HEAD MUSCULATURE 0F SPHINX MOTHS (LEPIDOPTERA: SPHINGIDAE) Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY RICHARD c. FLEMING ' 1968 LIBRAR‘X " ‘ Michigan Stave University THESIS This is to certifg that the thesis entitled Head Musculature of Sphingidae (Lepidoptera:$phingidae) presented by Richard C. Fleming has been accepted towards fulfillment of the requirements for Ph.D. Entomology degree in @4445} A 17% \ Major professor Da‘e May 9, l968 0-169 ABSTRACT HEAD MUSCULATURE OF SPHINX MOTHS (LEPIDOPTERA: SPHINGIDAE) by Richard C. Fleming An understanding of the myology of the adult head, as it occurs in the family Sphingidae is presented. All head muscles, exclusive of the intrinsic muscles of the proboscis, antenna, and labial palps, are described and compared for 15 species, representing 12 genera, and five subfamilies. Muscles are classified into groups and homologies are shown wherever possible.. Emphasis is placed on the role of head muscles used in the feeding process. It is established that some of the muscles associated with feeding are reduced in number and/or size in mothgfgfaghe subfamily Smerinthinae and of the genus Ceratomia of the subfamily Sphinginae. These moths are no longer capable of feeding activity. It is further established that, on the basis of the evolution of head musculature, the subfamily Smerinthinae is farthest removed from the hypothetical sphingid ancestral type. HEAD MUSCULATURE OF SPHINX MOTHS (LEPIDOPTERA:SPHINGIDAE) By Richard C. Fleming A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Entomology 1968 Q5/7112? ACKNOWLEDGEMENTS To all those individuals, Dr. Roland Fischer, Dr. Frederick Stehr, Mr. Julian P. Donahue, Mr. John Newman, and Mr. M. C. Nielsen, who aided in providing the many specimens needed in this study, I am deeply grateful. Considerable appreciation is expressed to Professors Gordon Guyer, Frederick Stehr, and Rollin Baker, of the Michigan State University staff, who criticized and offered their helpful sug— gestion during the preparation of this manuscript. And especially to Dr. Roland Fischer, of the Michigan State University staff, whose con- stant encouragement, worthwhile guidance, and thoughtful stimulation made this work possible, and without whose help it would not have reached fruition, I shall be ever thankful. A special note of thanks is due my mother, Mrs. Helen A. Fleming, for her considerable help in proof reading the original manuscript and to Mr. Duane T. Darling for his worthwhile advice in the preparation of the drawings. ii INTRODUCTION . METHODS . . . THE CRANIUM THE HEAD MUSCLES . MUSCLE DESCRIPTION . Head Muscles Head Muscles Head Muscles Head Mhscles Head Muscles in in in in in TABLE 0 0 OF CONTENTS Sphinginae . . Smerinthinae Macroglossinae . . . . Philampelinae . . . . . . Choerocampinae . . COMPARATIVE SUMMARY OF HEAD MUSCLES . . . DISCUSSION . SUMMARY . . . LITERATURE CITED . iii 15 21 24 27 32 34 36 38 43 52 55 Table II. III. LIST OF TABLES Pump Muscle Characteristics Number of Antennal, Labial Palp, and Proboscis Extensor Muscles Proboscis Lengths . . . . . iv 0 Page 40 41 42 LIST OF FIGURES Figures 1, 2, 3, 4: Generalized caranial muscles . 5: Paonias myops. ventral aspect of cranium . 6, 7: Manduca sexta. Cranial muscles . . 8, 9: Ceratomia undulosa. Cranial muscles . . . . . 10, 11, 12, 13: Ceratomia catalpae. Cranial muscles . 14, 15: Sphinx eremitus. Cranial muscles . . . . . l6, l7: Smerinthus geminatus. Cranial muscles . . . 18, 19: Paonias excaecata. Cranial muscles . 20, 21: Paonias myops. Cranial muscles . . . . . . 22, 23: Cressonia juglandis. Cranial muscles . . . 24, 25: Pachysphinx modesta. Cranial muscles . . . 26, 27: Haemorrhagia thysbe. Cranial muscles . . . 28, 29: Haemorrhagia diffinis. Cranial muscles . . 30, 31: Pholus satellitia pandorus. Cranial muscles . 32, 33: Ampeloeca myron. Cranial muscles 34, 35: Amphion nessus. Cranial muscles 36, 37: Celerio lineata. Cranial muscles Page 58 58 6O 6O 6O 62 62 62 64 64 64 66 66 66 68 68 68 INTRODUCTION It is surprising that, despite the great popularity of the Lepidoptera with amateur and professional entomologists so little work has been done on the morphology of the group. This is especially true of comparative internal structure. Scattered information con- cerning internal structure occurs in the literature and general text books, but comprehensive work has been sadly neglected. Probably the most significant study, which treats the com— parative morphology of feeding mechanisms in several families of moths and butterflies, is that of Schmitt (1938). More recently Ehrlich and Ehrlich (1962, 1963) have published papers that deal with the head musculature and thoracic musculature of butterflies respec— tively. The orientation of these papers is, however, taxonomic rather than morphological, their intent being to arrest the erroneous notion that non-skeletal features can be ignored with reference to the clas— sification of butterflies. The investigation reported here shall demonstrate that many species of sphinx moths do not feed as adults. General texts usually indicate that adult sphinx moths are nectar feeding insects. Ross (1965) states, ”The moths are extremely rapid fliers and feed on nectar.” Borror and DeLong (1964) say, ”Most of them feed much like hummingbirds, hovering in front of a flower and extending their proboscis into it...” Comstock (1924) states, ”As a rule they...have l 2 the habit of remaining poised over a flower while extracting the nectar..." Matheson (1951) comments, ”They are common visitors at flowers, sucking up the nectar with their long tongues.” While such statements are not completely untrue, generalization in this respect should be avoided for it appears evident that many adult sphinx