produced when hostile 
motivation is rather strong. The attack tendency must be stronger 
than the escape tendency in all or most silent Arch patterns, but 
probably relatively weaker than in actual overt attack. 

Both morphological types of Arch Postures, standing up and with 
both hands and feet grasping the perch, were common during silent 
Arch performances. 

(A relatively very few Arch Postures were accompanied by a few 
Gruff Grunts, Gulps, and/or “Squeaks” [see below]. These perform- 
ances appeared to be closely related to, perhaps nothing more than 
abnormal variants of, typical silent Arch performances. They may 
have been produced by motivation intermediate between that usually 
producing typical silent Arches and that producing the accompanying 
notes without Arches.) 

Silent Arch Postures seem to function as threat. The assumption 
of an Arch Posture by one individual during an intraspecific dispute 
usually induces retreat by its opponent. The sight of an Arch Posture 
apparently activates or stimulates the escape tendency (and probably 
suppresses or weakens the attack tendency) of the perceiving indi- 
vidual. This effect is probably partly due to the fact than an animal 
in an Arch Posture looks larger than usual. 

The Arch Posture is particularly interesting from a comparative 
point of view. The only other New World primate known to have 
a similar pattern is the howler Alouatta palliata. 

Part of the Arch Posture may have been derived from an intention 
movement of leaping. Even when it does not assume a semierect pos- 
ture, a Night Monkey usually “shifts” its body backward, to put more 
of its weight on its hind limbs, just before leaping for any reason in 
any circumstances; and this shifting tends to increase the normal 
curvature of the back. The straightening of the limbs in the Arch may 
have been derived from another source. Many other platyrrhines (e.g., 
Cebus spp.) tend to hold the limbs very straight and stiff when jump- 
ing up and down in rage. It is conceivable that the Night Monkey 
has retained the “‘stiff-leggedness” of such a pattern, even though it 
has lost the jumping component. At least, I have never seen similar 
straightening of the limbs by any other species (except Alouatta 
palliata) in other circumstances. 


OTHER VISUAL PATTERNS © 


An important negative feature of the hostile repertory of Night 
Monkeys is the comparative rarity of “displacement” activities, 7.e., 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 21 


activities that appear to be “irrelevant” or “extraneous” or “out of 
context.” Individuals of this species seem to perform such activities 
relatively much less frequently than individuals of many related 
species. 

The absence or extreme rarity of autogrooming in hostile situations 
has already been mentioned. 

There is only one pattern that may be partly equivalent to the “dis- 
placement” scratching of many other platyrrhines. Some of the cap- 
tive Night Monkeys on Barro Colorado Island sneezed repeatedly 
during some high-intensity disputes. In other situations, sneezing 
seems to be a cleaning or comfort movement of a peculiar type used 
to clear or clean out the nose. Sneezing is very common after feeding, 
presumably because animals tend to get food up their noses. As the 
Night Monkeys that sneezed so frequently during disputes had not 
been feeding immediately beforehand, it is at least conceivable that 
some or all of their sneezing was a direct result or consequence, in 
one way or another, of their hostility. 

(The whole problem of “displacement” activities by New World 
primates will be discussed in a later paper.) 

Night Monkeys defecate and urinate in some hostile situations. A 
captive but untamed individual may defecate and/or urinate when 
caught and picked up by a human being. This is usually accompanied 
by “Screams” (see below) and/or Gruff Grunts. Some or all of 
these defecation and urination patterns are probably produced when 
both attack and escape tendencies are strong, but escape is definitely 
predominant over attack. 

I do not think that the defecation and urination patterns of Night 
Monkeys subserve any signal function or help to repell predators at a 
distance. Wild individuals of many other species of New World mon- 
keys frequently urinate and/or defecate upon predators or potential 
predators passing beneath them in the forest; but none of the wild 
Night Monkeys on Barro Colorado Island was observed to do so. 

Vomiting may also be closely associated with the hostile behavior 
of Night Monkeys in certain circumstances. Some of the captive indi- 
viduals on Barro Colorado Island vomited occasionally, or performed 
the “heaving” motions that frequently precede vomiting, during or 
immediately after hostile reactions (escape movements, Swaying, 
and/or Gruff Grunts) to the approach of a human observer. 

Night Monkeys have fewer signal patterns designed to be perceived 
by the eye than many other New World primates; and most of the 
signal patterns of this type that they do have are comparatively crude 
or “gross,” produced by simple movements of the whole body and/or 


22 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


head. They lack the variety of facial expressions (e.g., frowns, baring 
of teeth, pursing of lips) which are important signal patterns in such 
platyrrhines as Cebus and Ateles species. They also lack the special 
elongated tufts and ruffs of hair on the head and around the face 
which are typical of many marmosets and tamarins, and which can be 
erected or depressed to convey signal information. The absence of 
such characters may be correlated with nocturnality. Adult Night 
Monkeys frequently become separated (by distances of at least several 
yards) from their mates and subadult young while feeding and moy- 
ing in the forest at night. In such circumstances, they probably can- 
not rely upon perceiving visual signals, especially small and complex 
signals, from their companions as frequently or as easily as can adults 
of other platyrrhine species which are diurnal or crepuscular (see also 
discussion on page 45). 

Andrew (op cit.) says that Night Monkeys open the mouth in 
threat. I never saw this except when sounds were uttered at the 
same time. 

The complex black-and-white head pattern of Night Monkeys 
(more complex than the corresponding patterns of any other New 
World primate) may provide a partial substitute for a variety of 
facial expressions, whenever Night Monkeys are close enough to- 
gether to perceive the pattern clearly. Because the black and white 
stripes and patches are convergent and divergent, and some of them 
are curved, a slight alteration of the position of the head relative to 
an observer will tend to alter the whole appearance or “gestalt” of the 
pattern (see the accompanying sketches). This must help to empha- 
size the signal effect of head movements. 


GRUFF GRUNTS 


Most of the vocal patterns of adult Panamanian Night Monkeys 
can be divided into eight main categories: Gruff Grunts, Resonant 
Grunts, Screams, Low Trills, Moans, Gulps, Sneeze-grunts, and 
Hoots. All except the last seem to be purely or partly hostile. 

Gruff Grunts are among the most common of the vocal patterns. I 
heard them uttered by both wild and captive individuals on Barro 
Colorado Island and by all the other captive individuals studied. The 
typical Gruff Grunts of almost all individuals were similar or identical 
in sound (to the human ear), low-pitched, moderately long, and mod- 
erately loud. Figure 4 is a sketch of a sound spectrogram of a single 
note of this type. 

(All the sound spectrograms illustrated in this paper are derived 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 23 


from recordings of captive individuals on Barro Colorado Island. The 
recordings were made by an Ampex 601, using an Electrovoice 666 
Variable D Cardioid microphone. ) 

Gruff Grunts usually are uttered singly or in short series of two 
to five notes. Each note of a short series seems to be essentially the 
same as a single note, and the successive notes may be uttered at 
slightly irregular intervals. Series are frequently repeated, but suc- 
cessive series generally are separated by relatively long pauses (much 
longer than the longest intervals between notes of a single series). 


0.2 0.4 
Fic. 4.—A typical Gruff Grunt, uttered by an adult. 


Based upon a sound spectrogram made by a Kay Electric Co. “Vibralyzer.” 

In this and all the following sketches of vocal patterns, the figures along the 
vertical axis indicate cycles per second and the figures on the horizontal axis 
indicate elapsed time in seconds. 


All Gruff Grunts seem to be uttered with the mouth almost or com- 
pletely closed; but they are usually, or perhaps always, accompanied 
by conspicuous inflation of the throat region, the throat swelling up 
just as a note begins, and then subsiding immediately as soon as the 
note ends (even when the note is part of a series). 

Both single Gruff Grunts and short series of such notes are uttered 
frequently during all sorts of hostile encounters, apart from partly 
sexual situations, during both brief and prolonged intraspecific dis- 
putes in the wild as well as in captivity, and as reactions to the 
appearance of a predator or potential predator. During intraspecific 
disputes, they are uttered by aggressive individuals before and after 


24. SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


attacking, by obviously frightened or defensive individuals before 
and after performing overt escape movements, and by individuals 
that perform both overt attack and escape movements as well as 
other displays. They are also uttered during some encounters between 
mates or potential mates (see below), but they are relatively (if not 
always actually) uncommon in such circumstances. These facts would 
suggest that all or most Gruff Grunts are produced by simultaneous 
activation of the attack and escape tendencies, and these tendencies 
alone. The range of motivation is probably quite broad. Some Gruff 
Grunts seem to be produced when the attack tendency is appreciably 
stronger than the escape tendency; others when the escape tendency 
is appreciably stronger than the attack tendency ; and still others when 
the two tendencies are effectively equal. The attack tendency probably 
is never as preponderant during any Gruff Grunt performance as 
during silent overt attack movements; and the escape tendency is 
probably never as preponderant during any Gruff Grunt performance 
as during silent overt escape movements or simple freeze or head- 
down patterns. The actual strength of the two tendencies involved 
may be quite different in different circumstances. Some of the single 
Gruff Grunts and short series of such notes during brief disputes 
probably are produced when both tendencies are actually weak. Some 
of the similar-sounding series of notes during prolonged and vigor- 
ous disputes may be produced when both tendencies are actually 
quite strong. The escape tendency may be relatively weaker during 
the least aggressive typical Gruff Grunt performances than during 
silent Swaying; or, if the relative strength of the two hostile tenden- 
cies is the same in the two types of performance, their actual strength 
may be greater during the Gruff Grunts than during silent Swaying. 
Similarly, the most aggressive Gruff Grunt performances may be less 
aggressive than typical silent Arch Postures or, perhaps less probably, 
produced when the total actual strength of the hostile motivation (the 
attack and escape tendencies together ) is less than in the silent Arches. 

Gruff Grunts certainly function as threat during intraspecific dis- 
putes. They probably do not have a regular signal function when pro- 
voked by the appearance of a potential predator. 

Some of the captive individuals on Barro Colorado Island were 
heard to utter single but frequently repeated soft notes which sounded 
as if they might be nothing more than weak or muffled versions of 
typical Gruff Grunt notes. Some of these were so soft that they were 
little more than audible expirations of breath. They were uttered by 
untamed individuals watching me approach, and were followed (when 
I came closer) by moderately vigorous escape movements, sometimes 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 25 


with Gulps and/or Sneeze-grunts. They may have been produced by 
weaker hostile motivation than typical Gruff Grunt performances and 
have contained a relatively stronger escape component. (The differ- 
ence in causation between these soft Grunts without Swaying and 
silent Swaying with typical Gruff Grunts was not clear.) 


RESONANT GRUNTS 


Resonant Grunt performances are much more complex in form 
than typical Gruff Grunt performances, and seem to be much rarer. 

All the Resonant Grunts heard were uttered in long series of at 
least 10 to 15 notes. The speed, loudness, and pitch of the notes 
varied considerably. 

The most frequently heard, and simplest, type of series was as fol- 
lows. The first notes were soft and low, much softer than typical 
Gruff Grunts but otherwise similar. Succeeding notes gradually be- 
came louder and louder throughout the first half of the series. By 
the middle of the series, they were often very loud indeed, much more 
so than any Gruff Grunts, and had developed a penetrating, resonating 
quality. They were still gruntlike (insofar as they retained a gutteral 
“rrrr’ sound), but they had also acquired an “ooo” sound slightly 
reminiscent of the Hoots (see below). All the notes during the first 
half of a series were always uttered exceedingly rapidly one right after 
the other, in an emphatic “pumping” rhythm. The intervals between 
successive notes were much shorter than the intervals between notes in 
typical series of Gruff Grunts. It is possible that the pitch of the 
successive notes gradually became higher during the first half of some 
of these Resonant Grunt series, but this change (if it actually oc- 
curred) was always slight. Once the climax of a series was reached, 
the next few notes remained essentially similar, loud, resonating, and 
rapid. Then, during the final quarter (approximately) of the series, 
the rhythm usually slowed down a little and, at least in some cases, the 
notes became successively lower and/or softer. 

Series of this type were heard only among the captive individuals 
on Barro Colorado Island. One male uttered 30 to 40 series during 
a violent dispute with neighbors in adjoining cages immediately after 
being reintroduced into a cage with which he was thoroughly familiar 
after an absence of several weeks. One female uttered one series 
toward the end of a long and violent dispute with a male to whom she 
had just been introduced. Another female, alone in her cage, uttered 
several series after being disturbed by a human being, just after the 
human being had left and disappeared from her sight. These three 
individuals uttered their series of Resonant Grunts at night ; and they 


26 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


certainly were not aware that their performances were being observed 
by a human being (from a blind). A very tame hand-reared male 
kept in a cage in the animal house, with a female to whom he was 
well adjusted, frequently uttered similar series at night and (more 
frequently) in the morning well after dawn, usually just as or just 
after an animal-keeper passed by his cage and/or some howler mon- 
keys (Alouatta palliata) in an adjacent cage uttered “barks” and/or 
“roars” (see below). 

The simple series of all these individuals except the solitary female 
were composed of essentially monosyllabic notes. At least some of the 
notes of the solitary female were definitely bisyllabic. Unfortunately, 
it was not possible to record any of these performances. The series 
of monosyllabic notes might be transcribed by something like 

“Roo roo roo roo... .” 
Coser se ie 
while the series of bisyllabic notes were something like 


” 


“00a T0038. TOOR, TOOA.< 6. 
itt. stir ir. tre 


In all these performances, the individual uttering the Resonant 
Grunts assumed a moderate or extreme Arch Posture as soon as it 
began a series and went out of the posture as soon as the series was 
ended. Most of these postures were identical with typical silent 
Arches, with both hands and feet grasping the perch. The male en- 
gaged in disputes with his neighbors, however, tended to stand up on 
his hind limbs when he assumed an Arch. During some or all of 
these standing postures, his arms hung downward as usual but his 
elbows went in and out in rhythm with the accompanying notes. I 
think that the elbows always went in as a note was uttered and out 
between notes. Night Monkeys have a large ventral air-sac communi- 
cating with the larynx (Hill, 1960). It is possible that the inward 
movements of the elbows helped to expel air from the sac and that 
the outward movements helped to draw air into the sac. This may 
have facilitated sound production. (The other individuals were able 
to utter similar and probably equally loud notes without similar move- 
ments; but they did not utter as many series of notes in a short 
period of time.) 

All these series of Resonant Grunts were immediately preceded by 
more or less prolonged periods of intense activity, the performing 
animal running and leaping madly about its cage and uttering many 
Gruff Grunts, Gulps, and Sneeze-grunts (and sometimes Moans and/ 
or Low Trills). 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 27 


The tame captive male occasionally uttered series of Resonant 
Grunts which contined to accelerate throughout. In some cases, these 
series ended in a “bubble” of very, very short rapid notes. In other 
cases, they accelerated to such an extent that the terminal notes “ran 
together,” to form a continuous or semicontinuous and moderately 
load “Roar.” Both types of accelerated series were uttered in the 
same circumstances as the simpler series of the same male, in associa- 
tion with the same other patterns. 

Dr. John H. Kaufmann (personal communication) heard one or 
more wild Night Monkeys on Barro Colorado Island utter Roars 
which may have been the culmination of similar accelerated series of 
Resonant Grunts. These patterns were uttered during a prolonged 
dispute between two pairs or family groups feeding, or attempting to 
feed, on a palm tree in fruit. Unfortunately, the postures and move- 
ments accompanying the notes were not observed. I heard two series 
of Resonant Grunt-like notes apparently uttered by wild Night Mon- 
keys on Barro Colorado, again without being able to see the animals 
while the calls were uttered. The first series was composed of from 10 
to 15 moderately loud and long bisyllabic grunts, all essentially similar 
and very low in pitch. The second series followed shortly after the first, 
and included many more notes. The notes were monosyllabic, and the 
series rose and fell in pitch at least twice. The general effect was 
almost singing. These series were probably uttered by one or both 
members of a mated pair, just before retiring to their sleeping hole 
at dawn. I could not determine what (if any) external stimuli pro- 
voked the performances. 

With these few and enigmatic exceptions, Resonant Grunts were 
not heard to be uttered by wild Night Monkeys on Barro Colorado 
Island. It is possible that I missed many Resonant Grunt perform- 
ances (see below), but I was out in the forest frequently enough to 
establish that such performances must at least be relatively very rare 
on Barro Colorado. 

The association of Resonant Grunts with Arch Postures, their 
similarity in sound to undoubtedly hostile Gruff Grunts, and their 
occurrence during some disputes, would indicate that they are at least 
partly hostile. Although some Resonant Grunts show a slight resem- 
blance to Hoots, and the latter are undoubtedly sexual (see below), 
none of the individuals observed to utter Resonant Grunts appeared to 
be sexually aroused at the time. They did not perform any overt 
pairing or copulatory reactions. This would suggest that all Resonant 
Grunts may be purely hostile. The rarity of Resonant Grunts, the 
complex forms of the series of such notes, and their close association 


28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


with periods of intense activity and obvious excitement would suggest 
that they are very high intensity patterns, produced when the total hos- 
tile motivation is very strong. They probably are higher intensity on 
the average than either typical Gruff Grunts or typical silent Arch 
performances. They certainly appear to be much more closely asso- 
ciated with, or confined to, periods of intense excitement than either 
of the latter patterns. Approximately half of the Resonant Grunt 
performances of the male engaged in disputes with his neighbors were 
followed immediately by his performing overt attack. The remainder 
were followed by a variety of ambivalent movements and/or other 
vocal patterns, all or most of which probably included an escape com- 
ponent, but never by actual, complete, overt escape. It may also be 
significant that the only captive individual that uttered Resonant 
Grunts while obviously aware of being watched was the very tame 
hand-reared male, the only individual that was not at all afraid of 
human beings. These facts would indicate that Resonant Grunts are 
at least as aggressive, on the average, as the most aggressive Gruff 
Grunts or (more probably) silent Arch patterns. (In other words, the 
preponderance of the attack tendency over the escape tendency may 
be as great in Resonant Grunt performances as in silent Arch postures, 
although the actual strength of both tendencies is greater in the former 
than in the latter.) The escape component may be almost as minimal 
in some Resonant Grunt performances as in some overt attack pat- 
terns, especially the “deflected” attacks. 

The complete Resonant Grunt performances of the captive male 
engaged in disputes with his neighbors appeared to function as threat ; 
but it was impossible to distinguish between the signal effects of the 
notes themselves and those of the accompanying Arch Postures. 

The Resonant Grunt performances are reminiscent of the most 
spectacular vocalizations of some other platyrrhine monkeys. In par- 
ticular, they sound much like some calls of titi monkeys, Callicebus 
spp., and the howler Alouatta palliata. The former have not been 
studied in sufficient detail to permit discussion of probable homologies 
and analogies; but the situation is clearer with respect to the latter. 
The Roar at the end of some series of Resonant Grunts seems to be 
strictly homologous with the loud and prolonged vocalization of Alou- 
atta spp. which is responsible for the vernacular name of the genus. In 
the case of A. palliata, this is the pattern which Carpenter (1934) 
calls “type 1 vocalization” and describes as “a voluminous barking 
roar,” and which Altmann (1959) calls the ‘‘roar or howl, type Al.” 
It often sounds like nothing more than a much amplified and pro- 
longed version of the Roar of the Night Monkey, and is frequently 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 29 


preceded by barking or loud grunting noises which may be largely or 
completely homologous with typical Resonant Grunts. At low inten- 
sities of motivation, male A. palliata may utter series of similar or 
identical sounds (probably the notes that Altmann calls “male bark, 
type Cl” and “incipient male bark, type C2’) without ever breaking 
into an actual roar. Some of the lowest intensity and softest versions 
of such series, when heard at a distance, are remarkably difficult to 
distinguish from the simplest series of Resonant Grunts heard close 
up. 

The resemblance between the two patterns is so great that the tame 
male Night Monkey had been uttering Resonant Grunts occasionally 
for several weeks before I realized that they were not being uttered 
by the howlers in the adjacent cage. 

It has already been mentioned that the Resonant Grunt perform- 
ances of this tame Night Monkey were sometimes uttered during or 
immediately after roars and barks by the adjoining howlers. Possi- 
bly they were directly released by the sound of the latter. If so, this 
would help to explain why this Night Monkey tended to utter Reso- 
nant Grunts in the morning rather than at night. 

The apparent facilitation of Resonant Grunts by the roars and 
barks of howlers might be additional evidence that the patterns are 
related to one another. 

(Although individuals of A. palliata may assume Arch Postures 
while uttering certain types of barks, they apparently do not usually 
assume sttch postures while uttering roars or the barks most closely 
associated with roars. ) 

The roars and barks of A. palliata are certainly hostile. Individuals 
of this species tend to associate in bands; each band seems to have its 
own home range or territory ; and members of different bands usually 
utter roars and/or barks whenever they come close together (see 
Carpenter, 1934, and Collias and Southwick, 1952). If the Resonant 
Grunts of Night Monkeys are provoked by similar stimuli, the rarity 
of such notes in the forest on Barro Colorado Island may be partly 
due to the rarity of encounters between different pairs and family 
groups (see page 10). The roars of A. palliata apparently also func- 
tion as long-range proclamations of territorial ownership (see Alt- 
mann, op. cit.), and are uttered quite regularly apart from close-range 
or face-to-face encounters with rivals or neighbors. It seems very un- 
likely that the Resonant Grunt performances of Panamanian Night 
Monkeys can serve a similar function with any appreciable frequency. 

(It is possible, however, that some of the same or closely related 
patterns may do so in other populations of Night Monkeys. Hill, 


30 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


1960, cites several descriptions and transcriptions of calls uttered 
by South American Night Monkeys which may be Roars or typical 
Resonant Grunts or similar notes. There are enough records of such 
calls to suggest that they may be common among South American 
populations. It is conceivable that the typical Resonant Grunts and/or 
the Roars are becoming “obsolescent,’ in process of disappearing, 
among Panamanian Night Monkeys.) 

Some of the Resonant Grunt performances of Panamanian Night 
Monkeys may function as “triumph ceremonies” in certain circum- 
stances. The performances by captive individuals after being dis- 
turbed or irritated, when the cause of the disturbance was leaving or 
had left, appeared to be “‘proclamations” of the fact that the perform- 
ing individuals had defended their territories successfully. 

(It is possible that Night Monkeys have some sort of “territorial 
motivation,’ a tendency which is satisfied by the possession of a terri- 
tory, in addition to the ordinary attack tendency (see page 53). This 
might be involved in the causation of some Resonant Grunt perform- 
ances. If so, it might help to explain why Resonant Grunt perform- 
ances in which escape seems to be minimal do not always lead to 
immediate attack. ) 

Intermediates between typical Resonant Grunt and typical Gruff 
Grunt performances do occur, but apparently are always rare. I 
heard such intermediates only three times. All took the form of series 
of notes. Each note was similar to an ordinary Gruff Grunt in loud- 
ness and tone, but the successive notes of each series first rose and 
then fell in pitch in much the same way as in typical Resonant Grunt 
performances. One series of this type, including a great many notes, 
occurred in the same dispute in the forest on Barro Colorado Island 
in which typical Resonant Grunts also were heard. It was not possible 
to see the postures and movements accompanying this performance. 
A similar series was uttered by one member of a captive pair on 
Barro Colorado when disturbed by some stimulus (possibly an oce- 
lot?) outside its cage. This was uttered from a crouch posture. Sev- 
eral shorter series were uttered by the mate of the captive male who 
uttered typical Resonant Grunts while engaged in disputes with his 
neighbors. She usually started a series just after the male had begun 
a series of Resonant Grunts, or just after he had finished—thus pro- 
viding a peculiar ‘“‘echo”’ to his performances. Her series were uttered 
from typical Arch Postures, with both hands and feet grasping a 
perch. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 31 


SCREAMS 


The captive Night Monkeys on Barro Colorado Island and at Iqui- 
tos usually uttered loud Screams when caught and handled by a 
human being. (I presume that all Night Monkeys are capable of 
uttering similar or identical sounds ; but none of the other individuals 
studied was caught and handled.) These Screams were always high 
pitched but wavering. Many of them were quite prolonged. Figures 
5 and 6 are sketches of sound spectrograms of two typical Screams. 


