I 706792 6 ——— —<—<—<—S ee = ee, Te eee 6 | Ny 3 Ua. ar Tarenre LIBRARY Digitized by the Internet Archive in 2007 with funding from Microsoft Corporation htto://www.archive.org/details/colorationinpoli00enteuoft St a f ee ee ee er cl ee COLORATION IN POLISTES - BY WILHELMINE M. ENTEMAN qo ad * PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON NOVEMBER, 1904 CARNEGIE INSTITUTION OF WASHINGTON PUBLICATION No. 19 WASHINGTON, D. C. JUDD & DETWEILER, PRINTERS e/*93 CONTENTS. Page Tntroduction cr sie cicteiiers's PO RC UOC OOUS OA C08 COG 0 CRU. CIC Re Ane Ode Er CaO Ate Ti General ACCOM Ib es ea oe eea oe ch or Ticy area cralarace Shans e cheee tale ee rede oe als BLES OS 9 Individual variation in the Aaer DaGtecitiot VONStES! wen en cme see newts sisi II TUESLES OGITALES: GLESSO Ue S55 sb svaug 48 RT ie aie Oe OH OUR AE Obie Res Sees ae 12 CREPE PY DE OP NATIANOU iis stare Aafacs-ain o's ujs-orviasals sicla seve. wieceyb ence vine Gough Be 12 Distribution of material among the various types..........eeee scene 14 Degree of variability in males and females ...............0ceeeeeees 15 Nature-of the Variation in a single nest. is si ies ss ies Se aase annie vos 16 Relation of color pattern to the time of emergence..............000- 20 Correlations between pattern of abdomen and other color characters. 21 SIDI 5ch yee OP RO ee, ing es Rear Rare ee em RITE 32 POLSIES Var TAL IS GNU TOLALEG SPECleSs «fc 5::/ajchaied ach sree wo wle Gis Sv aaa are ears ecRS 33 Polistes vartatus aud POMSIES PANTPOS. 6 vee siccc asians vind 40d ae a HOR Owes 33 POUSTES UATIALUS ANG OOLISLES A UIA Chora. c otis Oats) Olle Get uicle bee 35 General conclusions concerning relations of Polistes variatus........ 37 Warations iotner species of -POlstesi) 5 4c:odsuis es autes sis Sone etal 08 38 PNCLEOLANESTUDIPINOSHS LDC. c Siig ok kee hows a oemek Meee wl Sey ek 39 RHC CUP OLOIUS CY DE mae cyte sealers Le va ene he aiiera ts hits hee heels ness 40 MPtOGenests. OF LHe -COlOt DAGLETE a sss cis- sau as ses 5 a SO vine eee PAE owe Hw 42 Development of the pattern in Polistes variatus .............. oti al ovehetre 42 Development of the pattern in Polistes Pallipes....ccccccccccccccccecees 43 Pherred= browse SPeCies wai ce,seia1c seta his a's tainio ais cheucs oucyouette covets: dssyalataerapeiaseldie- ats ye ier 44 Physical:and chemical-nature of the pigment, 06 35.0.6 cies. sleus,s's oi aie Rees ears 45 Chemiical’examination of the pigments]. . AS B. G: We NOU retin eter 25 4o 25 be) ° Loti 2etcass psienars fo) 16 18 52 14 The distribution of males and females in the foregoing classes is ‘shown as follows: Class en ek Be ea | ee MAG ac 25 28 I | ° ° Lot 1 females, c.focc ee fo) 12 24 | Io ° MALS: ih2752 8 fe) 15 I2 | 6 re) Lot2 femaless:2s;0:4: fe) I 6 | 46 14 From this it is apparent that there is an approach to sexual dimor- phism in the markings of the metathorax. This holds true for the whole collection of over 1,200 specimens. For a given condition of tthe abdomen the metathorax of the male is invariably more melanic COLORATION IN POLISTES. 31 than that of the female. This expresses itself both in the width of the median yellow stripes and the condition of the lateral yellow stripes. As might be expected, there is close correlation between these two markings of the metathorax. The metathorax is the only part of the thorax which varies con- spicuously in P. variatus or in its two allies, P. aurifer and P. pallipes. There is an irregular spot on the mesopleura which varies slightly in size, shape, and position, its most usual form being a triangle elon- gated in an antero-dorsal and postero-ventral direction. There is also slight variation in the width of the borders of mesothorax, scutellum, and post-scutellum. In general the yellow zone of the last is about three times as broad as that of the scutellum, and, further, the yellow is slightly more conspicuous in the light-colored than in the dark-colored lots; but the differences are not important in specific relations, so they may, for the purposes of this research, be disregarded. CORRELATION BETWEEN MARKINGS OF APPENDAGES AND MARKINGS OF ABDOMEN. There is conspicuous sexual dimorphism exhibited in the ventral sur- face of the thorax and the coxe of the appendages. In the males these surfaces are predominantly yellow; the coxee have dark borders, and occasionally the most posterior pair are black or bear traces of black. In the females the corresponding surfaces are black, the coxee bor- dered with yellow, and occasionally the ventral surfaces variegated with yellow. For the dorsal surface of the coxe the conditions are reversed. In each there is an external yellow border and a yellow spot near the articulation with the trochanter. In the males the former is usually narrow, leaving the greater part of the area black. In the females it is broad, almost covering the dorsal surface. In both sexes the dark pigment occurs in the proximal end of the femur, the distal end of the tibia, and the terminal tarsal joints. The rest of the appendage is a dull, brownish straw-color. From these points the pigment spreads, so to speak, in a varying degree in the femur distally and in the tibia proximally, but in all cases the joint between the two, the knee, remains light colored. In the males, however, the color is confined to the outer (postero-dorsal) surface of the appendage. In the females, it is most extended on the inner (antero-veutral) surface of the femur, but occurs also on the postero- dorsal surface. In the tibia the males agree with the females in hav- ing the pigmental area on the outer surface. In the determination of the correlation between the light and dark character of the body and that of the appendages, the males and females are considered separately. 32 COLORATION IN POLISTES. Lot 1 (females) dark ; Lot 1 (males) dark ; Lot 2 (females) light ; Lot 2 (males) light. Class A, appendages have a large amount of dark pigment, Class B, appendages have a moderate amount of dark pigment. Class C, appendages have a small amount of dark pigment. The distribution of females and males in the classes is as follows : Females. Males. Class A Class B, Class C.} Class A] Class B.| Class C. | — a 74 7 32 Neither of these tables indicates close correlation between the pig- mentation of the body and that of the appendages. The condition of the latter tends to be mediocre, whatever the condition of the body. This is what the examination of any collection of species leads us to expect. The appendages are never conspicuously colored, and usually furnish the least important of the color differences used in the separation of the species. The pattern is always the same as that de- scribed for P. variatus and varies simply in the depth of pigmentation and its spread from the areas indicated. The foregoing examination reveals, therefore, a positive correlation between (1) the pattern of the dorsal and ventral surfaces of the abdo- men; (2) the amount of dark pigment in the abdomen and the pat- tern of the clypeus; (3) the pigmentation of the abdomen and that of the metathorax; (4) the pigmentation of the body and that of the appendages. These correlations are graphically expressed in figs. 12 to 21. The metathorax shares in the tendency toward sexual dimorphism exhibited in the ventral surfaces of the body and the appendages. We thus clearly see that the coloration of Polistes varies as a whole and along certain definite lines. Summary.—The color pattern in P. variatus, then, varies according to the varying encroachment on the yellow hypodermal areas of the darkly pigmented zones. In the given collection that variation is continuous for both males and females, but the latter adhere more closely to the typical or modal condition. Again, the types of mark- ing for the members of the individual colony appear to cluster about that possessed by the female founder of the colony, and the kind and extent of their variation from this type of marking is dependent, in a measure at least, on environal conditions. COLORATION IN POLISTES PL. Wl E. B. MEEK, DEL. HELIOTYPE CO., BOSTON. SPECIES OF POLISTES Fig. 33. P. aurifer. Fig. 34. P. pallipes. Fig. 35. P. variatus. Fig. 36. P. Carolimis. Fig. 37. P. fuscatus exilis. COLORATION IN POLISTES. 33 The color characters of the wasp vary as a whole—there is marked positive correlation between the dorsal and ventral surfaces of the abdomen and between these surfaces and all other areas exhibiting a tendency to vary. Finally, the species varies in a definite direction ; on the one hand toward the xanthic aurifer of the West, on the other toward the pallipes of the East. POLISTES VARIATUS AND RELATED SPECIES. The most marked affinities of P. variatus, as has just been indicated, are for the species pallipes and aurifer. POLISTES VARIATUS AND POLISTES PALLIPES. In the case of fallipes the trend of variation is toward the type as described by Lepeletier rather than any of the fourteen varieties given by De Saussure (Description of Species, p. 79). This is significant. P. pallipes (P1. III, fig. 34) has justly heen considered one of the most variable of the species of Polistes. Inspection of any large collection of the species easily reveals the presence of most of the varieties described and usually of some others which have not been described. Eighty specimens collected from a single patch of golden rod at Willow Grove, Pa., exhibit such great diversity of marking that it might almost be said that no two are alike. The kind and extent of this variation will be further considered in the section devoted to the geographical distri- bution of the species in the United States. The resemblances between ?. fallifes and P. variatus are closer when we consider the representatives of P. pallipes as it occurs farther north. We will therefore proceed to the study of nest variation in specimens of P. pallipes from Long Island. About 500 specimens of this species, including ontogenetic stages, were secured at Cold Spring Harbor in the summer of 1900. Our study here, however, will be confined to 200 specimens, 100 of which were collected at random and 1oo directly from the nest. Various types of color marking for the abdomen were selected and the material grouped about them. It was found that the random col- lection forms a series beginning with a form possessing on the sides of the second abdominal segment large ferruginous areas which are con- fluent with the broad ferruginous border. The yellow margins are more or less obscure, but tend to persist in the first segment. ‘This type of marking represents the most xanthic condition for the collec- tion. Inthe next stage the spots and borders gradually decrease, owing to the encroachment of the black, the narrow yellow border still vary- ingly present. From this we pass through a stage where the ferrugi- S 34 COLORATION IN POLISTES. nous is entirely absent to a large class where the yellow borders are entirely obsolete excepting in the first segment. In the maximal melanic condition the abdomen is entirely black. Accordingly we find the material distributed in the following classes: (1) Large amount of ferruginous, borders variable, 13 females; (2) Ferruginous reduced, borders variable, 40 (29 females, 11 males); (3) Ferruginous absent, yellow present in more than first segment, 8; (4) Ferruginous absent, yellow present only in first segment, 32 (31 females, 1 male) ; (5) Abdomen entirely black, 7. The collection of 100 P. pallifes from a single nest shows the same stages, but in addition a number possess a prominent lateral spot which is yellow or yellow edged with ferruginous. If wecall these two types —r1 and o respectively, and preserve the values for the remaining classes as above defined, we find the nest variations distributed among the seven classes as follows: Classes. —I fe) I 2 2: 4 5 Number of males....... 2 14 fo) BQ Ash OP Sale Save tial eeeystetegs Number of females...... fe) fe) I 8 6 26 be) Number of specimens.| 3 14 I 40 6 26 Io Both collections give curves with two modes, and the modes fall in the same classes, namely, 2 and 4. Moreover, the males again show themselves more aberrant than the females. The diversity is so great that we might think we were dealing here with two species did not the single nest give us exactly the same types of coloration as does the random collection, with an even greater range of variability. In this species again we find the ventral surface of the abdomen more melanic than the dorsal. Classes 4 and 5 are entirely melanic. Class 3 has narrow, interrupted, yellow borders on the anterior seg- ments. In class 2 ferruginous is often present in the form of a small dot at the postero-lateral angle of thesegment. In class 1 ferruginous is usually present, but always in smaller amount than on the dorsal side. Markings of the thorax.—As should be expected from the slight amount of yellow in the abdomen, we find the yellow of the thorax correspondingly reduced. The lateral spots of the metathorax are entirely absent and the median stripe greatly narrowed and sometimes obsolete. These median stripes vary somewhat in width, but in very COLORATION IN POLISTES. 35 few cases are they as wide as in the modal condition of vaviatus. In many of the more melanic specimens the metathorax is entirely black. Clypeus.—This, as always, is uniformly yellow in the male. In the female it offers all the variations from a uniform reddish ferruginous, through ferruginous with a central black spot of varying size, to an entirely black clypeus. In only 4 specimens does yellow appear, and here at the apical angle. Where the amount of ferruginous in the rest of the body is large, the clypeus is prevailingly ferruginous ; where small, the clypeus is prevailingly black. The similarity between the color markings of these two species does not necessarily indicate that they should accordingly be considered one species. This is apparent from the following considerations : Toward the south we find the varieties of P. pallipes just described merging into a lighter series of forms, which, however, differ markedly from the lighter forms of vaviatus found in Wisconsin. ‘The fact of this convergence of two distinct species toward a common form along two different lines has an important bearing on the question of the influ- ence of climatic conditions on the species and the various lines of migration which they have taken. This will be considered in detail in the following subdivisions. POLISTES VARIATUS AND POLISTES AURIFER. Related to P. variatus through its xanthic trend is the species P. aurifer, which is most sharply differentiated from the former in Cali- fornia, but whose representatives in Colorado are easily confused with those of P. variatus (see Pl. III). We have already seen how the latter species might pass into P. aurifer, and it would be interesting to learn just what degree of correspondence exists between the nest variations of the two species. Unfortunately no large series of variations for P. aurifer have been available. Through the kindness of Dr. Heath, of Leland Stanford Junior University, I am in possession of one brood of P. aurifer from San José and a number of broods of Polybia flavitarsis. The last named is a species of social wasp occasionally found in south- eastern and southwestern United States, which in its type of marking closely resembles P. aurifer. Nest variations in P, aurifer.—This series of 20 specimens of nest variations of auvifer is interesting by comparison with representatives of the species occurring to the north and south of San José. They are materially lighter than the former and darker than the latter. They also show some slight variations among themselves. The lateral spot of the second abdominal segment is always large, quadrangular, with the axis in the main axis of the body and its pos- terior edge broadly confluent with segmental border (Pl. III, fig. 33). 36 COLORATION IN POLISTES. In 12 specimens this spot is touched with ferruginous, usually on its anterior angle or antero-lateral border. The size of the spot and degree of confluence varies, the extremes being represented by one specimen where the spot reaches the anterior portion of the segment, leaving only a narrow black line between itself and the preceding seg- ment, all but meets its fellow in the middle line, and is confluent with the segmental border almost the whole of its posterior extent ; and another specimen where a black zone 2 mm. in width intervenes be- tween the lateral spot and the anterior border of the segment. This zone attains a width of 1 mm. between the spots, and further back separates them from the segmental border for a distance of 2 mm. The ventral side is marked similarly to the dorsal, but in most cases it is not so xanthic, thus illustrating the general relation already de- scribed as existing between the dorsal and ventral sides of ?. variatus. The width of the middle stripes of the metathorax also varies. ‘The lateral stripes are entirely absent in this nest, though this is not the case with all the members of this species. The clypeus here is uni- . formly yellow, only occasionally tinged with light ferruginous. An examination of a large series of P. aurifer from the West clearly shows the relation of this pattern to that already described for P. va- riatus. ‘Taking the nest variation just considered as a basis, we see that this species tends to vary in the direction of P. vaviatus. The varying size of the yellow areas, the tingeing of these areas with fer- ruginous, in a word, the same general type of correlation, is the most striking feature of this relation. The nest collection is coherent and exhibits greater unity than could easily be found to exist among any 20 specimens in a random collection. COMPARISON BETWEEN NEST COLLECTION AND COLLECTION AT RANDOM. The depth of melanism for nest collection is about midway between that observed for southern California and Oregon. In the former case the body color is predominantly yellow, the darker areas being greatly restricted and represented in brown or red-brown. In the latter case the yellow areas are greatly restricted, the darker areas spreading and becoming more nearly black in hue. Specimens from this part of the United States are identical with the type of P. variatus. All the specimens from Idaho, Utah, Montana, and Nevada which I have seen are in general similar to those from corresponding lati- tudes along the Pacific coast; but this section of the country is but meagerly represented in the general collections, and I have no doubt that more careful study would reveal a great degree of local variation for this species. COLORATION IN POLISTES. 