moths are not morphologically equipped with those modi- fications necessary for feeding. Field observations, strongly reinforced by the internal and external morphological evidences to be presented here, substantiate the proposal that probably none of the Smerinthinae nor moths of the genus Ceratomia of the Sphinginae feed as adult insects. This work presents the myology of the head as it occurs in the family Sphingidae. The mechanisms involved in the feeding process, classification of the head muscles into logically arranged groups, muscle homologies, and the similarities and differences of muscles between 15 species, 12 genera and five subfamilies are discussed. It represents the first comparative study of several species of moths in one family. METHODS Freshly obtained moths were placed into Kahle's fluid where they were left for at least 48 hours in order to insure complete infiltra— tion and fixation of the tissue. The insects were removed from the Kahle's fluid and placed in 80 per cent ethyl alcohol to which had been added a few drops of glycerin. Mbths prepared in this manner will remain acceptable for dissection virtually indefinitely. The compound eye of the left side was removed to reveal the proboscis extensor and antennal muscles. In species where the former muscles reach a high degree of development, the cranial proboscis ex- tensor muscle was removed so that the tentorium and antennal muscles could be more easily observed. In many cases the number and position of the antennal muscles could be determined from this dissection, but in some cases it was found necessary to remove the antennal muscles of the left side, the brain, and the sucking pump so that the con— figuration of the antennal muscles (of the right side) could be ob— served from mesal (internal) View. In a few doubtful cases, portions of the tentorium, the antennal muscles, and basal part of the antenna were dissected out as a unit. A head was placed face upward in the dissecting pan and scales were removed so that the facial sutures could be observed. Frequently, a cleared head was placed next to the head to be dissected, as an uncleared head did not always show the facial features clearly. Most 3 4 of the frontoclypeal sclerite was removed, revealing the dilator muscles of the sucking pump. The nature and position of the sucking pump, the salivary duct, the brain, the subesophageal ganglion, and other internal structures were discerned by way of a sagittal cut completely through the cranium. By tilting the cranial half slightly the muscles of the labial palps could be observed. Details of the maxillary parts, labial parts, and other fea- tures of the ventral aspect of the cranium were obtained from cleared and uncleared crania without dissection. Microdissection is delicate and time consuming work and hence it was not practical, although it might have been desirable, to dis— sect long series of each Species. In most cases at least five or more heads of any species were used for study. With one or two ex- ceptions, little individual or sexual variation was found in the crania of any species. At least two drawings, a side View of the cranium with the eye removed, and a frontal view with a major portion of the facial sclerites removed, are presented for each species. The former illustrates the proboscis extensor muscles and the antennal muscles, while the latter shows the positions of the sucking pump dilator muscles and such cranial features as sutures, inflections, mandibles, position of the anterior tentorial pits, and the labrum. A generalized drawing represents a sagittal section of the cranium and illustrates intrinsic pump muscles, pump dilator muscles, 5 the salivary duct, the muscles of the salivarium, and the position of the brain and subesophageal ganglion. Another generalized drawing represents a ventral View of the cranium and shows the ventral sclerites and the position of the muscles of the labial palps. Features not pertinent are omitted from the drawings which are somewhat diagramatic and designed to aid the reader in a quick and accurate interpretation of the material discussed in the text. Drawings of the same species are in approximate scale to one another, but drawings of different species are not. Histological slides, using standard paraffin sectioning methods, were made of the sucking pumps of Manduca sexta and Paonias myops to reveal the structural nature of that organ. Dissections of the digestive tract of these species were also made. Action of muscles was determined by micromanipulation and deduction. If a technique could be worked out, observations of muscle action with living specimens might prove most meaningful. No such attempts were made in the course of this study. THE CRANIUM Few definitive statements concerning the musculature of the sphinx moth head can be made without some appreciation of the integu- mental anatomy of the head. The sphinx moth cranium is similar in most respects to other lepidopterans and a fair understanding of its morphology may now be attained. Differences between the crania of the five subfamilies of sphinx moths are minor. The literature contains several partial discussions and illus- trations for various groups of Lepidoptera, the most comprehensive being the work of Schmitt (1938). Other noteworthy works include those of DuPorte (1946), Madden (1944) on Manduca (=Protoparce) sexta, Short (1951) on Dilina tiliae, DuPorte (1956) on Danaus plexippus and Manduca quinquemaculata, Ehrlich (1958) on Danaus plexippus, Ehrlich (1960) on Epargyreus clarus, and Michener (1952) on Saturniidae. Other works that deserve attention, but do not limit themselves to the crania of Lepidoptera are Snodgrass (1935, 1947, 1960), Ferris (1943b), Cook (1944), and