6,000 
4,800 
3,600 
2,400 


1,200 


Fic. 5.—A relatively short Scream, uttered by an adult. 


Based upon a spectrogram by a “Vibralyser.” 
Only the first harmonic is shown. There certainly was at least one other 
present. 


All the observed Night Monkey Screams were uttered with the 
mouth wide open, but they were not accompanied by any conspicuous 
inflation of the throat region. 

All the captive animals struggled violently when first picked up. 
Most of the struggling movements were obviously attempts to escape ; 
but many of the animals also tried to bite their captors whenever 
possible. The animals usually uttered many Screams during the initial 
struggles, when both their escape movements and attempts to bite were 
most vigorous. If the animals were held firmly for several minutes, 
however, they usually calmed down, gradually, and their struggles sub- 
sided. At this stage, they generally stopped Screaming and started to 
utter Gruff Grunts and/or Gulps. 


32 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


These facts would indicate that the Screams were produced when 
hostile motivation as a whole was very strong, when the escape tend- 
ency was much stronger than the attack tendency (although probably 
less preponderant than in the silent overt escape and alarm patterns). 
The actual strength of the hostile motivation as a whole during 
Screams was almost certainly greater than in any Gruff Grunt pat- 
tern in which escape is predominant. 


02, 04 06 08 Lo 12 14 Le 1.8 2.0 
Fic. 6.—Two long Screams, uttered by an adult. 


Based upon a spectrogram by a Kay Electric Co. “Sona-graph.” 
Only the first harmonic is shown. There was at least one other with each note. 
In this and all the following sketches which include more than one note, the 
interval(s) shown between notes are the actual interval(s). 


Screams may be adapted to frighten or startle a predator, and thus 
give a screaming animal a better chance to escape. 


LOW TRILLS 


Night Monkeys utter a variety of sounds that can be described as 
Trills. 

The most common sounds of this type uttered by captive adults 
on Barro Colorado Island were “Low Trills.” A single Low Trill 
was a “bubbling” series of short low-pitched notes, uttered very 
rapidly one right after the other but still distinguishable by the human 
ear. All or most of the individual notes in a series were slightly plain- 
tive in tone, and sounded somewhat like abbreviated versions of simple 
Moans (see page 36). The number of notes in a single series varied 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 33 


from three to at least twelve. The successive notes tended to rise 
slightly in pitch. Sketches of sound spectrograms of three more or 
less typical Low Trills are shown in figures 7, 8, and 9. 


2,400 
1,200 
960 
720 eer ae 
480 eB tet dh dh 
240 vi v 
0.6 1.2 
Fic. 7—A Low Trill, uttered by an adult. 
Based upon a spectrogram by a Kay Electric Co. “Missilyzer.” 
960 
720 
= 
Ee 
480 Lam Ef 
i= 
240 Sie, ? id 


2.0 4.0 
Fic. 8.—One Moan, followed immediately by one Low Trill, uttered by an adult. 


Based upon a spectrogram by a “Missilyzer.” 


All Low Trills were uttered with the mouth closed or nearly closed. 
They were not accompanied by any conspicuous inflation of the throat 
region. 

They were uttered in several different situations. They were com- 


34 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


mon during the initial reactions between strangers, previously un- 
acquainted individuals meeting one another for the first time. In 
such circumstances, they could be uttered by any one or all of the 
individuals involved. When the previously unacquainted individuals 
were of opposite sex, females tended to utter Low Trills more fre- 
quently than males. Almost all individuals meeting a stranger tend 
to be very cautious at first, moving in a hesitant and apparently 
slightly alarmed manner. Females usually are even more hesitant 
than males. The Low Trills during initial “greetings” often were 


960 


aye 


720. 


mnnyes 


480 


240 


ner 


4.0 


Fic. 9—One Sneeze-grunt, followed immediately by one Low Trill, uttered 
by an adult. 


Based upon a spectrogram by a “Missilyser.” 


uttered in close association with, immediately before, or during “social 
sniffing” performances (see page 4/7). Similar notes were uttered 
frequently by single individuals immediately after being released in 
new cages, as they began to explore their quarters in a cautious man- 
ner. Some individuals also uttered such notes when they went down 
to get food from dishes on the ground, even when they were in familiar 
quarters and/or there were no other Night Monkeys nearby. Captive 
Night Monkeys tend to be slightly “nervous” whenever they have to 
come down to the ground, presumably because they would never do 
so under natural conditions. Although many Low Trills during “greet- 
ings” were eventually followed by overt, unritualized, hostile, and/or 
sexual behavior (depending upon the age, sex, and condition of the 
individuals involved), they were not usually uttered during or after 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 35 


either actual fighting or copulatory reactions. The only prolonged 
and vigorous dispute during which I heard such notes uttered was an 
encounter between a male and a female that had fought with one 
another previously, been separated for several days, and then were 
brought together again. 

These facts would suggest that many or all of the Low Trills were 
partly hostile reactions, produced when the escape tendency was 
stronger than the attack tendency, but effectively counteracted by 
some other nonhostile tendency, such as sex or feeding, or a non- 
hostile tendency plus attack. The nonhostile tendencies involved prob- 
ably were not direct causes of the Low Trills in the same way as the 
escape tendency, as different nonhostile tendencies were involved in 
different situations while the sounds themselves remained essentially 
the same in all situations. 

Some of the circumstances in which Low Trills were uttered were 
similar to those in which many other species perform appeasement 
displays. The Low Trills may have had a similar function; but I 
could not distinguish their exact signal effects from those of associated 
patterns such as social sniffing. 

Two full-grown Night Monkeys at the National Zoological Park 
in Washington, D. C., and some apparently full-grown individuals in 
the forest on Barro Colorado Island were heard to utter Trills remi- 
niscent of typical Low Trills in sound but higher pitched, more 
mechanical or metallic in tone, and without any plaintive or moaning 
quality. 

The significance of these patterns was difficult to assess, as the 
captive individuals in the National Zoological Park were observed 
briefly and the wild individuals on Barro Colorado uttered their notes 
in obscure and/or complex social situations. 

The National Zoological Park animals were a male and a female. 
I was told that they had been placed together in the same cage only 
a few days before my observations. They uttered their high-pitched 
Trills whenever they came face to face. Many of their Trills were 
followed by social sniffing, without any (other) overt indications of 
hostility. I did not hear them utter anything more closely similar to 
the Low Trills of the captive adults on Barro Colorado. This might 
suggest that their high-pitched Trills were equivalent to the latter, 
produced by the same or very similar motivation and serving the 
same function(s). 

Unlike the Zoo animals, the wild Night Monkeys that uttered high- 
pitched Trills were also heard to utter typical Low Trills during the 
same social encounters. 


36 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


I think that the high-pitched Trills of all these animals were very 
similar to, or even identical with, the High Trills of young Night 
Monkeys in sound (see page 69). 

It is possible, therefore, that the apparently full-grown individuals 
heard to utter such Trills were really immature, in spite of their size, 
or had reverted to juvenile or infantile behavior for some reason (see 
discussion, page 58). It is also conceivable that high-pitched Trills 
are the only pattern of this type in the population from which the 
National Zoological Park animals were obtained. 

As indicated in the accompanying figures, the notes of Low Trills 
are usually or always slightly higher in pitch than simple Moans. This 
might be evidence that Low Trills have been derived, in ontogeny 
and/or phylogeny, from a combination of typical Moans and High 
Trills. 

Some or all Low Trills seem to have a slightly “liquid” quality, 
somewhat reminiscent of Gulp notes. There may be some sort of 
special relationship between the two types of patterns, either direct or 
indirect via the Gruff Grunt patterns (see both the accompanying 
figures and the comments below). Gulps, Moans, and Low Trills are 
all similar in some aspects of motivation. 


MOANS 


Many of the wild Night Monkeys on Barro Colorado Island and 
almost all the captive individuals at the National Zoological Park and 
at Iquitos as well as on Barro Colorado were heard to utter brief, 
soft, plaintive-sounding Moans (see figures 8 and 10). 

Like the Low Trills, the simple Moans were uttered with the mouth 
apparently closed and were not accompanied by conspicuous inflation 
of the throat. 

All or almost all the animals uttered Moans more frequently than 
Trills on the average. They uttered Moans in all the situations in 
which Low Trills were characteristic, and in other circumstances as 
well. Moans were relatively rare during the initial “greetings” be- 
tween strangers and during the exploration of strange surroundings ; 
but they were uttered frequently by individuals in familiar social and 
physical environments. They were the most common notes uttered 
by particularly tame (hand-raised) captive animals on Barro Colorado 
Island when I came within a few feet of their cages. At such times, 
they were uttered by animals approaching me in an apparently friendly 
or “curious” manner, as well as by animals performing slight retreat 
movements or intention movements. They were also uttered by mated 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 37 


animals in captivity on Barro Colorado Island, if the mates were 
well adjusted to one another, when one individual rejoined its mate 
after being separated from it for a few seconds or minutes (the indi- 
viduals that uttered Moans in such circumstances were long past the 
stage of uttering Low Trills as “greetings” ). Moans were not usually 
directed by one individual toward its mate in other circumstances, 
1.¢., after they had been together for more than a few seconds. (The 
only exceptions were a few Moans uttered during the first parts of 


480 


) 


alls 


aye 


2.0 4.0 
Fic. 10.—One Moan, uttered by an adult. 


Based upon a spectrogram by a “Missilyzer.” 


long precopulatory sequences. The animals that uttered these notes 
did not utter more Moans immediately before or during the actual 
copulations.) Some captive individuals on Barro Colorado Island 
uttered Moans while they watched other individuals fight without 
becoming involved in the fights themselves. A few individuals at both 
Iquitos and Barro Colorado kept in cages with monkeys and marmo- 
sets of other species but with no other individuals of their own species 
uttered Moans regularly as they moved about their cages in an almost 
perfectly relaxed manner. The other vocal patterns with which Moans 
were associated most frequently were Low Trills, Gulps, and Sneeze- 
grunts ; 7.¢., an individual that uttered Moans was quite likely to utter 
one or more of these other sounds immediately before and/or after 
the Moans. Both the Gulps and the Sneeze-grunts seem to contain 


38 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


an escape component, at least at times (see below). Moans also 
were associated with Gruff Grunts, but relatively rarely. 

It is probable, therefore, that Moans are similar to Low Trills in 
being produced when both hostile and one or more types of nonhostile 
motivation are activated simultaneously; but the hostile component 
seems to be relatively (and probably actually) weaker in the Moans 
than in all or most Low Trills. 

Moans may also function as appeasement. 


“1 OCOMOTORY NOTES’—GULPS AND SNEEZE-GRUNTS 


These two types of notes are among the most enigmatic and puz- 
zling of Night Monkey vocal patterns. 

Gulps are very common, probably even more so on the average than 
Gruff Grunts. They were uttered by all the adult individuals studied 
during the present investigation. 

A single Gulp is usually moderately loud and always rather “liquid” 
sounding, rather like a human belch. Sketches of the sound spectro- 
grams of two typical Gulps are shown in figure 11. The mouth remains 
apparently closed during the utterance of all or most Gulps; but 
every Gulp is accompanied by a brief swelling of the throat (to all 
appearances the same as the inflation accompanying Gruff Grunts 
described above). 

Gulps are usually uttered in series. A single uninterrupted series 
seldom includes more than five notes, and series of only two or three 
notes are quite common; but series are often repeated rapidly one 
right after the other, with only brief pauses between successive series. 

An overwhelming majority of all the Gulps heard were uttered by 
individuals engaged in vigorous locomotory activities. They were 
uttered during most periods of really active running and leaping in 
almost all social circumstances. 

One type of performance was particularly remarkable. As noted 
above, Night Monkeys generally are most active just after sunset 
and just before dawn. During the predawn periods, the captive ani- 
mals kept in large cages on Barro Colorado Island usually ran and 
leapt around their cages rapidly and almost continuously for a half 
hour or more. Single individuals alone in their cages, mated individ- 
uals in cages with their mates, and. individuals kept together in larger 
groups, all tended to behave in much the same way at such times. 
They seldom interrupted their running and leaping to feed or drink 
or groom themselves, and the individuals that had companions in their 
cages seldom performed unmistakable hostile or sexual patterns. They 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 39 


all tended to utter many Gulps during such periods, but relatively few 
or no other notes (except Sneeze-grunts—see below). Most of the 
wild Night Monkeys on Barro Colorado Island also uttered many 
Gulps and relatively few obviously hostile or sexual calls just before 
dawn. 


960 
720 
480 


240 


2.0 4.0 
Fic. 11—Two Gulps, uttered by an adult. 


Based upon a spectrogram by a “Missilyzer.” 

It will be seen that these notes are rather similar to the Sneeze-grunt shown 
in figure 9. There are, however, a few definite differences. The Sneeze-grunt 
has a range of frequencies between 0 and approximately 650 cycles per second 
(c.p.s.), all starting at exactly the same time and having predominant amplitudes 
and longer signals around 100 c.p.s. and 260 c.p.s. The Gulps begin with high 
frequencies around 850 c.p.s., and drop rapidly in frequency to fairly equal 
amplitude signals between approximately 240 and 600 c.p.s. 


The causation and function(s) of this intense (and often apparently 
“purposeless’”’) activity were difficult to identify. It is possible that 
Night Monkeys, like some other mammals that normally range over 
large areas (e.g., many canids—see Lorenz, 1952), have a strong 
“internal” tendency to perform locomotory movements, a tendency 
expressed by overt activity even when the other immediate needs of 
the animals (e.g., food, water, and sex) can be fully satisfied with- 
out moving about very much. 

If so, many or all Gulps may be expressions of the same “inde- 
pendent” locomotory tendency. They may be produced when the loco- 


40 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


motory tendency is too strong to be completely expressed by running 
and leaping movements alone. 

There are other indications, however, that some or all of the Gulps 
may be at least partly hostile. Although most overt escape and 
simple alarm patterns are silent (see page 13), Gulps are often uttered 
before and/or after such reactions. Very occasionally, Gulps are 
uttered by individuals that are not running or leaping at the time. One 
captive individual on Barro Colorado, for instance, uttered a particu- 
larly loud and rapid series of Gulps while sitting, frozen, in a crouch 
posture immediately after overt escape. It is also my impression that 
the wild individuals in the forest tended to utter relatively few Gulps 
when they were apparently unaware of being observed. The fre- 
quency of their Gulps sometimes increased greatly at the instant they 
seemed to become aware of the presence of an observer—even when 
their other activities continued unchanged. 

It is possible, therefore, that some or all of the Gulps may be pro- 
duced by a combination of locomotory and escape tendencies, usually 
when the latter is relatively and/or actually weak. 

The Gulps may exemplify a complex evolutionary change. 

Although they are probably partly hostile in motivation, they do not 
appear to be hostile in function. (They certainly do not function as 
warning signals. I have never seen a Night Monkey respond to the 
sounds of Gulps by performing overt escape.) They may well function 
as “contact notes.” They may help to keep mated individuals and 
members of the same family group together under natural conditions. 
The utterance of Gulps by one individual should enable its companions 
to tell exactly where it is (especially as Gulps are very penetrating, 
even when not very loud). 

No other species of platyrrhine monkey with which I am familiar 
utters similar notes with similar frequency in similar situations.” This 
would suggest that Gulps, as adult contact notes, are highly specialized. 
If so, they probably have been derived from some purely hostile 
Grunt pattern. (They are similar to Gruff Grunts in some aspects of 


2 Adult individuals of some other species, e.g., some tamarins, may utter “lost” 
notes or calls when completely isolated; but such notes or calls are seldom or 
never uttered by individuals moving about near other individuals of the same 
species in the same way as the Gulps of Night Monkeys. 

Adult howler monkeys of the species Alouatta palliata may utter some notes 
which help to maintain contact between the members of a band (Carpenter, 
1934, and Collias and Southwick, op. cit.) ; but such notes seem to be uttered 
relatively much less frequently than the Gulps of Night Monkeys, and may be 
produced by rather different motivation. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 4I 


form—see accompanying figures—and gruntlike patterns are wide- 
spread among many species of New World monkeys. ) 

It should be mentioned, in this connection, that the Gulps of Night 
Monkeys are uttered in almost exactly the same social circumstances 
as the contact notes of some species of Panamanian birds, and that 
the latter seem to have had a similar evolutionary history, also being 
derived from purely hostile patterns (Moynihan, 1963a). 

The development of Gulps as contact notes may be another adapta- 
tion to nocturnal habits. Night Monkeys probably cannot keep in touch 
with their companions by sight as easily as can other platyrrhines. 

(While recording the sounds of captive Night Monkeys on Barro 
Colorado Island, a few short and single “twanging” noises were heard, 
through the earphones of the recording machine. Unfortunately, these 
noises were always heard at times when the animals were not under 
observation. They probably were not mechanical in origin [at least, I 
could not find any object in the animals’ cage which looked as if it 
could be used to produce such noises] ; but nothing similar was heard 
in other circumstances. To my ears, the noises sounded as if they 
might be Gulps uttered in very close proximity to the microphone. 
This theory does not, however, seem to be supported by sound 
spectrograms [see figure 12]. The significance of these noises remains 
problematical.) 

One captive Night Monkey at Iquitos uttered series of gulping 
notes which had a distinctive hooting quality (rather as if “ooo” 
sounds were superimposed upon the ordinary Gulps of Barro 
Colorado animals). These notes were uttered while the animal was 
moving about its cage in a perfectly normal manner. They may have 
been a characteristic pattern of the local population of Night Monkeys, 
or an individual peculiarity, or (possibly) intermediates between 
typical Gulps and typical Hoots. 

Both the wild and captive animals on Barro Colorado Island were 
heard to utter Sneeze-grunts. 

A single Sneeze-grunt sounded like a single sneeze superimposed 
upon a single, loud, typical Gruff Grunt. The sound was always sharp 
and abrupt. Most Sneeze-grunts were also at least slightly nasal in 
tone (each note might be transcribed by something like “Aaanh’’). 
Figure 9 includes a sketch of a sound spectrogram of a more or less 
typical Sneeze-grunt. 

Sneeze-grunts were usually uttered singly. Sometimes two were 
uttered together in quick succession ; but they never occurred in longer 
series like Gulps and Gruff Grunts. 

They were uttered with the mouth closed or nearly closed, and 


42 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


apparently were not accompanied by conspicuous inflation of the 
throat region. 

They were almost always much less common than Gulps, but were 
uttered in the same types of situations, usually by individuals moving 
about very rapidly and energetically. They were frequently uttered in 
close temporal association with Gulps. In such cases, the most common 
arrangement was two or three Gulps, followed by a single Sneeze- 


2,400: 


1,200 


0.6 1.2 
Fic. 12.—A single “twanging” noise heard while recording adults. 


Based upon a spectrogram by a “Vibralyzer.” 


grunt, followed by two or three more Gulps, followed by another 
Sneeze-grunt, etc. 

Sneeze-grunts also were uttered quite frequently in close associa- 
tion with Moans and Trill patterns. In such cases, the notes were 
usually arranged in “doublets” of one note of each type: A single 
Moan followed immediately by a single Sneeze-grunt, or a single 
Sneeze-grunt followed immediately by a single Low or High Trill, or 
(most frequently) a single Sneeze-grunt followed immediately by a 
single Moan. These arrangements were stereotyped enough to suggest 
that they may have become partly ritualized per se. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 43 


It seems probable that the Sneeze-grunts were produced by some 
combination of locomotory and escape tendencies similar to that 
producing Gulps. It was noticeable, however, that Sneeze-grunts were 
most likely to be uttered just before or just as an individual made a 
long leap, while Gulps were relatively more frequently associated 
with shorter leaps and running movements. Similarly, when wild 
Night Monkeys were encountered in the forest, they sometimes 
uttered Gulps when approached in the dark, switched to Sneeze-grunts 
when a light was put on them, and then retreated in overt escape. 
Such facts would suggest that the Sneeze-grunts may have been 
produced when the escape tendency was stronger than in all or most 
Gulps. 

The Sneeze-grunt is the only Night Monkey pattern similar to the 
ordinary Warning Notes of many other New World monkeys in 
form. It resembles the latter in being loud, sharp, and usually single. 
It may well be phylogenetically related to the Warning Notes of other 
species, and may have functioned in the same way in the ancestors of 
the Night Monkey, alerting other individuals of the same species to 
possible danger in the environment and inducing them to escape. It 
does not, however, seem to function in this way at the present time. 
Its signal effect is obscure at present ; but it may be another contact 
note like the Gulps. 


DISCUSSION 


The preceding list includes almost all the typical vocal patterns of 
adult Night Monkeys.* It may be useful, therefore, to add some gen- 
eral comments about their vocal repertory as a whole. 

Adult Night Monkeys do utter vocalizations which are more or 
less obviously intermediate between otherwise distinct types of notes. 
Some of these intermediates have been mentioned above; others in- 
clude intermediates between Moans and Gruff Grunts, between Gruff 
Grunts and Gulps, and between Sneeze-grunts and Gruff Grunts. 

Although some of these intermediate vocalizations are uttered with 
appreciable actual frequency, they seem to be comparatively rare, and 
are uttered relatively less frequently than intermediate notes in the 
repertories of adults of some related species, e¢.g., species of Saimiri, 


3 Some brief experiments with an ultrasound detector would indicate that 
neither the adults nor an infant kept in captivity on Barro Colorado uttered any 
calls or notes completely inaudible to the human ear. In this respect, Night 
Monkeys seem to differ from at least one other species of New World primate, 
the Pygmy Marmoset (Cebuella pygmaea). 


44 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Cebus, and Afeles. In these latter species, all or most of the major 
adult vocal patterns seem to intergrade with one another through 
nearly continuous series of common intermediate notes. 

Such differences in organization of the vocal repertory presumably 
are correlated with different ways of functioning. 

When an adult of a Cebus species is motivated by three or four 
different tendencies (e.g., attack, escape, sex, and gregariousness), 
it may be able to express all the tendencies simultaneously by uttering 
a single ambivalent vocal pattern, more or less perfectly intermediate 
between the patterns which would be produced by each one of the 
tendencies (or pair of tendencies) if activated alone. A signal of this 
type must contain a relatively large amount of information. This in 
itself must be advantageous. But such signals may have disadvan- 
tages as well. It may be difficult for the receiver of a signal of this 
type to perceive or “decipher” all the information contained in the 
signal. Complex signals are probably more easy to misunderstand 
than simple signals containing less information. In these adult monkey 
repertories, it is also probable that the frequency of intermediate notes 
reduces the contrast between different signals. These features may 
induce momentary confusion in the individual(s) toward which 
the signals are directed. Even when confusion is avoided, they must 
tend to reduce the speed of the response to the signals. This is prob- 
ably very disadvantageous. There is evidence that selection always or 
nearly always tends to favor increased rapidity of social responses 
(Moynihan, 1963b). 