37 Passing eastward, Colorado, Kansas, and Nebraska furnish all imag- inable transitions between the auvifer type and the typical variatus, by which is meant the modal condition for the species in Wisconsin. Specimens from southern Colorado and lower altitudes stand very close to P. aurifer. Farther north and in higher altitudes the black and ferruginous areas tend to increase until in all the specimens from Kansas and Nebraska, which in collections are variously labeled auvifer and variatus, the corresponding pattern and line could be paralleled in my series of nest variations from Wisconsin. GENERAL CONCLUSIONS CONCERNING RELATIONS OF POLISTES VARIATUS. In summing up our observations on the related species we are led to the conclusion that the genus Polistes is represented in the North Central States by the very variable species P. variatus. Its variations exhibit in the main, two trends: (1) Toward a conspicuously black and yellow form with yellow areas large and well defined; (2) toward a * black, ferruginous and yellow form, with the yellow areas largely obscured by ferruginous or black. Southwestward representatives of the species merge into forms con- sonant with the first or xanthic trend of variation ; eastward and south- eastward they pass insensibly into forms more and more ferruginous, or black and ferruginous. It is hardly probable that we have in P. variatus a primitive species which has differentiated in two directions, but, as we shall see from the study of the geographical distribution of the species, P. auvifer and P. pallipes are two originally distinct species which, from the course of their migrations northward, have come together in the Mississippi Valley, and by their commingling produced a species having in some measure the characters of both. Related to these three species are a number of species whose validity I am not prepared todeny ; but the representatives of these species, as I have studied them in the large and comprehensive collections at my disposal, have generally exhibited among themselves much slighter range of variability than were in many cases observed in the nest collections of P. variatus and P. pallipes. These species are on the one hand P. anaheimensis from California, which is plainly identical with the aurifer of southern California, and P. minor, fuscatus, fuscatus insta- bilis, fuscatus exilis, and metrica. Until further detailed study can be made on the nest variations and distribution of these species, we will consider them in the same general group with P. pallifes, which species they approach, in general, more clearly than they do P. varviatus. 38 COLORATION IN POLISTES. VARIATIONS IN OTHER SPECIES OF POLISTES. The three species just considered, together with their allies, com- prise the principal representatives of the genus in the United States. Occurring scatteringly throughout southern United States are three other types. These are, first, a prominently xanthic type, whose prin- cipal representative is P. flavus and which occurs in the arid region of Arizona and New Mexico; second, a more or less uniformly colored melanic type, represented by P. annularis and P. canadensis and scat- tered through southern United States ; and, third, a series of reddish- brown forms occurring in southern and southeastern United States, comprising a great assemblage of forms, which will be named in fol- lowing paragraphs. Concerning the first, no detailed study of individual variation has yet been made in this species. Its coloring is cinnamon brown, with large metameric spots confluent with the broad yellow borders. The few specimens examined show variation in the depth of the brown coloring and its area of distribution. Concerning the second, some doubt has been expressed by De Saussure as to the reality of the dis- tinction between the two species axnzularis and canadensis. Collections for the United States exhibit considerable variation in the extent and distinctness of their markings. They differ from each other mainly in the possession by aznularis of a prominent terminal border on the first abdominal segment ; in canadensis all the segments may have a narrow terminal border, and many transitions exist between this condition and a uniform dark fuscous or black coloring forthe abdomen. Examina- tion of any series of specimens which vary thus shows that the borders of the posterior segments tend to disappear first, and even after the borders of all the other segments have become obsolete the most ante- rior one is found to remain distinct and prominent. This tendency is particularly striking in the variable species of /ineatus and pallipes, and from this we see how such a species as aznularis might arise by the fixing of a slight individual variation. The third class, consisting of red-brown forms, comprises a great number of species occurring in the Gulf and South Atlantic States. They are especially well represented in Texas, where they have been collected and studied by Mr. E. T. Cresson. It will be convenient to separate these forms into two great groups, which differ mainly in point of size, and which we will denominate respectively as the fexanus-rubi- ginosus and the carolinus type. They will be considered separately. COLORATION IN POLISTES. 39 THE TEXANUS-RUBIGINOSUS TYPE. By this name we designate a number of large, handsome forms which pass into one another by innumerable transitions. The prevailing color varies from alight golden red toa deepred-brown. Some of the species are nearly uniform in coloring, while others are conspicuously orna- mented with yellow, and yellowand black. The ornamentation of the abdomen and face is of the same general nature as that already described, but the thorax displays several new characters. I have succeeded in securing only one colony of rudiginosus from Florida, but the series is well represented in all the general collections I have studied, and through the kindness of Messrs. Melander and Brues I am in posses- sion of a large number collected at Austin, Tex. CONSIDERATION OF VARIATIONS AT RANDOM. P. texanus is the lightest member of the series. It has a reddish-yellow color, varied with yellow in the borders and the metameric spots ; metathorax with two or four variable yellow stripes, post-scutellum all yellow or more or less obscured by yellowish red, lower side in general less yellow except in the male, where it has the yellow pattern character- istic for the male of variatus. The size and uniformity of the yellow areas varies greatly, as does also the depth of the yellowish red coloring. Occasionally black appears in the anterior part of the anterior abdominal segments. /. rubiginosus represents the uniform red-brown extremes of the series. It is the most commonly known member of the group and was first described by Lepeletier as being entirely clear reddish yellow. But in the various collections studied the name has been made to include specimens prevailingly uniform in their coloring, the tone ranging from very light to very dark reddish brown. Yellow occasionally persists in the anterior segments of the abdomen, and black appears to a varying degree in the thorax and the anterior portion of the abdominal segments. The various patterns which the red and brown may assume in the meso- metathorax are illustrated in Plate II, figs. 30, 31, and 32, and their relations will be made clear by the study of the ontogenesis of the color pattern. Between these two species as extremes occur three species, de//icosus, perplexus, and generosus, which in the order named form a series with increasing melanic tendency ; de/licosus has its markings somewhat more obscure than ¢exanus and a somewhat deeper general coloring. Ac- cording to Cresson, who had 7 specimens, it is ‘‘smaller, less robust, and darker in coloration than fexanus, to which it is closely allied.’’ The ornamentation of the abdomen is different. The yellow margins 40 COLORATION IN POLISTES. are somewhat narrowed, while both above and below the yellow lateral spot may be entirely obscured by the darker color. Black occasionally appears in the thorax in the position illustrated in Plate IV, fig. 38. P. perplexus was first described by Cresson, who remarks that it may prove to be only a male variety of vwbiginosus. In general coloration it easily passes into del/icosus. It has much more black in the meta- thorax than de//icosus, but the amount varies; yellow borders may be greatly obscured or entirely lacking, and black appears in variable amount in the anterior portion of the abdominal segments. P. generosus completes the transition to rudiginosus (Pl. IV, fig. 39). The yellow is usually entirely absent, but it may persist as a faint line bordering the prothorax and metathorax. The mesothorax varies through patterns similar to stages 22, 23, and 24, Plate II; scutellum and post scutellum are rusty red; metathorax varies through a form which is rusty red with a single median line through those with red persisting in two or four zones to a uniform black, and black is again variable in the anterior portion of the abdominal segments. NEST VARIATIONS. Six specimens of P. generosus taken from a single nest show variation in the depth of red-brown, in the prominence of the border of the first abdominal segment, and in the thoracic markings. This species tends to be uniformly very dark red-brown, but three zones in the mesothorax approach black in their depth of coloring, while in one specimen whose development was watched the degree of melanism is more marked still and is represented in Plate 2, fig. 32. From the foregoing, and especially from the nest variations just con- sidered, it is apparent that we have here a series which in the general nature of the variability of the abdomen greatly resembles its northern relatives, but we have added here another variable zone, viz, that of the mesothorax, which in its type of variation is identical with the Pennsylvania representatives of P. pallipes, and relates itself, as we shall see in the following chapter, by its ontogeny, to the color pattern in all species. THE CAROLINUS TYPE. The species included here are smaller and less robust than those be- longing to the /exanus-rubiginosus type, and embrace the very smallest representatives of the genus (PI. III, fig. 36). In coloration and type of marking they greatly resemble the foregoing, except that the yellow is never so prominent as in the /exanus extreme of the series. The variations in the random collection from Florida have been de- COLORATION IN POLISTES. 4I termined and will be considered in detail in the section devoted to the geographical distribution of the species. They are in general along the same lines as those just described for the /exanus-rubiginosus type. One collection comprising .21 specimens from a single nest (species undetermined) shows variation in general coloration from a light red- dish brown to a rich ferruginous. One specimen, a male, is conspicu- ous for the rich red coloration of the mesothorax and the second abdominal segment. The yellow borders are usually prominent, be- coming obsolete from behind forward. In one specimen they were entirely absent. The reddish-black zone of the abdomen is variable, sometimes absent in all segments, sometimes present so as to obscure wholly all the visible portions of the segments posterior to the second ; there is a shallow triangle in the second segment, which varies from a mere line somewhat dilated in the middle aspect to an area covering one-third of the segment. Median yellow stripes are usually present in the metathorax (in three specimens obsolete), but these stripes are on a background which is obscured to a variable degree with black. Six specimens from a single nest, undetermined but intermediate in size between the foregoing and rudbiginosus, show in general the same variations. In all but two specimens the yellow is entirely absent in the thorax, but is present in the metathorax in the faint border of the postscutellum. CONCLUSION. From our study of individual variation in the genus, we con- clude that all the species of Polistes are remarkably variable in their color markings; but the various species exhibit differences in their degree of variability. These variations are continuous, proceeding by insensible stages along certain definite paths. The conspicuously vary- ing region of the body differs somewhat in the different types, depend- ing primarily on the degree of melanism of the form; but in spite of this there is in all species observed marked positive correlation between the variations of the conspicuously varying region and all other variable regions of the body. 42 COLORATION IN POLISTES. ONTOGENESIS OF THE COLOR PATTERN. DEVELOPMENT OF THE PATTERN IN POLISTES VARIATUS. The pupa is at first of a uniform creamy white color, which, before any trace of the darker pigmentation appears, deepens into a flesh color. About eight days before the emergence of the wasp the first indications of the color pattern appear in the form of dull brown traces onthe tegule. Almost contemporaneously with this there appear three posteriorly converging areas in the mesothorax (Pl. IJ, fig. 19) and two faintly tinted squarish areas in the scutellum. The coloration of the mesothorax may now advance until nearly the whole surface is suffused, before pigment appears in any amount over the rest of the body (Pl. II, figs. 20, 21.) Next a dark line appears at the base of the post-scutellum, and the metathorax becomes dark- ened at three points. The first of these has the form of a J and is at the anterior end of the median longitudinal groove ; the other two are cir- cular spots at either side of this triangle. At these points the pig- ment gradually deepens and then spreads posteriorly and laterally until the whole metathorax is suffused except the zones occupied by the longitudinal stripes. It will thus be seen that the two median yellow stripes commonly present occupy positions between the middle and the two lateral pigmented areas, while the two lateral stripes occupy zones. external to the lateral pigmented areas. ; Meanwhile pigment has appeared at the base of the anterior abdom- inal segments (Pl. II, fig. 22). In the first segment the area is quad- rangular and gradually fades out posteriorly. In the second it takes the form of a shallow triangle of varying altitude. In addition, there is a series of dark metameric spots which correspond in position with the spiracular muscles, and in the second segment there is a second transverse band just posterior to the middle zone and coinciding with the zone where pigment first appears in the third segment. The pigmented areas have at first a light brownish-drab color. ‘This. gradually deepens, the areas where the colors first appeared remaining at first darker than the margins of the pigmented area. ‘The deepening hue here becomes fuscous, and finally in many cases almost ‘entirely black. In the segments posterior to the second, the pigmented triangle is shallower and its base extends entirely across the anterior aspect of the segment. In these zones the pigment gradually deepens, and from them it spreads to suffuse the whole integument excepting those zones which are to be yellow in the adult. The latter retain the original COLORATION IN POLISTES PL. IV —. B. MEEK, DEL. HELIOTYPE CO., BOSTON. SPECIES OF POLISTES Fig. 38. P. flavus. Fig. 39. P. generosus, Fig. 40. P. rubiginosus. Fig. 41. P. canadensis. Fig. 42. P. Carnifex. COLORATION IN POLISTES. 43 creamy flesh color until the pupal skin is cast and the young imago dries out preparatory to leaving its cell. The yellow then appears. uniformly over these areas and gradually deepens to the adult color. The pigmentation of the ventral side of the abdomen does not quite keep pace with that of the dorsal. Here, as on the dorsal side, the pigment first appears at the anterior edge of the segment, but the lat- eral spots are larger than on the dorsal side, and pigmentation proceeds more by the centrifugal spreading from these spotsthan by a posterior spreading from the anterior zones. Plate II, fig. 28, illustrates the condition of the ventral side in a stage already shown for the dorsal side in Plate II, fig. 24. Plate II, fig. 29 represents the corresponding condition in the male. On the head the coloration of the vertex and front keeps pace with that of the mesothorax. The first indication is in the form of a spot in the region of the ocelli (Pl. II, fig. 19). In the male the face re- mains uniformly flesh-colored until pigmented with yellow. In the female the clypeus shows a faint central stain at first, and this may spread until the whole surface is darkened. The appendages remain unpigmented even after the definite pattern of the body has been clearly indicated. Color first appears here as faint stains near the articulations. Next, brownish lines appear wher- ever the pigmentation is to be deepest in the adult, and from these regions there is a lateral spreading until the permanent color pattern is attained. In the zones where individual variation occurs we find the suffusion last to appear. For instance, in the metathorax of specimens whose markings are reduced to two we find the regions where the yellow lateral stripes usually occur to be the last to show pigmentation. The same is true in the case of the lateral spot of the second abdominal segment. This area is sometimes pigmented, and when it is, it is the very last to become so. The pigmentation is brought about by the spread of the pigment posteriorly from the original triangle and ante- riorly from the transverse zone until these two areas come together in the middle zone of the segment. There is then a centripetal spread until the area is wholly suffused. DEVELOPMENT OF THE PATTERN IN POLISTES PALLIPES. In the dark variety of P. pallipes characteristic of New England, the early developmental stages, as studied in a great many specimens, are identical with those described for P. variatus. The only difference between the two species is that the typical P. variatus stops, so to speak, at an earlier stage than does P. pallipes, which advances to the 44 COLORATION IN POLISTES. extreme melanic condition of P. variatus, aud in most cases passes far beyond it. The relation between the two is illustrated in Plate II, figs. 24 to 27. THE RED-BROWN SPECIES. The pigmentation of P. generosus, and of a similar though smaller red-brown species, was also studied. Here again there is remarkable resemblance between the steps of the process and that already described for variatus. The pupa at first is of a uniform cream color, which deepens to flesh color, and later, while the pigmentation is progress- ing, acquires a distinct orange cast (PI. II, fig. 32). The zones of the thorax where the pigment first appears are identical with those of P. variatus, with this exception, that they are more restricted laterally (Pl. II, fig. 30). Asthe process goes on, the coloration in these zones may remain deepest, and thus give rise to the pattern so usual to the thorax of these forms. In the abdomen we find the pigmentation of the anterior segments in advance of the posterior ones. It occurs in the form of an irregular surface on the first segment, a triangle with an acute apex at the anterior end of the second segment, and posterior to this a broad band. Although the pigmented zones of the following segments do not appear until a little later, they nevertheless spread rapidly enough to complete the uniform coloration by the end of the pupal stage. Sometimes, however, the deeper pigmentation of these zones persists in all, but more usually it persists only in the anterior segments, thus giv- ing rise to the red-brown abdomen characteristically varied with black. Hence we see that, aside from differences in tinge shown by the vari- ous types, the origin and steps of the process of coloration are essen- tially the same in widely differing species. There is good evidence for that persistence in the adults of different species, of different ontoge- netic states which plays so important a part in the coloration theory of Eimer. But this at best would be but a partial explanation of the phenomenon. I have therefore attempted to see whether I could not, by an histological and chemical examination of the pigment, arrive at any causal relation between the resulting color pattern and the condi- tions under which pigmentation takes place. COLORATION IN POLISTES. 