Matsuda (1965). The cranium of the lepidopterous head is a rather simple structure and only a few sutures are present. Figures 1-4 represent a generalized View of that structure as it occurs in the Sphingidae. The clypeus (Fig. 1) forms an area on the lower part of the face and is not clearly distinct from the frogs, This fact has given rise to considerable debate in the literature as to which part represents the 6 clypeus and which part the frons. While it is not the purpose of this paper to engage in a lengthy discussion concerning this point, it is of practical significance to indicate the present consensus. For many years the View held by Snodgrass (1935) and others that the facial sclerites may be interpreted properly by examination of cibarial and pharyngeal muscles, using them as landmarks, was, and still is, popular. Thfs, in the lepidopterous cranium, the area anterior to the suture that extends between the antennal sockets is the clypeus, as the cibarial muscles originate from it. The muscles posterior to the frontal ganglion are clearly the pharyngeal muscles, and their origin falls on a sclerite posterior to the suture between the antennae, and this would be the frons. Short (1951) follows this interpretation. In terms of generality the view is a solid one but the possibility of exceptions must be recognized. Certain doubt, however, has been cast upon the advisability of using muscles as landmarks. Cook (1944) indicates that the cranial muscle origins are relatively independent of ectodermal structures and origin, alone, cannot be used for interpretation. Origins of muscles, positions of sutures, and fixed points must be taken into considera— tion in relation to morphological possibilities. DuPorte (1956) main— tains that, Muscles are purely functional units and their mechanical efficiency depends in large part on their point of origin in the skeleton. With changes in the form or in the direction of growth of the organs they must often shift their origins even if this involves crossing a secondary inflection. Another interpretation adopted by Michener (1952), based on DuPorte (1946) holds that the trilobed structure, usually considered 8 the labrum, is, in reality, the reduced clypeus of the lepidopterous head, and the area posterior to it the frons, with the cibarial muscles shifted to that structure. This interpretation does not seem very likely. Ferris (1943b) questions the validity of the word "frons" in any insect, indicating that it means only a facial portion of the antennal segment and does not designate any separate morphological element. The interpretation of DuPorte (1956) and the terminology used by him will be followed largely in this paper. It should be pointed out that in his 1956 paper he rejects the interpretation he set forth in 1946. He states (1956) that in Manduca two sets of cibarial muscles (muscles 1 and 2_in the present work) originate from what is clearly the clypeal region, one pair (muscle 3) originates from what is prob- ably the frons but could be interpreted as the clypeus, and another pair (muscles 4_and'5) originate from an area that cannot be inter- preted as the clypeus. I have considered that area from where muscles l_and 2_originate the clypeal area, that area posterad of it and extending to the suture connecting the antennal sockets, the frontoclypeal area (Fig. 1), and that area just posterad of the suture, the frontal area (Fig. 1). The transfrontal suture (Fig. 1) extends between the antennal sockets. The laterofacial suture (Fig. 1), a continuation of the transfrontal suture, extends ventrad from the antennal sockets to the clypeolabral suture which is a mesal extension of the laterofacial suture on either side. A transclypeal band (Fig. l) meets the 9 extension of the laterofacial suture at a point even with the base of the anterior tentorial pits which lie in the laterofacial suture just above the outer edges of the labrum. The laterofacial suture is in- flected within the cranium as the laterofacial inflection (Fig. l) to which certain cranial muscles attach. In Celerio lineata the in- flection is considerably larger than in other species examined. In all species studied the bulging compound eyes extend below the base of the cranium. The crania in Sphinginae, Choerocampinae, and Philampelinae are rather elongate and large in relation to the rest of the body. In Macroglossinae the head is somewhat smaller in relation to the rest of the body, and in Smerinthinae it is decidedly smaller and less elongate than in any other subfamily. The three lobed labrum is comparatively small. (Fig. l). The longer outer lobes represent the pilifers (Fig. l, 4) while the inner lobe represents, in part, the epipharynx (Madden, 1944, and others). Little variation exists in the labrum among sphingid species but it is noteworthy that in Celerio lineata, on the lateral edges of the pilifers and firmly united with them, are small, elongate structures which were not present in any other sphinx moths studied. They are lightly sclerotized at their proximal end where they unite with the pilifers. The mandibles (Fig. 1, 4) are fixed. They are rather prominent in some sphinx moths and freed from the cranium, except at the base, in some species. They are considerably reduced in Smerinthinae. It is of historical interest that at one time the pilifers were considered the mandibular bursa. Rothschild and Jordan (1907) indicated the 10 incorrectness of calling these structures mandibles. They termed the true mandibular remnants ”Wangenfortsatz." (Genal ”cheek" processes.) In sphinx moths the proboscis may be very highly developed or degenerate. It may range in length, depending on species, from two to 90 millimeters (Table III). Mbths of the genus