An adult Night Monkey motivated by a complex combination of 
tendencies usually does not utter intermediate notes, but rather a 
series of different types of notes, each one of which expresses only 
one or two of the tendencies. Thus, for instance, a wild adult sur- 
prised by a disturbing stimulus may utter a rapid “jumble” of Gruff 
Grunts, Moans, Gulps, Low Trills, and Sneeze-grunts, not a single 
vocal pattern expressing the complete combination of attack, escape, 
locomotory, and (possibly) other tendencies by which it is motivated. 
This sort of message may be able to convey as much information as 
a single note expressing all the tendencies simultaneously, if the ar- 
rangement of the different notes is not random and if the receiver’s 
reaction to the first note does not interfere with its reactions to the 
later notes; but it is certainly produced with greater physical effort 
and, more important, more slowly. The principal advantage of this 
sort of message presumably is that each one of its constituent notes, 
being an expression of only one or two tendencies, is a comparatively 
simple signal and, therefore, difficult to misunderstand. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 45 


The principal differences between these two types of platyrrhine 
vocal repertories may be summarized as follows. In one type, the 
information is coded very rapidly in an economical form but may be 
difficult to read, or difficult to read rapidly. In the other type, the 
information is coded less rapidly and efficiently but can be read more 
easily and/or more rapidly. 

The relative importance of the advantages and disadvantages in- 
herent in each type of repertory may be different depending on the 
circumstances and environments. The type of repertory possessed 
by adult Night Monkeys may be another (partly indirect) adaptation 
to nocturnality. 

Adults of Saimiri, Cebus, and Ateles can avoid some of the dis- 
advantages inherent in their type of vocal repertory by combining the 
information received from auditory signals with information derived 
by another method of perception. They are all thoroughly diurnal, 
and must usually be able to see other animals and objects in their 
environment quite clearly at considerable distances. As noted above, 
both Cebus and Ateles species have a wide variety of facial expres- 
sions which function as signals. They also have many other display 
postures and movements involving other parts of the body; and so 
does Saimiri. Any adult individual of these species hearing calls or 
notes from another individual of the same species usually will 
perceive visual signals and/or receive visual clues from the physical 
environment at the same time. These visual aids should enable the 
individual to grasp the significance of any vocal pattern, even when 
the latter is difficult to decipher or ambiguous in itself. 

As adult Night Monkeys usually receive less visual information 
(because they are nocturnal and frequently become separated from 
one another and have fewer visual displays), they are more dependent 
upon auditory signals and more likely to misinterpret any possibly 
ambiguous vocal patterns. It must be absolutely essential for them 
to have a system of vocal patterns which are easily and immediately 
comprehensible, whatever the inevitable disadvantages of such a sys- 
tem. 

It is probably significant, in this connection, that infant and juvenile 
Night Monkeys utter relatively more intermediate notes than do 
adults (see below). The vocal signals of infants to their parents 
may be supplemented by tactile stimuli, as the infants are always 
carried by their parents. Even when no longer carried, juveniles may 
be able to provide important supplemental visual information because 
they tend to remain very close to their parents. 

To my knowledge, the only Old World monkey in which this aspect 


46 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


of behavior has been studied is the Rhesus, Macaca mulatta. Accord- 
ing to Rowell and Hinde (1962), the vocal patterns of this species also 
intergrade through frequent intermediates. Rhesus Monkeys are 
diurnal, and have many visual displays, including many different facial 
expressions (Hinde and Rowell, 1962). 

These facts would suggest that there may be a general rule among 
monkeys that species or classes of individuals largely dependent upon 
auditory signals for the regulation of their social behavior tend to have 
discrete, sharply delimited vocal patterns, while species or classes of 
individuals less dependent upon auditory signals tend to have inter- 
grading vocal patterns. 

The major vocal patterns of adult Night Monkeys not only are 
discrete but also sound quite different from one another to human 
ears. This is true of patterns that look somewhat similar in sound 
spectrograms, e.g., Gulps and Sneeze-grunts, as well as of patterns 
that look very different. There is every reason to suppose that the 
contrasts between different types of patterns are equally obvious to 
the Night Monkeys themselves. It is quite evident from their behavior 
that the hearing of Night Monkeys is at least as good as that of 
human beings. 

Contrasts in sound may facilitate rapidity of comprehension when 
messages are composed of many notes of different types. 

The probable importance of this feature is emphasized by the fact 
that all the associations between different types of notes that are 
particularly common and/or seem to be ritualized per se are associa- 
tions between notes that contrast with one another in a particularly 
striking manner. In some cases, e.g., Gulps and Sneeze-grunts, the 
associated notes seem to serve similar functions. In other cases, e.g. 
Sneeze-grunts and Moans, they seem to serve quite different func- 
tions. Associations between notes that are somewhat less conspicuous- 
ly contrasting in sound, e.g., Moans and Low Trills, seem to be less 
frequent and/or less close. 

It is conceivable that information conveyed by contrasting stimuli 
may tend to produce stronger reactions (1.¢., may appear to be more 
“emphatic” to the receiver) than the same information conveyed by 
stimuli with less contrast; but this would be very difficult to prove. 

In spite of the obvious differences between the hostile vocal patterns 
of adult Night Monkeys, there are some morphological resemblances 
between many or most of them. This is certainly true of Gruff Grunts, 
Resonant Grunts, and Sneeze-grunts. A few similarities between 
Gulps and both Gruff Grunts and Sneeze-grunts have already been 
mentioned. It may be convenient to refer to all these patterns, together, 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 47 


as the “Grunt Complex.’ The low-pitched Moans and Low Trills may 
be related. It is possible that all the patterns of this complex have 
been derived, in whole or in part, from a single pattern or a single 
group of completely intergrading patterns (perhaps comparable to 
the “basic agonistic sound” of the Rhesus Monkey described by 
Rowell, 1962) in the course of evolution. 

The hostile vocalizations of adult Night Monkeys seem to be pro- 
duced when motivational conflict is stronger, 7.e., when incompatible 
tendencies are more nearly equal in strength, than when many or 
most of the nonvocal hostile displays are produced. A similar ar- 
rangement occurs in some species of birds (see, for instance, 
Moynihan, 1962b). It may be characteristic of most higher ver- 
tebrates. 

Some other aspects of the vocal behavior of Night Monkeys will 
be discussed below, in connection with the Hoot and Squeak patterns. 


SOCIAL SNIFFING 


This is the most common of the “greeting” patterns. In captivity 
(at least), it is almost always performed whenever any two pre- 
viously unacquainted individuals meet one another for the first time, 
and it also occurs in certain other social circumstances. 

It may be initiated by either one of the individuals involved, or by 
both simultaneously. A sniffing animal simply approaches another, 
stretches its neck forward a little, and smells the other, usually for 
at least several seconds. When sniffing is initiated by one individual, 
the individual being smelled remains motionless at first, but usually 
responds by sniffing in return if the first individual continues long 
enough. Sometimes two individuals will sniff one another alternately. 
More often their sniffing overlaps in time, even when they do not 
begin simultaneously. 

Most sniffing—both mutual sniffing and sniffing by only one in- 
dividual—begins “‘nose to nose.” An individual usually starts by 
bringing its face as close as possible to the face of its ‘‘partner” in 
the performance. It also may sniff into the armpit or under the arm 
and/or at the perineal region of its partner. This is relatively (if not 
actually) rare as an initial reaction, but it is very common immediately 
after nose-to-nose sniffing. Occasionally, two individuals will sniff at 
one another’s perineal regions simultaneously. This is done less fre- 
quently by Night Monkeys than by some tamarins (for which it is 
the usual form of social sniffing). 

It seems likely that this behavior is guided by, or orientated toward, 
special scent glands. The skin glands of Night Monkeys have been 


48 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


described by Hill, Appleyard, and Auber (1959) and Hanson and 
Montagna (1962). Apocrine glands are largest in the face and 
genital areas. Lateral to the alae of the nose they are intermingled 
with large sebaceous glands. There are also larger and more spe- 
cialized glandular organs in the sternal region and at the base of the 
tail. It is possible that individuals which appear to be sniffing in the 
armpits of their “partners” are really trying to get at their sternal 
organis. 

When complete strangers are introduced to one another, they may 
repeat social sniffing whenever they come close to one another during 
the first few hours (or nights) after the introduction. This behavior 
may disappear, gradually, as the animals become familiar with one 
another. Only in the case of males and females that become mated 
to one another does social sniffing remain frequent in certain situa- 
tions. Mates often perform social sniffing immediately before copula- 
tion and related patterns (see below). 

Social sniffing was performed by all the captive animals observed, 
at Washington and Iquitos as well as on Barro Colorado Island. It 
seemed to be performed with approximately equal frequency by all 
individuals in similar circumstances in captivity. 

Very tame, hand-reared individuals in captivity on Barro Colorado 
Island sniffed at human beings occasionally, especially if they were 
familiar with the person involved but had not been visited by him or 
her for some time. 

Although social sniffing may be followed by high-intensity hostile 
behavior as well as (or instead of) copulatory reactions and may be 
accompanied by a variety of vocalizations, such as Trills and Squeaks, 
it is probably purely investigatory in itself. The first part (at least) 
of social sniffing seems to be essentially the same as the sniffing of 
food described above. 


SEXUAL BEHAVIOR AND ASSOCIATED OR 
RELATED PATTERNS 


The term ‘‘sexual behavior” will be used in a broad and general 
sense throughout the following pages to include all behavior patterns 
characteristically directed by adult individuals of one sex toward 
adults of the other sex, or performed by adults of one sex in the 
presence of the other. 

The causal factors responsible for the performance of sexual ac- 
tivities by Night Monkeys are particularly difficult to identify, even 
tentatively, for several reasons. 

In many species of higher vertebrates, all or most sexual activities 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 49 


seem to be produced by two main types of motivation: A pairing 
tendency, which is satisfied by the presence of a mate, and a copulatory 
tendency, which is satisfied by the performance of a successful copula- 
tion (or ejaculation in the case of male). Night Monkeys may have 
the same two tendencies; but this could not be confirmed, during the 
present study, because of the practical difficulty of observing the 
usual long-term sequence of sexual reactions under natural conditions. 

In many other species of New World monkeys, it is possible to 
distinguish between the sexual tendencies and a gregarious tendency, 
which is satisfied by association with other adult individuals of the 
same species irrespective of sex. This cannot be done in the case of 
the Night Monkey, simply because adults of this species do not usually 
associate with any other adults of the same species except their own 
mates. It is conceivable, however, that the complex of factors produc- 
ing the sexual behavior of Night Monkeys includes components 
homologous with the gregarious tendencies as well as the strictly 
sexual tendencies of related species. 


HOOTS 


Both wild and captive adult Night Monkeys on Barro Colorado 
Island uttered Hoots. So did a subadult individual, who seemed to be 
behaving in an essentially adult manner, at Iquitos. Among the 
captive individuals on Barro Colorado Island, Hoots were uttered by 
both males and females. The individual Hoots of all these animals 
sounded similar or even identical (apart from some variation in loud- 
ness) to my ears. 

They were low-pitched, moderately prolonged, and at least moder- 
ately loud. Each individual Hoot sounded to me like a single note. I 
seem to have been misled, however. A sketch of what sounded to me 
like three Hoots is shown in figure 13. It will be seen from this that 
each one of the apparently unitary Hoots was actually compound, 
composed of one or two relatively long sounds, followed by two to 
five short sounds, with only very brief intervals between them. 

In spite of this discrepancy, it will be convenient to refer to each 
hooting pattern that sounded to me like a single note as a “Hoot” or 
“a note.” The briefer sounds making up a Hoot, in this sense, will be 
referred to as “components of a Hoot.” Using these terms, it can 
be said that the intervals between the components of a Hoot were 
always much shorter than the intervals between successive Hoots or 
between a Hoot and any other type of call or note. 

(The Hoots sketched in figure 13 were uttered by an adult female. 
Unfortunately, these were the only Hoots uttered by an adult of 


50 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


which it was possible to make spectrograms. They seem to have been 
somewhat different from the Hoots of a young male sketched in figure 
20. I think that the differences between the Hoots of these two in- 
dividuals were due to the difference in their age, not their sex.) 

The adult Night Monkeys on Barro Colorado Island uttered Hoots 
singly and in short series of two, three, and four notes. Two was the 
most common number. The Iquitos animal apparently always uttered 
Hoots in series of three or four. 

The postures and movements accompanying Hoots were observed 
only in captive individuals. The Barro Colorado adults uttered their 
Hoots while sitting in a perfectly normal unritualized posture, hind 


0.2 0.4 0.6 0.8 1.0 1.2 1.4 1,6 1.8 


Fic. 13.—Three Hoots, uttered by an adult female. 


Based upon a spectrogram by a “Sona-graph.” 
The harmonics accompanying these notes are not shown. 


limbs flexed, back slightly curved (as in all sitting postures), and 
hands firmly grasping the perch. They usually looked more or less 
straight forward throughout their performances. The mouth was 
opened slightly as each Hoot was uttered, and apparently closed be- 
tween the successive notes of a series. The throat and upper breast 
were inflated with each note, and subsided between notes. The Iquitos 
animal sat in a similar posture but looked almost straight downward 
throughout its performances. This may have been a low intensity 
indication of a head-down pattern (the animal had not been in its 
cage very long, and it may have been somewhat shy). A single adult 
female on Barro Colorado looked downward briefly (and perhaps 
lowered her whole head) as she uttered Hoots in otherwise typical 
series, but she always looked up (and perhaps raised her head) be- 
tween each note of a series. 

These captive individuals were always (with one exception—see 
page 51) alone in their cages, or at least the only individuals of their 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 51 


own species in their cages, at the times that they uttered Hoots. Any 
adult that had been Hooting regularly every night always (again with 
the same single exception) stopped Hooting as soon as another Night 
Monkey was put in the same cage. The Hoots stopped irrespective 
of whether the animals fought with one another and/or performed 
copulatory reactions or did neither. (I might add that Hoots were 
not uttered during disputes between individuals in different cages, 
even “long-range” hostile reactions between individuals in cages quite 
far apart from one another.) 

As far as I could tell, all the wild individuals heard to utter Hoots 
were also alone at the time. 

These facts would suggest that the Hoots of adult individuals 
are produced when a gregarious and/or a pairing tendency is thwarted 
by the absence of a suitable companion. 

Although I was never able to observe responses to Hoots in the 
forest on Barro Colorado Island, it seems likely that their primary 
function, when uttered by unmated adults under natural conditions, 
is to attract potential mates. If so, they are essentially similar to cer- 
tain types of bird vocalizations which have been called “song.” (An- 
other resemblance to most “song” is that an individual uttering Hoots, 
i.e., “wanting” a companion, apparently always remains stationary, 
instead of going in active search of a companion. It has already been 
mentioned that the captive individuals sat during their Hoots. Wild 
individuals frequently moved from tree to tree between series of 
Hoots, but they were never heard to move during the utterance itself.) 

The single case of Hooting by an adult in the same cage with an- 
other individual of the same species, cited above, occurred immediately 
after a male and a female had been brought together again after a 
long separation. They were probably mated before, but may have 
become incompatible during the period of separation. The female 
uttered many Hoots during the first few nights after the animals 
were brought together again; but all her Hoots were unusually soft 
and probably low intensity. 

The Hoots of Night Monkeys sound very much like some notes 
of owls. There is, in fact, one species of owl on Barro Colorado 
Island which utters hooting notes that are almost identical (to human 
ears) with those of Night Monkeys (except for the fact that they are 
always or almost always uttered singly). It is possible that notes of 
this type have acoustic properties which are particularly advantageous 
in tropical forests at night. 

It seems probable that the pitch, at least, of almost all the char- 


52 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


acteristic vocal patterns of adult Night Monkeys is a specialized 
adaptation to such conditions. Apart from the Screams (and the 
Squeaks, if they are a “normal” pattern of adults—see page 58), all 
the adult calls and notes are comparatively low in pitch. The vocal 
repertory of adult Night Monkeys is lower, on the average, than the 
adult repertory of any other New World primate with which I am 
familiar, except the howlers of the genus Alouatta. It is certainly 
much lower than the repertory of any other New World monkey of 
comparable size. All other factors being equal, a low-pitched sound 
will carry farther than a high-pitched sound of equal energy value 
at the starting point. In all circumstances, the energy of high-pitched 
sounds will be used up more rapidly than that of low-pitched sounds. 
Another property of low-pitched sounds may be particularly signifi- 
cant in a forested environment. The relatively long sound waves of 
low-frequency noises will be less affected, i.¢., deflected or scattered, 
by obstructions such as twigs and branches than the shorter waves 
of high-pitched sounds. Low-pitched sounds may be particularly 
useful to adult Night Monkeys because they are largely dependent 
upon auditory signals for long-distance communication. 

It must be significant, in this connection, that the high-pitched 
Screams seem to be the only purely hostile vocalizations of adults 
that are always short-range signals. Similarly, the only Squeaks that 
may be a regular part of adult behavior are uttered only when two 
individuals are in actual physical contact with, or in very close prox- 
imity to, one another (see below). 


RUBBING 


The only individuals observed to perform this pattern were the 
captive adults on Barro Colorado Island. 

In the most complete and vigorous Rubbing performances, an in- 
dividual would flex its hind legs, press its perineal region against a 
branch, and then rub the whole region against the branch with rapid 
side-to-side movements. Generally there were two or three move- 
ments in each direction. The tail was usually lifted at the same time, 
and frequently twitched from side to side. The gland at the base of 
the tail may have brushed against the branch during these movements, 
but I was never able to make sure of this. 

Low-intensity Rubbing consisted of nothing more than one or two 
sideways “swipes” at a branch, sometimes very poorly orientated. 

All Rubbing was usually silent. 

Some individuals had favorite Rubbing sites—particular spots 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 53 


where they performed Rubbing much more frequently than elsewhere. 
All or most of these sites were places where there was a slightly 
projecting irregularity on the surface of a branch, or even the stump 
of a projecting twig. 

As an intraspecific reaction, Rubbing was performed almost ex- 
clusively during encounters between males and females, usually en- 
counters in which either one or both individuals also performed 
overt (and often high-intensity) hostile and copulatory patterns. Low- 
intensity Rubbing might be performed by either one or both indi- 
viduals during such encounters; but complete and vigorous Rubbing 
generally was performed only by females. Most male Night Mon- 
keys tend to be slightly dominant over females. It may be significant, 
therefore, that the only male observed to perform high-intensity 
Rubbing during an intraspecific encounter was obviously subordinate 
to the female with which he was associated at the time. He would 
seem to have been playing a feminine role. Although usually sub- 
ordinate, the females that performed high-intensity Rubbing were not 
very frightened. They usually did their Rubbing as they approached 
males, to initiate either overt hostility or precopulatory behavior (or 
both). 

Some very tame individuals performed low-intensity Rubbing 
movements when approached by human beings. A moderately tame 
individual (almost certainly a male), in a cage with its mate and 
young, repeatedly performed vigorous and complete Rubbing after 
being approached by a human being, as the latter walked away. 

These facts would suggest that Rubbing is usually produced by 
some combination of hostile and sexual tendencies. (Some other 
“friendly” social tendency may take the place of sex in some cir- 
cumstances.) Possibly some territorial motivation is also involved. 
(It is not yet clear how the motivation of Rubbing differs from that 
of Moans or Low Trills. There is not enough evidence to permit a 
choice among the several possibilities that come to mind.) 

Rubbing is probably related, phylogenetically, to the “territory 
marking” patterns of many other mammals (which are very similar 
in form). Night Monkeys may spread some odoriferous substance on 
the branches that they rub against. But I have never seen a Night 
Monkey smell or sniff at the Rubbing site of another individual in 
an unusually attentive manner. 

The Rubbing by the individual with its mate and young appeared 
to be a “triumph ceremony,” proclaiming the successful defense of 
its territory, in much the same way as some Resonant Grunt per- 
formances. 


54 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Hill (1960) says that captive Night Monkeys may moisten the 
palms of their hands and the soles of their feet with urine, and 
he compares this habit to the similar pattern of nocturnal Lorisoids 
which has been interpreted as a method of marking territory. I never 
saw anything like this in any of the Night Monkeys that I studied. 
It may be confined to other populations of the species and/or a result 
of special conditions in captivity. 

I might add, in this connection, that I have never seen such behavior 
performed by Tufted Capuchins (Cebus apella) kept in captivity on 
Barro Colorado or in the Zoo at Lima, Peru, although Nolte (1958) 
reports it as common among some individuals of the same species 
kept in captivity in Rio de Janeiro. This sort of behavior may be 
extremely variable among Platyrrhini. 


COPULATIONS, ALLOGROOMING, AND ASSOCIATED PATTERNS 


Like many other New World primates, Night Monkeys seem to 
perform overt copulatory reactions rather infrequently under natural 
conditions. I never saw copulations by the wild individuals on Barro 
Colorado Island. Copulations are also at least relatively rare among 
captive individuals. The only captives observed to perform copulatory 
reactions or closely related patterns were the ones on Barro Colorado 
Island. Even here, the copulatory reactions of mated individuals kept 
together continuously were not only rather infrequent but also rapid, 
brief, and inconspicuous. The only really elaborate and conspicuous 
sequences of copulatory patterns were performed by males and fe- 
males coming together for the first time or rejoining one another after 
being separated and kept in isolation for several weeks or months. 

The most simple forms of apparently successful copulations are 
simple indeed. In such cases, the male and female tend to approach 
one another silently. The male usually sniffs at the female’s genital 
or perineal region, either immediately or after sniffing at her face. 
The female may sniff at his perineal region while he sniffs at hers. 
He then mounts her from the rear, clasps her around the upper part 
of her body with his arms, and maneuvers his body into a position 
from which he can insert his penis into her vagina under her tail (she 
may lift her tail, or shift it sideways, as he does so, but this move- 
ment is always very slight). After insertion, the male usually makes 
three or four pelvic thrusts. Ejaculation apparently occurs only 
during the last thrust, which sometimes is sustained relatively long. 
The whole lower part of the male’s body may quiver visibly during 
this last thrust. The male then dismounts. Sometimes his dismounting 
seems to be accelerated by the female turning her head to look at him. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 55 


Apart from this, the female seems to remain essentially passive 
throughout the whole performance. Both individuals generally are 
silent throughout. The whole performance seldom or never takes more 
than a couple of minutes. Sometimes it is much briefer. There is no 
regular postcopulatory display ; although it is not uncommon to see 
a female sniff at the male’s perineal region, briefly, immediately after 
he dismounts. 

Simple and rapid copulations of this type seem to be performed 
only when both the male and female are highly motivated and 
thoroughly familiar with one another. 