45 PHYSICAL AND CHEMICAL NATURE OF THE PIGMENT. Reference has already been made to certain differences existing between the yellow and the darker pigments of the exoskeleton of Polistes. The latter are incorporated in the outer layers of the chiti- nous integument of the insect, the underlying or embryonic cuticle remaining unpigmented. ‘The former is deposited in and among the hypodermal cells. (See text figs. 5 and 6.) This yellow is not evenly distributed all over the inner surface of the chitin, but is confined to the areas where the chitin has remained in any degree transparent. There is also a relation between the thick- ness of the yellow hypodermal layer and the depth of pigmentation of the chitin overlying it; the more transparent the chitin the thicker the deposit of hypodermal yellow. Thus in the modal condition of P. variatus, where the lateral abdominal spot is yellow with a ferrugi- nous margin, we find the hypodermal yellow thickest under the yellow area of the spot. The ferruginous effect is due to a slight deposit of the dark pigment in the chitin, leaving it more or less translucent and allowing a thin layer of the underlying yellow to show through. At the margin of the ferruginous area the yellow gradually thins out, and, finally, where the cuticular pigment is dense, disappears entirely. The more deeply pigmented areas of the chitin have a distinct rela- tion to underlying structures ; they represent places of attachment of the muscles to the integument. This is better expressed by saying that the areas where the pigment first appears in the pupa represent the position and attachment of developing muscles in the manner already described by Tower (23) for the Coleoptera. In the thorax of the pupa there are three great bundles of muscles. One is median and longitudinal in its direction, and on either side of this there is a large lateral bundle which lies in a dorso-ventral direc- tion. The first is inserted anteriorly in the mesothorax just below the median pigmented region, the other two at either side below the lateral ' pigmented areas. The posterior end of the median longitudinal muscle is attached below the three primitively pigmented areas of the meta- thorax ; some fibers are also inserted in the scutellum. In the abdomen the powerful muscles are attached along the anterior edge of the respective segments. An extra large and powerful muscle, which is used in the movement of the abdomen as a whole, is inserted below the triangular area which so constantly appears on the surface of the second abdominal segment. The segmental dots also probably represent the position of the spiracular muscles. 46 COLORATION IN POLISTES. In the appendages, again, the position of the pigmented areas is inva- riable, though there are differences in the extent of these areas. They always mark, however, the attachment of the leg muscles. . Microscopic examination of the chitin shows it to be striated in two directions and perforated at intervals with openings through which pass spines. These latter have protoplasmic cores which communicate directly with the nerves of the insect ; they thus probably function as tactile organs. The transverse striations represent the various stages of growth of the chitin. This is generally supposed to be an exudation from the hypodermal cells, and to develop by successive depositions on the out- side of the chitin already there (see Bibliography, 16 and 23). ‘These transverse striations are usually more or less wavy and suggest a flow of material which is in a somewhat plastic condition. The perpendic- ular strictions which sometimes appear in the cross-sections of dry chitin are thought to represent minute perforations in the chitin, which in the adult are empty or filled with air. Their significance has been hitherto unknown, unless they aid in the interchange of gasesintheinsect. The following description of their relations to other structures in a pupa of P. variatus, together with fig. 5, may suggest something further con- cerning the functions of these perforations. This section is through the anterior abdominal segment, whose sur- face has already attained a considerable degree of pigmentation. ‘The chitin is thick, and shows very decided transverse striations. "The hypo- dermal layer is several cells in thickness, but interposed between it and the chitin is a structureless zone, which nevertheless stains deeply in hematoxylin. This deeply stained substance may be traced along the numerous exceedingly fine perpendicular striations to within a short distance of the deeply pigmented surface. From this it will appear that these striations do in reality represent very numerous fine pores, through which is exuded a substance which has been in relation with, and per- haps elaborated by means of the hypodermal cells. Fig. 6 represents a section through the mesothorax of the pupa. From this it will be seen that the pigmented chitin presents a coarsely granular appearance, which gradually disappears as we pass toward the less deeply pigmented portion at the right of the section. In propor- tion as the granular character disappears, a striated condition takes its place. Again, the original granular condition of the pigment is largely lost in the adult chitin, although it is here plainly indicated by the irregularity of the surface and striations. I interpret these appearances in the following manner: The pig- ‘mented chitin is produced at various points of the surface in the form COLORATION IN POLISTES. 47 of large oval granules. These by their disintegration gradually spread over contiguous areas, the pigment meanwhile diffusing into the sur- rounding unpigmented chitin which has likewise been deposited from the hypodermal cells. From the relation existing between the pigmented areas and the muscles, it is evident that the pigmentation is influenced by the meta- bolic process accompanying the development of muscle. Needham (18) considers that the nuclei of the disintegrating cells of the fat body pass directly over into the nuclei of the developing muscles. Nothing is definitely known concerning the nature of the chemical process involved here; but it would seem probable that in such a region substances would be produced which, through the medium of the hemolymph, would affect the action of the hypodermal cells, modifying the product of their activities, and at the same time performing the function of excretion for the organism. CHEMICAL, EXAMINATION OF THE PIGMENT. Chitin has long been known as one of the most resistant of organic compounds. It withstands boiling in alcohol, ether, alkalies, and dilute acids; consequently the colors which it contains are extremely difficult of isolation and analysis. The strong acids, such as HNO,, H,SO,, and aqua regia, which dissolve the chitin, completely oxidize the pigments and render the determination of their nature entirely out of the question. The color changes, however, which the pigmented chitin undergoes when boiled with acids and alkalies are extremely suggestive and give some clue, I think, to the probable constitution of these pigments. If the chitinous integument of P. variatus be boiled with yellow nitric acid or aqua regia, it becomes oxidized, and meanwhile passes through the following series of color changes: The dark areas become pro- gressively lighter, passing through deep red-brown, red-brown, orange- red, orange, orange-yellow, and pale yellow. Finally, just before total decomposition, the chitin becomes perfectly transparent and colorless. Treatment of this acid-solution of chitin and pigment with am- monia throws down an orange-brown crystalline precipitate. If, how- ever, the transparent areas of chitin be similarly treated, no color changes will be produced. ‘Treatment of a solution of such areas with ammonia produces a transparent precipitate. When this process of oxidation is interrupted at any stage and the mass thoroughly washed in water, the color of the stage persists in- definitely. Further, when the process of oxidation is interrupted at any stage, as just described, and the mass is treated with ammonia or 48 COLORATION IN POLISTES. some other alkali, the reverse series of color changes sets in, the color passing through increasingly darker shades until the original blackish hue is again reached. | Boiling with nitric acid any of the species in which the pigment is lighter in shade than it isin P. vaviatus (such as P. flavus, where the pat- tern is in varying shades of cinnamon brown and yellow, or P. ineatus, which is prevailingly red-brown) causes the same succession of colors, beginning in every case at the stage corresponding with the adult coloration. Thus the pale brown chitin of favus passes through yel- low and pale yellow before becoming transparent, and the red-brown /ineatus through reddish orange, orange, yellow, and pale yellow. Conversely, if the light cinnamon-brown skeleton of P. flavus be boiled in potassic hydrate, the pigmented areas approach the dull- brown coloration characteristic for the more melanic species ; but the color is not so dark, since the pigment in ?. flavus is less concentrated. Similarly the light red-brown areas of P. dineatus may be changed to dark red-brown and finally to black by boiling with ammonium sulphide. Tests were also made on the nature of the yellow hypodermal pig- ment. If this layer of pigment be scraped off and boiled with dilute potash it becomes dull brown in color. If dried skeletons of wasps whose color pattern is varied with yellow be boiled with (NH,) 2S and then carefully dried again, the areas which were originally yellow will be found to have changed to a deep red-brown. The foregoing reactions would indicate that the various colors dis- played by Polistes are related to one another by slight differences in chemical composition; that the darker ones are derived from the lighter by a process of reduction, and the lighter from the darker by a process of oxidation. Further, the ontogenetic stages would indicate some such relation. In P. variatus I have sometimes observed a uniform pale yellow con- dition before the darker pigment began to appear in its characteristic areas, and P. rudbiginosus passes through a yellow, orange, and reddish- orange stage before assuming its final red-brown color. The manner in which the pigment is deposited in the chitinous lay- ers bears out this idea. I have as yet made no attempt to derive the yellow hypodermal pigment from the pupal hemolymph; but it is undoubtedly so derived, though probably indirectly, through the inter- vention of the hypodermal cells. Moreover, it has been shown by Pap- penheim (19) that in all series of color compounds yellow has always the simplest chemical composition, and that the color deepens through orange to brown and black, or through green to red and violet with COLORATION IN POLISTES. 49 an increase in molecular complexity. It might therefore easily be the case that the chitin elaborated in the neighborhood of certain develop- ing structures is influenced by the metabolism of these structures, so as to vary slightly in its state of oxidation or reduction. So far, however, we have been able to learn nothing as to the nature of the color compounds of this genus or of their possible relations to known organic colorcompounds. A clue to this is, however, furnished by the well-known glacial acetic acid test. If a mixture of one part sulphuric acid and four parts glacial acetic acid be added to a fatty substance containing any of the yellow azo- compounds, the whole heated to boiling, and well shaken, the acid mixture will develop a beautiful deep wine color. This is a reliable test for the minutest traces of the yellow azo-colors. When a little of the hypodermal yellow of Polistes was treated in the way just described, the characteristic color was well developed. The same was true when the pupa of P. generosus in its orange-yellow stage was similarly treated. Both colors also gave the sulphuric acid test for azo-colors; that is, the solution of the pigment in HCl or H,SO, acquired a rosy tinge upon the addition of NH,OH, and this deepened when further diluted with water. The azo-color compounds are a group belonging to the benzene series and exhibiting various shades of yellow, orange, brown, red, scarlet, and indigo. Benzene azo-benzene, C,H;N,C,H,, is regarded as the pro- totype of the group. Compounds of the type are produced by the action of mild reducing agents, such as alcoholic potash, on the cor- responding nitro-bodies. 2C,H,NO, + 4H, = C,H, N,C,H, + 4H,0. From this by various other processes of nitration and reduction the other members of the series may be derived. Of the basic primary azo-compounds, amido azo-benzene, chrysoidin, and phenylen brown show a gradation in shade from yellow through orange to brown. Among the acid azo-colors a regular gradation of shade is alsoshown. With the increase in molecular weight, the lowest members of the series are orange, the highest scarlet of an increasing shade of blueness. Primary azo-colors, as a rule, dissolve in strong sulphuric acid with a red or orange color, while secondary azo-colors, as a rule, give solutions a violet, blue, or green color. We will now examine more closely the series represented by aniline yellow, chrysoidin, and phenylen brown. These are among the best and earliest known azo-compounds. The first is known chemically as amido azo-benzene and is repre- 4 50 COLORATION IN POLISTES. sented by the formula C,H,;N,C,H,NH, ; it is benzene azo-benzene, in which one hydrogen atom has been replaced by the group NH,. It is very slowly soluble in hot water ; moderately so in alcohol and ether. Its salts are decomposed by water and dye wool yellow. Chrysoidin, the second member of the series, is known as diamido- azo-benzene and is represented by the formula C,H,N,C,H,(NH,),. It may be regarded as derived from the foregoing by the substitution of another amido group for the H atom. It isa reddish orange amor- phous powder very slowly soluble in water, yielding an orange solution which turns red upon addition of HCl. Itis reduced at 150° to aniline and tri-amido-benzene, which is the third member of the series, or phenylen brown. Phenylen brown is a brown amorphous powder represented by the formula C,H,N,C,H,(NH,),. It is one of the constituents of the well- known dye Bismarck brown. This may also be derived from benzene- azo-benzene by nitration and reduction. A large number of tests were made upon these substances and par- allel changes noted in the pigments of Polistes. Thus phenylen brown, when treated with yellow nitric acid or aqua regia, passes through exactly the same color changes as does the brown pigmented chitin, and the yellow solution finally obtained upon addition of NH,OH | yielded an orange-brown crystalline substance. ‘The orange chrysoidin passes through similar changes. Conversely, the aniline yellow upon boiling with (NH,)2S dissolves and yields, upon cooling, a light red- brown powder. Further treatment with the same reagent gives a powder of a deeper tint. Chrysoidin may be darkened in the same manner. In fact, all the pigments of Polistes which were examined show great similarity in their chemical reactions with the compounds of this series. A spectroscopic examination was made of the pigments in question. If the pigmented integument of Polistes be boiled in concentrated H,SO, or HCl, the greater part of it will pass into solution. If now this solution be decanted off and diluted with water, it is said that the chitin is precipitated, leaving the pigment in the solution. When the solution is thus diluted, a very dark brown precipitate is thrown down, and the liquid retained is of a yellowish-brown color. I do not pre- sume to know what actually happens in this case, but since an equal amount of transparent or very slightly pigmented chitin (it is difficult to obtain absolutely transparent chitin), when boiled in the acid and diluted, throws down a similar black residue and yields a solution which is very much lighter in color, I think it safe to suppose that we have in the dilute acid mentioned above, pigment in solution. COLORATION IN POLISTES. 51 Bismarck brown and vesuvin brown were also treated with concen- trated H,SO,. In neither case did the dye wholly dissolve. Dilution produced a rich golden-red solution, threw ‘down a deep red-brown residue, and finally turned the solution a light reddish yellow. As the solution was further diluted, it approached the color of the solutions of Polistes already described, until a stage was reached where the color was identical with the solution of dark chitin of P. rubiginosus in H,SQ,. I pulverized the skeletons of two specimens of P. “meatus in a mor- tar and afterward treated them with dilute HCl. The pieces of chitin remained intact, but a colored solution was obtained which was similar to a solution of Bismarck brown in HCl. Following is a comparison of the results obtained by the spectro- scopic analysis of the pigments of Polistes and the yellow and brown azo-compounds. Solution of dark P. rubiginosus in dilute H,SO, cuts off almost all the trays from the violet to the blue end of the spectrum, leaving only a slight thin line near the green. There is an indication of absorption bands in the yellow. Solution of Bismarck brown in H,SO,, same degree of dilution, cuts off same end of spectrum and to exactly the same degree. The absorption band in the yellow is more marked. The solution is also slightly more orange yellow than that of the wasp pigment, which inclines slightly to brownish. Solution of the pigment of P. /ineatus in dilute HCl cuts off a zone extending from the violet to the blue end of the spectrum, also the red end, and gives a marked absorption band between the orange and the yellow zones. Solution of Bismarck brown in HCl is yellower than that of /ineatus, cuts off the ends of the spectrum to the same degree, and has an absorption band in the yellow. Solution of P. vari- atus in H,SO, is intermediate in orange-yellow tint between the solution of P. rubiginosus and Bismarck brown in H,SO, and gives a spectrum similar to that belonging to these two. A further series of tests was made both upon the yellow hypodermal color of the adult and the developing colors of the pupa, but no reac- tion was obtained that indicated these to be proteid in their nature or related to uric acid, as has been found to be the case in Pieridie (Hop- kins, 13). Boiling in moderately concentrated HNO, did not change the color of the yellow hypodermal pigments. This treatment imparts to proteid a yellow flocculent appearance, and subsequent treatment of both the proteid and the yellow pigment in HNO, with NH,OH gives them a tich orange color. This is called the xantho-proteid reaction. 