Manduca and the closely allied genus H§£§e_possess the most strongly developed proboscises of all North American sphingids. In Smerinthinae and a few Sphinginae (Ceratomia and Lapara) it is reduced and probably non—functional. Forbes (1948) noted the reduction of the structure in these groups and Rothschild and Jordan (1907) recognized the reduction of the proboscis in some Smerinthinae and the genus Ceratomia. Tillyard (1923) showed that the proboscis in Lepidoptera is derived from the galeae. Each half of the proboscis, where it is functional, forms the sides of a tube, the lumen of which is continuous with the body cavity through the stipes. Burgess (1880a) indicates each proboscis unit is rendered flexible by a series of fine rings separated by a membrane. Schmitt (1938) notes that in butterflies and higher moths the rings are made up of many small, flat circles of hard cuticula. The food channel is lined with similar rings but they have only about one third the width of the outer rings. Burgess (18803) suggested that the coiling of the proboscis is effected by the action of muscles passing obliquely between the rings. Berlese (1910) confirmed this finding. It was not until 1938 that the functional mechanism of the uncoiling action of the proboscis was described by Schmitt, although the sug— gestion that blood pressure might be the uncoiling agency was first suggested by Snodgrass (1935). Schmitt (1938) describes the activity 11 as dependent on the action of the proboscis extensor muscles (9 and 19_in the present paper) which insert on the flat, mesal sclerite of the stipes. A valve arrangement lies between this flat sclerite and the outer, tubular part of the stipes. As the muscles pull the stipes upward the valve closes, and the tubular part becomes a closed cylinder. In this manner pressure is exerted upon the blood in the stipes cylinder, and as the stipes is closed at its proximal end, the blood is forced outward through the stipes toward the proboscis. The posterior proboscis extensor muscle (11) attaches to the stipes near the base of the proboscis. Contraction of this muscle influences pressure on the blood within the stipes cylinder and also raises the base of the proboscis unit, bringing it in close contact with the stipes and effecting a tight seal with the functional mouth. Blood is forced out into the lumen of the proboscis, causing it to unroll. Schmitt points out that in many Lepidoptera the stipital cylinder may be modified, but the principle is invariably the same. He further states that the musculature concerned with extension of the proboscis con- sists of three basic pairs, but that in a large number of species one or two pairs may be absent. Functional maxillae always have at least two pairs. A series of careful observations and experiments by Eastham and Eassa (1955) cast doubt on the ”inflation" theory of Schmitt. They show that in the butterfly Pieris brassicae proboscis extension is the result of the contraction of oblique muscles within the proboscis coupled with the formation of a closed haemocoele in that structure. Further they contend that a fold at the base of both 12 galeae prevents the passage of fluid from the galea to the head and that the stipital aperture is constructed so that a closed galeal haemoccoel can result. In light of their findings it is apparent that more investigation of the extension mechanism of the proboscis is necessary. In sphinx moths, as in most other Lepidoptera, the stipes (Fig. 4) shows clear division into two parts; a folded, tubular lateral part, and a flattened mesal part. The stipes is reduced in Smerinthinae. The Eardg_(Fig. 4) is a small, immovable sclerite bordered laterad by the gena, anterad by the stipes, and posteromesad by the labial sclerite.(H5.4) Arising on the stipes, near the proboscis base, are the tiny, one segmented maxillary palps (Fig. 4). Forbes (1948) states that the maxillary palps are absent in sphinx moths. The labium (Fig7—4) is composed of a sclerite that extends from the well developed hypostomal bridge (Fig. 4) to the base of the proboscis. Much of it, especially the lateral parts, is only lightly sclerotized, but the area of heaviest sclerotization varies with dif- ferent species. It is bordered by the gena at the level of the three segmented labial palps (Fig. 4), and by the cardo and the stipes anterior to the labial palps. The most highly developed parts of the tentorium are the anterior arms. (Fig. 3). There are no dorsal arms. Schmitt (1938) indicates that dorsal arms do not occur in any adult Lepidoptera. In most cases, among Sphingidae, the anterior arms are rather straight. 13 In the Philampelinae, feeding members of the Sphinginae, Macroglossinae, and Choerocampinae they possess ventral and dorsal raised portions that afford broad muscle attachment surface for the powerful proboscis ex- tensor muscles and antennal muscles respectively. Many of these in- sects also possess a lateral ridge on the anterior arms that affords even more muscle attachment surface as well as extra structural strength to the arms. In the Smerinthinae only the dorsal raised portion of the arm is present. It is noteworthy that this group does not possess all three pairs of proboscis extensor muscles as found in species with fully functional mouth parts. Only the posterior proboscis extensor muscle remains, and that is considerably reduced. The ventral swellings of the anterior arms are very slight in the genus Ceratomia in the Sphinginae which do not have strongly developed proboscis extensor muscles. In the smerinthin species Cressonia juglandis the anterior arms have a distinctive horizontal ”S” shape not present in other sphinx moths examined. (Fig. 23). In all sphinx moths the anterior arms