A pattern frequently associated with less simple copulation attempts 
is Allogrooming. One of the individuals grooms the other, or they 
groom one another mutually. When only one individual performs 
Allogrooming, it may be either the male or the female. It is usually 
the male when the Allogrooming occurs in very close association with 
apparently successful copulation attempts. The form of the Allo- 
grooming movements seems to be the same in both sexes. 

One individual wishing to groom another generally begins by 
sniffing. The “groomer” usually sniffs most attentively at the sides, 
armpits, or perineal region of the “groomee.” The actual grooming 
follows immediately after the sniffing. It is accomplished by use of 
both hands and teeth. The use of the teeth seems to be most important 
in Allogrooming (unlike autogrooming). The grooming individual 
nibbles at the fur and/or skin of the individual being groomed, pre- 
sumably separating matted hairs, removing dead hairs and patches of 
dead sloughed skin, and also any small arthropods that may be 
present (both ticks and chiggers are common, at times, on Barro 
Colorado). The hands are used both to “curry” or comb the fur of 
the individual being groomed and, even more frequently, to separate 
the fur so that the grooming individual can get at the base of the 
hairs and the skin with its teeth. During some Allogrooming, the 
groomer uses its hands to push or pull the groomee into a position 
or posture to facilitate the grooming. 

The groomer may groom any and all parts of the groomee, in 
almost any sequence; but it usually devotes most of its attention to 
the sides, back, crown and back of the head. The two animals gen- 
erally sit side by side during Allogrooming, facing in the same or 
opposite directions, and the groomer reaches the groomee from the 
side and/or the rear (see figure 14). 

Allogrooming may lead to mounting. This “grooming mounting” 
may intergrade with the type of mounting that occurs during copula- 
tions, and intermediate performances are common, but in its typical 


VoL. 146 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 


Fic. 14.—Positions and postures sometimes assumed by mated individuals during 


Allogrooming. 


The 


lowering of the front part of the body by the latter is probably a balancing 


the individual on the left. 
reaction, #0f¢ an indication of a head-down posture. 


is grooming 


The individual on the right 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 57 


form it is quite distinct. It occurs when the groomer wants to groom 
areas on the far side of the groomee. In its attempt to get at these 
areas, the groomer may climb up on the back of the groomee. It 
usually climbs up from the side, not from the rear as in the case of 
a male mounting a female during a typical copulation attempt. A 
grooming individual also tends to keep its head pressed down into 
the fur of the groomee. (A male generally does not lower his head 
as much during copulation). The only individuals observed to per- 
form grooming mountings were males; but this may have been largely 
due to the fact that most of the females observed were slightly sub- 
ordinate to their males and somewhat reluctant to handle them 
roughly (there is no obvious reason why females should not be able 
to perform such mountings in suitable circumstances). 

If the individual being groomed does not object to the process 
(and rough or awkward Allogrooming may provoke disputes—see 
below), it remains passive, sitting quietly. Only rarely does it shift 
position voluntarily to present particular areas that it wants to have 
groomed. 

Night Monkeys do not seem to have any special “soliciting” posture 
to invite (initiate) either copulation or Allogrooming (unless sitting 
quietly by the side of another individual can be considered such). 

Some Allogrooming performances are quite silent. Others are ac- 
companied by Squeaks and/or a variety of wholly or partly hostile 
notes (see below). Hostile notes are relatively very rare except during 
the obviously “unsuccessful” performances which break up in open 
disputes. 

Allogrooming may occur immediately before and/or after both 
successful and unsuccessful copulation attempts. It may also occur 
apart from overt copulation attempts, but only relatively rarely. Al- 
most all the individuals observed to perform such grooming were 
known to be in reproductive “mood,” copulating with the individuals 
they groomed on succeeding and/or preceding nights, even if not 
always immediately before or after the actual grooming. 

These facts would suggest that the copulatory tendency may be at 
least one of the causal factors producing Allogrooming in all cir- 
cumstances. 

The detailed relations or interactions between Allogrooming and 
overt copulatory movements are rather complex. In spite of their 
frequent association, and their apparent similarity in motivation, the 
two types of behavior seem to be mutually exclusive, to some extent, 
on a short-term basis. Allogrooming appears to serve as a partial 
“substitute” for copulation. It is very common to see a male groom 


58 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


a female briefly before mounting and performing an apparently 
successful copulation; but he will almost certainly not copulate suc- 
cessfully, or at least will not copulate successfully immediately after 
the Allogrooming, if he grooms her vigorously for more than a few 
seconds. 

This would suggest that the performance of prolonged Allogroom- 
ing tends to “consume” an appreciable amount of copulatory motiva- 
tion, 1.e., raises the releasing threshold of overt copulatory behavior 
in at least many circumstances. 

Allogrooming may have several functions. Its successful per- 
formance may help to reduce the actual or potential hostility between 
male and female. The close physical contact involved may also pro- 
vide some direct sexual stimulation. 

Either one or both of these effects may be very rapid. They may 
be more rapid but less long sustained and/or have less potential 
strength than the effect of raising the threshold for copulatory be- 
havior. If there are such differences in speed, and possibly strength, 
between the contradictory effects of Allogrooming, this would ex- 
plain why successful copulations frequently follow brief grooming 
and rarely follow prolonged grooming. 

Repeated Allogrooming may also have the relatively very long-term 
effect of strengthening the pair bonds between mates. 

If the Allogrooming of Night Monkeys does function in these 
different ways, it is basically similar to many “precopulatory” patterns 
of other species, e.g., the Allogrooming of many other mammals and 
the Food-begging and Allopreening of many birds. 

The penis of a male Night Monkey may be erected before he mounts 
during copulation, but it is not “shown off” to the female. This 
behavior pattern does not seem to be a ritualized display like the 
penile erection of Squirrel Monkeys, Saimiri, described by Ploog 
and MacLean (1963). 

The only other patterns which must be considered in connection 
with the sexual behavior of Night Monkeys are Squeak notes. These 
seem to be typical of young animals and infants, but are uttered also 
by adults in certain circumstances. 

Many of the Squeaks uttered by adults may be reversions to, or 
persistent remnants of, juvenile behavior ; and may be produced by 
some or all of the same wide range of causes as the Squeaks of 
juveniles (see below). Squeaks are certainly uttered more frequently 
by adult individuals that have been raised in complete or partial isola- 
tion from other individuals of their own species than by adults raised 
in the wild under natural conditions; and the retention of juvenile 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 59 


patterns by adults is the sort of phenomenon which might be expected 
to be produced by the abnormal circumstances of captivity. (lf 
nothing else, young animals raised in captivity tend to regard the 
human beings that take care of them as their parents, and often 
retain this attitude after reaching maturity, long after they would 
have become independent of, or separated from, their parents in the 
wild.) 

Other Squeaks uttered by adults may have a different significance. 
Many of the captive adults on Barro Colorado Island uttered frequent 
Squeaks during sexual reactions. This occurred often enough to 
suggest that Squeaks may be a normal part of the adult repertory, a 
normal accompaniment of sexual behavior even if not of any other 
adult activity. 

Squeaks were uttered by both adult males and females. They 
were uttered most frequently by individuals approaching their mates, 
and just beginning to perform Allogrooming, but were relatively much 
less frequent during the later stages (when Allogrooming was in full 
swing and most vigorous). Squeaks also were relatively rare during 
copulations and overt disputes between mates. 

As far as I could tell by ear, all the Squeaks uttered by adults were 
similar to, or identical with, the “pure” Squeaks of juveniles in form. 

The function(s) of adult Squeaks in sexual situations were obscure. 
They did not seem to produce any overt, positive, response in the in- 
dividuals toward which they were directed. They may, conceivably, 
have functioned as some sort of appeasement pattern. One might 
expect that any behavior pattern that is performed most frequently 
by infants and juveniles would tend to reduce hostility (especially 
aggressiveness) in any adult toward which it is directed, even when 
it is performed by another adult. 


EXAMPLES 


Some of the different sequences and combinations of Allogrooming, 
copulatory, hostile, and other patterns during different types of en- 
counters between males and females may be illustrated by extracts 
from my notes made at the times of observation. 

The first is an example of the behavior of experienced males and 
females. 


February 4, 1962. Z male and X female. These two animals are mates. They 
were separated several days ago. The male has been kept in the cage in which 
the pair lived previously. The female has been kept in another cage (out of 
sight and hearing of the male) by herself. 

7:40 p.m. Female put back in cage with male. Male is in sleeping box at the 


60 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


time. Female rushes straight toward him. He comes out of box. The two ani- 
mals sniff at one another’s perineal regions. Then the male mounts the female, 
from the rear. No hesitation or awkwardness. Copulation. Apparently suc- 
cessful. Lasts at least 20 seconds. Then the male dismounts. No postcopulatory 
display. Male walks away, female follows. She utters two Low Trills as she 
does so. 

7:44. The two animals sniff one another’s perineal regions again. And again. 
Silent. No Allogrooming or attempts to mount. Then sit side by side. One or 
both utter(s) a few Squeaks. Then male does begin to groom female. From 
side. Then mounts her again, from rear. One or both animal(s) utter(s) 
Squeaks as he mounts. Copulation. Female resists, but not violently. She just 
turns around to look at male. He dismounts, walks away. Female immediately 
assumes Arch Posture—and utters a series of Gulps! Then relaxes. Follows 
male around cage, 7:50. Then both animals retire to sleeping box .. . 

7:56. Male is grooming female in box. This is accompanied by lots of sneez- 
ing by one or both animals .. . 

8:02. Both animals come out of box. Sit side by side, facing in opposite 
directions. Female utters a few Squeaks. Then male begins to groom her. 
His movements are gentle. He devotes most of his attention to her lower back 
and rear. Female continues to utter Squeaks. Goes quite limp while male 
grooms. Possibly half asleep? Then she suddenly utters loud Squeaks, moves 
away from male. Goes into box. Male follows, continues Allogrooming in 
DOK us 

8:17. Female apparently startled or annoyed by something. Bounces out of 
box, crouches in far corner of cage. Utters long series of Gulps. Male comes 
out of box, runs toward female, utters one Sneeze-grunt as he does so. Female 
runs back to box. Performs one burst of Rubbing movements in midst of run. 
Back into box. Male follows, resumes Allogrooming again .. . 


The following extract illustrates the behavior of an inexperienced 
male. 


January 28, 1962. O male and C female. This male has been hand reared, 
in partial isolation, ever since the age of approximately 14 months. Now full 
grown. Since being taken from his parents, he has had only brief contacts with 
other Night Monkeys. He is not familiar with C female. 

7:00 p.m. Female let loose in male’s cage. Male leaps away, freezes and 
crouches in far corner. Female explores cage, uttering Low Trills as she does so. 
Approaches male. The two animals sniff one another, nose to nose. Male utters 
Squeaks. Female moves away, resumes exploring with Low Trills. Approaches 
male again. He sniffs at her perineal region and utters more Squeaks. She 
moves away again. Explores with Low Trills. Approaches male again. He 
utters more Squeaks. She moves away again... 

7:06. Female approaches male. He suddenly clasps her back, from the side, 
but does not mount. She breaks away and resumes exploring. Then approaches 
male again. A vicious fight breaks out. Male apparently the initiator, striking 
at the female with his hands. Lots of running, chasing, and striking during 
fight. Both animals quite silent throughout. Then they separate. Male retires 
to corner and assumes Arch Posture. Female walks around, not very rapidly, 
uttering many Gulps and a few Sneeze-grunts... 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 61 


7:18. Male starts to approach female. She assumes an Arch Posture, starts 
to advance toward him. He retreats. . 

This sort of inconclusive behavior continues until 9:10, when I decide to stop 
observations. Then, happening to pass the cage at 9:50, I find the animals 
copulating! Resume observations. 

Copulation apparently successful. Male dismounts. Then mounts again imme- 
diately. No preliminary display by either animal. Copulation again apparently 
successful. Pelvic thrusts at intervals of at least several seconds. Then male 
dismounts. No postcopulatory display. Male walks away a couple of feet. 

Back to female almost immediately. Approaches her from side. Sniffs at 
her side and utters Squeaks. Climbs up onto her back, mounting from side. 
Once mounted, he seems to be trying to shift around into a copulatory position. 
Gets half way there, but female turns around to look at him, and he slides off. 
Then walks away again... 

10:01. Male approaches female again. From side. Sniffs at her side and 
utters Squeaks. Half mounts, from the side. Then he begins to groom the 
female, while he remains sprawled sideways across her back. His grooming 
movements are extremely vigorous, even violent, and very awkward looking. 
Female just sits. Male eventually slides off. Remains sitting beside female, one 
hand resting on her back, and utters many Squeaks at irregular intervals. Fe- 
male eventually walks away. Male follows... 

10:10. Male grooms female steadily for several minutes. Reaching her from 
side. He repeatedly turns her half way over on her back, apparently trying to 
get at something under her arm on the far side. Both animals silent. Finally 
the female breaks away. Male follows her. Uttering Squeaks. Then mounts 
her from the rear. Tries to copulate. Copulation attempt probably unsuccessful. 
Only one or two pelvic thrusts. Both animals silent. Then male slips off ... 

10:19. Male approaches female again. Silently. Mounts her immediately, 
from rear. Copulation. Apparently successful. Long sustained, with many pelvic 
thrusts. Male finally dismounts. Sits beside female. She sniffs at his flank 
region. He immediately starts to groom her. Grooming lasts only a few seconds. 
Then male mounts again, from the rear. Copulation. Apparently successful. 
Many pelvic thrusts; the last one very long sustained and accompanied by quiver- 
ing of the rear part of the body. Male slips off. Immediately starts to groom 
female again. This Allogrooming is much more prolonged and violent than the 
Allogrooming immediately before the last copulation. 

10:22. Male stops Allogrooming momentarily. Then starts again. Grooming 
movements even more violent than before. Again and again he twists the female 
over on her back. No attempt to mount. Female silent; male utters occasional 
Squeaks. Then the male stops Allogrooming and walks away. 

He goes back to the female almost immediately. Mounts her from the front, 
climbing over her head. Lies on her back, his head over her tail, his tail hang- 
ing in her face! He slips off after a few seconds, and the two animals sepa- 
rates. 

This sort of behavior, including many copulation attempts (all apparently 
unsuccessful) and much violent Allogrooming by the male, continues spas- 
modically as long as I continue to watch the animals, until 11:00 p.m. 


During this encounter, it was obvious that the male had a strong 
copulatory tendency, but he seemed to have difficulty in orientating 


62 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


his copulatory movements, and he was certainly somewhat hostile to 
the female at first. (His awkwardness persisted for at least 5 nights. 
But shortly afterwards he was observed to perform sexual reactions 
quite smoothly.) 

The last extract illustrates the behavior of pairs when the female 
is dominant over the male. 


January 24, 1962. P female and Z male. These animals may have had some 
previous sexual experience; but they are not acquainted with one another. They 
have been kept isolated in separate cages for several days. 

6:45 p.m. The male is let loose in the female’s cage. The female rushes 
to him. They sniff one another, nose to nose. The female utters one Low Trill 
during sniffing. Then they separate. The male runs around the cage, possibly 
frightened. The female moves around more slowly. Neither animal is following 
the other. One or both always assume(s) an Arch Posture when they come 
close to one another. 

6:50. The two animals settle down to a routine, which does not change 
very much for almost three hours. 

The male retires to a corner of the cage, and sits quietly in a crouching freeze 
posture whenever he is left alone by the female. Sometimes his freeze includes 
a trace of a head-down component. After an hour or so, when the female 
starts to become less active (see below), he occasionally closes his eyes, briefly, 
while continuing to sit in the same posture most of the time. 

The female approaches the male repeatedly. When she approaches him slowly, 
or at a normal running pace, she usually sniffs him. She usually begins by sniffing 
at his face, and then sniffs his side and/or his perineal region. When she sniffs 
at his side, she often appears to be trying to smell under his arm or in his 
armpit. Sometimes she seizes his hands, arms, or head, in order to pull him 
into the right position to facilitate sniffing the area she is interested in at the 
time. Sometimes she grooms the male immediately after sniffing. This occurs 
first during her third approach. It becomes increasingly frequent as the evening 
wears on. 

Sometimes the male moves away after the female begins to sniff at him. 
Sometimes he just sits still. Most frequently, he sniffs in return. In most cases, 
he sniffs more briefly than she does. He often appears to be unwilling to sniff, 
but unable to resist the stimulus provided by the female. He usually sniffs her 
perineal region after sniffing her face (the animals are then nose to tail). 

Most of this sniffing and Allogrooming is accompanied by Squeaks by the 
female. One or both animals also utter(s) a few Low Trills and many Moans. 
The Moans are uttered most frequently around 8:50-9:00 p.m., presumably when 
the animals are becoming accustomed to one another. 

Sometimes the Allogrooming develops into a dispute. When the female 
grooms too energetically. The male strikes at her with his hands. Silently. 
When the male does this, the female always responds by “bombarding” him. 
She runs away from him, uttering many Gulps and a few Sneeze-grunts, turns, 
faces him, pauses a second, peforms a few Rubbing movements, and then leaps 
at him. Usually she leaps just above him, landing on the side of the cage an 
inch or less above his head. Her tail usually slaps into his face as she lands. 
Occasionally she crashes into him full force with her whole body. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 63 


She usually performs several leaps one right after the other, before resuming 
slower approaches and Allogrooming... 

The male does not respond to the female’s leaps as strongly as might be 
expected. Usually he just sits quietly throughout. Occasionally (especially dur- 
ing the first hour) he assumes an extreme Arch Posture just as the female 
starts to leap. Once or twice he strikes at her, but he never chases her .. . 

8:45. The male starts to relax a little. Occasionally starts to follow the 
female when she goes down to the floor of the cage to get food... 

9:45. The two animals are sitting side by side. Then the female goes down 
to the floor, picks up a piece of fruit, climbs up to a branch, about three feet 
from the male, and starts to eat. The male walks toward her slowly. Sniffs at 
her perineal region. Then suddenly mounts her, from the rear. Slightly awk- 
ward. The female does not lift her tail, so the male has to shift his body in 
order to insert his penis from the side. Copulation. Apparently successful. 
Several pelvic thrusts at irregular intervals. The female leans downward and 
forward during the copulation, still clutching food in her hand. Then the male 
dismounts. Sits beside female, facing in the same direction. Then turns around, 
sniffs at her perineal region again, mounts her again, and copulates again. Again 
apparently successful. Then he dismounts and sits beside the female as before. 
She calmly finishes eating her food, turns around, and sniffs his perineal region. 
Then both animals sit quietly ... 


BEHAVIOR OF YOUNG ANIMALS 


Three baby Night Monkeys were born, to different parents, in 
captivity on Barro Colorado Island. One of these was left with its 
parents, but died when approximately 2 months old. The second 
was taken from its parents at the age of 14 months, and raised by 
hand. The third was born only 9 weeks before this paper was written, 
and was left with its parents. Several other infants were bought in 
the Panama market, and hand raised successfully in captivity. Many 
of the other individuals kept in captivity on Barro Colorado Island 
were secured when they were still subadult. A few subadult in- 
dividuals were observed in the wild on Barro Colorado. 

The infants born in captivity on Barro Colorado Island were not 
watched very much while they remained with their parents, as it was 
feared that prolonged observation might disturb the parents unduly 
and cause them to abandon or neglect the young. It was only possible to 
note a few aspects of the parent-infant relationship. 

For the first few days after birth, the infant is always carried by 
the mother, clinging to her breast or belly. Later it usually rides on 
the back of a parent. 

Male Night Monkeys seem to play an important role in raising 
their offspring. Before they were 3 weeks old, two of the infants 
born in captivity on Barro Colorado Island were being carried by 


64 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


their fathers most of the time when they were not actually being 
suckled by their mothers. The third infant was certainly carried 
by its father much of the time, but I could not determine exactly how 
much, simply because the father and mother were very difficult to tell 
apart (without catching them in the hand). This was the infant that 
was observed most frequently. It was first seen being carried by the 
father at the age of 9 days. This was also the first time that it was 
seen riding on the back of a parent. (It seems probable that these two 
developments were causally related. But carrying on the back is not 
a purely masculine pattern. The same infant sometimes rode on its 
mother’s back before and/or after being suckled.) I was informed 
by a keeper at the National Zoological Park that a baby Night Monkey 
born to a captive pair there, and raised by its parents, also usually 
was carried by the father. 

This parental behavior is interesting from a comparative point 
of view. Infants and young apparently are always carried by their 
mothers, as long as they are carried at all, in such genera of New 
World primates as Alouatta and Ateles (Carpenter, 1934 and 1935). 
The only other platyrrhines in which the father is known to carry 
the young regularly are the marmosets and tamarins (Sanderson, 
op. cit., and Hill, 1957). 

There is at least one published record of a male Titi Monkey, Calli- 
cebus, carrying young (L. E. Miller, quoted in Allen, 1916) ; but it 
is not known if the habit is regular in the species. 

Hill’s (1960) statement that Allen (in the same paper) cited evi- 
dence to show that male Saimiri frequently carry young is appar- 
ently erroneous. 

The carrying of young by males may be primitive among Platyr- 
rhine and/or an indication of close phylogenetic relationship between 
Night Monkeys and marmosets and tamarins. (Night Monkeys re- 
semble marmosets and tamarins in a number of other characteristics. ) 
In any case, such behavior must be adaptive. Its principal advantage 
may be the sharing of labor. When the male carries the young 
part of the time, neither parent will become as exhausted and “run- 
down” (and therefore susceptible to predators and disease) as the 
female would be likely to become if she did all the work herself. (Even 
if the male has the harder job because he carries the young most of 
the time, he can support the strain better than his mate simply because 
he does not have to produce milk. Males may get less tired also because 
they are usually, or perhaps always, slightly larger and more powerful 
than their mates.) 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 65 


In all probability, the advantages of such behavior can be secured 
only when the sex ratio is effectively one to one and the pair-bonds 
between mates are close and long sustained. Species of Alouatta and 
Ateles probably cannot afford such behavior because they usually live 
in groups in which there are more females than males. There may 
occasionally be more infants than adult males in such groups. The 
looseness and brevity of pair-bonds among howlers must also tend 
to prevent the establishment of continuous bonds between males and 
infants. Infants must be fed by their mothers at more or less fre- 
quent intervals, and regular transfers from one parent to the other 
might be difficult to arrange when the parents do not stick together. 

Thus, the type of parental behavior shown by Night Monkeys in 
captivity would seem to provide strong support for the hypothesis 
that males and females of this species usually form pair-bonds which 
are at least moderately long sustained under natural conditions (see 
page 12). 

Like all baby monkeys, infant Night Monkeys have very strong 
clutching tendencies and try to keep in close physical contact with a 
parent at almost all times. The only exceptions occur when an infant 
moves from one parent to another before and after feeding (see next 
paragraph). When clutching a parent, an infant usually uses both 
hands and feet to hold on to the parent’s fur, and may also keep its 
tail pressed against the parent’s body. The infant born in captivity 
that was observed most frequently was very inactive, aside from feed- 
ing, during the first few days of its life. By the time it was 3 weeks 
old, however, it sometimes moved around most energetically on its 
parent’s back. 

My observations would suggest that an infant usually shifts from 
one parent to the other, at feeding time, under its own power. The 
parents do not lend a hand to lift or support the infant during the 
transfer unless it should start to fall. 