52 COLORATION IN POLISTES. The similarity between the two reactions with NH,OH would point in the direction of the derivation of the pigment from proteid. It is possible that the partial oxidation occasioned by treating the proteid with HNO, and thus producing the yellow flakes is analogous to the process by which the yellow pigment is derived from the protoplasm. Similarly the further reaction by NH,OH may be paralleled in the processes of reduction by which the darker pigments are derived. There is no reason why compounds of the benzene series should not occur in these pigments, and especially if these are of the nature of excretory products; for phenol (C,H,OH) is a normal constituent of the urine. Other aromatic compounds occurring in the body and con- sidered to be products of proteid decomposition are indol or benzo-pyrol (C,H,N) and skatol or B-methyl-indol (C,H,CH,NH). These undergo union with the oxy-group, and are further paired with H,SO, before being excreted as ethereal sulphate and ethereal skatoxyl sulphuric acid. Chitin (C,,H,,N,O,,) itself is considered to be an amido-derivative of a carbohydrate, with the formula 2(C,,H,,0O,,). The reducing agent is anitrogenous body, glucosamine, which is an amido derivative of grape sugar (C,H,,O,— OH+NH, = C,H,,NO,;+ 3H). If the pigments occurring in Polistes are related to one another by so small a difference as the addition of one or two NH, groups, and the hypodermal pigment is an elaboration from the protoplasm of the hypo- dermal cells, it may readily be imagined how slight differences in metabolism in certain regions of the body, combined, perhaps, with certain external influences, such as humidity and temperature, may produce the range of variation in color and amount observed in the darker pigments of these species. COMPARISON WITH THE COLOR PHENOMENA OF BUTTERFLIES. The succession of colors here is the same as that observed in the developing color pattern of butterflies (Mayer, 17). The first colors to appear in the ontogeny and the phylogeny of this group are the yellows and drabs, which are also the simplest in molecular constitu- tion ; the last to appear are the more complex blues and purples. All. of these colors are displayed by the various azo-compounds. ‘The reac- tions which Mayer gives suggest similarities with these compounds. If such a relation does exist, we are perhaps nearer an explanation of the striking phenomena of coloration observable here than we have hereto- fore suspected. ———— ee eee ee re. eee, oes iw] ahh Ab: G8 68 &S LG 10! 2S0! 60! ofl! fll T 19 T T T a 3 2 T ee ps T ‘o so — AW ? 00g OOF O0& 00% OOLOS O L SO]MI JO aTBOG “ea a ee saystjod jo uoynquysip burye.ysnytl SGUIVIS GHLINI AHL 40 dvW Sec sia aa 7 Bn Vaae stip jnuuy snavjy - snsourbiqny seujourg snyeuea = Sadie JOpUNy "A s2 5 @ [© o) (+) BPs] ( .*) See] Sec) Fn) | here? ' a | at ee Ves ne aNa9a1 721 " Gzl zl ‘A Blvd S31LSI10d NI NOILVYO1O9 COLORATION IN POLISTES. 53 GEOGRAPHICAL DISTRIBUTION OF TYPES OF COLOR MARKING. DISTRIBUTION IN THE UNITED STATES. THE TYPES. The species represented in the United States are so numerous and varied in their minute details and the characterizations so difficult to follow that for the purposes of this description the various species will be grouped under a number of heads according to affinities determined in the study of individual variation. For the detailed description of the species the reader is referred to the systematic treatment of the genus given at the close of this paper. Each group is named from its best known and most widely distributed representative, wherever this could be determined. The grouping is as follows: 1. Annularis : annularis, canadensis, comanchus, navajoe, apachus, carnifex. 2. Aurifer: aurifer, anaheimensis, variatus. 3. Pallipes : pallipes, metrica, nestor, fuscatus, fuscatus exilis, fuscatus instabilts. 4. Carolinus : carolinus, americana, crinitus, crinitus lineatis. 5. Rubiginosus : rubiginosus, perplexus, generosus, bellicosus, texanus, flavus. The annularis group runs parallel in its general coloration with the pallipes and aurifer groups. It is separated from them mainly by differences in size, although in this matter the smaller representatives of canadensis easily pass into the large representatives of jpallipes. Moreover, the larger members of the azzularis group tend to be less conspicuously colored than in general do the pallifes and aurifer groups. The conspicuous yellow or yellow and brown spots are uniformly absent, and the borders if broad are not so bright a yellow. The group, then, is characterized by great size and dull brown or brown and black coloring ; yellow is present in the borders, but restricted or often slightly suffused by the darker pigment. Thus in canadensis all the borders may be lacking or extremely narrow, while in anxnularis the border of the first abdominal segment alone persists as a more or less conspicuous ring. In the remaining members of the group the borders are broad, but suffused and darkened by the cuticular pigment. The aurvifer group is intermediate in size and coloring. It is black or dull brown, with conspicuous yellow borders and metameric spots, which blend to form large areas, especially in the more xanthic mem- bers of the species aurifer and anaheimensis. The metathorax is COLORATION IN POLISTES. e marked with two or four parallel yellow stripes. Variatus is the most melanic member of the group and connects it with the pal/zpes group. The pfallipes group includes all those melanic species of intermediate or smaller size whose ornamentation is a rich ferruginous, more or T 18 vi ng 6 FIG. 18.—Relation between latitude and melanism, second abdominal segment taken as measure of melanism. Lot I. 100 specimens collected at random near Gotha, Florida. Lot 2. 100 specimens collected at random near Willow Grove, Pa. Lot 3. 100 specimens collected at random near Cold Spring Harbor, Long Island, N. Y. Segment light red-brown with a yellow margin. Segment light red-brown, yellow margin obscure. Segment red-brown, with black triangle at base. Segment fuscous (nearly black), red-brown in the form of large oval spots. Class 4. Segment fuscous (nearly black), red-brown spots becoming obsolete, Class 5. Segment black, some yellow still present at the margin. Class 6. Segment entirely black. Broken line represents distribution of material from Gotha, Fla.; continuous line, distribution of material from Willow Grove, Pa.; dotted line, distribution of material from Cold Spring Har- Class o. Class 1. Class 2. Class 3. bor, N. Y. One square equals ten specimens, less varied with yellow. In pattern it displays all the variation dis- played in the two preceding groups, except that ferruginous tends everywhere to displace the yellow. The less melanic forms occurring in the Middle Atlantic States connect themselves readily with the more melanic members of the following group. COLORATION IN POLISTES. 55 In the cavolinus group coloration is conspicuously reddish brown. Yellow appears in the borders, but rarely in the metameric spots, and black, if present, is confined to the ventral surface of the female and limited areas in the mesothorax and metathorax and anterior part of the abdominal segments. In range of pattern it corresponds very closely with the darker members of the rudzginosus group, from which it is separated mainly by difference in size. The rudiginosus group comprises the largest members of the genus, and in pattern presents all transitions from the light reddish-yellow texanus, with conspicuous yellow ornamentation, to the uniformly dark red-brown rudiginosus. It is noticeable that here, as in the canadensis group, the largest members tend to be least conspicuously colored ; f#avus, which perhaps stands more alone than any other species, con- nects the most xanthic members of this group with those belonging to the aurvifer type. DISTRIBUTION OF TYPES. Plate V represents the distribution of these types in a map of the United States. From this it will be seen that the aurizer type occurs throughout western United States from Vancouver to Lower California. Eastward and westward it is replaced by variatus, which is the pre- vailing species for the Plains and the States just east of the Mississippi River. A line passing through Texas, Colorado, Utah, and Oregon roughly marks the transition zone of these two species. The pallipes group is well represented through all the eastern portion of the United States from Maine to Georgia and from the Atlantic Ocean to the Mississippi River. The aznularis group is scattered throughout southern United States, the more melanic members, such as canadensis and annularis, being best represented in the Gulf States; the less melanic, such as zavajoe and comanchus, in Arizona and New Mexico. The carolinus group occurs in the Gulf and the South Atlantic States as far north as North Carolina. P. rubiginosus reaches its highest devel- opment, both in numbers and species, in Texas. It occurs also in Florida and the other Gulf States and scatteringly as far north as Penn- sylvania, southern Ohio and Indiana, Kansas, Colorado, and westward to Arizona and New Mexico. In the last two States this group is rep- resented almost entirely by the xanthic flavus, which is to be regarded as the xanthic extreme of the series, while in Kentucky, Ohio, and Illinois it is represented by the dark rudiginosus, the melanic extreme of the same series. The most striking feature in this distribution is the increase in mel- 56 COLORATION IN POLISTES. anism as we go northward along the Alantic and the Pacific coasts. On the Atlantic side the light red-brown cavolinus type passes insen- sibly into the lightest members of the fad/ifes group, and this group dis- plays an increasing melanism as we proceed, reaching its culmination in the uniform dark-colored fallipes of New England. On the Pacific side the tendency is illustrated in the increase in the amount of black in the species aurifer, which finally in Washington merges into the variety of variatus characterized by narrow yellow borders and small metameric spots. This melanic tendency is discussed in detail in the following sections. RELATION BETWEEN LATITUDE AND AMOUNT OF DARK PIGMENT IN POLISTES. In this determination use was made of three lots of 1co specimens each, collected at random at Gotha, Fla.; Willow Grove, Pa., and Cold Spring Harbor, Long Island, N. Y. ‘This material, which probably represents three or four different species, falls into the fol- lowing classes with respect to the second abdominal segment : Class 0, segment light red-brown with a yellow margin. Class 1, segment light red-brown, yellow margin obscure. Class 2, segment red-brown with black triangle at base. Class 3, segment fuscous (nearly black), red-brown in the form of large oval spots. Class 4, segment fuscous (nearly black), red-brown spots becoming obsolete. Class 5, segment black, some yellow still present at the margin. Class 6, segment entirely black. The distribution of the material from the three localities in these classes is as follows: EIOPOAG 55 wees es | hack 10 82 4 Tis (Pesteveneel eenta Pennsylvaniays, 2: |53 10 6 27 56 8 BO ean Ons lat Gea cra Pema cale aeeers 2 21 39 36 2 The curves on fig. 18 represent the condition graphically. From . this it is apparent that there is a steady increase in the proportion of melanic specimens as we pass northward. The change does not affect all the specimens equally, but the mode is shifted one stage toward the extreme of melanism for each collection from Florida northward. COLORATION IN POLISTES. o7 CORRELATION BETWEEN PIGMENTATION OF SECOND ABDOMINAL SEGMENT AND THE SEGMENTS POSTERIOR THERETO. In general, the melanism in the more posterior segments advances at least one stage over that of the second segment—that is, when a specimen would be put in class o, with reference to the second segment, reference to the remaining segments would place it in class 1; the yellow margin would be obsolete; when the second segment belongs 19 : 3 4 5 6 Ie. 19.—Correlation between pigmentation of the second abdominal segment and that of the metathorax ; Ioo specimens from Cold Spring Herbor, N. Y, Classes for the abdom- inalsegment same as in determination figured in fig. 18. Classes for metathorax: 1. Red-brown, with or without parallel yellow stripes. 2. Red-brown, with black in the median furrow and perhaps in two points at the anterior end of the red zones which border the furrow laterally. 3. Black and red-brown about equalin amount. This condition is brought about by thespread of the black from the points above indicated. 4. Black extended to obscure all but two red-brown lines, which occasionally are tinged with 5. Entirely black, with occasional slight traces of yellow in the zone of median stripes. yellow. Distribution of material with respect to second abdominal segment represented by continuous line; distribution of material with respect to metathorax represented by dotted line. to class 1 the remaining ones belong to class 2—that is, they havea line of pigment at the base. In class 2 the remaining segments have a broader band or a deeper black triangle than has the second. They rarely have the red-brown in the form of a large oval spot. This condition is represented graphically in figs. 27, 24, and 25. In fig. 27 the curve for the second segment of the Long Island material 58 COLORATION IN POLISTES. is transferred from fig. 18, and the corresponding curve constructed from the data for the remaining segment is expressed alongside. Figs. 24 and 25 illustrate a similar relation in the specimens from Pennsyl- vania and Florida. The coloration in the thorax varies along the general lines already laid down for the red-brown and fallifes types. Certain regions appear to be more stable than others. Thus the prothorax is prevailingly red-_ brown for classes 0, 1, and 2, and prevailingly black in classes 4, 5, 6, and 7, class 4 showing in several cases a limited red-brown area or a slight red-brown stain on the posterior dorsal border. Class 3 is em- 20 cd ‘ . \ ‘ . ¥ / ’ . s Ss mC) > oe re -: “ 2 ay oe uk 2 3 4 5 6 1 h r j Fic. 20.—Expressing similar relation as in fig. 19 for material from Willow Grove, Pa. Classes the same as in figs. 18 and I9. phatically transitional in the females, being all black or all red-brown, or having the two colors in varying proportions, while the males show the usual yellow, with red-brown or black or both. The same tendency is expressed in the mesothorax. It is prevailingly red-brown or red-brown with a slight median black line in classes 0, 1, and 2, and, with one exception, is entirely black in classes 4, 5, and 6. Pennsylvania, however,.forms an exception to the condition for class 2. Here the mesothorax is often variegated in the fashion already described at length for the zexanus-rubiginosus type. Inclass 3 the mesothorax is either all red, or all black or variegated in the manner just described. The scutellum and post-scutellum also show variation in pattern, but for the purposes of this comparison we will confine attention to the variations in the metathorax. COLORATION IN POLISTES. 59 With respect to the pattern of the metathorax, the specimens fall into five classes : Class 1. Red-brown, with or without parallel yellow stripes. Class 2. Red-brown, with black in the median furrow, and perhaps in two points at the anterior end of the red zones which border the furrow laterally. Class 3. Black and red-brown about equalin amount. This condi- tion is brought about by the spread of the black from the points above indicated. ; : t | H ! ; Z K \ wy i “ Gis teen fc eee ea Ce wos en x 1 2 3 4 Len for) i Fic. 21.—Expresses similar relation as in fig. 19 for material from Gotha, Fla. Class 4. Black extended to obscure all but two red-brown lines, which occasionally are tinged with yellow. Class 5. Entirely black, with occasional slight traces of yellow in the zone of the median stripes. The distribution of material in these classes is shown in the following table. Classes according to the second abdominal segment are arranged horizontally, class 6 being included with 1, and o with 5, according to the metathorax, vertically. 60 COLORATION IN POLISTES. Class. I 2 3. 4 5 Mea raptopie tees ate elena Ge ec ia euall ara akeuera lias sesieleueatieers 7 ear ee ere OS ON IER II 69 Ads etait lara tacks Bia, eo iare stat hione cis cisterns ° 21 4 5 | eer yee RCE CHA AYE I 16 27 y ee ere SP casas sg kane oispelelete cae ioe I 521 540 40 41 POtAE Ss ats “one se 19 LUE | wor 48 41 From Florida......... 12 OA taas as i fe) From Pennsylvania... . 7 25 | 58 8 3 From Long Island..... ° oR bo 39 38 CORRELATION BETWEEN MARKINGS OF SECOND ABDOMINAL SEGMENT AND CLYPEUS OF FEMALE. With respect to coloration of the clypeus, the material falls into five classes : Class 1. Red-brown with prominent yellow border. Class 2. Red-brown with yellow persisting occasionally at tip. Class 3. Red-brown with black spot or dot. Class 4. Black and red-brown in about equal amounts. Class 5. Entirely black. Classes according to the markings of clypeus are arranged vertically ; those according to the markings of abdomen are arranged horizontally. Class. Ie 2 a: 4. 5. 1 ee ED a 3 2 peed es err onl Wes 7 Si CE OE ero 6 7 Arc} erate al apcrsiees Bins fe iocitd a's, Ghasapaicey tien stetiate Jail sve le severe 62 19 I2 3 Banas te arerey shacale a tive 6 ea ea | ec afononete)| Banepa ters 14 13 Be) [RSAC EES vers ec rea 2 I2 22 } POCAL, So ckceeiies 9 71 41 37 35 ba foc g(t: Wa aeea Seng Ara 5 66 2 bey Lae oa Pennsylvania.......... 4 5 23 Ae Rene Tonge lsishd peo oc Wisc non le aes | 16 32 35 COLORATION IN POLISTES. 61 From these determinations it is apparent that a positive correlation exists between the degree of melanism of the abdomen and that of the thorax and the clypeus. The curves shown in figs. 22, 23, and 26 represent these relations between the clypeus and abdomen graphically. This same tendency toward melanism in the northern forms is observed in the rubiginosus group. Here the lighter species, such as texanus and bellicosus, predominate in the narrow coastal region, while the group, so far as it is found to the northward, is represented by the dark uniformly colored rudiginosus. Just as striking, perhaps, as the relation between melanism and lati- tude, is the prevalence in the arid regions of Arizona and New Mexico of the yellow species, P. flavus and P. aurifer, and even P. navajoe. 2 i . ‘ . 7. a . bo = . oS S | | | FIG. 22.—Relation between melanism of second abdominal segment and Cclypeus of female. Cold Spring Harbor collection. Fic. 23.