are united posterad by a rather narrow tentorial bridge. Where the tentorial bridge and the tentorial arms come together, one finds, in the postoccipital region of the cranium, the posterior tentorial pits. The posterior tentorial arms are evidently incorporated into the flared area where the pits enter the postoccipital suture. In the Smerinthinae, when viewed from below, the hypostomal bridge, labial palps, labium, cardo, and stipes are shifted anterad so the 14 tentorial bridge is clearly visible. (Fig. 5). This condition is not present in any other sphinx moth subfamily. In all sphinx moths good structural support is rendered to the ventral and posterior part of the cranium by the well developed hypo- stomal bridge. The small tentorial bridge probably offers little help in this respect. It should be noted that in a comparatively thin structure like the cranium, a danger of buckling exists. This is especially true in certain insects, including sphinx moths, with powerful cranial muscles. Short (1951) indicates that the inward inflection of the transfrontal suture (his ”epistomal ridge”) safeguards against this buckling. Other ridges and inflections, including the postoccipital ridge (Fig. 2) and laterofacial inflection, as well as the tentorium and the curvature of the cranium itself, render sufficient resistance to buckling. THE HEAD MUSCLES The literature on the musculature of the head of sphinx moths is not extensive, and, until the present work, no attempt has been made to compare the morphology of several species of the family. Berlese (1910) described and illustrated to some extent the muscles of the head of Sphinx convolvuli. Snodgrass (1935) described and illustrated some of the sucking pump muscles of "a sphinx moth.” Schmitt (1938) illustrated and discussed head muscles in several species, including Darapsa pholus, Haemorrhagia thysbe, and Smerininthus geminatus. Short (1951) illustrated some of the head muscles of Dilina tiliae. DuPorte (1956) pictured sucking pump muscles of Manduca quinquemaculata. Matsuda (1965) illustrates some of the head muscles of a sphinx moth. The following muscle group discussion and description includes all muscles of the sphinx moth cranium exclusive of the intrinsic muscles of the proboscis and the antennae. Muscle numbers indicate suspected and obvious homologies between the species. Muscle dif- ferences are pointed out in Tables I and II, in the figures, and in the discussions of the subfamilies. Figures 1-4 represent a gen— eralized condition. Dilator muscles of the sucking pump (Muscles l) 2, 33 fl, ) This group of muscles is the principal set used in expanding the sucking pump. They are better developed in species with strongly 15 16 developed feeding mouth parts than in those with degenerate mouth parts. Muscle 2_may be absent. Muscles of the wall of the sucking pump (Muscles 6, 7) Muscle §_is really a complex of several muscle bands that obviously constrict the pump. The oral valve muscle, 2) apparently constricts the oral opening and probably keeps ingested juices from escaping when they are forced into the digestive tract. While Z_may be considered one of the intrinsic pump muscles, it is always dis— tinct from the rest. It is similar in all Sphingidae. Histological sections of the pumps of Paonias myops and Manduca sexta clearly showed the muscular nature of most of that organ. Schmitt (1938) established that the muscles in Danaus menippe are arranged in two double-layered groups, one group transverse and the other longitudinal. My own studies did not clearly demonstrate this arrangement to be the case in Sphingidae, but there is little doubt of the several layered nature of the pump walls. Since dilator muscles, themselves, contribute to some of the pump musculature I am in agreement with Schmitt that some of the intrinsic pump muscles could have been derived from them. The sucking pump of most Lepidoptera is rather well developed, although it may be considerably reduced in non-feeding species. One of the first descriptions of the pump was offered by Kirbach (1883) using vanessa 32, Burgess (1880b) and Kellogg (1893) considered the same subject in the monarch butterfly. But the morphology of the pump was not understood until later. Snodgrass (1935) stated that 17 no definite statement can be made as to the morphology of the sucking pump of Lepidoptera without further study, but he indicates that the pump includes at least the buccopharyngeal region of stomodaeum. It should be noted that the dilators of that organ (muscles 1:5) are inserted on it, both anterad and posterad of the frontal ganglion. This phenomenon is helpful in determining which part of the pump is cibarial and which pharyngeal. Schmitt (1938) offers good evidence that at least part of the pump is composed of the cibarium, that structure defined by Snodgrass (1935) as, The food pocket of the extraoral or preoral mouth cavity between the base of the hypopharynx and the under surface of the clypeus. Schmitt points out that in orthopteroid insects a pair of muscles that compress the labrum originate on the anterior wall of the labrum and insert on the epipharyngeal wall. If the small lobe between the pilifers is the epipharynx, and it seems to be, this pair of muscles as it occurs in some Lepidoptera (apparently absent in sphinx moths) would indicate that the cibarium would then form the anterior section of the pump. Schmitt offers other evidence that the cibarium is included in the pump, based on the structure of its floor. He notes that at the base of the salivary meatus, in numerous generalized insects, there is a cup-like depression into which products of the medial salivary duct are poured. This depression, the salivarium (Fig. 2) is supplied with three pairs of muscles, the dorsal pair arising on the suspensory sclerite of the hypopharynx. 