The infants raised in captivity apart from their parents quickly 
learned to accept human beings as foster parents. They appeared to 
be almost perfectly satisfied to be carried by a human being, espe- 
cially if they could ride on the head and clutch hair. Somewhat 
more slowly and reluctantly, they also learned to accept turkish towels 
as parent substitutes. An infant conditioned in this way usually spent 
all its time clutching its towel (unless it had to leave the towel to 
feed or drink). 

Except when specifically stated otherwise, the behavior patterns of 
infants and juveniles described hereafter in this section were per- 
formed by young animals hand reared in captivity, after being sepa- 


66 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


rated from their parents. All these individuals were at least 14 months 
old. 

A perfectly content infant, with all its desires satisfied, is silent 
and motionless (and usually asleep). Sounds and movements are 
apparently always indications of dissatisfaction. When an infant is 
slightly dissatisfied, it may begin either movements or sounds first, or 
both together. (It is my impression that an infant is most likely to 
move without vocalizing when it “knows” in advance that it can re- 
move its dissatisfaction by its own efforts.) The first movements may 
be nothing more than slight readjustments in the position of the limbs 
or other parts of the body. At higher intensities of dissatisfaction, the 
infant usually goes straight to whatever individual or object it ex- 
pects will be able to satisfy its need (e.g., a food dish when it is 
hungry) or, if this is not possible, makes vigorous searching move- 
ments. 

The hand-reared infants apparently never moved about just for the 
sake of moving. They did not seem to have any trace of the independ- 
ent locomotory tendency of adults. All their movements appeared to 
have an immediate “practical” end in view apart from the mere per- 
formance of the movements. 

By far the most common vocal patterns of infant and young juve- 
nile Night Monkeys are Squeaks and Squeaklike notes. A typical 
“pure” Squeak is brief, not very loud, and much higher in pitch than 
any of the Grunt Complex patterns. Sketches of sound spectrograms 
of pure Squeaks are shown in figures 15, 17, and 18. Each pattern of 
this type sounds like a single note to human ears. The sound spectro- 
grams, however, indicate that at least many of the pure Squeaks are 
actually partly or wholly compound, composed of two or three sounds 
separated by very short intervals or two or three partly distinct 
“syllables.” It will be convenient to refer to such patterns in terms 
similar to those applied to the equally compound Hoots described 
above. A pattern that sounded to me like a single note will be called 
“a Squeak” or “a note.” The various sounds which spectrograms 
may show to be included in a single Squeak, in this sense, will be 
called “components of a Squeak.” 

All the accompanying sketches of vocal patterns by an immature 
individual are based upon recordings of a single male who was exactly 
45 days old at the time of recording. 

The pitch of a single Squeak seldom or never remains level through- 
out. In most cases, the pitch first rises and then falls. When a Squeak 
is composed of two more or less distinct components, the rise usu- 
ally extends throughout the first component and the fall throughout 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 67 


the second component. The rate of rise or fall may vary within the 
same component. When a Squeak is composed of three more or less 
distinct components, the first usually rises, the second usually rises 
or remains approximately level, and the third usually falls. More 
rarely, the pitch may rise or fall more or less steadily throughout a 
whole Squeak, or may first fall and then rise. 


0.2 0.4 0.6 0.8 1.0 1.2 14 1.6 


Fic. 15.—Four typical “pure” Squeaks, uttered by an immature. 


Based upon a spectrogram by a “Sona-graph.” 


Some Squeaks are much louder than usual without being prolonged. 
Patterns of this type tend to extend over a particularly wide range 
of pitch (see figure 16). 

In some Squeaks with both rising and falling pitch the rising part 
is loudest ; in others the falling part is loudest. 

The Squeaks of infants and young juveniles intergrade with both 
Screams and High Trills. 

The “pure” Screams of infants and young juveniles sound similar 
to adult Screams, but are slightly softer and perhaps slightly plain- 
tive in tone. The most extreme Screams of the young animals that I 
studied sounded as if they were approximately as long as the adult 
Screams shown in figure 5, and only slightly higher in pitch on the 
average. It is possible that infants and juveniles never utter Screams 
of as long duration as the adult patterns shown in figure 6; but I 


68 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


never subjected any young individual to the stresses which would 
be expected to produce the highest intensity type of Screaming. 


11,000 

10,000 
9,000 
8,000" 
7,000 
6,000 
5,000 
4,000 
3,000 
2,009 


1,000 


0.2 0.4 0.6 0.8 1.0 


Fic. 16.—A single very loud Squeak, uttered by an immature. 


Based upon a spectrogram by a “Sona-graph.” 
There were at least three more harmonics with this note that are not shown 
in the drawing. 


Patterns intermediate between pure Squeaks and pure Screams are 
variable, but always more or less intermediate in length and sometimes 
slightly higher in pitch, on the average, than the lowest Squeaks. They 
also tend to fluctuate, rise and fall slightly in pitch, in much the 
same way as pure Screams, but the number of fluctuations is less 
than in the latter. Sometimes an intermediate pattern retains a trace 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 69 


of the slight (inaudible to human ears) division into separate com- 
ponents which is characteristic of many Squeaks. Figure 17 includes 
a pattern of this type. The intergradation between pure Squeaks and 
pure Screams seems to be complete, insofar as infants and young 
juveniles utter every possible type of morphologically intermediate 
pattern. 


10,800 r S 
= ae i. 
> am 

6,000 z 
a 
— 


PBS 


a 


0.2 0.4 
Fic. 17—A “pure” Squeak, followed by a longer note more or less perfectly 
intermediate between a Squeak and a short Scream. Uttered by an immature. 


Based upon a spectrogram by a “Vibralyzer.” 
Compare with the Scream shown in figure 5 and the Triils shown in figure 19. 


“Pure” High Trills are conspicuously compound. To human ears, 
each High Trill sounds as if it were composed of three or four 
Squeaks uttered in rapid succession. Most sound spectrograms convey 
the same impression. A sketch of a more or less typical pure High 
Trill is shown in figure 18. The successive notes in a single High 
Trill are seldom exactly the same pitch. In most cases, the first one 
or two notes are largely or completely rising and the last one or two 
notes are largely or completely falling. 

A remarkable feature of the vocal repertory of young Night Mon- 
keys is that the arrangement of components within a single Squeak, 
1.€., the sequence of changes in pitch and the relative distance be- 
tween high points and low points is sometimes similar to the arrange- 
ment of successive Squeaks within a single High Trill. Thus, the 


7O SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


form of a pure High Trill may be essentially similar to that of a pure 
Squeak, only on a much longer time scale. The sequence of rises and 
falls in pitch within a single short Scream may be equally similar to 
the arrangement of the corresponding features within both a single 
Squeak and a single High Trill. 


0.2 0.4 0.6 0.8 1.0 1.2 1.4 
Fic. 18.—A series of four notes uttered by an immature. 


Based upon a spectrogram by a “Sona-graph.” 
This series sounded, to me, like one “pure” Squeak followed immediately by 
a short High Trill. As a whole, it is comparable to the series shown in figure 15, 
which sounded as if it were composed of four separate Squeaks. 


Therefore, although most High Trills appear to be, or sound as if 
they were, accelerated series of several Squeaks, the same morpho- 
logical effect might be produced by slowing down a single Squeak in 
such a way that the individual components within it become more dis- 
tinct and widely separated from one another. Similarly, the effect of 
a short Scream could be produced either by slowing down and length- 
ening a single Squeak, while maintaining or strengthening the con- 
nections between its components, or by letting a series of several 
Squeaks “run together.” In other words, many or most High Trills 
could be interpreted as discontinuous series of either Squeaks or com- 
ponents of Squeaks, while many or most short Screams could be 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 71 


interpreted as continuous series of either Squeaks or components of 
Squeaks. (These relationships are difficult to describe or explain 
verbally, but I think that they will become clear if the accompanying 
drawings of sound spectrograms are studied in detail.) 

In view of these facts, it is not surprising that the patterns inter- 
mediate between pure Squeaks and pure High Trills appear to be 
somewhat heterogeneous. Some intermediates are moderately rapid 
series of a few obviously distinct notes, apparently series of Squeaks 
which are not accelerated as much as the components of pure High 
Trills. Others are brief patterns which sound rather like single 
Squeaks with faint “rattling undertones.” These may be Squeaks in 
which the internal components have become more distinct from one 
another than in typical pure Squeaks. The two intermediate types 
intergrade with one another. It is my impression, in fact, that the 
intergradation between pure Squeaks and pure High Trills is as 
complete as the intergradation between Squeaks and Screams. 

Infant and juvenile Night Monkeys also utter many patterns that 
seem to be intermediate between Screams and High Trills. These 
are all more or less prolonged, discontinuous, and wavering. Figure 
19 is a sketch of two patterns of this type (somewhat nearer to pure 
High Trills than to pure Screams). At least equally common are 
patterns that appear to be intermediate between High Trills, Screams, 
and Squeaks. These are similar to the intermediates between High 
Trills and Screams but shorter. 

It is evident, therefore, that the patterns which have been called 
pure Squeaks, pure Screams, and pure High Trills in the preceding 
discussion are merely the extreme points of a continuum. This whole 
group of patterns may be called the “Squeak Complex.” 

It seems probable that all the vocal patterns of both adult and young 
Night Monkeys can be included in either this complex or the Grunt 
Complex. 

The mouth is opened to a moderate extent during all or most notes 
of the Squeak Complex (see figure 22). 

Infants raised apart from their natural parents utter Squeaks and 
intermediate notes closely similar to pure Squeaks very frequently 
whenever they are not clutching a foster parent or parent substitute 
(if they are not “distracted” by food or drink). A tame infant which 
has been silent while being carried by a human being will always begin 
to utter Squeaks (with or without other notes—see below) as soon 
as it is lifted away. It may also start to move in an obvious attempt 
to follow and rejoin the human being. Its Squeaks may become louder 
and more rapid if the human being then disappears from sight. 


72 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


These facts would suggest that many or most of the Squeaks 
uttered by infants are produced when their tendency to keep in physi- 
cal contact with a parent is thwarted. 

This may not, however, be true of all their Squeaks. Infants 
clutching a foster parent or parent substitute may also utter Squeaks 
just before shifting the position of their limbs of their own accord, 
i.e., before they can have lost contact. Once, an infant which had 


0.2 04 0.6 0.8 1.0 1.2 14 1.6 1.8 


Fic. 19—Two loud trilling patterns, intermediate between “pure” High Trills 
and Screams, but most similar to the former. Uttered by an immature. 


Based upon a spectrogram by a “Sona-graph.” 

Each trilling pattern is a series of four notes or groups of components. Each 
of these series seems to be similar to the series shown in figures 15 and 18, and 
to the components of the single Squeak and the intermediate between a Squeak 
and a Scream shown in figure 17. 

Additional harmonics, up to at least 18,000 c.p.s., accompanied these trilling 
patterns, but are not shown in the drawing. 


been quite silent while riding on my head suddenly began to utter 
Squeaks when rain began to fall, in spite of the fact that it continued 
to clutch me tightly. 

Such incidents would suggest that infants may utter Squeaks when- 
ever they become “uncomfortable” or feel “frustrated” in any one of 
several different ways. If so, their Squeaks can be considered a gen- 
eralized distress reaction, strictly comparable to the “distress calls” 
of many young birds. 

Of all the other platyrrhines I know, ony infant tamarins of the 
genus Saguinus utter distress notes as frequently as infant Night 
Monkeys in similar circumstances. 

Under natural conditions, it seems likely that the parents of an 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 73 


infant Night Monkey would respond to its Squeaks by trying to make 
it more comfortable, e.g., by helping it to readjust its position or by 
feeding it. 

Unfortunately, this could not be checked by observation of the 
infants raised by their own parents in captivity. The parents of both 
infants seemed to be attentive and conscientious, and the infants were 
silent most of the time, presumably because they were seldom suffi- 
ciently uncomfortable or thwarted in the right way to induce vocaliza- 
tions. I did hear them utter a few Squeaks, quite like those of the 
hand-reared infants. They uttered these notes in a variety of circum- 
stances, e.g., while moving around on a parent’s back, while trying to 
suckle, and (once) after falling off a parent’s back. In most cases, 
the infants stopped vocalizing almost immediately, before the parents 
reacted. They apparently managed to achieve satisfaction by their 
own efforts. The infant that fell was retrieved by a parent, but I 
could not determine if this parental act was a response to the infant’s 
Squeaks and/or to the scrambling movements that it made at the 
same time. 

The fact that high-pitched notes will not carry as far as low-pitched 
notes has already been mentioned. It seems highly probable that the 
Squeaks of infants, like those of adults, are primarily short-range 
signals. Under natural conditions, they are probably almost always 
uttered by infants at least fairly close to their parents. 

It is my impression that the Squeaks of both adults and young are 
slightly “ventriloquial.” In the dark, I found it difficult to tell exactly 
where Squeaks were coming from. Their source was more difficult 
to locate than that of any other vocal pattern of the species (with 
the possible exception of some High Trills). 

These features may be particularly advantageous because individ- 
uals uttering Squeaks may be so intent upon the activity in which 
they are engaged, or so distressed, that they may become less alert 
than usual to outside stimuli and fail to note the approach of a pos- 
sible predator. 

Infants may utter Squeaks (and/or closely related short interme- 
diate notes) singly or in unaccelerated series (1.e., series that are not 
at all Trill-like) of up to seven or eight notes. Longer series presum- 
ably are produced by greater distress than shorter series. Series are 
frequently repeated with only brief intervening pauses. 

Infants also utter many series of notes which include both brief 
Squeaks or Squeaklike notes and longer Screams or Screamlike notes. 
The arrangement of notes in such series is quite variable. One com- 
mon arrangement is three or four long notes followed immediately 


74 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


by two or (less frequently) three short notes. Another common 
arrangement is four or five short notes followed immediately by a 
single long note. Other and more or less intermediate arrangements 
also occur, but seem to be somewhat less common. 

Infants may “settle” on one particular type of mixed series, and 
repeat it without variation in form for appreciable periods of time, 
even when there is reason to believe that the strength of their motiva- 
tion should be changing slightly. They sometimes seem to become 
“stuck in the same groove.” This would suggest that the utterance of 
one type of series may facilitate repeat performances of the same 
series, but does not facilitate, and may even discourage, subsequent 
performances of different types of mixed series composed of the 
same notes in different sequence. 

Some of the Screams and intermediates between Screams and 
Squeaks uttered by infants may contain a hostile component like the 
Screams of adults. The infant born and hand raised in captivity on 
Barro Colorado Island uttered many Screams and related intermedi- 
ate notes during the first few days after being taken from its parents, 
a period during which it also performed an appreciable number of 
overtly and unmistakably hostile reactions. Infants also tend to utter 
Screams when handled somewhat roughly. But other Screams and 
intermediate notes are almost certainly not hostile. It was very com- 
mon, for instance, to hear a captive infant which had accepted human 
beings as foster parents utter many notes of this type in immediate 
association with both Squeaks and Hoots (see page 75) when left 
alone. These notes were not accompanied by any trace of overtly 
hostile movements or Gruff Grunts, and both the Squeaks and the 
Hoots were certainly provoked by the thwarting of the infant’s de- 
sire to be with its foster parents. It seems likely, therefore, that 
at least many of the patterns intermediate between Squeaks and 
Screams are similar to the former in being generalized distress reac- 
tions. This, and the complete intergradation between the two extreme 
types of patterns, would suggest that the Screams of infants are little 
or nothing more than the highest intensity form of their Squeaks. 
(It would not, in fact, be necessary to give the patterns different 
names if they were not so distinct, in both form and causation, when 
uttered by adults). 

The captive infants uttered High Trills when approaching their 
food dishes and feeding, and also when I lifted them up and brought 
their faces close to mine. This would suggest that their High Trills 
were produced by the same motivation as some or all of the Trills of 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 75 


adults and/or by the same factors as their own Squeaks plus an 
added component of alarm or escape. 

Some of the brief Squeaks or Squeaklike patterns of infants have 
a slight “gulping” quality. They may be the source from which the 
Gulps of adults are derived in the course of ontogeny ; but they cer- 
tainly are not distinctive or well segregated when uttered by infants. 

Infants utter Hoots occasionally. To human ears, these sound very 
much like slightly softer versions of the usual Hoots of adults; but 
sound spectrograms indicate that they are sometimes (at least) also 
higher pitched and more broken up (see figure 20). They are usually 


6,000 


Fic. 20.—Two Hoots, uttered by an immature male. 


Based upon a spectrogram by a “Sona-graph.” 


uttered in series of two or three notes, apparently always when an in- 
fant has become separated from its parents or parent substitute. Like 
adults, infants usually or always sit still while uttering Hoots. Their 
Hoots seem to be high-intensity patterns, produced when their desire 
to be with a parent is stronger, or more strongly thwarted, than when 
Squeaks or related intermediate notes are uttered. Thus, for instance, 
an infant suddenly separated from its foster parent or parent substi- 
tute may utter Hoots, intermediates between Squeaks and Screams, 
and more or less pure Squeaks at first, then stop uttering Hoots but 
continue the other notes, then stop the Screamlike notes but continue 
Squeaks, and finally fall silent, as it gradually becomes accustomed 
to being alone. Similarly, an infant accustomed to being carried 
almost steadily by a foster parent is apt to utter relatively more Hoots 
and fewer Squeaks than an infant used to being carried only occa- 


76 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


sionally, when both are separated from their foster parents in similar 
circumstances and with the same degree of abruptness. 

(It is interesting that the vocal patterns uttered by infants who want 
to attract or join their parents are similar to those of adults who want 
to attract or copulate with their mates. This might suggest that the 
sexual tendencies of adults develop from the infantile tendency to keep 
in contact with parents.) 

The lowest intensity Hoots of infants (uttered toward the end of 
a period during which Hoots have become progressively less frequent) 


5,000 
4,000 | 
Ny { 
3,000 
2,000 lauyy 
cp gle 
ae Wy AL 


0.2 0.4 0.6 0.8 1.0 
Fic. 21—A single “Brrrrp” sound, uttered by an immature. 


Based upon a spectrogram by a “Sona-graph.” 
This drawing does not show all the harmonics present. 


are relatively very soft and somewhat reminiscent, to human ears, of 
the Moans of adults. The two types of patterns may be related 
ontogenetically. 

The infant whose calls were recorded was heard to utter a few, 
rather low-pitched, rattling sounds (see figure 21). These sounds 
were relatively very rare, and were not heard to be uttered by other in- 
dividuals. It seems probable that they were more or less “abnormal” 
variants of some more common pattern. They sounded, to me, as if 
they could be intermediate between High Trills and Gruff Grunts ; 
but spectrograms suggest that they were related to Hoots. They may 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN Fi. 


have been incompletely formed Hoots, uttered without proper ad- 
justment of the vocal apparatus. 

That this sort of variation is possible is indicated by the behavior 
of howler monkeys. Male howlers usually utter lengthy roars at dawn. 


Fic. 22—An infant uttering Squeaks. 


This shows the characteristic shape of the mouth when opened most widely 
during vocalizations of the Squeak Complex. 


When fully developed, each roar sounds absolutely continuous. But 
males of the species on Barro Colorado (Alouatta palliata) seem to 
take considerable time to “warm up,” and their first attempts to roar 
are apt to come out as wooden-sounding rattles (these may be the 
sounds that Altmann, op. cit., describes as “pops’’). 

Infants utter Gruff Grunts which sound quite like the correspond- 
ing notes of adults and occur in similar hostile circumstances. They 


78 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


also try to bite when handled, and perform other overt unritualized 
attack and escape patterns like those of adults as soon as they have 
developed adequate coordination of the limbs. 

Changes in behavior between the infantile stage and maturity were 
not studied in detail, as it was not considered advisable to do much 
experimental work with the young animals being raised in captivity. 
(In any case, there are indications that at least some patterns develop 
at different rates in captivity and in the wild—see above.) Intermittent 
observations of the captive animals did, however, reveal the following. 

The young animals gradually lost their desire to be carried by a 
parent. This process seemed to be completed by the time that they 
were approximately half grown (at the age of 6 or 7 months). By 
the time that they were a year old, they usually objected violently to 
being touched by the human beings whom they had regarded as parents 
earlier. 

It was my impression that all or most of the signal patterns of 
young animals were essentially the same as those of adults by the time 
that they were half grown, with the exception of the strictly sexual 
patterns and the Squeaks. The young animals continued to direct 
many Squeaks toward human beings with whom they were familiar 
until they were approximately 1 to 14 years old. All or most of the 
later Squeaks seemed to be essentially friendly “greeting” patterns. 
Many of them were closely associated with social sniffing. 

The obviously subadult individuals observed in the forest on Barro 
Colorado Island also tended to utter more Squeaks than adults. 

Two immature Night Monkeys at Iquitos performed a variety of 
signal patterns in addition to the Hoots already mentioned. They 
were observed to perform Swaying and to utter typical Gruff Grunts 
and Screams like those of adults on Barro Colorado Island. They 
also uttered many Squeaks like those of young animals of similar age 
(or stage of development) on Barro Colorado. 


PREVIOUS DESCRIPTIONS OF NIGHT MONKEY 
BEHAVIOR PATTERNS 


A number of signal patterns of Night Monkeys have been described, 
more or less briefly, in previous publications. Some of these descrip- 
tions have been cited above. It may be useful, however, to mention 
some others, try to identify and classify the described patterns accord- 
ing to the terminology used in this paper, and discuss some apparent 
discrepancies and problems. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 79 


Hill (1960) summarizes several accounts by earlier observers and 
quotes some of their transcriptions of vocal patterns. The notes 
variously transcribed as “oo-00-00,” “bu-bu-bu,” and “boom” may be 
the same as the notes I have called Hoots. The notes transcribed as 
“chip-chip-chip,” “chuip-chuip,” and “kweep-kweep” presumably are 
Squeaks, The notes transcribed as “‘urr-urr” may be Gruff Grunts. 

Sanderson (op. cit.) says that a Night Monkey that he kept in 
captivity uttered “grrrrrrmph” notes as a sign of contentment or 
pleasure. His transcription would suggest that the notes were Grunts; 
but, if so, his interpretation of their significance is almost certainly 
wrong. He may have been misled by the fact that captive Night 
Monkeys sometimes utter many Gruff Grunts without extreme or im- 
mediately recognizable attack and escape movements, after they have 
learned that such movements do not produce the desired results, i.e., 
after they have learned that they cannot get out of their cages or force 
their captors to retreat permanently. 

Sanderson mentions a variety of other sounds, all or most of which 
are unidentifiable. He also says that no two individual Night Monkeys 
have the same repertory of sounds. This is not only extremely im- 
probable per se, but is not borne out by my own observations. All 
Panamanian Night Monkeys probably utter almost exactly the same 
types of sounds, although there may be slight differences (in pitch, 
length, or loudness) between the equivalent sounds of different indi- 
viduals, and different individuals may utter the same sounds in slightly 
different situations, presumably because they have had different 
histories. (All the more or less distinctive notes and calls which were 
heard uttered by only one individual were extremely rare; and it 
seems likely that further observations would have shown that they are 
also present in the repertories of other individuals.) 

The most extensive published descriptions of Night Monkey calls 
are by Andrew (op. cit.). His account is based upon observations of 
two individuals in the laboratory. Unfortunately, he does not say 
where his animals came from or describe the conditions in which they 
were kept. Asa result, it is difficult to interpret some of his findings. 