—Relation between melanism of second abdominal segment and clypeus of female. Willow Grove collection. These three species stand in the relation of albinic extremes to the three groups—rudbiginosus, aurifer, and canadensis—and appear to have been produced by similar conditions in these three different groups of the genus. Finally, I have been impressed with the mixed condition of the species for the upper half of the Mississippi Valley. The Missis- sippi Valley is emphatically a transition zone between the aurifer type of the West and the pad/ifes type of the East. Specimens gathered in this valley fall into the species aurifer and variatus or pallipes and metrica, according as they incline to the western or the eastern type of coloring ; but, as I have shown (pp. 12-33), representatives of both types are repeatedly obtained from the same nest; also in different parts of this area the species vary in a fashion parallel to their vari- ation in other parts of the country. Thus the divergence in southern Illinois is, on the one hand, toward the aurifer type with a large 62 COLORATION IN POLISTES. amount of yellow, and, on the other, toward the rufous representatives of the pallipes type characteristic of the Middle Atlantic States, while in southern Wisconsin the divergence is toward the more melanic forms of both the aurifer and pallifes types, characteristic of Oregon and the New England States. At Rockford (north central Illinois) the con- ditions are intermediate between the two just described. I suggest that the name /. vaviatus be retained for all these forms, and regard the species as having been derived by the hybridization of two originally distinct lines of development of southern origin, which in the course of their migrations along the eastern and western sides no or a) | | : ia 3 1 2 4 5 Fic. 24.—Relation between melanism of second abdominal segment and remaining segments similarly expressed for 100 specimens from Gotha, Fla. Fic. 25.—Similar relation for material from Willow Grove, Pa. Classes same as in figs. 22 and 23. of the continent have come together in the Mississippi Valley. This view is in accordance with the opinion of T. D. A. Cockerell: The conclusion is justified that the central region fauna was practically stamped out during the glacial epoch, and that the present fauna is derived from the boreal faunz which survived to the east and to the west and the southern fauna which survived in Mexico. This view seems to be supported by a consideration of the present distribution of species as well as by geological evidence. This, in the main, accords with the lines of diffusion as far as made out for other orders of insects. Webster (28), in treating of the dis- tribution of the chinch bug, makes out two principal lines of diffusion, one along the Atlantic seaboard and thence inland, the other up the 63 COLORATION IN POLISTES. A third minor trend is Mississippi Valley from the Gulf of Mexico. along the Pacific coast. He also states it to be his opinion that this same course of migration has been taken by various other species of Coleoptera and Lepidoptera. 27 26 Ce as Troan 1m ‘ ‘ G ‘ é ‘ ’ ’ te oO} {Q- ~*~ . . MS ‘ . . . . oe ee eee Fic. 26.—Correlation between pigmentation of second abdominial segment and clypeus of females in collection from Gotha, Fla. Classes o and 1, fig. 24, are combined to form Class 1, Classes 5 and 6 are combined to form Class 5, and distribution of material in these five classes is represented by the continuous line. With respect to markings of the celypeus material falls into the following classes : 1. Red-brown with prominent yellow border. 2. Red-brown, yellow persisting occasionally at tip. 3. Red-brown, with black spot on dot. 4. Black or red-brown in about equal amounts. 5. Entirely black. Distribution of material represented by dotted line. FIG. 27.—Showing correlation between melanism of second abdominal segment and remaining segments for collection from Cold Spring Harbor, New York. (See figs. 24 and 25.) DISTRIBUTION IN EUROPE. In Europe some entomologists recognize four species of Polistes— 1. €., biglumis, pictior, gallicus, and diadema ; others recognize only the D. Sichel, whose premature death prevented the publica- two latter. tion of detailed observations on this group, proposed (20) to reduce 64 COLORATION IN POLISTES. the indigenous species to a single one, P. gadlicus, of which the rest should be varieties or subvarieties. This change in classification he based on the following considerations: (1) The slight value of color characteristics in the Vespidz for the purposes of classification, since in the yellow and black insects the two colors alternate very irregularly with each other; (2) in a great series of Polistes taken on the wing the transitions presented are so numerous and so insensible that it is often impossible to tell where one variety begins and another leaves off; (3) several nests reared with a single female of P. gadllicus nevertheless produced specimens of P. diadema, (4) colonies of Polistes which pre- sented numerous transitions between P. gadllicus and P. diadema. Siebold (26), whose opinion is certainly entitled to great considera- tion because of his long and careful study of the genus under natural conditions, believed that Europe possesses one species, P. gadlicus, which is very variable, but only one of whose varieties, P. diadema, can be con- sidered at all distinct. I quote from his work at length, since it gives not only the best available definition of the species and its varieties, but a detailed account of their distribution over Europe and northern Africa: Nach meinen Erfahrungen lassen sich die in Mitteleuropa einheimischen Polistes- Formen in zwei Racen unterbringen, welche von vielen Entomologen als zwei selbststandige Arten, namlich als Polistes gallica Lin. und Polistes diadema Latr. festgehalten werden. ‘ Polistes gallica gehort nicht allein ganz Siideuropa an, sondern findet sich auch noch in Nordafrika und in den an das Mittelmeer stossenden Landern Asiens bis nach Persien hin vor, wird aber auch in Mitteleuropa sehr verbreitet angetroffen. Hier in Mitteleuropa geht nun mit dieser Polistes gallica eine Ver- anderung in der Farbung und Zeichnung vor, wobei die gelbe Farbung am Kopf, Thorax und Abdomen durch Ausbreitung der schwarzen Farbe in einer Weise verdrangt wird, dass sich verschiedene Entomologen nach althergebrachter Sitte veranlasst sahen, solche in Farbe und Zeichnung abgeanderte Formen als beson- dere Arten zu betrachten und mit besonderen Artnamen zu bezeichnen. Da sich bei diesen Varietaten nicht bloss die gelbe und schwarze Farbung in ihrer Ausbreitung verandert zeigt, sondern da sich auch das ganze Naturell und die Gewohnheiten ganz anders aussern, so halte ich es fur angemessen, auf diese Varietaten um so mehr aufmerksam zu machen, als bei den Beschreibungen dieser abweichenden Polistes-Formen von dem verschiedenen Benehmen und den verschie- denen Eigenheiten dieser Varietaten die Entomologen fast gar keine Notiz genom- men haben. Zur leichteren Uebersicht der zu erwahnenden Polistes-Varietaéten muss ich vorerst noch darauf hinweisen, dass an der Polistes gallica und ihren Varietaten im Allgemeinen drei Hauptfarben zu unterscheiden sind, namlich als Grundfarbe Schwarz, als Schmuckfarbe Citronengelb, zu welchen beiden Farben an den Fihlern und Beinen noch Rostyelb hinzukommt. Wahrend nun diese Farben in ihrer Vertheilung und Ausbreitung auf das mannigfaltigste abandern, macht sich an allen noch so verschiedenen Varietaten, mit Ausnahme der oft sehr auffallenden Grossen-Unterschiede der ganzen Insecten, weder in den Umrissen noch in der Sculptur der einzelnen Korperund Gliederabschnitte dieser Wespen eine irgend auffallende Abweichung bemerklich. * * * * Bei einer Musterung COLORATION IN POLISTES. 65 von verschiedenen siideuropdischen, afrikanischen und asiatischen Individuen dieser Polistes gallica kommt man sehr bald zu der Ueberzeugung, dass unsere mitteleuropdische Form der Polistes gallica unter dem Ejinflusse der siidlichen klimatischen Verhaltnisse ihr buntes Kleid durch eine immer breiter sich ausdehn- ende Vergelbung allmahlich so verandert, dass dadurch die in den siidlichsten Gegenden von Europa sowie an der afrikanischen und asiatischen Kiiste des mit- telmeeres fast ganz gelb auftretenden Formen der Polistes gallica zur aufstellung neuer Arten verfiihrten. Herrich Schaffer hat wirklich zwei solche sehr stark gelb gefarbte Polistes-Mannchen als besondere Art unter dem Namen Polistes ital- tca und nachher als Polistes pectoralis charakterisirt. Icherkenne in dem von ihm abgebildeten Mannchen nichts anderes als eine Polistes gallica mit derselben vor- herrschenden citronengelben Farbung, welche Savigny in der Abbildung von Polistes gallica aus Aegypten angedeutet hat, wohin auch die von Guerin als Polistes gallica (var. Lefebvrei) aus Aegypten abgebildete Wespe gehort. Ebenso ist auch Polistes bucharensis Erichson’s nach der Beschreibung zu urtheilen, nichts anderes als eine sehr stark vergelbte Form der Polistes gallica, was schon von Saussure, ganz richtig bemerkt worden ist. An solchen siidlichen Varietaten verbreitet sich die citronengelbe Farbe besonders als Einfassung des Hinterrandes des Prothorax ausserordentlich stark, zuweilen in einer Weise, dass die Schultern ganz gelb erscheinen und die beiden gelben Seitenpunkte des ersten Abdominal- Segmentes mit dem gelben ebenfalls verbreiteten Quersaume des Hinterrandes zusammenfliessen. Die beiden sehr breiten gelben Seitenflecken des zweiten Ab- dominal-Segmentes lassen durch ihr sehr haufiges Zusammenfliessen mit dem breiten gelben Hinterrandssaume nur eine geringe schwarze Farbung auf diesem zweiten Hinterleibsegmente tibrig. Alle diese gelben Zeichnungen nehmen an den mannlichen Individuen noch um vieles mehr tiberhand, wobei das Gelb der Brust zuweilen mit dem der Schultern sogar zusammenfliesst. Solche stark ver- gelbte Varietaten habe ich aus Aegypten, Tunis und Algerien, aber auch aus Pa- laestina vor mir. Was nun die in Mitteleuropa verbreitete Polistes gallica var. diadema latr. betrifft, so zeigt sich hier die citronengelbe Farbung und Zeichnung durch die starkere Verbreitung der schwarzen Farbung mehr und mehr verdrangt wobei das Schwarz in einer Weise iiberhand nehmen kann, dass dadurch ebenfalls wieder die Aufstellung von besonderen Polistes-Arten veranlasst wurde. Wahrend die gelben Flecke, Striche und Segment-Einfassungen sich sehr verkleinern und verschma- lern, wodurch bei den Querbinden des Hinterleibs die vorderen Ausbuchtungen derselben meist verschwinden, nimmt zugleich als charakteristisches Merkmal dieser Varietat die schwarze Farbe an den Fithlern eine grdssere Ausbreitung ein, indem sie sich von der Riickenseite des grossen Fiihlerschaftes auch iiber den Rucken der sonst rostrothen Glieder der Geisel bis zur Fiihlerspitze hinsieht. Auch bei dieser Varietat lassen die beiden Geschlechter eine Verschiedenheit in der Farbung und Zeichnung wahrnehmen. Ich darf es nicht unerwahnt lassen, dass auch nach meinen Erfahrungen nicht bloss innerhalb dieser beiden Racen Fol. gallica und Pol. diadema mancherlei Varietaten vorkommen, sondern dass auch beide Racen durch Varietdten in einander iibergehen. Ich habe so viele dieser Ueberginge in meinen Handen gehabt, dass ich mir dartiber, wie Sichel, ebenso klar geworden bin: eine Trennung beider Racen als zwei besondere Arten lasst sich nicht durchfiihren: kein einziges Merkmal, welches man als specifisch fur die eine oder andere Art feststellen wollte, wurde sich als stichhaltig ausweisen. 5 66 COLORATION IN POLISTES. Beide Racen, /. gallica und P. diadema, zeigen sich iibrigens in Mitteleuropa auf eine sehr verschiedene Weise vertheilt, und fast will es mir vorkommen, als ob beide Racen nicht gerne neben einander wohnen, ja, als ob das Vorhandensein der einen Race in einer Gegend das Vorkommen der anderen Race daselbst ausschliesst. So fand ich zu Freiberg im Breisgau, zu Ueberlingen am Bodensee und zu Erlangen in Mittelfranken nur /ol. gallica, also die Race mit rein gelbem Clypeus der Weibchen, wahrend ich in Hohenschwangau, Starenberg, Miinchen, Tegernsee, Berchtesgaden, in Kufstein, bis hinauf nach Innsbruck, ferner bei Miiggendorf in der frankischen Schweiz immer nur /o/. diadema, also die Race mit schwarzem Querstreif auf dem gelben Clypeus des Weibchens angetroffen habe. Wahrschein- lich liebt die Race P. gallica offene, warmere und niedriger gelegene, weil diadema kithlere Gegenden als Aufenthalt vorzieht. Nach einer brieflichen Mittheilung, welche ich Herrn Gerstacher aus Berlin verdanke, wurde von demselbem P. diadema in den Alpen bei Kreuth bis gegen 4000 Fuss und auf dem Stilfser Joch sogar bis gegen 6200 Fuss hoch angetroffen. Es stimmt dies freilich nicht mit dem in Miinchen so haufigen Vorkommen dieser Race zusammen. Vielleicht ist die sehr hohe Lage Minchens die Hauptursache, dass ?. diadema in hiesiger Gegend so gut gedeiht. Jedenfalls ist die P. diadema eine Gebirgbewoknerin. Beide Racen unterscheiden sich wesentlich durch die ganz verschiedene Auswahl der Localitaten, an denen dieselben ihre Nester aufbauen. Wa&ahrend Pol. gallica ihr Nest dem Sonnenlichte entzieht und unter Dachern und Gebdlk verstecht ‘anlegt, klebt die Pol. diadema ihr schutzloses Nest ganz frei der Sonne ausgesetzt an die steilen Wande von Mauern, Zaunen und Felsen. From the foregoing description it is evident that Europe possesses two species, or perhaps better a species and its variety, which in their coloring, habitat, and range bear to each other exactly the same rela- tion as P. aurifer and P. variatus do in our own country. This is extremely suggestive when considered in connection with the repre- sentatives of the genus in eastern and southern Asia. A Russian en- tomologist, Radowezkowsky (20), in 1871 maintained that he had estab- lished incontestably that P. gallicus, diadema, geoffroyi, pictior, biglumis, and ducharensis were only varieties of a single species, P. gallicus, the only one existing in Europe and throughout Africa and Asia. He pushed his conclusions still further. Among Hymenoptera that were sent him from the Himalayas he found a nest with 38 specimens of P. chinensis, and was greatly astonished to observe among them all transitions of coloring between this species and P. gadlicus. ‘The tran- sitions to P. chinensis are effected in the following manner: The chapron, which is black above, with a median yellow line, and bor- dered, becomes by degrees yellow, with two black points on the disc, or even entirely black. The second abdominal segment is often reddish at the base, as is the end of the abdomen. This color is reduced little by little at the base of the second segment to two spots, which become very small and often even imperceptible in the specimens from Oriental Siberia. This relation suggests the transitions by which variatus passes into pal/ifes in northern and eastern United States, and the COLORATION IN POLISTES. 67 closeness of relation between the two is confirmed further by Frederick Smith. In the Trans. Ent. Soc. for 1873 Smith states that P. chinensis, of Hakodadi, Hiogo, Shanghai, Hongkong, and Siberia, is not in his opinion specifically distinct from P. diadema, Europe. This species, furthermore, is related to P. hebyveus, of China and Japan. He states that the Japanese and Chinese form of the latter species differ greatly in coloration from any seen in India ; but the palest colored examples of the species from Japan very closely resemble the darkest ones from India. Smith’s description of these species is very inadequate, but inspection of the figures given by De Saussure (24) shows that the coloration of species for this region is very similar to if not identical with that of the padlipes and carolinus types of the United States. We thus arrive at the following conclusions: Comparing America with the continent of Eurasia, we find in both, distributed throughout their western parts, a similar form, whose coloration is in the main black and yellow. In America it is called P. aurizer ; in Europe, ?. gallicus. ‘Throughout the eastern parts of the two continents we find another series of very similar forms, whose coloration is mainly red-brown, varied with black and yellow. In America this is repre- sented by the fallipes and carolinus types; in Asia by P. chinensis and P. hebreus and their varieties. Along both eastern and western coasts of these continents the forms show marked increase in melanism as we proceed northward, and both types come together in the interior of the two continents by means of numerous transition forms. (See Pl. VI.) I do not know that this relationship between the two continents has beeii made out with any degree of detail for any other group of animals, but as long ago as 1859 Asa Gray (9) pointed out the general accord between the flora of Japan and eastern America. Ina lecture delivered before the students of Harvard University in 1878 he expressed the relationship as follows : Let me recall to mind the list of kinds of trees which enrich the Atlantic coast but are wanting at that of the Pacific. Now almost all of these recur in more or less similar but not identical species in Japan, North China, etc. Someof them are likewise European, but more are not so. Extending the comparison to shrubs and herbs, it more and more appears that the forms and types which we count as pe- cculiar to our Atlantic region, when we compare them as we first naturally do with Europe and with our West, have their close counterparts in Japan and North China. DISTRIBUTION IN SOUTH AMERICA, De Saussure (24) has described and figured a large number of species of Polistes for South America, but his specification as to locality is oftentimes so loose that I rely mainly-on the collection from Brazil made by Mr. H. H. Smith, to which I had access through the 68 COLORATION IN POLISTES. courtesy of Mr. Fox, and on the distribution of species as given in the latter’s paper (8). Mr. Fox adheres in the main to De Saussure’s classification, but he has created several species in addition to those given by De Saussure. From this collection and Mr. Fox’s account of the genus, it becomes evident that by far the greatest number of species approach the cavolinus and the pallifes types in their coloration. ‘These types are also most widespread in their distribution, and so far as the peculiarities of habitat can be made out, the cavolinus type is characteristic for the tropical zone, the pallipes for the temperate zone. ‘The latter also passes by many transitions into the canadensis type ; but we find nowhere the extreme development either in size or coloration represented in such members of the rudbiginosus group as rubiginosus, texanus, and flavus. Neither are the xanthic forms of aurifer and carnifex to be found here. This is interesting when we reflect that we do not have in the region from which the Polistes were collected any of the extreme climatic condi- tions to which these developments of Polistes seem to correspond. ‘The most suggestive species among these is P. versicolor, occurring widely distributed from Cayenne in the north to Chapada on the south. It appears to be a form transitional between P. aurifer and P. carolinus, and may be regarded as an auvifer in which the yellow is reduced, the usually black areas represented in the main by red-brown, while the black is confined to the ventral side of the thorax, the metathorax, and a triangular area on the second abdominal segment. The 100 specimens in the H. H. Smith collection were very variable in the depth of red-brown, the amount of black, and the size and distinctness of the yellow markings, and altogether the species presents a plasticity which might be molded either into the red cavolinus of moist, tropical Florida, or into the yellow aurifer of the dry region of southwestern United States. These relations will be considered further in connec- tion with the laws of color differentiation in the genus. DISTRIBUTION IN AFRICA. This continent falls into two divisions, northern and southern. ‘The former, including the Barbary States and Egypt, is the home of P. ga/- licus, which corresponds to P. austfer of our country. The latter sub- division, possessing in general uniform climatic conditions, approaches South America in its representation of species. The pallipes type occurs in tropical Africa and at the Cape of Good Hope. The carvolinus type occurs in Senegambia and meridional Africa, and evidently merges into the pallifes type ; but since the data for this continent are slight, it is hardly profitable to consider the differentiation here in further i i ae be” COLORATION IN POLISTES. 