0f the three pairs this is the only pair that typically occurs in Lepidoptera. These muscles, 8, l8 originate on the floor of the sucking pump in Lepidoptera, indicating that the anterior part of the floor is derived from the hypopharynx and therefore this portion of the sucking pump belongs to the cibarium. The floor of the pump is quite heavily sclerotized in the Sphingidae. The insertions of muscles 4_and 5, posterior to the frontal ganglion, indicate that that part of the pump is the pharynx, as it is in certain other insects including Hymenoptera. Muscles of the salivarium (Muscle 8) This muscle evidently exerts some control over the release of salivary secretions from the salivarium. It is very weakly developed in non-feeding sphinx moths. Proboscis extensor muscles (Muscles 2, 19, ll) Muscles 2_and 19_raise the stipes and the role they play in proboscis extension has been previously described. Muscle ll_is in— volved with the creation of blood pressure within the stipes and the raising of the proboscis base. The maximum number of proboscis extensor muscles found in adult Lepidoptera is three pairs. In all sphinx moths with functional mouth parts this is the case, and muscle 2_is frequently the largest muscle in the head. It may obscure most of the antennal muscles from lateral view. In sphinx moths with degenerate, and apparently non— functional mouth parts, muscles 9_and lQ_may be reduced in size or absent. Musc1e_ll is always present but may be reduced in size. Antennal muscles (Muscles 12, 13, 14, 15, lg) In the Sphingidae four or five antennal muscles may be present. 19 When only four are present, muscle 1§_drops out. Elevation of the antennae is accomplished by the action of muscles 11_and 12, while depression is accomplished by muscles 14, 15, and 16, The positions and configuration of the antennal muscles represent a highly func— tional, well balanced system. As in all groups of Lepidoptera homologizing of antennal muscles in sphinx moths is not easy and I recognize the fact that I may be in error with some of the interpretations here presented, especially with reference to the Smerinthinae in comparison with other subfamilies. Muscles 11, 13, and 1§_would seem obviously homologous between all smerinthin sphinx moths, and in fact, between all sphingid species regardless of subfamily. Muscles 14_and 15, however, are less ob— viously homologized between smerinthins and other sphinx moths. Cook (1944) presents evidence for generalities usually accepted in considering musculature. The generalities suggest that insertions of muscles are always consistant and never shift their morphological re— lations and are, as far as known, absolutely dependent upon structures which they move. Thus muscle homologies may be judged best on the basis of their insertions. The antennal muscles considered to be homologous in the present work were judged on the basis of their insertions and relative posi- tions. Both criteria, although not infallible, should, in this case be valid, as the well balanced muscle system of the antenna could not have its muscle components shifted very much and still operate. The position of the antennae themselves do not shift in sphinx moth species. 20 Muscles of the labial palps (Muscles_11, 18) —_ Muscle 1Z_apparently moves the palp outward and depresses it, while muscle_1§ elevates that structure. Schmitt (1938) indicates that the presence or absence of labial palps muscles is a variable situation among lepidopterous families, but that the number is never more than two. My observations agree with this. Schmitt further states that sphinx moths may have one or two palp muscles and one is the usual case. The present study in— dicates that most sphinx moths have two muscles per palp, (Table II), although muscle 1§_may be extremely reduced and easily overlooked. While it is assumed that the origin of muscle 1Z_is on a part of the labial sclerite, it is certainly possible that this area rep— resents an anterior extension of the hypostomal bridge. There is no clear way of demarking the posterior edge of the labium, and the origin of the palp muscles cannot be used as a guide for determination of the sclerotized area. Schmitt (1938) showed that muscle 1Z_(his depressor muscle of the labial palp) may originate on the labial sclerite or the hypostomal bridge in Lepidoptera. Ehrlich and Davidson (1961) indicate that this muscle originates on the labial sclerite in Danaus plexippus. MUSCLE DESCRIPTION 1. Anterior cibarial dilator muscle (Figs. 1, 2) This unpaired but short and broad muscle arises on the lower portion of the clypeal area of the cranium and inserts on the cibarial portion of the sucking pump, anterad of other cibarial dilator muscles. 2. Medial cibarial dilator muscle (Figs. 1, 2) This paired, well developed muscle is frequently divided into two parts. It arises laterad of the mid line on the clypeal region of the cranium, posterad of 1_and inserts on the cibarial part of the sucking pump, opposite of its point of origin. 