All or most of the patterns that he calls “twitters” would seem to be 
varieties of what I have called Squeaks. So, in all probability, are 
the notes that he calls ‘guinea-pig squeaks.” (It may be worth men- 
tioning that none of the Squeaks or Squeaklike notes of the individuals 
kept on Barro Colorado appeared to have “traces of clicks” like those 
which Andrew describes as being superimposed upon twitters.) The 
patterns which Andrew calls “‘trills’ may be “High Trills” according 


80 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


to the terminology used here. (His account would suggest that the 
individuals that he studied were either young or adults which had 
retained juvenile characteristics.) The patterns which he calls “waver- 
ing squeaks,” “‘sharp calls,” and “booms” are not precisely identifiable, 
although the latter two (at least) would appear to belong to the Grunt 
Complex. 

One further comment may be added. There may be some geo- 
graphic variation in the form of some vocal patterns of Night Mon- 
keys ; but it seems unlikely that the differences between the patterns of 
different populations or subspecies are as great as might be inferred 
from some of the published accounts. 


SUMMARY 


This is the first in a series of papers on the social signals and some 
other behavior patterns of New World primates. 

Night Monkeys are moderately small. Under natural conditions, 
they are purely arboreal and nocturnal. In Panama, at least, they are 
not very gregarious. They are seldom found in groups larger than a 
single family of two adults and one young, and even mated individuals 
may stray some distance apart from one another. 

The hostile behavior of adult Night Monkeys includes unritualized 
attack and escape movements and a variety of ritualized displays. A 
few of these displays are special postures and movements, 1.e., visual 
signals, but the great majority are notes and calls, 7.¢., auditory signals. 
The sexual behavior of adults includes olfactory and tactile signals in 
addition to unritualized patterns and a few auditory signals. 

Some of the most distinctive features of this display behavior seem 
to be direct or indirect consequences of, or adaptations to, nocturnality. 

Adult Night Monkeys have fewer visual displays than any other 
platyrrhines whose behavior has been studied. The few that they do 
have are relatively crude, produced by simple movements of the whole 
head and/or body. Some of their visual displays are less exaggerated 
in form that the homologous patterns of related species. They do not 
have any facial expressions, or erectile tufts or ruffs of hair around 
the face which could be used in signaling. It seems likely that they 
have lost, or failed to develop, an extensive and complex system of 
visual signals simply because they frequently cannot see one another 
clearly in the forest at night. 

As partial compensation, they utter “contact notes’ more frequently 
than other platyrrhines. These may help to maintain social cohesion 
between the adult members of a family group in the dark. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 81 


The whole vocal repertory of adult Night Monkeys is composed of 
discrete units, nine or ten distinctly different types of notes and 
calls. These patterns do not intergrade with one another to any 
appreciable extent. Intermediates between different types of notes 
and calls are comparatively (if not always actually) rare. Complex 
messages are given in the form of series of different notes or calls, 
each one of which contains part of the message, not in the form of a 
single intermediate or ambivalent note or call containing the whole 
message in itself. This type of vocal repertory is quite different from 
that of any other monkey whose behavior has been described. It is 
also quite different from the repertory of infant and young juvenile 
Night Monkeys. It may be an adaptation to ensure that vocal mes- 
sages cannot be misinterpreted, even when they are not accompanied 
by any relevant nonvocal information. In many circumstances, adult 
Night Monkeys must have to react to, and rely upon, vocalizations 
alone. 

There may be a general rule, among all monkeys, that species 
or classes of individuals that are largely dependent upon auditory 
signals for the regulation of their social behavior tend to have 
discrete, sharply delimited vocal patterns, while species or classes 
of individuals that are less dependent upon auditory signals tend to 
have intergrading vocal patterns. 

Most vocal patterns of adult Night Monkeys are very low pitched. 
They are lower on the average than those of any related species of 
similar size. As low-pitched sounds should carry farther than high- 
pitched sounds, this may be another adaptation to ensure that vocal 
messages are as clear as possible. The only high-pitched vocal patterns 
of adult Night Monkeys are short-range signals. 

Some peculiar negative features of the behavior of adult Night 
Monkeys may be correlated with their slight degree of gregariousness. 
Unlike adults of many related species, they seldom perform “Allo- 
grooming” (the grooming of one individual by another) except in 
copulatory situations, or perform redirection attacks upon other in- 
dividuals of their own species. They also seem to lack vocal patterns 
whose primary or only function is to warn other individuals of possible 
danger in the environment. 

Other distinctive features of the species include: Care of the young 
by the male (this may be possible only because the sex ratio is one 
to one and pair-bonds are close and long sustained) ; comparatively 
frequent use of the hands during fighting (possibly because the 
canine teeth are small); the apparent absence of any tendency to 


82 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


jump up and down and break off branches in rage (this may be 
correlated with the small size of the species); and the apparent 
absence of “displacement” scratching or grooming. 

Most of the vocal patterns of infants are high pitched and com- 
pletely intergrading. Infants may be able to afford such behavior be- 
cause they are always in close contact with their parents. The effects 
of their vocalizations may be supplemented and reinforced by tactile 
and visual stimuli. 

Some of the display patterns of Night Monkeys are particularly 
reminiscent of howler monkeys (Alouatta), titi monkeys (Cal- 
licebus), and/or tamarins (Saguinus). 


ACKNOWLEDGMENTS 


I am greatly indebted to many people for assistance in studying the 
vocal patterns. In particular, Dr. W. John Smith helped with the 
recordings, made the spectrograms on the “Sonagraph,” and contrib- 
uted many useful suggestions and criticisms. D. K. North and Miss 
J. Arnold provided technical assistance. The other spectrograms were 
made by Martin S. Brewer ; I must thank Mrs. Helen Hayes for help- 
ing to arrange this. Dr. Brian Patterson and Dr. A. S. House gave 
helpful advice and explained some puzzling features of the spectro- 
grams. Dr. J. D. Pye very kindly provided assistance and equipment 
in an effort to detect ultrasonic notes or calls. 

I am also grateful to Dr. Theodore H. Reed and J. Lear Grimmer 
for facilitating work in the National Zoological Park, to Charles 
Hawkshead for permitting me to observe his animals in Iquitos, to 
Richard W. Thorington, Jr., for useful information on the skin 
glands of Night Monkeys, and to Dr. John H. Kaufmann and 
David Fairchild, 2d, for assistance in the field and in the laboratory 
on Barro Colorado Island. 


BIBLIOGRAPHY 


ALLEN, J. A. 
1916. Mammals collected on the Roosevelt Brazilian Expedition, with field 
notes by Leo E. Miller. Bull. Amer. Mus. Nat. Hist., vol. 35, 
pp. 559-610. 
ALTMANN, Stuart A. 


1959. Field observations on a howling monkey society. Journ. Mamm., 
vol. 40, pp. 317-330. 


NO. 5 BEHAVIOR OF THE NIGHT MONKEY—MOYNIHAN 83 


ANnprEw, R. J. 
1963. The origin and evolution of the calls and facial expressions of the 
primates. Behaviour, vol. 20, pp. 1-109. 
Bastock, M.; Morris, D.; AND Moyninan, M. 
1953. Some comments on conflict and thwarting in animals. Behaviour, 
vol. 6, pp. 66-84. 
BENNETT, C. F., Jr. 
1963. A phyto-physiognomic reconnaissance of Barro Colorado Island, 
Canal Zone. Smithsonian Misc. Coll., vol. 145, No. 7. 
BIEGERT, JOSEPH. 
1961. Volarhaut der Hande und Fiisse. In “Primatologia,” ed. H. Hofer, 
A. H. Schultz, and D. Starck, II/1:3. Basel. 
CABRERA, ANGEL. 
1957. Catalogo de los mamiferos de America del Sur. Rev. Mus. Arg. Cie. 
Nat. “Bernardino Rivadavia,” vol. 4, No. 1, pp. 1-307. 
CABRERA, ANGEL, AND YEPES, JOSE. 
1940. Historia natural Ediar ; mamiferos sud-americanos. Buenos Aires. 
CARPENTER, C. R. 
1934. A field study of the behavior and social relations of howling monkeys 
(Aloutta palliata). Comp. Psych. Monogr., vol. 10, No. 2, pp. 1-168. 
1935. Behavior of red spider monkeys in Panama. Journ. Mamm., vol. 16, 
pp. 171-180. 
Cottras, NICHOLAS, AND SOUTHWICK, CHARLES. 
1952. A field study of population density and social organization in howling 
monkeys. Proc. Amer. Phil. Soc., vol. 96, pp. 143-156. 
CuLten, J. M. 
1963. Allo-, auto- and hetero-preening. Ibis, vol. 105, No. 1, p. 121. 
Enopers, R. K. 
1935. Mammalian life histories from Barro Colorado Island, Panama. Bull. 
Mus. Comp. Zool., vol. 78, pp. 385-502. 
Hanson, GLorta, AND MontTaGNa, WILLIAM. 
1962. The skin of primates. The skin of the owl monkey (Aotus trivir- 
gatus). Amer. Journ. Phys. Anthrop., vol. 20, No. 4, pp. 421-429. 
HERSHKOVITZ, PHILIP. 
1949. Mammals of northern Colombia. Preliminary report No. 4: monkeys 
(Primates), with taxonomic revisions of some forms. Proc. U.S. 
Nat. Mus., vol. 98, pp. 323-427. 
1958. A geographic classification of neotropical mammals. Fieldiana: zool- 
ogy, vol. 36, No. 6, pp. 581-620. 
Hinz, W. C. Osman. 
1956. Behaviour and adaptations of the primates. Proc. Roy Soc. Edin- 
burgh, vol. 66, pp. 94-110. 
1957. Primates. III. Pithecoidea, Platyrrhini. Edinburgh. 
1960. Primates. IV. Cebidae, Part A. Edinburgh. 
Hitt, W. C. Osman; AppLeyarD, H. M.; AND Auser, L. 
1959. The specialized area of skin glands in Aotes Humboldt (Simiae Pla- 
tyrrhini). Trans. Roy. Soc. Edinburgh, vol. 63, part 3, pp. 535-551. 


84 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Hinpe, R. A., AND RowEtt, T. E. 
1962. Communication by postures and facial expressions in the rhesus 
monkey (Macaca mulatta). Proc. Zool. Soc. London, vol. 138. 
pp. 1-21. 
LorENz, Konrap. 
1952. King Solomon’s ring; new light on animal ways. New York. 
Marter, P. 
1956. Behaviour of the chaffiinch. Behaviour, Supplement 5, pp. 1-184. 
MoyniHan, M. 
1955. Types of hostile display. Auk, vol. 72, No. 3, pp. 247-259. 
1962a. The organization and probable evolution of some mixed species flocks 
of neotropical birds. Smithsonian Misc. Coll., vol. 143, No. 7, 
pp. 1-140. 
1962b. Hostile and sexual behavior patterns of South American and Pacific 
Laridae. Behaviour, Supplement 8, pp. 1-365. 
1963a. Display patterns of tropical American “nine-primaried” songbirds. 
III. The green-backed sparrow. Auk, vol. 80, No. 2, pp. 116-144. 
1963b. Inter-specific relations between some Andean birds. Ibis, vol. 105, 
No. 3, pp. 327-339, 
Noite, ANGELA. 
1958. Beobachtungen tiber das Instinktverhalten von Kapuzineraffen (Cebus 
apella L.) in der Gefangenschaft. Behaviour, vol. 12, pp. 183-207. 
Prooc, D. W., AND MacLean, P. D. 
1963. Display of penile erection in squirrel monkey (Saimirt sciureus). 
Animal Behaviour, vol. 11, No. 1, pp. 32-39. 
RoHEN, JOHANNES W. 
1962. Sehorgan. In “Primatologia,’ ed. H. Hofer, A. H. Schultz, and 
D. Starck, II/1:6. Basel. 
RowELL, T. E. 
1962. Agonistic noises of the rhesus monkey (Macaca mulatta). Symp. 
Zool. Soc. London, No. 8, pp. 91-96. 
Rowe Lt, T. E., Aanp HInpg, R. A. 
1962. Vocal communication by the rhesus monkey (Macaca mulatta). Proc. 
Zool. Soc. London, vol. 138, pp. 279-294. 
SANDERSON, IvAN TERENCE. 
1957. The monkey kingdom; an introduction to the primates. New York. 
Scott, JOHN PAUL. 
1958. Animal behavior. Chicago. 
Srvpson, GEORGE GAYLORD. 
1961. Principles of animal taxonomy. New York. 
ULLRICH, WOLFGANG. 
1961. Zur Biologie und Soziologie der Colobusaffen (Colobus guereza cau- 
datus Thomas 1885). Zool. Garten, vol. 25, pp. 305-368. 


; 
| 
: 
{ 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NUMBER 6 


~ 


A REVISION OF THE. 
AMERICAN VULTURES 
OF THE GENUS CATHARTES 


By 
ALEXANDER WETMORE 


Research Associate, Smithsonian Institution 


(PusiicaTion 4539) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
AUGUST 14, 1964 


= 


ae 


wal 


’ 
} 
} 
y 


+ 


— 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NUMBER 6 


mosey LSION OF Ee 
AMERICAN VULTURES 
Sion GENUS CATHARTES 


By 
ALEXANDER WETMORE 


Research Associate, Smithsonian Institution 


(PusticaTion 4539) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
AUGUST 14, 1964 


PORT CITY PRESS, INC. 
BALTIMORE, MD., U. S. A. 


m REVISION OF THE AMERICAN VULTURES 
OF THE GENUS CATHARERS 


By ALEXANDER WETMORE 
Research Associate, Smithsonian Institution 


Turkey vultures, found widely throughout the Americas, though 
easily recognized in life or when freshly killed, pose many difficulties 
in identification when preserved as museum specimens. The color 
differences of the bare head and upper neck that separate the species, 
and in the case of the red-headed group some of the subspecies, 
change soon after death to a dull hue discouragingly similar in all. 
My personal interest in these birds began in 1920 when I first en- 
countered the yellow-headed vulture in life in the Chaco of Argentina 
and Paraguay, and in the report on that expedition I ventured to 
publish a synopsis that covered what I had been able to learn of the 
genus as a whole (Wetmore, 1926, pp. 86-91). The subject has 
remained one of intriguing interest, in large part because of its diffi- 
culties, and I have continued to examine birds of the genus whenever 
possible. A preliminary account of the yellow-headed group has been 
covered in another study (Wetmore, 1950, pp. 415-417). The account 
that follows is based on data from several hundred museum skins, 
in addition to many observations on living individuals that I have seen 
and have collected during my expeditions in tropical regions. 

In these studies I have been indebted to many individuals, among 
whom I should mention especially Mr. J. D. Macdonald and other 
authorities of the British Museum (Natural History) for privileges 
in connection with their collections, in particular the material that 
had been studied by Harry Kirke Swann. Dr. G. Rokitansky of the 
Naturhistorisches Museum in Vienna made arrangements that, in his 
absence, allowed me to examine the series of yellow-headed vultures 
in that institution. The collections in the American Museum of 
Natural History and the Chicago Natural History Museum have been 
of repeated assistance. I have to thank especially Dr. George H. 
Lowery, Jr., for the loan of specimens that included two of the new 
form described in this paper. Dr. Emerson Kemsies, in charge of 


SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 146, NO. 6 


2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


the bird collections in the Museum of the University of Cincinnati, 
very kindly made special arrangements that allowed study of the 
large series of vultures in the Herbert Brandt collection under his 
charge. Curators of other collections have aided in allowing examina- 
tion and, when necessary, in the loan of skins for comparison with 
the series in Washington. 

The account that follows is a summary of the characters and distri- 
bution of the species of the genus, with their geographic races. 


CATHARTES AURA (Linnaeus): Turkey Vulture 


Bare head and neck in adult birds in life red, in one race lined 
narrowly across the back of the cranium with yellowish or greenish 
white; skin of the side and front of the neck near its base smooth, 
without caruncles ; upper hindneck in adult bare. 

Found in the Americas from the northern temperate area of North 
America through the tropics to the temperate regions of the far south 
where these birds range to Patagonia, Tierra del Fuego, and the Falk- 
land Islands. 


CATHARTES AURA SEPTENTRIONALIS Wied 

Cathartes septentrionalis Wied, Reise Nord-Amer., vol. 1, 1839, p. 162. (Fox 
River, near New Harmony, Indiana.) 

Cathartes aura carolinensis “Townsend” Friedmann, U.S.Nat.Mus. Bull. 50, 
pt. 9, Sept. 29, 1950, p. 44. (Nomen nudum; lapsus calami for Cathartes aura 
septentrionalis as listed by C. W. Townsend, Mem. Nuttall Orn. Club, no. 5, 
Aug. 1920, p. 97.) 


Characters —Borders of lesser wing coverts paler than in other 
races, averaging wider, and therefore more prominent; distal sec- 
ondaries with paler brown borders and tips; size large, wing 509- 
545 mm. 

Measurements—Males (43 specimens), wing 509-545 (526), tail 
250-288 (267), culmen from cere 22.7-26.8 (24.6, average of 40), 
tarsus 61.7-73.0 (66.0) mm. 

Females (35 specimens), wing 518-552 (535), tail 255-292 (275), 
culmen from cere 23.7-27.2 (25.3, average of 29), tarsus 61.7-73.0 
(66.0) mm. 

Range.—Breeds in eastern North America from eastern Minnesota 
(Itasca County, rarely), central Wisconsin (Oconto County, one 
record), south-central Michigan, southern Ontario, central New York, 
southwestern Massachusetts, and Connecticut, south through eastern 
Iowa, Missouri, and Arkansas to Louisiana, the Gulf coast, and 
southern Florida: Intergrades with C. a. meridionalis in Minnesota, 
Kansas, Oklahoma, and eastern Texas. 


No. 6 REVISION OF AMERICAN VULTURES—-WETMORE 5 


Winters from the Ohio Valley, central Maryland (rarely in the 
intervening mountains), and New Jersey south to southern Texas 
(Rio Grande City), the shores of the Gulf of Mexico east to 
southern Florida, and the southeastern Atlantic coast. 

Recorded casually in southern Arizona (Pima County), Quebec, 
New Brunswick, Nova Scotia, Labrador, Newfoundland, eastern 
Massachusetts, Vermont, New Hampshire, and Maine: Accidental 
in Bermuda (one record, December 1853). 

Remarks—The paler appearance of the wing coverts, due to the 
broad, light brownish-gray edgings of the individual feathers, sepa- 
rates this form from the western race meridionalis, in which many 
individuals are of equal size. Measurements of septentrionalis from 
birds taken during the breeding season indicate a cline from the 
smallest in Florida to the largest in the northern area of the range. 
The smaller individuals in the resident group in southern Florida 
are within the upper limits of the size range of Cathartes aura aura, 
but all that I have seen have the paler margins of the wing coverts of 
septentrionalis. The large northern individuals move in winter 
throughout the south to the southern limits of the form. 

Through the kindness of Dr. W. J. Breckenridge I have had the 
loan of specimens from Minnesota which indicate that the birds of 
the small group in Itasca County, in the northeastern part of the State, 
while intermediate toward the western form, are nearer septentrion- 
alis. The same is true of material from Douglas County in north- 
eastern Kansas. These points serve to indicate a general border area 
between the eastern and western forms. 


CATHARTES AURA MERIDIONALIS Swann 


Cathartes aura meridionalis Swann, Syn. Accipitres, pt. 1, Sept. 28, 1921, p. 3. 
(Santa Marta, Province of Magdalena, Colombia.) 

Cathartes aura teter Friedmann, Proc. Biol. Soc. Washington, vol. 46, Oct. 26, 
1933, p. 188. (Riverside, California.) 


Characters —Edgings of the lesser wing coverts definitely darker, 
browner, and somewhat less in extent, so that they are less prominent 
than in C. a. septentrionalis; distal edgings and tips of secondaries 
averaging very slightly darker ; size large, but with the maximum and 
average less than in septentrionalis. 

Measurements.—Males (25 specimens), wing 487-528 (509), tail 
237-268 (253), culmen from cere 22.2-26.6 (24.5), tarsus 60.6-65.1 
(63.7) mm. 

Females (16 specimens), wing 495-526 (511), tail 245-272 (259), 
culmen from cere 24.0-26.3 (25.2), tarsus 62.5-67.6 (64.9) mm. 


4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Range.—Breeds in western North America from southern British 
Columbia, central Alberta, south-central Saskatchewan, and southern 
Manitoba south through California (except the lower Colorado 
River Valley) to southern Baja California, south-central Arizona, 
south-central New Mexico, and south-central Texas; east to south- 
western Minnesota, western Iowa (Audubon), and central Kansas. 

Winters from California and Nebraska southward, moving in 
migration through Central America in vast flocks: Some continue 
beyond Panama to South America from Colombia (Santa Marta; 
Rio Guatipori, 3,000 meters elevation, in the Sierra Nevada de Santa 
Marta; El Tambo, Cauca), and central Venezuela (Caicara), south 
to Ecuador (Monji), the Paraguayan Chaco (Orloff), and southern 
Brazil (Salto Grande, Rio Paranapanema, Sao Paulo). 

Casual in Florida (Merritts Island; Cape Sable). 

Remarks.—Swann, in an early review of the turkey vultures 
(1921, pp. 3-4) held that the birds of North America, Central 
America (with the exception of Isla Cozumel), and the West Indies 
were alike, and, therefore, listed them as Cathartes aura aura (Linn- 
aeus). He described the Cozumel bird as C. a. insularis on supposed 
smaller size and also separated the populations of western South 
America from Colombia to northern Chile and Argentina under the 
name of Cathartes aura meridionalis, which he listed as “subsp. nov. 
[nom. nov. Cathartes aura aura (Linn.) ed. 1, et auct. plur. Type 
loc. sugg. Colombia.|” For this he selected as type a bird in the 
collections of the British Museum (Natural History). This is a 
specimen originally in the Salvin and Godman collection, B.M. no. 
87.5.1.11, , taken at Santa Marta “U.S. of Colombia,” by F. A. A. 
Simons, February 27, 1879. A label in small script, written by the 
collector, with ink that has faded until some words are illegible, reads 
in part as follows: “No. IV.I Sta Marta Sex:g*. 2 N.V. aura. Con- 
sidered a great boon to the town, as they keep the place clear of all 
smelling meat, etc. Flesh about head fine pinky flesh color giving it 
the appearance from a distance of a fine red head. February 27, 
1879.” 

On examination I have found that this type is an adult bird with 
light brown edgings on all the wing coverts, and the following meas- 
urements: Wing 525, tail 266, culmen from base 24.7, tarsus 66.3 
mm. It is obviously a migrant from North America and as evidently 
one from the western part of the continent. The name meridionalis 
Swann must therefore replace the later teter described by Friedmann, 
who was the first to note that the western race was distinct. It will 


NO. 6 REVISION OF AMERICAN VULTURES—WETMORE 5 


be observed in the winter range of this bird as given above that I have 
seen another specimen of this race taken at 3,000 meters elevation 
in the Sierra Nevada de Santa Marta (U.S.N.M. no. 386705), in ad- 
dition to others from Venezuela, Ecuador, Paraguay, and Brazil. 