69 detail. Madagascar is of interest in possessing a form which resembles canadensis, but has a smaller head. DISTRIBUTION IN AUSTRALIA AND THE EAST INDIES. Australia possesses ausifer, pallipes, and the transitional versicolor, but I have been able to learn nothing further of the relation of these three species. The types of the East Indies have the same relation to the. Polistes of Japan as do those of the Antilles to Florida and the Atlantic States. There is the same predominance of the smaller forms marked with red-brown and yellow, and also the occurrence of a large reddish-brown carnifex with suffused yellow borders. _ We may sum up the features of the geographical distribution of the genus, so far as we have been able to determine them, as follows: The southern extension of the eastern and western continents each possesses two great classes of forms, one which may be designated as belonging to the canadensis type, the other including a series of forms which in their coloring and pattern range from the mode of aurifer to that of carolinus, with a prevailing condition distinctly be- tween the two. These classes appear to be fairly well separated in point of size as well as color pattern. They extend northward to a varying degree. The former is the more limited both in its range and its number of species; on the northern hemisphere it does not extend far beyond the thirty-fifth parallel. The latter ranges about 20 degrees farther northward, and is represented by a great number of species, which differ both in size and coloration. Along the opposite coasts of the two continents we observe for the latter class two distinct trends of development ; the one along the east- ern coast verging toward a red-brown form with colors tending toward suffusion, the other along the western coasts toward black and yellow coloration with sharply delineated areas. Both exhibit increasing melanism as we pass northward. ‘These points are represented graph- ically in Plate VI. 7° COLORATION IN POLISTES. GENERAL LAWS GOVERNING COLOR DIFFERENTIATION. We have seen that in all Polistes the pattern is produced by the con- currence of two colors: (@) the hypodermal color which occurs in the areas unoccupied by the darker pigment of the chitin, (4) the pigment of the chitinous cuticula. The former is always yellow, varying only slightly for the different species. The latter varies greatly from a shade only slightly darker than the hypodermal yellow to a shade and concentration which approaches black in its effect. The final effect of black is produced by a concentration of at least two kinds of cuticular pigment: (a) dull brown in color, (6) red-brown in color. These two kinds of pigment by their varying amount produce infinite gradation of hue in the genus, and there is good evidence that they are related to each other and to the hypodermal yellow by slight differences in. chemical composition. With reference to the distribution of these colors in the integument of Polistes the following laws have been formulated : (1) The arrangement of the yellow areas is perfectly definite for all representatives of the genus, the yellow when present always occur- ring in the borders or in the form of metameric spots. They are thus related primarily to the external organization of the body, and this relation is most forcibly expressed in the abdomen of the wasp. (2) The dark areas are related to the internal structure of the body, for the regions where the color of the chitin occurs most constantly in the different species and earliest in the ontogeny overlie the attach- ment of the principal muscles of the body. (3) As far as individual variations occur among the progeny of the same parents, they are continuous, the more aberrant forms proceeding by insensible stages in definite directions from the parent type of mark- ing as a modal condition. (4) The cause of the variation among the individuals springing from the same stock is to be sought in slight differences in environ- mental condition. . (5) The same law holds true for the representatives of a species for a given locality. Here again the variations are continuous and fall into various trends, determined by the predominance of certain envi- ronmental conditions. These trends are impressed as it were upon certain types which are held more or less distinct by inbreeding in the nests—1t. ¢., tendency toward segregation in very limited areas. (6) For the larger areas of distribution the pattern and coloring are again the result of the impress of climatic conditions on the particular type or types which have come to occupy the area. COLORATION IN POLISTES. 71 Evidence tending to confirm these last three laws is further elabo- rated in the following paragraphs: Chief among the environmental. influences which play a réle in the color differentiation of Polistes are temperature and humidity ‘aken in coniunction, although temperature appears to affect mainly the amount of the pigment, and moisture its kind and degree of suffusion. The colder the climate, and the more prolonged the pupal period, the greater the amount of pigment produced ; but whether that pigment shall spread uniformly over the surface or be restricted to definite * areas depends also on other conditions, one of which appears to be humidity. This may Be imagined to affect the mobility of the pig- ment as it is being deposited. When the moisture is abundant it is safe to suppose the pigment to be more mobile, and the chitin, hard- ening more slowly, allows it to spread farther from its center of dis- tribution. When slight in quantity the pigment does not spread so far; so, with the same amount of pigment deposited, we may have under dry conditions numerous small dark areas interspersed with large areas of hypodermal yellow—. ¢., P. aurifer—or under moist conditions a uniformly-colored light species, such as P. carolinus of Florida. Moisture also appears to affect the color of the pigment. Abun- dance of moisture gives the pigment a redder color, scarcity a duller brown color (cf. P. aurifer and P. carolinus). Further, in P. flavus we have the apparent effect of a hot, dry climate on a form allied to P. texanus, into which it passes by transitions, and in P. zavaioe and P. carnifex the effect of similar conditions on the canadensis type. Finally the increase in depth of melanism as we go northward is evidence for the effect of temperature, but the fact that the melanism is more marked along the coast than inland indicates that moisture is also a factor here. On the eastern portions of both the old and new worlds the melanism is, as it were, impressed on a rich red-brown form that has developed in the humid regions of the East Indies and the West Indies and Florida, which is practically one with the latter so far as climatic conditions are concerned. In the western portions the melan- ism affects the type peculiar to the arid zones of southwestern United States and northern Africa. This hypothesis is further borne out by observation in individual variation in a number of instances, and in a striking manner also by the parallel types of coloration observed in genera related to Polistes. 72 COLORATION IN POLISTES. INDIVIDUAL VARIATION UNDER VARYING EXTERNAL CONDITIONS. In several specimens of ?. variatus that were kept in a glass tube for the purpose of learning the time and manner of pigmentation, the pupal period was prolonged much beyond the usual length. This was probably owing to thecold, the average temperature being 18° to 24°C., and therefore far below that under which Polistes usually develops. The imagines in this case were smaller and decidedly more melanic than any obtained from the same nest or even the same region. The yellow was restricted to the merest indication of borders on the ab- dominal segments and two very narrow lines on the metathorax. The rest of the surface was intensely black. One specimen from a nest of generosus obtained from Florida, and under similar observation also came out much more melanic than its congeners. The temperature was in both these cases greatly reduced, for not only was the weather extremely cold for the season (September), but the manner of the keeping of the pupa favored radiation greatly over that usual to the pupa in the nest. The results of a number of experiments in which the external conditions were varied point in the same direction. Our experiments were crude and not at all extensive, owing to the diffi- culty of obtaining material at the time they were carried out. Because of this inadequacy too much stress will not be laid on those results, but the results, as far as any were obtained, are in accord with the hypothesis. A brief account will herewith be given. Two nests from which a number of imagines had previously been obtained under nor- mal circumstances were broken each into four pieces and subjected to the following conditions: (a) Warm and saturated (35° to 42° C.); (6) warm and dry (35° togo° C.); (¢) cool and saturated (about 18° C.); (d@) cool and dry (about 18° C.). The state of saturation was produced by placing a dish with water in a closed case containing the fragment of nest, the dryness by the use of CaO. From this material not more than four imagines were in any case obtained, with five or six pupz just ready to emerge. Where the conditions were warm and saturated the lateral spot was large and of a reddish orange color, shading at the center into reddish yellow; where the conditions were cold and dry, although the specimens were from the same nest, the three wasps that emerged after the experiment had been in progress eight days were very black, with narrow, well- defined borders. One specimen had a small yellow lateral dot in the red segment, another a small brown dot, while in the third the second segment was unmarked. Where the conditions were warm and dry, four specimens just ready to emerge showed broad yellow borders and COLORATION IN POLISTES. 73 prominent brownish or brown and yellow lateral spots, and where cool and saturated the coloration was produced by a diffused dull-brown pigment present in large quantity. Here we see that moisture tends to a suffused coloration with tendency to a reddish cast of pigment, dry- ness to a duller brown. Cold produces great amount of pigment, heat little——and thus the tendency is toa restricted and dull brown color or a diffuse and dull brown color, according as moisture is lacking or abundant. TYPES OF COLORATION IN RELATED GENERA. The most striking confirmation of these tendencies, however, is afforded by the colors displayed in other genera of the Vespide. - Polybia, the slender thread-waisted social wasp, from which Polistes has, in all probability, sprung, is peculiar to the tropics of the New World and reaches its highest development in South America. It presents all the color variations displayed by the Polistes of South America, but only two forms have made their way into the United States. These are P. cubensis of Florida and P. flavitarszs of California. Inits color and markings the former corresponds exactly tothe Polistes of Florida, while the latter is identical with P. aurifer of California. The genus Icaria is the representative of Polybia in the Old World. Here the yellow forms, socialistica and cabeti, occur in Tasmania; the melanic form, guttatipennis, is obtained from Senegal. Madagascar has two green species peculiar to itself, and as this is the only instance of the occurrence of green throughout the family, it certainly speaks for the effect of isolation in originating and maintaining a species. Be- lonogaster, which is the only other social wasp besides Vespa found in the Old World, is peculiar to Africa. Itis a notable fact that in its type of marking it follows very closely the Polistes of the various sections. Thus 2. junceas, which is found in equatorial Africa, resembles P. pallifes in its coloration. 2. griseas, filventris, madecassa, and guerinii all resemble canadensis ; the first two forms occur in Senegal, the two latter in Madagascar. In this connection it will be remembered that the only Polistes found in Madagascar is a canadensis having a slightly smaller head than the normal. Inthe East Indies zzdzcus belongs to the carolinus type. Among the Eumenidez, which are the solitary wasps from which the social wasps had their origin, we find good illustration of the same general laws, although the habits and instincts here are so various that we would hardly expect this concord. Septentrional Eumenide are usually of a black color varied with yellow. I have not studied the variations in all these black and yellow 74 COLORATION IN POLISTES. species, but at any rate along the Pacific coast they show the same gradations in coloring as does Polistes. Thus species of the genus Ancestroceris from California and Nevada have the characteristic black and yellow pattern, and in the species from Vancouver the yellow areas are greatly reduced. In the genus Pterocheilus, d¢A/agtata shows the yellow borders broad, and conspicuous yellow spots, while morrisonii from Nevada shows the yellow areas somewhat reduced. An extensive series of Eumenide (undetermined) in the Philadelphia Museum and a large number of Odynerii from California illustrate this same law. On the Florida side the genus Slossonz exhibits the cavolinus type of marking; the same is true of Humenes smithi. De Saussure (24) gives the type with yellow metameric spots as characteristic for the United States and Europe, and further adds that ‘‘ besides this, there is produced in the southern United States another type mixing with this, but characterized by a preponderance of red and suffused yellow.’’ These two forms thus bear the same relation to each other that /. variatus and P. carolinus do. Tropieal Africa, again, shows a tendency toward uniformity of species along with uniformity of climatic conditions, while the East Indies offer an apparent anomaly by possessing, in addition to the orange and black carolinus type,-one which resembles auvifer as much as it does versicolor. We may thus consider the great trend of development in the genus Polistes as determined by general climatic conditions, while the slighter differences are due to slight differences in environmental conditions, operating of course on slight initial differences in the types for various nests. Keeler (14) supports these ideas in his evolution of land birds. He gives his testimony that moist seacoast climates produce rather darker and duller hues than normal. ‘‘ The maximum of sunlight and moisture together forms the most pure and brilliant coloration, sunlight without moisture has a tendency to burn and bleach, while moisture without sunlight produces darker and duller colors.’’ In any given region the coloration, while adhering to one general mode, may change to a varying extent in certain directions, owing to the differences in station of various forms and their comparative isolation from one another. Here the inbreeding in the same nest or neighboring colonies would tend to accentuate a particular line of variation. COLORATION IN POLISTES. 75 CONSIDERATIONS WITH RESPECT TO VARIOUS THEORIES OF EVOLUTION After the discussion in the preceding section little need be said of the bearing of the facts on theory. While the evidence from the study of individual variation is decidedly in favor of the occurrence of numerous almost imperceptible differences and against discontinuous variation, I have been able to find in the environmental condition no adequate factor for the selection of such variations which might lead to the evolution of divergent species. The observations on the devel- opment of the color pattern in the various species are in accord with Eimer’s theory of orthogenesis—that is, the color pattern for both the individual and the species may be regarded as arising from the tendency of the organism to pause at some particular stage of the process of pigmentation—but this tendency is due not to any inherent property of the organism, but rather to the differing environal conditions to which it issubjected. These conditions affect the metabolism of the insect in definite directions, and through it the products of metabolism, in this instance the kind and amount of pigment. Specific differentiation springs from individual variation, and both are in large measure due to external conditions. Such characters appear to belong to the class of auto-adaptive characters. Study of the specific distinction of many other orders prove such auto-adaptive characters to be of wide occurrence in the animal world. The following passage is quoted from J. A. Allen’s (1) Influence of Physical Conditions in the Genesis of Species: We find among the mammals and birds of the United States three strongly marked phases of color-differentiation among representatives of the same species, characterizing the three most strongly marked climatal regions—a bright, strongly colored form east of the Great Plains, a pallid form over the dry central region, and a deeply colored fuscous form over the rainy, heavily wooded region of the northwest coast. Examples of this differentiation are afforded by apparently all the species whose habitats extend entirely across the continent, the several local forms being in some species only more strongly marked than in others. Among mammals illustrations are afforded by different species of squirrels, hares, mice, lynxes, deer, etc. ; and among birds by six or eight species of sparrows, a number of woodpeckers, several fly-catchers, thrushes, and warblers, the meadow lark, various hawks, owls, etc. Generally these several geographical forms were orig- inally described as distinct species, and many of them are still thought worthy of recognition by varietal names As intermediate links began to be discovered, they were at first looked upon as the result of hybridity between the supposed distinct species whose characters they respectively combined, but eventually such links were found to be too frequent and too general over the areas where the habitats of the several forms come together to render such a supposition longer tenable, it finally appearing evident that they were only the connecting forms between merely local races or incipient species. 