3. Posterior cibarial dilator muscle (Figs. 1, 2) This paired, flattened, heavy, frequently divided muscle is variously developed depending on species but it is always the largest of the cibarial dilator muscles. It may originate on the laterofacial inflection, the antennal ridge, or a point on the cranium just posterad of the transfrontal ridge, or a combination of these. It inserts on the dorsomesal region of the cibarial portion of the sucking pump. 4. Lateral pharyngeal dilator muscle (Figs. 1, 2) A paired, rather slender muscle that arises on the cranium posterad of the transfrontal suture, posterad and laterad of muscle 3_and inserts on the pharyngeal part of the sucking pump. 5. Medial pharyngeal dilator muscle (Figs. 1, 2) This small muscle may be paired, unpaired, or absent. It originates on the 21 22 frontal part of the cranium posteromesad of 4_and inserts on the pharyngeal part of the sucking pump, mesad of the insertion of 4, 6. Intrinsic pump muscles (Fig. 2) These heavy muscle bands are arranged in layers and, in conjunction with the dilator muscles, themselves, contribute to the walls of the sucking pump. 7. Oral valve muscle (Fig. 2) This well developed muscle extends across the anterior part of the pump. 8. Salivarium muscle (Fig. 2) This paired, usually small muscle extends from the hypopharynx to the salivarium. 9. Cranial proboscis extensor muscle (Fig. 3) This large, fan shaped muscle arises on the laterofacial part of the cranium along the laterofacial inflection and inserts on the flat, mesal sclerite of the stipes, laterad of other proboscis extensor muscles. It is absent in some species. 10. Anterior proboscis extensor muscle (Fig. 3) This fre- quently powerful muscle arises on the lateral and ventral surfaces of the anterior arm of the tentorium and inserts on the mesal sclerite of the stipes, just mesad of the insertion of 2, It may be absent in some species. 11. Posterior proboscis extensor muscle (Fig. 3) This well developed muscle arises on the mesal surface of the anterior arm of the tentorium and inserts at a distal point on the stipes, mesad of 19, 12. Anterior antennal levator muscle (Fig. 3) This large fan- shaped muscle arises on the dorsal or dorsolateral surface of the 23 anterior arm of the tentorium and inserts on the inner, lateral part of the scape. 13. Posterior antennal levator muscle (Fig. 3) This muscle, of similar shape, but always smaller than the other antennal muscles, arises on the dorsal surface of the tentorium, posterad of other antennal muscles, and inserts on the inner, posterior part of the scape. 14. Anterior antennal depressor muscle (Fig. 3) This strongly developed, fan-shaped muscle originates on the dorsolateral surface of the anterior arm of the tentorium and inserts on the inner, antero— lateral part of the scape. 15. Posterior antennal depressor muscle (Fig. 3) This moder- ately developed, fan-shaped muscle arises on the dorsal surface of the tentorium and inserts on the inner, anterior part of the scape. It may be absent in some species. 16. Mesal antennal depressor muscle (Fig. 3) This moderate to well developed, fan—shaped muscle arises on the dorsomesal surface of the anterior arm of the tentorium, mesad of other antennal muscles, and inserts on the inner, mesal part of the scape. l7. Anterior palp muscle (Fig. 4) This rather small muscle originates on the labial sclerite and inserts along the proximal, inner surface of the first segment of the labial palp. l8. Posterior palp muscle (Fig. 4) This muscle, which is usually smaller than_11, arises on the hypostomal bridge and inserts on the proximal, inner surface of the first segment of the labial palp. It is absent in some species. 24 Head Muscles in Sphinginae Species selected for morphological examination included Manduca sexta (Johansson), Ceratomia undulosa (Walker), Ceratomia catalpae (Boisduval), and Sphinx eremitus (Hfibner). Field observa— tions show that fit sexta and S, eremitus, as well as other moths in those genera, are active feeders as adults. 0n the other hand no member of the genus Ceratomia has ever been seen taking food. While obvious homologies exist between the head muscles of the Sphinginae, it is noteworthy that one genus, Ceratomia, shows con- siderable divergence from the basic pattern found in the subfamily. The divergence manifests itself primarily as a strong reduction in the size of some of the muscles. In general the reduction of the size of muscles corresponds with the apparent non-feeding habits of the species of Ceratomia. Dilator muscles of the sucking pump In general the dilator muscles are well developed in all feeding species examined. A figure in a paper by DuPorte (1956) of _M. quinquemaculata shows considerable development of these muscles in that species. I strongly suspect that all feeding members of this subfamily have strong dilator muscles. In the genus Ceratomia the muscles are only moderately developed and, unlike other moths considered in the present study, there is notable individual variation in both extent and number of subdivisions of some of these muscles. 25 1, (Figs. 6, 8, 10, 14) In M, EEEEE this muscle is much more highly developed than in the other members of the Sphinginae examined. 2, (Figs. 6, 8, 10, 14) Clearly divided into two parts (13, SS) only in.S, eremitus. A slight to moderate tendency toward sub— division is evident in the other species. Less highly developed in S, catalpae and S, undulosa than in fig §E§£§_and S, eremitus. S, (Figs. 6, 8, 10, ll, 12, 14) A muscle divided into several parts depending on the species. In.SL §E§E§.it is not divided and is very strongly developed. In S, undulosa it is only moderately developed and undivided. There is some individual variation in this muscle in this species, it being less heavily developed in some individuals than in others. In S, catalpae the muscle may be divided into two parts (S3, SS), but in many individuals it is not. The variations illus— trated represent the most noteworthy ones. The muscle showed other conditions not readily categorized. In some individuals it was moderately developed, while in others it was less robust with only a few strands. The tendency to split into several parts was greater in some specimens than in others. In no other sphinx moth was muscle_S so variable as in S, catalpae. In S, eremitus the muscle, well developed, was divided into three parts, (S3, SS, SE). 