As indicated under C. a. septentrionalis, intergradation with that 
form in the upper Mississippi Valley is shown in birds from Minne- 
sota where those from Dawson County in the west-central part of 
the State are intermediate, but are nearer meridionalis, as are speci- 
mens farther south from central Kansas and west-central Oklahoma 
(Mt. Scott, Comanche County). Allocation of the breeding birds 
from southeastern Kansas and eastern Oklahoma from present infor- 
mation is uncertain. 

The diagnosis and measurements given above have been taken from 
birds presumed to be on or near their breeding grounds. Cathartes 
a. meridionalis shows the same cline of steadily increasing size from 
south to north as is found in the eastern race. To the south there is 
no sharp break between this form and C. a. aura. In fact, the type 
specimen of teter Friedmann, from Riverside, about 50 miles east 
of Los Angeles in southern California, is on the borderline between 
the two in size and color. 

Occasionally, birds from arid regions taken late in summer show 
fading in the color of the wing coverts so that they appear lighter 
than normal. At first glance these may suggest septentrionalis but 
on comparison with specimens of that race in similar stage of plum- 
age are definitely browner. First fall birds of meridionalis, and also 
of typical aura, often show narrow, grayish-white edgings on the 
middle wing coverts. 


CATHARTES AURA AURA (Linnaeus) 
Vultur Aura Linnaeus, Syst. Nat., ed. 10, vol. 1, 1758, p. 86. (Veracruz, 
México.) 
?[ Aquila] nudicollis Ritter, Naturhist. Reise Westind. Insel Hayti, 1836, p. 155. 
(“Geyer mit nackenden Halse’: No further description. ) 
Cathartes aura insularis Swann, Syn. Accipitres, pt. 1, Sept. 28, 1928, p. 3. 
(Isla Cozumel, Quintana Roo, México.) 


Characters.—Similar in color to C. a. meridionalis, but smaller, 
with shorter wings and tail. 

Measurements——(Taken from birds assumed to be on or near their 
breeding grounds.) Males (21 specimens), wing 462-495 (478), 
tail 226-249 (238), culmen from cere 22.8-25.1 (22.6), tarsus 58.8- 
64.5 (62.4) mm. 


6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Females (12 specimens), wing 471-495 (482), tail 231-251 (241), 
culmen from cere 22.7-25.9 (24.1), tarsus 58.6-66.5 (62.5) mm. 

Range.—Breeds from the lower Colorado River Valley (Riverside 
Mountain, Calif.), southern Arizona (Yarnell, Yavapai County; 
Vail, Pima County; Bisbee, Cochise County), southern New Mexico 
(San Luis Mountains), and southern Texas (Delaware Creek; Lim- 
pia Creek; Chisos Mountains; Starr County) south into Sonora, 
Chihuahua, and Coahuila (southern limits on the Mexican tableland 
uncertain), through the Caribbean slope of tropical and subtropical 
México, and Central America to Honduras, probably to central Costa 
Rica; the Bahama Islands (Mangrove Cay, Grand Bahama, Abaco, 
Andros), Cuba, Isle of Pines, Jamaica, and southwestern Puerto 
Rico (introduced) : On the mainland intergrades on the north with 
C. a. meridionalis. 

In migration and winter south through Panama to Darién (Jaqué), 
including the Pearl Islands (Isla San José). 

Remarks.—lIt is evident that the division in two races, meridionalis 
and typical aura, appears arbitrary, with a considerable area of over- 
lap. However, such separation seems required in view of the dis- 
parity between the populations with small size of the far south and 
those decidedly larger of the north, the range in the wing in males 
being from 462 to 525 mm. and in females from 471 to 526 mm. 
Smaller size is coupled with tropical and lower subtropical zone 
range, against the mainly Sonoran and Temperate Zone distribution 
of the larger, northern birds. It is probable that there are other 
intangible factors of difference involved that have not been evident 
in examination of museum specimens. 

The criterion for the size limits assigned to the southern sub- 
species has been fixed through measurements of specimens from the 
resident birds of the Greater Antilles, where there is no confusion 
through the periodic invasion of northern migrants as is the case in 
the continental breeding range. The dimensions of the Antillean 
group then have been the yardstick used to outline the breeding range 
assigned to the subspecies aura on the mainland. Division between 
this group and meridionalis comes near the boundary between México 
and the United States, with the smaller southern form penetrating a 
short distance to the north of this line. The southern limit indicated 
is tentative since no specimens definitely known to be on their breed- 
ing grounds have been seen from southern Central America. I have 
taken winter migrants in eastern Darién so that this race may range 
to Colombia. 


NO. 6 REVISION OF AMERICAN VULTURES—WETMORE 7 


Some specimens from Cuba have the edgings on the wing coverts 
slightly paler than normal. It is possible that this may be due to a 
factor of septentrionalis relationship (in which this paler color is 
definite) from the population that is found in nearby Florida. 

Immature birds in aura, like those of meridionalis, sometimes have 
the middle coverts edged lightly with grayish white. 


CATHARTES AURA RUFICOLLIS Spix 


Cathartes ruficollis Spix, Avium Spec. Nov. Brasiliam, vol. 1, 1824, p. 2. 
(Interior of Baia and Piaui, Brazil.) 

Oenops pernigra Sharpe, Cat. Birds Brit. Mus., vol. 1, 1874, p. 26. (South 
bank of the River Amazon, about 100 miles above the Rio Negro, Brazil.) 

Cathartes aura “ITllig.” d’Orbigny, Voy. Amér. Mérid., vol. 4, Oiseaux, 1834, 
pp. 38-42, pl. 1, fig. 3. (Amérique méridionale et s’étend méme dans l’Amérique 
du nord. Aprés l’avoir perdu de vue au 28.¢ degré de latitude sud dans la 
province de Corrientes, nous ne l’avons plus retrouvé que dans la Patagonie, 
au 41.¢ degré.”) 

Cathartes orbignyi Sztoleman, Ann. Mus. Polonici Hist. Nat., vol. 4, no. 3, 
Dec. 1, 1925, p. 322. (Based on d’Orbigny, cit. supra.) 


Characters.—Definitely blacker above and below than the northern 
subspecies C. a. septentrionalis, C. a. aura, and C. a. meridionalis ; 
under surface of body decidedly black; borders of wing coverts very 
dark brown, darker than in awra; in life, head and neck dull red, 
with several distinct transverse yellowish white or greenish white 
lines across the posterior surface of the crown and the nape; adult 
usually with an irregular area of yellowish white in the center of the 
crown. 

Measurements.—Males (18 specimens), wing 476-508 (490), tail 
235-265 (254, average of 17), culmen from cere 21.9-24.3 (23.2, 
average of 16), tarsus 60.0-64.9 (62.4, average of 17) mm. 

Females (21 specimens), wing 475-509 (491), tail 235-264 (247), 
culmen from cere 22.3-26.6 (23.7, average of 20), tarsus 60.4-68.0 
(63.8, average of 20) mm. 

Range.—Throughout the tropical zone in Panama; on the Pacific 
slope from near the Costa Rican boundary in western Chiriqui (in- 
cluding Isla Coiba, Isla Taboga, and the Archipiélago de las Perlas), 
and on the Caribbean side from the Canal Zone eastward ; across north- 
ern Colombia (specimens seen from Cordoba, Antioquia, Bolivar, 
Magdalena, and Guajira); and from northern Venezuela south 
through South America east of the Andes to northern Argentina 
(Formosa, Chaco, Santa Fé), Uruguay, and southern Brazil; west 
to southeastern Colombia, eastern Peru, and eastern Bolivia: North- 
ern limit in Central America uncertain, probably in Costa Rica. 


8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Remarks.—The blacker body color, darker brown edgings on the 
wing coverts, and the head markings in life, where the dull red (plain 
in the three northern subspecies) is variegated by cross lines of 
yellow to whitish or greenish yellow across the back of the crown 
and the hindneck, with addition of an ivory-colored area in the center 
of the crown, readily identify this distinct race. I became familiar 
with the differences in plumage markings in my first observations of 
these turkey vultures in the field in South America, but it was not 
until April 1940 that I noted the interesting colors on the head in 
a bird taken in the foothills of the Serrania Macuire in the Guajira 
Peninsula, northeastern Colombia. My first report outside South 
America was of an immature individual shot in 1944 on Isla San 
José in the Gulf of Panama, which I identified by plumage characters 
as ruficollis. At the time I believed that this bird was a wanderer 
from breeding grounds in Colombia. As further field work made me 
familiar with these vultures in Panama, additional records have 
served to establish ruficollis as the breeding form across the isthmus 
on the Pacific slope from Darién on the Colombian border to the 
western province of Chiriqui, where I have recorded it within a few 
miles of the Costa Rican boundary. On the Caribbean side I have 
identified it from the Chagres Valley at Gamboa and Juan Mina in 
the Canal Zone, and in the San Blas from Mandinga, Armila, and 
Puerto Obaldia. The resident form to the west through the provinces 
of Colon and Bocas del Toro on this slope remains to be established. 

There seems little doubt that ruficollis will be found in Costa Rica, 
and it may range beyond far to the north on the Pacific slope. While 
there are no specimens to prove this, van Rossem (1946, pp. 180-181) 
has reported the head color in two male turkey vultures that he shot 
on March 14, 1946, but did not preserve, on Isla Lechuguia (also 
called Isla de los Burros) off Topolobampo, northern Sinaloa, as 
follows: ‘Head, neck, cere (including encirclement of nostrils), 
about ‘Carmine’ or ‘Eugenia Red’; extreme lower bare portion of 
neck at juncture with feather line yellowish orange, the color mostly 
concealed and obvious only on examination ; transverse corrugations 
across crown between eyes and small tubercles on preocular region, 
ivory white; transverse corrugations of hind crown, nape, and sides 
of head grayish blue (about ‘Deep Green-blue Gray’).” Van Rossem 
explains that ‘“‘not having a color chart at the time, the color terms 
in quotes are an approximation based on field notes.” With this in 
mind, it is evident that the description is similar to the condition 
found in Cathartes a. ruficollis. The measurements that van Rossem 


NO. 6 REVISION OF AMERICAN VULTURES—-WETMORE 9 


gives of wing, 490 and 500 mm., and of tail, 244 and 250 mm., also 
agree with those of that race. 

Several from the Chaco of Paraguay in the Brandt collection at 
the University of Cincinnati are somewhat larger than the usual 
measurements of this race and also show lighter coloration—grayish 
brown to grayish white—on the outer webs of the distal ends of 
the secondaries. Otherwise these birds agree with ruficollis in blacker 
body color and darker coloration of the neck ruff. They appear to 
represent a population intermediate toward jota to the west. The 
differences are quite distinct and if found to have a broad enough 
distribution may warrant recognition by name. This, however, may 
be established only with more information, since from present data 
it is not certain that part or all of these larger birds may not be cold 
weather migrants from some Andean area to the west, and, therefore 
intermediates between ruficollis and jota. 

In my earlier review (1926, p. 89) the name ruficollis was estab- 
lished as the proper designation for this race. Among the synonyms 
listed above I have examined the type of Oenops pernigra Sharpe in 
the British Museum. The specimen has an original label that states 
“Collected by A. R. Wallace. 1851, Upr. Amazon.” Another tag 
reads “South bank about 100 miles above the Rio Negro.” The bird 
is a typical example of ruficollis. The next name in the synonymy, 
Cathartes orbignyi Sztolcman, is based, as indicated, on an account 
of Cathartes aura by d’Orbigny. While this, in part, is not specific, 
the head colors in the description, and in the accompanying colored 
plate, are those of ruficollis. The plate however shows three lines 
of prominent, rounded caruncles on the side of the neck at the base 
so that part of d’Orbigny’s account may refer also to one of the 
species of the yellow-headed group. 

It is of interest to observe that Azara (1802, p. 27) in his account 
of the Acabiray also describes in detail the head colors of ruficollis. 


CATHARTES AURA JOTA (Molina) 


Vulcur [sic] Jota Molina, Sagg. Stor. Nat. Chili, 1782, pp. 265, 343. (Chile.) 

Cathartes occipitahis Sztoleman, Ann. Zool. Mus. Polonici, vol. 4, no. 4, Dec. 1, 
1925, p. 319. (Huambo, 1,100 meters elevation, 80 kilometers southeast of 
Chachapoyas, Pert.) 


Characters——Secondaries margined in variable amount with light 
gray, in some these edgings extended to the greater coverts, and in 
a few to the outermost middle coverts; slightly browner above and 
below than ruficollis; size large. 


Io SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Measurements—Males (12 specimens), wing 514-546 (521), tail 
252-271 (262), culmen from cere 22.6-26.1 (24.1), tarsus 62.4-71.0 
(65.7) mm. 

Females (6 specimens), wing 511-535 (523), tail 250-281 (260), 
culmen from cere 23.7-25.3 (24.4), tarsus 64.0-68.7 (66.1) mm. 

Range.—From southern Colombia (where it appears to intergrade 
with C. a. ruficollis) south through the Andes, and the adjacent 
valleys, in Ecuador, Pert (Lago Junin), and Bolivia (Cochabamba ; 
Choro) to southern Chile (Angol; Temuco; Puerto Montt; Estre- 
chos de Magallanes), western and southern Argentina (Tucuman ; 
La Rioja; western Mendoza; General Roca; Rio Negro). 

Remarks.—In central Colombia this form appears to intergrade 
with C. a. ruficollis to produce a bird of smaller size, and browner, 
less deeply black color on the body. Some specimens that come from 
the interior of the Chaco in northwestern Paraguay also are similar 
to these supposed intergrades in size and color. 

Molina’s description has been allocated to this interior form so 
that the designation by Swann (1921, p. 4) of Concepcion, on the 
coast of Chile, as type locality is erroneous. The northern and also 
the eastern limits of this subspecies remain to be clearly assigned. 


CATHARTES AURA FALELANDICA (Sharpe) 


Catharista falklandica Sharpe, Ann. Mag. Nat. Hist., ser. 4, vol. 11, no. 62, 
Feb. 1873, p. 133. (Berkeley Sound, East Falkland, Falkland Islands.) 


Characters —Similar to C. a. jota, but grayish edgings on wings 
usually more extensive; size smaller. 

Measurements.—Males (5 from the Falkland Islands), wing 485- 
508 (499), tail 250-265 (254), culmen from cere 23.5-25.1 (24.1), 
tarsus 60.5-71.7 (67.1) mm. 

Males (7 from coastal areas and islands, Ecuador to Chile), wing 
460-507 (485), tail 220-257 (240), culmen from cere 22.7-24.8 (23.9, 
average of 6), tarsus 62.0-66.5 (65.3) mm. 

Females (3 from Falkland Islands), wing 505-510 (508), tail 
258-272 (265), culmen from cere 23.7-26.9 (24.8), tarsus 65.3-69.0 
(66.8) mm. 

Females (3 from islands off Ecuador and Pert), wing 466-478 
(474), tail 223-225 (224), culmen from cere 23.4-25.7 (24.6), tarsus 
65.5-69.6 (68.1) mm. 

Range.—Falkland Islands ; north along the western coast of South 
America to Chile (Isla Mocha, Penco), Pertt (Talara, Islas de Chin- 
cha), and Ecuador (Isla Jambeli, Isla La Plata). 


NO. 6 REVISION OF AMERICAN VULTURES—WETMORE II 


Remarks.—This race and jota are less black on the lower surface 
than ruficollis. While birds in the range outlined are similar in 
grayish-white markings on the wings to jota, and so have been in- 
cluded under that name by recent authors, they differ definitely in 
smaller size. The coastal area that they inhabit from Chile northward 
appears to be invaded by wandering individuals of the inland race 
jota, which has led to confusion in understanding. It may be supposed 
that the population is one adapted to maritime conditions influenced 
by colder oceanic waters. 

Those found along Pertti and Ecuador appear smaller than those of 
the Falklands and may prove separable as another form when more 
is known about them. 


CATHARTES BURROVIANUS Cassin: Yellow-headed Turkey Vulture 


Bare area of front and side of neck with numerous small papillae 
or caruncles, mainly low down toward the feather line, in lesser 
number toward the base of the head ; prominent in freshly killed birds, 
visible in museum specimens as small, wartlike processes; hindneck 
feathered to base of cranium; in life, head and neck yellow and 
orange, varied by prominent blue markings bordered more or less 
with green on the crown. Color of dorsal surface dull black, with 
feathers tipped prominently with fuscous and fuscous-black ; brown- 
ish black below; metallic bluish or greenish sheen less extensive than 
in the larger species with similar head colors. Development of the 
neck papillae begins in young birds as soon as they are on the wing. 

Until I collected specimens in western Panama in 1948, which led 
to the proper allocation of Cassin’s ancient name burrovianus of 1845 
(Wetmore, 1950, pp. 415-417) this species was believed to be re- 
stricted in range to South America. In the abundant material now 
available from the region between northeastern México and northern 
Argentina two geographic races distinguished by differences in size, 
may be recognized. 


CATHARTES BURROVIANUS BURROVIANUS Cassin 


Cathartes Burrovianus Cassin, Proc. Acad. Nat. Sci. Philadelphia, vol. 2, no. 8, 
March-April 1845, p. 212. (Near Veracruz, Veracruz, México.) 


Size small; wing 432 to 459 mm. 

Measurements—Males (10 specimens), wing 432-455 (445.4), tail 
195-225 (207.2), culmen from cere 19.6-23.3 (21.3), tarsus 51.1-59.1 
(57.1), approximate width of central rectrix 42-49 (46.3) mm. 


i2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Females (12 specimens), wing 444-459 (449.3), tail 193-230 
(206.8), culmen from cere 21.2-24.0 (22.1), tarsus 56.2-60.0 (58.0), 
approximate width of central rectrix 43-49 (46.3) mm. 

Range.—Eastern México, in Tamaulipas (Tampico; Lomas del 
Real), Veracruz (Veracruz; Alvarado), and Tabasco (Miramar; 
Villa Hermosa); British Honduras (Belize); Honduras (Puerto 
Lampira) ; Pacific slope of Panama from Chiriqui (David) through 
Veraguas (Sona), Coclé (Rio Hato; Anton), Herrera (El Rincon; 
Pesé; Santa Maria), and Los Santos (Pedasi) to eastern Province of 
Panama (La Jagua; Rio Chico) ; Colombia from Atlantico, Magda- 
lena (Santa Marta; Gaira), and Guajira (Maicao) south locally to 
the upper Rio Cauca, upper Patia Valley in Cauca, the upper Magda- 
lena Valley in Huila, and, in the northern Ilanos, to northern Meta 
(Quenane) ; northwestern Venezuela, in Zulia (Encontrados) and 
Falcon (El Planchon). 

There is some seasonal variation in depth of color of the head. An 
adult female near breeding stage shot on March 24, 1961, at La Jagua, 
Panama, had the iris orange-red; center of the crown indigo, in an 
irregular triangle with the apex forward and the base behind, 
bordered narrowly on either side by pale greenish blue; side of head, 
including the loral area and the base of the mandibular rami, bright 
orange; the bare foreneck, including the prominent caruncles, dull 
orange ; back of the head dull blue, crossed by three irregular rows of 
caruncles which are dull orange; bill dull ivory white; crus dull 
yellowish white; front of tarsus dull greenish gray, rest dull white; 
toes fuscous black; claws fuscous. In another female, also adult but 
in resting stage, taken at the Ciénaga Macana, Herrera, March 1/7, 
1948, the iris was red; center of crown and spot in front of the eye 
dull bluish gray; cere, forepart of crown to center of eyes, nape, 
back of head, and neck to throat dull orange-red; lores greenish 
yellow ; sides of head from posterior loral space back around the eye 
and ear, including the area below the gape, bright orange. A com- 
panion bird taken at the same time was slightly duller. 

In México and Central America the yellow-headed vulture tends 
to frequent lowland marshy areas that often are difficult of access. 
As it is far less common than the red-headed turkey vulture, found 
everywhere, it may be overlooked since the two are closely similar 
except in the head color, which is seen only under favorable condi- 
tions. Now that attention has been attracted there have been recent 
reports that have added considerably to the details of range. Dr. and 
Mrs. Richard R. Graber (1954, pp. 165-166) collected one on July 21, 


NO. 6 REVISION OF AMERICAN VULTURES—WETMORE 13 


1953, in Tamaulipas, 8 miles north of Tampico. They recorded 
others in this general area south of Altamira, and one earlier, on 
June 19, at Lomas del Real, 30 miles north of Tampico, near the 
Gulf of México. Recently Col. L. R. Wolfe has sent me the skin of 
an immature bird taken on March 23, 1964, 18 miles south of Alva- 
rado, in Veracruz. 

Specimens lent to me by Dr. George H. Lowery, Jr., from the 
Museum of Zoology, Louisiana State University, include five from 
Teapa, Miramar, and Villahermosa, Tabasco, taken in 1959 and 1960; 
one from Belize, British Honduras, August 26, 1960; and two from 
Puerto Lampira, Honduras, February 5, 1963. In my work in 
Panama I have found the yellow-headed vulture in the lowlands of 
the Pacific slope from near the Costa Rican boundary to the Province 
of Panama near the mouth of the Rio Bayano, 25 miles east of the 
Canal Zone. 

In view of the known occurrence of these birds in southern Tamau- 
lipas the possibility that they may range farther north should be kept 
in mind. In the early period when the identity of Cassin’s Cathartes 
burrovianus was not clear Dresser (1865, pp. 322-323) reported a 
small vulture seen near Brownsville, Tex., that he thought might be 
this species, but to date there has been no record of it. 

The outline of the range of this race in Colombia is taken mainly 
from discussion by Lehmann and Dugand. Personally I have ex- 
amined specimens from Bolivar (Simiti), Atlantico (Laguna de 
Guajaro), northern Magdalena (Santa Marta), and Guajira 
(Maicao). 


CATHARTES BURROVIANUS URUBITINGA Pelzeln 
Cathartes Urubitinga Pelzeln, Sitzungsb. math.-naturw. KI. Akad. Wiss. Wien, 
vol. 44, pt. 1, 1861, p. 7. (Forte do Rio Branco= Forte Sao Joaquim, Rio 
Branco, Brazil.) 
Cathartes burrovianus dugandi Lehmann, Mus. Hist. Nat. Univ. Cauca Nov. 
Colombianas, no. 3, Dec. 1, 1957, p. 120. (Caicara, Bolivar, Venezuela.) 


Larger, wing 457-509 mm. 

Measurements—Males (27 specimens), wing 457-502 (475.6), 
tail 205-238 (216.0), culmen from cere 20.5-24.7 (21.9, average of 
25), tarsus 56.5-68.8 (60.8), approximate width of middle rectrix 
43-50 (46.9) mm. 

Females (18 specimens), wing 461-509 (484), tail 204-236 (219), 
culmen from cere 20.8-23.4 (20.7, average of 17), tarsus 53.2-64.0 
(60.0), approximate width of middle rectrix 43-51 (46.2, average of 
13) mm. 


14 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Range.—From the llanos of southeastern Colombia, central and 
eastern Venezuela (Caicara, Bolivar; Cantaura, Anzoategui), British 
Guiana (Abary, Georgetown), and Surinam (near Paramaribo) 
through Brazil and Paraguay to northern Argentina (Las Palmas, 
Chaco; Mocovi, Santa Fé), and Uruguay. 