76 COLORATION IN POLISTES. The local races of any given region, as compared collectively with those of con- tinguous regions, and the manner of their mutual intergradations, point plainly to some general or widely acting cause of differentiation. This is indicated by the constancy of the results—so many species belonging to numerous and widely dis- tinct groups being similarly affected. Allen further asks: Will the fortuitous, spontaneous results of natural selection yield a satisfactory explanation of these phenomena, or must we seek some more uniform and defi- nitely acting cause? After an inquiry into the adequacy of natural selection in originating characters of this nature, he states his conclusions as follows: While there is perhaps little reason to question the correctness of these generali- zations [concerning natural selection], they have little bearing upon the question of the modification of species by the direct action of climatic conditions, but relate mainly to such unfavorable climatic influences as tend toward the extinction of species, or to the circumscription of theirranges. Indeed, the phenomena of varia- tion detailed in the foregoing pages were almost wholly unknown at the time the earlier editions of the ‘‘ Origin of Species ’’ were published, and have hardly as yet become the common property of naturalists. Gradual decrease in size southward in hundreds of species inhabiting the same continent, or a gradual increase or decrease in color in given directions on a similarly grand scale, are facts but recently made known, and have not as yet been very fully discussed by evolutionists of the purely Darwinian school. * * * * That varieties may and doarise by the action of climatic influences, and pass on to become species, and that species become, in like manner, differentiated into genera, is abundantly indicated by the facts of geographical distribution and the obvious relation of local forms to the conditions of environment. The present more or less unstable condition of the circumstances surrounding organic beings, together with the known mutations of climate our planet has undergone in past geological ages, points clearly to the agency of physical conditions as one of the chief factors in the evolution of new forms of life. The establishment of local races exhibiting some peculiar type in the general trends of divergence is to be explained on the principle of cumulative segregation. Observations on the mating habits of the species indicate that individuals belonging to the same colony tend to interbreed. It is also probable that there is extensive mating outside thecolony ; but since individuals tend to range only over a limited terri- tory, and further tend to occupy the same locality, and even the same nesting site, year after year, we have here all the conditions which favor the origination of nest individuality and local races. The same principle accounts for the maintenance of the distinction between such a species as ?. gallica of Europe and its variety, diadema. Von Siebold has shown that even where the ranges of these two overlap they occupy different stations, and consequently the changes which different environmental conditions may effect are not lost by inter- COLORATION IN POLISTES. a7 crossing, but are preserved and even accumulated by the persistent operation of the external causes. Finally, where the differences between two species occupying the same territory are mainly those of size, as is the case with the cavolinus and rubiginosus types of coloration, this may be regarded as illustrating another phase of the principle of segregation, that widely known as the principle of physiological isolation. We conclude, therefore, that specific differentiation in Polistes is in the main due to the influence of external conditions, and within the modes thus established slight variations are due to slight differences in environal conditions ; but these variations tend to cluster about more or less dis- tinct types, which are held thus distinct by the various kinds of isolation resulting from the habits of the insect. SUMMARY AND CONCLUSION. | (1) The colors of Polistes fall into two classes—hypodermal and cuticular. (2) The hypodermal coloring matter is yellow and is deposited in and among the cells of the hypodermis. (3) The cuticular color ranges from a shade scarcely darker than the hypodermal yellow, through varying tones of brown to a shade approaching black. It is due toa brownish pigment which is suffused through the chitinous integument. (4) In pattern two main types may be recognized: (a) The mz/tz- colorous, where the cuticular pigment occupies only limited areas and in varying quantity, leaving a more or less conspicuous yellow or yellowish ferruginous ornamentation; (4) the uzzcolorous, where the cuticular pigment extends beyond these areas, so that the whole surface is more or less suffused with the darker color. Numerous transitions exist between these types. (5) In the developing color pattern color first appears in certain areas of the uniformly flesh-colored groundwork, and later spreads laterally from these areas as centers until the pattern for the adult is attained. The yellow hypodermal color is the last to appear ; it is laid down in the areas which have been left free from the darker cuticular pigment. (6) The areas where the cuticular pigment first appears are con- stant for widely differing species. They represent the points of attach- ment to the cuticula of the more important muscles. (7) The various cuticular pigments are related to one another and to the hypodermal yellow by slight differences in chemical composition. 78 COLORATION IN POLISTES. (8) Spectroscopic analysis and numerous parallel reactions suggest a relation between the pigments of Polistes and a series of organic pigments known as the azo-color compounds. (9) The pigments appear to be elaborated from the protoplasm of the hypodermal cells, which in turn seems to be modified in a definite manner by surrounding metabolic processes, particularly those accom- panying the development of muscle. (10) These metabolic processes or the products elaborated by their agency are further affected by external conditions such as temperature and humidity, and thus varying climatic influences determine the coloration of the wasp. (11) The individuals of the same colony vary greatly among them- selves, but the variation is continuous and in certain definite directions which in some cases at least can be traced to the varying external conditions. : (12) There is extensive intergradation between the numerous species of the genus, which again bears a relation to climatic conditions of the region where the types occur. (13) In any given locality the race differentiation may be accounted for on the principle of segregation and slight differences in external conditions. (14) For the larger geographical areas the type of coloring is in the main due to climatic conditions in conjunction with cumulative segregation. DESCRIPTION OF-SPECIES: Polistes canadensis (Linn.). Head and thorax ferruginous brown; antennez ferruginous at their base, black in middle, orange at end; a little black on sides of thorax and in median dorsal furrow of metathorax ; metathorax with fine, transverse s'rie. Abdomen brown or blackish ; coxee and femora black, tarsi, ends of the femora and base of the tibiz, ferruginous. Wings deep brown, with brown reflections. Length 18 mm. Var. A.—The femora ferruginous, black underneath ; tibiz of the middle legs ferruginous. Var. B.—Body and legs entirely ferruginous. Var. C.—Posterior border of the first segment of the abdomen ferruginous, Hab.—Florida, Mississippi (Ashmead), Washington, D. C (Packard), Texas. P. annularis (Linn.). Black.—Head ferruginous, a line on the top of the clypeus and vertex black ; antenne ferruginous, black in the middle, orange at the end; the prothorax bor- dered with red ; tegulez, a point under the wings, two marks on the middle of the mesothorax, two points on the scutellum, and two on the metathorax, reddish ; postscutellum ferruginous yellow. Abdomen black, first segment bordered with yellow. Legs brown, with articulations yellow. Wings black. Length 18 mm. Var. A.—Thorax ferruginous, varied with black. Var. B.—Whole insect clear ferruginous. _ Var. C.—Insect brown, often confounded by the color with P. canadensis. Var. D.—Thorax and two marks on the side of the second segment reddish. Hab.—North America. Very common in the cotton belt of the South (Ash- mead), Brazil (Saussure). . P. crinitus (Felton). Female.—Head yellow, front and vertex ferruginous ; clypeus convex, rounded at the base; first joint of antenne ferruginous, the rest yellow, black above in the middle. Thorax black ; prothorax angular, bordered with yellow; scutellum and postscutellum yellow, separated by a black line ; tegulee and a line or mark under the wings rejoining the angles of the prothorax yellow ; metathorax smooth, with the median dorsal furrow very pronounced; articular valves yellow. Abdomen black, all the segments regularly and broadly bordered with yellow, the borders of the second and third very straight and preceded by a red band ; the last two fer- ruginous yellow. Legs black, ends of the femora, tibiz, and tarsi yellow, outer side of the hind legs black. Wings ferruginous, washed with brown. Var. A.—Body brown or red, instead of black ; borders of the abdomen somewhat fused with the brown or black ; antennz ferruginous with black in the middle. Var. B.—P. crinitus billardieri Sauss.—Posterior parts of the head yellow ; orbits and ends of the clypeus yellow. Thorax red; prothorax bordered with yellow in front and behind; scutellum bordered with yellow; metathorax black, with two yellow lines ; the first segment of the abdomen yellow on three sides, black in the middle, the rest reddish, narrowly bordered with yellow. Wings brownish. Var. C.—Mesothorax black, with two small yellow lines; metathorax with two yellow lines. Var. D.—Two yellow lines on posterior plate of metathorax and two on sides. HTab.—America. (79) 80 COLORATION IN POLISTES. P. crinitus lineatus Fab. Worker.—Head ferruginous yellow, the region of the ocelli and an irregular line under the antennz blackish ferruginous; antennz ferruginous yellow, the first two joints blackish ferruginous, as well as the upper part of several of the middle ones. Thorax ferruginous brown; two longitudinal lines on ‘the mesothorax, the borders of the tegule and prothorax, a mark under the wings, a line on the scu- tellum, one on the postscutellum, and a scalloped mark bilobed at the top on each side of the metathorax, all ferruginous yellow. Abdomen ferruginous brown, fwith the terminal borders of the segments ferruginous yellow ; this border extend orward on the side of the segments. Legs yellow, except the tips of the femoras and the outer ends of the middle and hind tibia, which are black. Wings ferru- ginous brown, with violet reflections. Length 16 mm. Hab.—America. P. aurifer Sauss. Female.—Black ; mandibles, clypeus, orbits, and a very open \/-shaped mark on the front yellow; antennz orange; the borders of the prothorax, a mark under the wings, the tegulz, the anterior border of the scutellum and postscutellum and two lines on the metathorax yellow. Abdomen yellow, base of the first two seg- ments black, the black excavating the yellow in the middle and sometimes forming there a design, the third occasionally black in the middle at its base. Legs black, ends of femora, tibiz, and tarsi bright yellow. Wings having a general golden tint or a ferruginous gray. Length 15 mm. Var. A.—End of the posterior tibiz black. Hab.—California, Colorado, Australia, Honolulu. P. fuscatus (Fab.). ‘*Fuscous, spotted with ferruginous; first segment of the abdomen with a yellow margin, second with two ferruginous spots. Length and magnitude of P. annu- farts ; antennz black, base ferruginous. Head fuscous, with the labium and man- dibles ferruginous. Thorax fuscous, with a ferruginous line on each side in front. Abdomen fuscous, the first segment with a yellow margin, second with a large ful- vous spot on each side. Feet variegated.’’ FTab.—America. P. fuscatus instabilis Sauss. Worker.—Insect dark red, a little rose colored; base of the clypeus yellow; vertex around the ocelli black ; antennz black in the middle ; prothorax bordered, its angles without spines ; underside of thorax and the sides of the metathorax and mesothorax entirely black ; metathorax smooth, or almost smooth; orbits often yellowish ; the borders of the prothorax, a large mark under the wings, anterior angles of the scutellum, postscutellum or only its anterior border, as well as two lunate spots on the metathorax, sulphur yellow ; on each edge of the metathorax on the sides, a yellow mark often connected with the mark on the posterior plate ; tegulez often yellowish ; base of the first two segments of the abdomen often black ; terminal borders of the first, second, and third segments yellow. Legs black, ends of the femora, tibize and tarsi yellow or red, posterior tibize black, except the base, which is yellow. Wings of a uniform gray tint. Length 14 mm. Var. A.—Mesothorax black, with two red marks; metathorax entirely red. Var. B.—Thorax entirely black ; first segment of the abdomen reddish black, bordered with yellow, the second red, its base black, its border glaucous yellow, yellow on the edges, a little black on the front of the border, the others black ; | COLORATION IN POLISTES. 81 the third and fourth bordered with greenish yellow, on each edge an irregular red marginal mark ; terminal segment reddish. Legs varied with red and yellow. Hab.—United States, Mexico. P. fuscatus exilis Sauss. Male.—Head and thorax black; face whitish yellow; posterior orbits and the space between the eyes of the same color ; mandibles sulphur yellow, with a black dot at their base; antennze’ black, ferruginous beneath, except at the end; the two edges of the prothorax bordered with sulphur yellow or ferruginous ; anterior border of the postscutellum and two lines on the metathorax sulphur yellow, as well as the articular valves. Abdomen black, all the segments bordered with sul- phur yellow, the second bearing on each side a round point of red, and the first, often two, yellow points in front of the angles of the border of this segment ; end of the abdomen often ferruginous. Legs ferruginous, under or posterior side of the femora black ; posterior legs a little brown ; front of the coxe yellow, which color prolongs itself onto the front of the thoracic segments placed above the coxe. Wings transparent, washed with yellow. Length 13 mm. Var. A.—Proth6érax and metathorax black; the first segment of the abdomen bordered with yellow. Hab.—North America. P. pallipes Lepel. Female.—Head black, with the mandibles, the convex clypeus, except the edge, a broad stripe behind the eyes, the sinus of the eyes with a line extending to the clypeus, ferruginous ; antennz black above, ferruginous beneath. Thorax black; posterior edge of the prothorax and the tegule ferruginous. Abdomen black, the first segment with a pale yellow terminal border ; all the coxee, the greater part of the femora, and the outer side of the tibize black, the rest of the legs ferruginous or dull yellow. Wings smoky brown ; some specimens have a small spot under the base of the fore wings, two points sometimes extended into a line on the front of the scutellum, a similar line on the postscutellum, and sometimes a longitudinal stripe on éach side of the median dorsal furrow of the metathorax yellow ; occa- sionally specimens are found in which all the abdominal segments above and beneath have a narrow terminal border; the clypeus is sometimes black and the markings yellow instead of ferruginous ; a few specimens have a ferruginous spot on each side of the second abdominal segment, and occasionally traces may be found on some of the other segments. Length 21 mm. Male.—Differs from the female in having the clypeus very flat, all the face, the breast and the coxe beneath, yellow; the front of the femora reddish yellow; the underside of the antennz yellow or reddish yellow. Length 20 mm. Worker.—Similar to the female, but somewhat smaller. Length 18 mm. M. de Saussure has given the following descriptions of varieties : Var. Worker.—Two yellow lines on the metathorax. Var. A.—Border of the tegulz indistinct, first segment of the antenne red ; postscutellum ornamented with two yellow points. Var. B.—Abdomen black, border of the second segment unmarked, the lateral spots red. Var. C.—Prothorax, clypeus, and border of the third segment of the abdomen red. Var. D.—The mark under the wings, two points on the scutellum yellow; meta- thorax with two yellow lines. Var. E.—Marks of the abdomen large, melting with the borders, which are red. Var. F.—AlIl the segments of the abdomen bearing a red mark on each edge. -6 82 COLORATION IN POLISTES. Var. G.—AIl the segments red, their base black, their borders more or less em- broidered with whitish yellow. Var. H.—Mesothorax and abdomen red; no yellow on the insect ; base of the second segment black. : Var. I.—Metathorax red, with two yellow lines; scutellum marked with yellow. Abdomen red. Var. J.—Abdomen chestnut, the first segment embroidered with yellow; two yellow points on the metathorax. Var. K.—Marks on the second segment of the abdomen white. Var. L.—Large ; abdomen ferruginous, with four quite large yellow borders ; metathorax with two yellow bands; antenne a little gray above. Ffab.—United States, Canada. P. metrica Say. Ferruginous, Abdomen black. Wings dark violaceous; body ferruginous ; antennz fuscous, first and second joints ferruginous beneath, last five or six joints fulvous beneath ; hypostoma with a few distant, short yellow hairs, not sericeous, at the middle of the tip a little prominent. Thorax with a black dorsal line abbre- viated behind, each side of which is an obsolete line confluent behind, exterior to which at the base is a black line attenuated before and abbreviated. Wings dark violaceous. Feet black, tibiae within, except posterior pair, knees, and tarsi yel- lowish. Abdomen black, first segment absolutely piceous each side and on the posterior edge, second segment also with obscure ferruginous on each side, some- times obsolete. Length over four-fifths of an inch. Hab.—United States. P. minor Beauv. e Worker.