An adult male that I collected on July 20, 1920, near Las Palmas, 
Chaco, in northern Argentina, had the iris red; side of the head and 
throat deep chrome yellow shading to olive-buff at base of bill; center 
of crown dark blue bordered on either side by dark green; a spot of 
dull slate blue beneath the nostrils; rest of bill cream buff; tarsus 
cartridge buff, shading to neutral gray on the toes. 

Division of this species into two geographic races necessarily is 
arbitrary since the measurements on which this is based show a cline 
from the smaller birds of Central America, northern Colombia, and 
northwestern Venezuela to the larger birds of the rest of South 
America. This seems justified, however, by the uniformity found 
within the limits assigned to each group, with a considerable difference 
between the smaller birds of the north and the larger ones of South 
America. 

The yellow-headed vulture was first reported in Colombia by 
F. Carlos Lehmann (1940, p. 461) under the name Guala de Cabeza 
Amarillo, Cathartes urubitinga Pelzeln, with specimens listed from 
Vaupés and Valle del Cauca. In April 1941, in the Guajira Penin- 
sula, Dr. Lehmann and I collected specimens of the yellowhead near 
Maicao (preserved by Lehmann) and later in April and early May, 
I saw them near Uribia, Nazaret, and Puerto Estrella. Dugand 
(1941, p. 54) listed additional records, based in part on specimens 
and observations by Dr. Lehmann that included the birds from 
Maicao. 

In another account Lehmann (1944, pp. 187-190) gave further 
details of occurrence that included the presence of these birds at an 
elevation of 900 meters in the upper valley of the Rio Patia, south of 
Popayan, a stream that flows through Narifio to the Pacific Ocean, 
and also reported these birds near Cali in the Department of Valle. 
He noted that they frequented marshy areas and that their main 
food was fish. In a further paper Dugand (1951, pp. 1-4) noted 
differences in size and listed the birds as two species under the names 
burrovianus and urubitinga. The following year Dugand (1952, 
pp. 1-4) discussed the question again and treated the two as geo- 
graphic races of burrovianus. It should be noted that in his account 
and list of material examined some of the specimens were wrongly 


NO. 6 REVISION OF AMERICAN VULTURES—WETMORE 15 


identified, as I have found that specimens under urubitinga include 
several Cathartes aura ruficollis. Part of his larger birds were the 
species that is described beyond. Lehmann (1957, pp. 118-121) gave 
a summary with further details divided under the two subspecies, 
and in this proposed to separate the two by a third race that he named 
Cathartes burrovianus dugandt, with the type a female in the Ameri- 
can Museum of Natural History from Caicara, Bolivar, Venezuela. 
With recognition of only two races in this species, this name falls 
under C. b. urubitinga. 

In June 1954, through the kind attention of Dr. G. Rokitansky 
I had the opportunity to examine Pelzeln’s type material in the 
Naturhistorisches Museum in Vienna. The specimens in the series 
include three adult and two immature birds collected by Joh. Natterer 
between 1817 and 1835. Though at one time four of these specimens 
had been mounted for exhibition, all were in a fair state of preserva- 
tion when their age as museum specimens is considered. Three, an 
adult male, adult female, and one of the immature birds, are labeled 
from “Forte do Rio Branco, Nord-Bras.,” a locality that corresponds 
to Forte Sao Joaquim, Rio Branco, in modern Brazil. This is accepted 
above as the type locality since the others are marked only “Brasilien,”’ 
as is an adult cotype in the U.S. National Museum, one of the original 
specimens studied by Pelzeln, received in exchange and entered in 
our catalog in 1864. My notes on the entire series fully substantiate 
the name urubitinga as applicable to the yellow-headed turkey vulture 
and to its southern race. 


CATHARTES MELAMBROTUS sp.nov.: Greater Yellow-headed Vulture. 


Characters.—With prominent caruncles on the neck, and head color, 
as in Cathartes burrovianus Cassin, but size definitely larger; tail 
decidedly longer with broader rectrices, the central pair especially 
much wider ; plumage entirely deep black with greenish and purplish 
sheen, without mixture of brown in the wing coverts, or elsewhere. 

Description Type, U.S.N.M. no. 483532, male (fully adult), 
collected by Pinney Schiffer at Kartabo, British Guiana, January 15, 
1930. Plumage deep black throughout, with an iridescent sheen, 
greenish in the main, but in part dull bluish with the light at certain 
angles; under surface of wings and tail dull dark brownish gray ; 
concealed down pure white. 

Measurements.—Males (9 specimens), wing 488-530 (505.7, aver- 
age of 8) ; tail 252-275 (264.4) ; culmen from cere 23.2-26.2 (24.6, 


16 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


average of 8) ; tarsus 68.5-75.1 (70.7) ; approximate width of central 
rectrix 59-70 (63.4) mm. 

Females (3 specimens), wing 510-512 (511, average of 2); tail 
272-285 (279), culmen from cere 23.9-25.5 (24.6), tarsus 69.3-72.5 
(70.9) ; approximate width of central rectrix 60-67 (64.3) mm. 

Type, male, wing 508, tail 272, culmen from cere 26.2, tarsus 68.5, 
approximate width of central rectrix 63 mm. 

A color photograph of a recently killed adult male, taken by John P. 
O’Neill at Tingo Maria, Pert, July 1, 1962, shows clearly that the 
iris was red; bill flesh color ; side of the head and throat deep yellow 
to light orange ; and the crown and a spot in front of the eye deep blue. 

Range.—From southeastern Colombia (Rio Vaupés), the Rio 
Orinoco in southern and eastern Venezuela (Isla Corocoro, Ama- 
zonas; Piacoa, Delta Amacuro), British Guiana (Kartabo; Rock- 
stone; Kamakuna), and Surinam (Keiserberg Airstrip; Wilhelmina 
Mountains), to eastern Pert (Rio Curanja, Loreto; Tingo Maria, 
Huanuco) and central Para in northern Brazil (Tauary on the Rio 
Tapajos; Tapara on the Rio Xingu). 

Remarks.—As my studies of specimens of the yellow-headed vul- 
tures progressed it became evident that there were occasional speci- 
mens that did not agree with the usual pattern of Cathartes burro- 
vianus in size and color. In fact, certain birds were definitely trouble- 
some in attempts to outline characters under which this species could 
be recognized. The first of these aberrant individuals was a skin in 
the American Museum of Natural History, an adult female of un- 
known locality that had died in captivity at the New York Zoological 
Society zoo on December 23, 1918. Presently I saw another of similar 
form in the Museum of Comparative Zoology, and later others in the 
Chicago Natural History Museum. At first I supposed that these 
might be the bird described briefly by Sharpe (1874, p. 26) as Oenops 
pernigra, but in due course when I saw this type it proved to be an 
individual of Cathartes aura ruficollis. There remained the possi- 
bility then that they were Cathartes urubitinga, named by Pelzeln 
from northern Brazil, but when I visited the Naturhistorisches 
Museum in Vienna I found that the type series were all individuals 
with duller color of burrovianus, and with the size of the southern 
population of that species listed above under Pelzeln’s name as 
Cathartes burrovianus urubitinga. It became obvious then that the 
larger birds represented a distinct group that has been overlooked. 
Since their known range is included within that of C. b. urubitinga 
they must be regarded as a distinct species. 


NO. 6 REVISION OF AMERICAN VULTURES—WETMORE 17 


For the extension of range to eastern Peri I am indebted to 
John P. O’Neill and Dr. George H. Lowery, Jr. It is probable that 
this species ranges also to northeastern Bolivia. 


KEY TO THE SPECIES OF THE GENUS CATHARTES 


1, Bare skin of the side and front of the neck at the base smooth; upper hind- 
neck in adult not feathered; bare skin of head and upper neck in adult birds 
in life dull red, usually plain, but in the subspecies ruficollis with several 
narrow lines of yellowish or greenish white across the back of the 
GUAUIUE Me Rena s Ae eee kk ssa ate ot eee Cathartes aura 


Bare skin of the side and front of the neck at the base with numerous 
caruncles, prominent in life; in museum specimens visible as small, wart- 
like projections; hindneck in adult feathered to near the base of the 
cranium; bare skin of head and upper neck in adult birds in life yellow and 
OLangelvaned bysmankingsioL bluests lesb. teni nee e) ans eae aries 2 


2. Back and wings with many feathers tipped and edged prominently with dull 
grayish brown; under surface more brownish black; definitely smaller, with 
tail shorter, not more than 240 mm. long; rectrices, especially the central 
Hate eSsathanwOa MM WIGS. a. seseriecusiciay eleeisielaerye - Cathartes burrovianus 


Plumage deep black throughout, with prominent greenish or bluish reflec- 
tions ; decidedly larger, with tail longer, 250 to 280 mm. or more; rectrices, 
especially the central pair, 59-70 mm. wide...Cathartes melambrotus sp. nov. 


Details of the characters that mark subspecies are given in the main text. 


LITERATURE CITED 


AZARA, FELIX DE 
1802. Apuntamientos para la historia natural de los paxaros del Paraguay 
y Rio de la Plata. Vol. 1, pp. i-xx, 1-534. 


Dresser, H. E. 
1865. Notes on the birds of southern Texas. Ibis, n.s., vol. 1, no. 3, July, 
pp. 312-330. 


DucGanp, ARMANDO 
1941. Adiciones a la lista de aves conocidas en Colombia. Caldasia, no. 3, 
Dec. 15, pp. 53-61. 
1951. Descubrimiento de Cathartes burrovianus Cassin en Colombia. Rev. 
Acad. Colombiana Cienc. Exact. Fis. Nat., vol. 8, no. 30, April, 
pp. 154-156. 
1952. Observaciones adicionales sobre Cathartes burrovianus y Cathartes 
urubitinga. Lozania (Act. Zool. Colombiana), no. 2, June 30, 
pp. 1-4. 
GRABER, RICHARD R., and GRABER, JEAN W. 
1954. Yellow-headed vulture in Tamaulipas, México. Condor, vol. 56, 
no. 3, May 21, pp. 165-166. 


18 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


LEHMANN V., F. CARLos 
1940. Contribucion al estudio y conocimiento de las aves rapaces de 
Colombia. Rev. Acad. Colombiana Cienc. Exact. Fis. Nat., vol. 3, 
no. 12, May-August, pp. 455-461, 2 pls. 
1944. Distribucion de Cathartes urubitinga en Colombia. Rey. Univ. del 
Cauca, no. 3, March-April, pp. 187-190, 1 pl. 
1957. Contribuciones al estudio de la fauna Colombiana XII. Mus. Hist. 
Nat. Univ. Cauca Nov. Colombianas, no. 3, Dec. 1, pp. 101-156. 
SHARPE, R. BOWDLER 
1874. Catalogue of the birds in the British Museum. Vol. 1, pp. I-XIV, 
1-480, 14 pls. (col.) 
Swann, H. Kirke 
1921. A synopsis of the Accipitres (diurnal birds of prey), Part I, Vultur 
to Acctpiter, Sept. 28, pp. 1-63. 
VAN Rosse, A. J. 
1946. Notes on distribution and color of the Mexican turkey vulture. Con- 
dor, vol. 48, no. 4, July 31, pp. 180-181. 
WETMORE, ALEXANDER 
1926. Observations on the birds of Argentina, Paraguay, Uruguay, and 
Chile. U.S. Nat. Mus. Bull. 133, Feb. 1, pp. I-IV, 1-448, 20 pls. 
1950. The identity of the American vulture described as Cathartes burro- 
vianus by Cassin. Journ. Washington Acad. Sci., vol. 40, no. 12, 
Dec. 15, pp. 415-417. 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NUMBER 7 
(Enp oF VoLuME) 


Mary Waux THalcott Fund for 
Publications in Botany 


A NEW SPECIES OF MARINE PENNATE 
_ DIATOM FROM 
HONOLULU HARBOR 


By 
PAUL S. CONGER 


Associate Curator, Division of Cryptogams 
Department of Botany, Smithsonian Institution 


(Pustication 4593) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
OCTOBER 23, 1964 


SMITHSONIAN MISCELLANEOUS COLLECTIONS 
VOLUME 146, NUMBER 7 
(Enp or VoLUME) 


fMary Waux CHalcott Fund for 
Publications tn Botany 


Pane Wy SPECIES OF MARINE PENNATE 
DIATOM FROM 
HONOLULU HARBOR 


By 
PAUL S. CONGER 


Associate Curator, Division of Cryptogams 
Department of Botany, Smithsonian Institution 


> 
Ne ° © 
HOS 
rite 


(Pustication 4593) 


CITY OF WASHINGTON 
PUBLISHED BY THE SMITHSONIAN INSTITUTION 
OCTOBER 23, 1964 


PORT CITY PRESS, INC. 
BALTIMORE, MD., U. S. A. 


Mary Waux Walcott Fund for 
Publications in Botany 


A NEW SPECIES OF MARINE PENNATE 
DIATOM FROM HONOLULU HARBOR 
By PAUL S. CONGER 
Associate Curator, Division of Cryptogams 


Department of Botany 
Smithsonian Institution 


(WitTH ONE PLATE) 


A RATHER DISTINCTIVE and interesting marine benthic epiphytic 
diatom from the bottom of Honolulu Harbor, Hawaii, was collected 
by Dr. R. E. Johannes of the Department of Zoology, University of 
Hawaii, and isolated and cultured by him for use in investigations 
on phosphorus metabolism, and as a source of food for amphipods 
which were being used experimentally. He submitted it to me for 
identification, and I am indebted to him for bringing it to my atten- 
tion. I am also indebted to Dr. David L. Correll, of the Division of 
Radiation and Organisms of the Smithsonian Institution, for carrying 
the diatom in culture for a few weeks. I required access to adequate 
fresh supplies for this study, because the diatom proved too delicate 
to allow satisfactory permanent preparations to be made. 

The diatom cultures well, multiplies rapidly, and is very hardy in 
artificial seawater culture medium. For these reasons it should be a 
very good species for investigational purposes and a good experi- 
mental form for wider use. Whether it will continue to thrive and 
can be maintained indefinitely away from supplies of fresh seawater 
remains to be seen. For all their hardiness under good conditions, 
these forms are very sensitive and demanding. 

It would also be desirable to make electron micrographic studies 
of it to determine its more intricate and finer structure, but I have 
not been in a position to do this. Because of the very great delicacy 
of the shell, the structure is not readily seen with the optical micro- 
scope. For this reason the electron micrographic studies would be 
helpful in its identification. 


SMITHSONIAN MISCELLANEOUS COLLECTIONS, VOL. 146, NO. 7 


2 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


Although it is not a particularly diminutive form in general dimen- 
sions, it is one of the most delicate ones I have had occasion to study. 
I have given it the name subhyalina to indicate its extremely tenuous 
and gossamer character. 


ACHNANTHES SUBHYALINA Conger, sp. nov. 


Plantae unicellulares; valvae breves breviter oblongae vel lineari- 
oblongae, interdum paullo apice constrictis, 5-104 longae, 3-4p latae, 
apice late rotundatae; valva superior cum pseudoraphe angustissima, 
recta, mediam valvae occupante, cum linea mediana transversa angus- 
tissima, nodulis terminalibus et nodulo centrali indeterminato, tota 
superficie valvae hyalina ; valva inferior similis ; chromatophori brun- 
nei, pyriformes, alterni vel oppositi. 

Habitat: In seawater of Honolulu Harbor, Hawaii, originally col- 
lected by R. E. Johannes. 

Frustules short- to linear-oblong, or long-rectangular with rounded 
ends, the latter type sometimes slightly, almost imperceptibly, con- 
stricted in face (valve) view; girdle view rectangular with rather 
sharp (or scarcely rounded) corners; valve surface flat and straight 
apically, sometimes slightly depressed in the center; valve mantle 
narrow, girdle zone two to six times as wide as valve mantle, with a 
lined appearance as if comprised of intercalary bands; end view of 
frustule square-rectangular with rounded corners; valve surface 
mildly convex transapically, with rounded margins; raphe a straight 
narrow line; valve with a median, narrow, transapical groove crossing 
it at right angles to raphe, in girdle aspect the groove, due to focal- 
depth refraction, with the appearance of a triangular or cone-shaped 
bright spot resembling a central nodule (believed a false optical 
effect) ; the slightly thickened corners of the valve end with the im- 
pression of terminal nodules in girdle view; valve surface markings 
cannot be resolved with the optical microscope. Valves 5-10 long, 
3-4 wide. 

Chromatophores in young, actively growing cells are bright orange- 
brown, more or less “‘tear-” or “pear-shaped,” with truncate ends, one 
in each end of the cell, with narrow ends toward the cell center, char- 
acteristically alternate, much less frequently opposed (that is, on the 
same side), occupying (estimated) one-third to one-half the cell vol- 
ume, the alternate arrangement giving frequently a twisting, sigmoid, 
scolio, or amphiproroid effect (actually not present). In older cells, 
the chromatin material is either duller, darker brown, or paler, and 
occupies more of the cell volume in a somewhat irregular pattern, but 


SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146, NO. 7, PLATE 1 


4 


weno 


LJ 


® 


Achnanthes subhyalina, sp. nov. 


(1) Single valves (girdle view); (2) complete frustule (girdle view); (3) 
valves (face view); (4) frustule (girdle view) with extra attached valve, 
sometimes seen; (5) abnormal valves; (6) end view of one frustule; (7) 
frustules (valve view) with chromatophores alternately arranged; (8) same, 
with chromatophores oppositely arranged; (9) spherical “resting” spore bodies 
with peripheral chromatin masses; (10) same completely filled by chromatin 
masses. 


NO. 7 NEW SPECIES OF MARINE PENNATE DIATOM—CONGER 3 


always leaves a central (transapical) stauros-like area, in both valve 
and girdle aspect. 

The shells are exceedingly delicate and gossamer-like and are not 
amenable to conventional microscopic preparation; they disappear 
completely in strong acid but withstand dilute hydrochloric and sul- 
furic acids, which turn the chromatin material green but do not 
digest it. The cells are very slightly silicified and are destroyed by 
incineration. No mounting is possible by conventional methods. 

The vegetative cell population contains occasional spherical, trans- 
parent bodies, peripherally pigmented with dense, essentially round, 
but more or less irregular, pigment masses over a quarter or less of the 
periphery of the sphere; the remainder of the cell is clear. The 
diameters of these spherical bodies range in size from about the length 
of the frustules to up to twice this length. Occasionally the whole 
sphere is filled with peripheral pigment bodies, obviously chromatin 
material similar to that of the diatoms, although no “‘shell’’ forms are 
distinguishable, or if present are collapsed. These pigment masses 
appear to be either residues of former cells or perhaps parts of poten- 
tial ones. They become quite numerous in old, stagnant, decadent cul- 
tures. (Whether they are “auxospores,” or some reproductive phase, 
or a protective or degradational resting body in senile and decadent 
cultures I am unprepared to conclude.) 

The cells are actively motile in new and healthy cultures, moving 
in a mostly linear course, with few reversals; the movement in a 
reversed direction is short (usually less than a cell length) before 
forward motion is again resumed. The rate of movement is about 
five to eight times the cell length in a minute. The cells in aging cul- 
tures, even though they may appear otherwise healthy, are slower, 
moving little or but a cell’s length in a longer time. 

The cells are very strongly adhesive to the substrate in an Erlen- 
meyer flask culture, making an even brown coating on the bottom of 
the flask, and require somewhat violent shaking to loosen them; in 
contrast, for instance, with Phaeodactylum tricornutum, which is 
either nonadhesive or readily stirred. Once detached from the sub- 
strate, they quickly form in dark brownish, free-floating aggregates 
or clumps that never adhere again to the bottom, but adhere strongly 
to one another. 

This diatom is probably one that migrates in its natural benthic 
environment in response to diurnal illumination, although there is no 
observational evidence of this. 

In young, healthy cultures among large populations there are no 


4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 


empty, “dead” cells (unpigmented frustules), which is the only con- 
dition in which they could be examined morphologically at all ade- 
quately. In quite old cultures, empty frustules, and occasionally sepa- 
rate valves, become more frequent. Empty frustules or valves are 
dim-whitish in appearance and almost invisible in water. This ‘“whit- 
ish” appearance of the diatom in water under ordinary full-field 
illumination suggests the advantage of “dark-field” illumination and, 
indeed, the latter (or “phase-contrast”) is a good way to bring out 
more prominently the obscure cell features. The diatoms are most 
readily located by the much greater visibility, in girdle view, of the 
false “central nodule” which can be picked up as a bright triangular 
spot, from which the rest of the cell outline can then be followed. 
Were it not for this the shells would not be easy to make out or would 
be overlooked completely. In valve aspect the raphe and transapical 
groove are the more easily seen features, appearing as moderately 
bright white lines. 

Although this diatom is necessarily described on fewer structural 
features than usual, it is felt that it should be readily identifiable from 
these features, and by the very characteristic “tear-” or “pear-shaped,” 
alternately arranged chromatophores, which afford it a rather con- 
spicuous and distinctive character. I have not been able to secure a 
separate view of the inferior valve, and so that has been hypothecated 
from the girdle view of the whole frustule. 

There is difficulty and uncertainty in making out even the generic 
status, although the diatom character is immediately evident and not 
at all to be questioned. The prevalence, range, and frequency of 
Achnanthes subhyalina are not likely soon to be determined. Its small 
size, frailty, and general obscurity make it a form not likely to be 
found by the conventional methods of examination of natural mate- 
rials that account for the discovery of most species of diatoms. It is 
unlikely to be found except when in quantity in isolated cultures, 
which suggests that there may well be many other such diminutive 
forms that have escaped notice due to the limitations of conventional 
procedures. On the other hand, the readiness and rapidity with which 
it grows and its evident hardiness suggest that it may be a widely 
distributed and abundant species. Because of its frailty and low 
degree of silicification, the shells are not likely to persist after death 
in the natural environment or to be recognized if they do persist. It 
must be observed in the living state for determination or recognized 
from dead shells in culture material. By present methods no perma- 
nent preserved “type” preparations, such as microscope slides, have 


NO. 7 NEW SPECIES OF MARINE PENNATE DIATOM—CONGER 5 


been possible. Material preserved in formalin, alcohol, or other liquid 
preservative is of uncertain and doubtful value. The living culture 
may best serve as confirmatory or “type” material. 

In the active, healthy cultures there is some range in the size and 
shape of the cells, and the size, shape, and arrangement of the chro- 
matophores, but this is well within the limits of expectation. In the 
large numbers of specimens observed the growth pattern is very con- 
sistent and typical, and the incidence of distorted or otherwise abnor- 
mal forms is exceedingly low. The generally healthy vigor of the 
species implies that it thrives under cultural conditions and adapts 
readily to them. The adaptability suggests it as a dependable and 
useful culture organism for many experimental purposes. The dis- 
covery of Achnanthes subhyalina suggests the importance of wide- 
spread “culturing” as a valuable exploratory method, as yet meagerly 
employed, for the recognition of many minute, obscure, and transient 
forms which have so far eluded detection and may continue to do so 
in the future without this method. It is more and more recognized 
that these watery, next to invisible, transitory forms may comprise a 
substantial, functionally important constituent of the micropopulation 
of the ocean. Hitherto they were a “blindspot” in our studies, which 
cannot afford to be overlooked any longer. They will be, at best, a 
tedious, difficult, and special study. 


‘ 
~