—Insect ferruginous ; clypeus yellow, with a ferruginous mark ; man- dibles yellow; orbits broadly bordered with yellow ; antennz black above at the middle, nearly to the end; the borders of the prothorax, tegule, and a point under the wings, the anterior border of the scutellum, the postscutellum and two large marks, which cover the posterior plate of metathorax, yellow. Abdomen short, oval, depressed, first segment broadly bordered with yellow, this border bearing a large square or tricuspid indentation ; the three following segments ornamented at borders with little festoons, rest of abdomen ferruginous. Legs ferruginous, knees, tibize, and tarsi yellow; posterior tibiae ferruginous at the end. Wings ferruginous, brown in the radial ; third cubital cell regularly lozenge-shaped. Male.—Antenne scarcely obscure above; borders of the orbits and the yellow of the clypeus forming a yellow Y on the front of the face. Hab.—Georgia, Louisiana, California, Texas. P. rubiginosus Lepel. Insect entirely clear reddish yellow ; a little black around the ocelli; antennze blackish above from the fourth segment up. Thorax bearing a golden down. Wings brown, with violet and golden reflections. Var. A.—Three black lines on the mesothorax, one on the metathorax. Length 18 mm. H2b.—Pennsylvania, Illinois, Georgia, Missouri (Murtfeldt), Texas. P. perplexus Cress. Male.—Ferruginous, strongly golden sericeous ; face and clypeus flat, dull yel- low-white, pale on orbits and above insertion of antenne ; mandibles and cheeks beneath more or less dull yellowish white ; antennze ferruginous, paler at base be- 7 COLORATION IN POLISTES. 83 neath, joints above more or less black, especially those at the apex ; collar more or less black ; prothorax generally more or less margined with pale yellow ; meso- thorax black, with two central, longitudinal, more or less distinct ferruginous lines ; scutellum sometimes divided centrally by a black line; metathorax with a broad, deep, longitudinal groove, finely and transversely striated, ferruginous, generally with three longitudinal black lines, which are sometimes subobsolete ; sometimes the metathorax is entirely ferruginous, except the groove, which is always black, sometimes entirely black, except a spot on the flanks and a stripe on each side of the groove; pleura generally yellowish beneath, more or less black laterally. Abdomen of different shades of ferruginous, strongly golden sericeous ; the three basal segments have frequently a more or less distinct, narrow yellow apical margin ; the base of the second, third, and fourth segments is more or less broadly black, sometimes obsoletely so, sometimes only the second or third or fourth is so marked ; ventral segments more or less marked with black at the base, the intermediate segments sometimes banded with yellow. Legs ferruginous, the four anterior coxze, femora, and tibiz beneath more or less pale yellowish, all the coxe more or less black above. Wings fuscous, darker along the costa and in the marginal cell; second and third submarginal with subhyaline streaks. Length. 18-21 mm. Hab.—Texas, Bermuda. P, generosus Cress, Male.—Head subsericeous, face long, dull luteous ; anterior orbits, a band above the antenne filling up the emargination of the eyes, and mandibles yellowish white or luteous ; venter and occiput black ; cheeks and a dot on each side be- hind the ocelli ferruginous ; clypeus longer than wide, flat, sparsely and finely punctured, apex angular; ridge between the antennz subtuberculate above; an- tennz long, black above, fulvous beneath, scape yellowish beneath, narrowly edged with fulvous above ; prothorax fulvo-ferruginous, black laterally, upper margin narrowly edged with yellowish ; mesothorax black, with two short ferruginous stripes on the anterior middle ; scutellum black, with two ferruginous spots ; post- scutellum ferruginous, narrowly margined with black ; metathorax transversely and rather coarsely wrinkled above, with a broad, shallow groove down the middle, a slender stripe on each side of the groove and a spot on each flank ferriginous ; pleura black; a ferruginous spot beneath the tegule and a luteous subangular mark behind the anterior coxe ; tegule ferruginous, fuscous at the base. Legs fulvous, yellowish beneath, sericeous; the four anterior coxze above, posterior pair entirely, and all the femora above, black. Abdomen subsericeous, fulvo- ferruginous, a longitudinal mark on the basal middle of the first segment; basal margin of second segment dilated above and suddenly dilated laterally ; the nar- row basal margin of the third and fourth segments black ; apex of the second and following segments stained more or less with fuscous ; apical segment blackish, rugulose ; venter black. Length 22 mm. Hab.—Texas. P. texanus Cress. Female.—Dull ferruginous, sericeous. Head yellow; vertex, occiput, and pos- terior margin of the cheeks ferruginous; clypeus subconvex, sparsely punctured, “ sometimes tinged with fulvous at the base; antennz entirely fulvo-ferruginous ; narrow margins of the prothorax, sometimes a band at the base of the scutellum, postscutellum more or less, two stripes on the disc of the metathorax, a spot or stripe on each side, a spot beneath the tegulz, a spot beneath the posterior wing, 84 COLORATION IN POLISTES. and the tegule more or less yellow; metathorax with a shallow median groove, transversely striated ; sometimes the mesothorax has two short discal lines. Abdo- men strongly pale golden sericeous, rather broad apical margin of all the segments, except the last, even on the first segment and more or less strongly sinuate ante- riorly on the remaining segments, an angular mark on each side of the first segment sometimes reduced to a dot, and a spot on each side of the remaining segments, larger on second and sometimes nearly confluent with the apical band, all yellow ; all the apical bands slightly interrupted or indented medially by a slender ferru- ginous line; venter ferruginous, more or less varied with yellowish ; sometimes the yellowish markings on the second and following segments are more or less obscure; anterior coxe beneath, a line on the outside of the four posterior coxe, four anterior femora beneath except base, tips of all the femora and outside of all the tibze and base of tarsi more or less yellowish, tips of tarsi fuscous. Wings yel- lowish fuscous, darker along the costa. Length 21 mm. Male.—Closely resembles the female ; the face flat as usual, whitish yellow, this color extending above the antenne and ona line with the emargination of the eyes; clypeus flat ; prothorax sometimes broadly yellowish laterally ; mesothorax occasionally with two short discal yellow lines; the markings of the abdomen vary considerably, being sometimes very distinct, with the apical bands broad and lemon yellow and the lateral spots round and whitish ; sometimes the apical bands are narrow and the lateral spots wanting, except on the second segment ; some- times the first segment has no lateral spot. FTab.—Texas. P. bellicosus Cress. Male and female,—Uniformly dull ferruginous, subsericeous. Female.—Clypeus, sides of face, a transverse line above the antennz, posterior orbits broader beneath, and the mandibles yellow; clypeus subconvex and sparsely punctured ; tegule, a line on the outer side of all the coxz, knees and four ante- rior tibize and tarsi beneath, the narrow apical margin of the abdominal segments more or less sinuate anteriorly, and sometimes a spot on each side of the first and second segments, all lemon yellow; mesothorax generally with a fine central longitudinal black line; metathorax finely and transversely wrinkled. Wings fuscous, or fusco-hyaline, yellowish along costa, subviolaceous. Male —Clypeus, face as far up and on a line with the emargination of the eyes, _ narrow posterior orbits broader beneath, and mandibles yellow; clypeus flat ; antenne long, pale beneath, dark above, scape yellowish beneath ; margins of prothorax, basal margin of scutellum, postscutellum, two longitudinal lines on disc of metathorax, sometimes a lateral spot beneath tegulz, space between four anterior cox, coxze beneath, all lemon yellow. Length 18 mm. ffab.—Georgia, Florida, Mississippi, Texas. P. variatus Cress. Female.—Strongly sericeous, black ; a transverse, subangular line above antenne ; posterior orbits broader beneath, and mandibles, except tips, yellow ; clypeus sub- convex, sparsely punctured, ferruginous, more or less broadly yellow at apex; cheeks sometimes ferruginous; antenne ferruginous, black above; prothorax ferruginous, black on extreme lateral corner and narrowly margined anteriorly and posteriorly with yellow; mesothorax and pleura, except yellow spot beneath tegulz, entirely black; scutellum and postscutellum ferruginous, margined at base with yellow; metathorax black, with four longitudinal yellow stripes, the lateral one sometimes slightly undulate, disc longitudinally sulcate and finely COLORATION IN POLISTES. 85 transversely wrinkled ; tegulz fulvous, margined with yellow. Abdomen strongly sericeous, fuscous or black, varied with dull ferruginous, especially on the second segment ; apical margin of each segment dilated laterally and slightly interrupted medially, and an irregular spot on each side, largest on the second segment yellow ; beneath ornamented much as above. Length 18 mm. Male.—Face and clypeus flat, and yellow as far up as and ona line with emargina- tion with the eyes, cheeks beneath broadly yellow; antennz long, yellowish beneath, apex black. Thorax without any ferruginouscolors ; sides of prothorax and pleura anteriorly beneath pale yellow; scutellum sometimes with only a lateral yellow dot at base; lateral stripes of metathorax sometimes wanting. Abdomen darker than in the female, sides of the second segment with a large ferruginous blotch inclos- ing a yellow spot; lateral margin of the first segment yellow; disc of the second, third, and fourth ventral segments with a large triangular yellow mark, apical mar- gins also yellow. Length 18 mm. Hab.—Texas. P. navajoe Cress. Female.—Head pale yellow, the venter, occiput, and tips of mandibles black ; a yellow spot on each side confluent with the orbits; clypeus rounded and ciliated anteriorly, sparsely punctured and pubescent; cheeks tinged with fulvous; an- tennz fulvo-ferruginous, the middle of flagellum black. Thorax velvety black ; . posterior border of the prothorax broader in front, tegule and scutellum ferru- ginous ; the mesothorax has sometimes a faint stain on each side; metathorax pubescent, obliquely striated, with a deep central longitudinal channel. Abdo- men longer than the head and thorax, sericeous, the first and second segments above and beneath, except apical margins, broader on the second segment, and the basal margin of the third segment above and beneath black ; the remainder lemon yellow ; anterior half of apical margin of the second gegment fulvous above, shading gradually into the yellow, sometimes entirely fulvous; third and remaining segments have a faint central longitudinal fulvous streak and a lateral spot of same color, that on third segment sometimes very distinct. Length 18-20 mm. Hab.—California, New Mexico, Arizona. P. flavus Cress. Female.—lLemon yellow ; base of clypeus, vertex, and cheeks tinged more or less with fulvous; clypeus subdepressed, subquadrate, sparsely punctured, each puncture giving out a short fulvous hair, apex angular, ciliated, the lateral angles with two teeth, the inner one the longest; tips of mandibles black; space between the antenne protuberant ; antenne entirely fulvous, darker above ; prothorax faintly tinged with fulvous, its posterior margin yellow; mesothorax fulvous, the incisures and a longitudinal black line on the disc abbreviated posteriorly, black ; a faint line on each side of the disc and the lateral margins yellow; metathorax with a deep central longitudinal channel; tegule with a median fulvous spot. Abdo- men bright lemon yellow, not longer than head and thorax, subsericeous, a central longitudinal streak posteriorly, a spot at the base of the first segment, a narrow line across the middle, slight stains on each side of the first and second segments at base, a transverse spot on each side near the apex of the second to fifth segments connected by a slender arcuated line (both above and beneath) fulvous; basal segment triangular, its apical breadth equal to its length. Legs faintly tinged with fulvous at base and on the tarsi. Wings varied with fuliginous, darker along the costa, a bright violaceous reflection ; base of both wings, a narrow longitudinal 86 COLORATION IN POLISTES. streak through the middle of the wing, and most of the second submarginal cell subhyaline ; stigma, costal vein from the stigma to the base, and the nervures at the base of the wing fulvous, the rest black. Length 18 mm. Hab.—New Mexico. P. anaheimensis Prov. Female.—Black with yellow spots ; clypeus, mandibles, front, orbits, an angular band above the antennz, and cheeks yellow; antennz beautifully honey yellow without any spot. Thorax black; anterior and posterior border of the prothorax, tegulze, spot on the sides, a line under the hind wings, the borders of the scutellum and postscutellum, and two longitudinal lines on the metathorax yellow. Abdo- men sulphur yellow; all the segments with a black band at the base; the band on the second and third segments is prolonged into a triangular point in the middle; the basal segment is black with the top yellow, and a lateral yellow spot contiguous * to the apical band. Feet yellow, coxce and two-thirds of the femora, with a spot on inside at end of hind legs, black. Wings smoky yellow, nervures brown. Male.—Abdomen almost entirely yellow, having only a narrow black line at the base of the segments. Length 17-18 mm. Ffab.—Anaheim, Cal. P, nestor (Fab.). Fuscous ; front yellow. Head fuscous, front and mouth widely yellow ; antennz black, ferruginous beneath. Thorax fuscous, with the anterior margin, two very’ fine oblique lines on the front, and two abbreviated lines under the scutellum yellow ; scutellum ferruginous. Abdomen ferruginous, a black spot at the base of each segment and the margin yellow. Feet ferruginous. Wings ferruginous; a little larger than Vespa marginalis. ffab.—North America, ‘ P. apachus Sauss. ‘‘Ferruginous, much marked with yellow. Abdomen subdepressed, ovoid ; segments bimaculate with sulphur-yellow ; mesothorax marked with two sulphur- yellow lines. Wings ferruginous.”’ Hab.—New Mexico and Sonora (Sauss.). P. carolinus (Linn.). * Saussure, who examined the type of Linnzeus, in London, states that the species which was originally put into the genus Vespa is a true Polistes, and is as long as Vespa crabro, His description is as follows: | ‘*Front yellow. Thorax ferruginous, with three black longitudinal lines. Abdo- men sessile, ferruginous ; fore wings blackish, hind wings hyaline.’’ Saussure further states that the species resembles P. bicolor, of South America, more or less, but with two yellow lines on the thorax and with the antennz brown. Hab.—Carolina. P. comanchus Sauss. rk Moderately stout ; margin of the clypeus rounded, not acute, dentate. Head ferruginous, vertex and middle of antenne black. Thorax black, bordered in front with ferruginous. Abdomen golden-yellow, the segments margined with yellow, base of the first black. FTab.—New Mexico. Io. Et. 12. 13. 14. 15. 16. BIBLIOGRAPHY. ; ALLEN, J; A. The influence of physical conditions in the genesis of species. The Radical Review. pp. 107-140. . BATESON, W. Materials for the study of variation, treated with especial regard to discon- tinuity in the origin of species. London and New York, pp. xvi + 598. > COCKERELE. 12 D:-A: The fauna of the mid-Alpine region of Custer coy Colorado. Trans. Am. Ent. Soc. . CRESSON, E. T. 1887. Synopsis of the Hymenoptera of America north of Mexico. Trans. Am. Ent. Soc. Suppl. vol., 1887, Philadelphia, pp. 350. iC RESSONS 1G be 1872-73. Hymenoptera texana. Trans. Amer. Ent. Soc., Iv, 1872-’73, « pp. 152-292. . DARWIN, CHARLES. 1859. The origin of species by means of natural selection. London and New York. . EIrMsErR, G. H. T. Die Artbildung und Verwandtschaft bei den Schmetterlingen. Jena, pp. 12-243; 4 Taf. . Fox, Wn. J 1899. Contributions to a knowledge of the Hymenoptera of Brazil, III (Vespidae). Proc. Ac. Nat. Sci., Phil., 1899. . Gray, A., and HooKER, J. D. 1881. The vegetation of the Rocky Mountain region and a comparison with that of other parts of the world. Bulletin of the United States Geo- logical and Geographical Survey of the Territories, vol. v1, No. 1, 1880. Guxick, J. T. 1893. Divergent evolution, through cumulative segregation. Journ. Linnean Soc., 1888, vol. xx, pp. 189-274. Rep. in Smithsonian Report for 1891. HAGEN, H. A. 1882. On the coior and pattern of insects. Proc. Am. Acad. Arts and Sci., PP 234-267. HENSLOW, GEORGE. 1888. The origin of floral structures through insect and other agencies. Internat. Sci. Series, vol. Lx111, New York, pp. x1 + 349. HopkKIns, K. G. -1896. The pigment of the Pieride: A contribution to the study of excre- tory substances which function in ornament. Philos. Trans. Roy. Soc. London, vol. 186, pp. 661-682. KEELER, C. A. 1893 Evolution of the colors of North American land birds. Occas. Papers of Cal. Acad. Sci., vol. 3. LEwIis, H. W. 1897. The Vespine of the United States and Canada. Trans. Am. Ent. Soc., XXIV, pp. 169-192. MAYER, A. G. 1866. The development of the wing scales and their pigment in butterflies and moths. Bull. Mus. Comp. Zool. Harv. Coll., vol. xx1x, pp. 209-236. (87) 88 17. 18, 19. 20. 21. 22. 23. 24. 25. 26. aii. 28. 29. COLORATION IN POLISTES. MAYER, A. G. 1897. On the color and color patterns of moths and butterflies. Bull. Mus. Comp. Zo6l. Harv. Coll., vol. xxx, No. 4, pp. 169-256. NEEDHAM, J. G. 1900. Some general features of the metamorphosis of the Weevil J/ono- myalius vulpeculus ———. Biol. Bull., Boston, pp. 179-I91. PAPPENHEIM, A. 1901. Grundriss der Farbchemie. Zum Gebrauch bei Mikroskopischen Arbeiten. Berlin, 1901, pp. 476. RADOWEZKOWSKY. Hor. Ent. Ross, VIII Bull., pp. 5, 6. ROMANES, G. J. 1897. Darwin and After Darwin, III. Post-Darwinian questions, Isolation and Physiological Selection. Chicago. ROUGET, A. Sur les coleoptéres parasites des Vespides. Dijon, Mem. Acad. Sci., 1, 1873, pp. 160-284. TOWER, W. L. Ig00. The development of the pigment and color-pattern in Coleoptera. Sctfence, n. s., vol. x1, No. 266, pp. 175-177. DE SAUSSURE, HENRI. 1852. Etudes sur le familledes Vespides. Paris, 1852-56. 2 vols., w. Supp. URECH, F. : 1893. Beitrage zur Kenntniss der Farbe von Insektenschuppen. . Zeitschr. f. wiss. Zool., Bd. 57, pp. 306-384. VON SIEBOLD, C. TH. E. 1871. Beitrage zur Parthenogenesis der Anthropoden. Leipzig, pp. 218. WALLACE, A. R. 1889. Darwinism. London and New York ; 16+ 494 pp. WEBSTER, F. M. 1898. The chinch bug; Its probable origin and diffusion, etc. U.S. Dept. Agri., Div. of Ent., Bull. No. 15. Washington. WEISMANN, AUGUST. 1872. Ueber den Einfluss der Isolierung auf die Artbildung. Freiburg. SS nae ‘MAMOUNAOM AO ANVN ' ‘aALVG *geqgstiTod UT WoTyeto[o) OSGOLT aLEL