4, (Figs. 6, 8, 10, 14) A muscle showing little variation among the species of Sphinginae. 2, (Figs. 6, 10, 14) Absent in S, undulosa. Unpaired in S, eremitus. Paired in S, catalpae and M, sexta. Muscles of the wall of the sucking pump S, (Fig. 2) These muscles are least highly developed in the 26 two representatives of the genus Ceratomia of all sphinx moths studied; a condition indicative of their loss of feeding ability. In S, E3533 and S, eremitus the intrinsic pump muscles are well developed. Histo- logical study of the pump of SS §g§£§_clearly indicated the functional possibilities of the structure as a pumping organ. 2, (Fig. 2) Similar in all Sphinginae. Muscles of the salivarium S, (Fig. 2) A muscle weakly developed in S, undulosa and S, catalpae and rather strongly developed in_M. sexta and_S. eremitus. In fact, for moths used in this study, it was most highly developed in M. sexta. Proboscis extensor muscles 2, (Figs. 7, 9, 15) In M, §g§£§_and S, eremitus an extremely large muscle obscuring most of the antennal muscles. Size was con— siderably reduced in S, undulosa and altogether absent in S, catalpae. 19, (Figs. 7, 9, 13, 15) A rather powerful muscle in.gh §g§£§_ and S, eremitus, and much less powerful in S, undulosa and S, catalpae. _11. (Figs. 7, 9, 13, 15) In M,_§gx£§ and S, eremitus this muscle is quite extensive, but most of it is hidden from lateral view by the anterior tentorial arms themselves and the other proboscis extensor muscles. In S, undulosa and S, catalpae the muscle is less highly developed. It is noteworthy that the cranial proboscis extensor muscles show a significant reduction in size in members of the genus Ceratomia. 27 Proboscis extension would prove difficult and not very efficient, if indeed, it would be possible at all. Antennal muscles 1}, (Figs. 7, 9, 13, 15) Similar in all Sphinginae. 1S, (Figs. 7, 9, 13, 15) Similar in all Sphinginae. 13, (Figs. 7, 9, 13, 15) In S, undulosa and M, §g§£§_this muscle arises mesad of 11, while in S, catalpae and S, eremitus it arises laterad of 13, In S, catalpae it originates much farther anterad of the other antennal muscles. Such a wide separation of this muscle from the others was not found in any other moths con— sidered in this study. 15. (Figs. 7, 15) Absent in S, catalpae and S, undulosa. 16. (Figs. 7, 9, 13, 15) Similar in all Sphinginae. Muscles of the labial palps 11, (Fig. 4) Except for minor size variation, similar in all 5 Amhinginae. 18. (Fig. 4) Not present in S, catalpae. Head Muscles in Smerinthinae Species studied were Smerinthus geminatus (Say), Paonias excaecata (Smith & Abbot), Paonias myops (Smith & Abbot), Cressonia juglandis (Smith & Abbot), and Pachysphinx modesta (Harris). I have never observed any species of this subfamily feeding as adults. While the head musculature of the Smerinthinae shows variation within species of the group, homologies between the five species 28 examined are readily apparent. Even when compared to other sub- families, homologies are clear, but so are certain modifications which distinctly set this subfamily apart. The obvious non—feeding habits of these moths, and the corresponding reduction of feeding mouth parts are reflected to a significant extent by modifications of in— ternal head components, including muscles. An internal dissection of the digestive tract of S, myops dis- closed the fact that the crop is absent, another indication that it is a non-feeder. The lack of a crop was in sharp contrast to the condition found in_M. sexta (Sphinginae) whose digestive tract reveals a large crop existing as a diverticulum of the posterior part of the stomodaeum. Dilator muscles of the sucking pump The muscles of the sucking pump show a strong tendency toward subdivision. The muscles are always reduced in size as compared to other subfamilies (except the genus Ceratomia of the Sphinginae). But in some species the pump muscles are not so reduced as to rule out their ability to still perform the function of dilation of the sucking pump. The reduction in size of the dilator muscles is greatest in S, juglandis. While it is felt that the size reduction of the dilator muscles may be a significant factor in the lack of feeding ability, it is the presence of reduced mouth parts and re— duction or absence of the proboscis extensor muscles that indicate that these species are not capable of feeding. 1, (Figs. l6, 18, 20, 22, 24) Similar in all Smerinthinae. 29 2, (Figs. 16, 18, 20, 22, 24) Divided into two parts in all species examined. Except in S, juglandis, where S§_and SS_are of equal size, S§_is smaller than SE, S, (Figs. l6, 18, 20, 22, 24) Divided in all species except _S. excaecata where the division is not complete. In S, juglandis, S, modesta and S, geminatus the muscle is divided into three parts (SE, SE, 22). In S, juglandis and S, modesta the origin of the parts is similar, but S, geminatus shows some differences in this regard, as indicated by the figures and muscle summary. In_§..mygp§_the muscle is divided into two parts, (SE, SS). It is apparent that in all sphinx moths this muscle is the most powerful and well developed of the sucking pump muscles. In the Smerinthinae it is not so heavily developed as in certain members of other subfamilies. In S, juglandis the poor development of this muscle is especially notable. 3, (Figs. 16, 18, 20, 22, 24) Similar in all Smerinthinae, but more highly developed in S, modesta than in the rest. 2, (Fig. 16) Absent in all Smerinthinae examined except _S. geminatus where it is small and paired. Muscles of the wall of the sucking pump S, (Fig. 2) Histological study of the pump of S, Ezgpg, which may be considered typical of the Smerinthinae with reference to the pump, demonstrated the muscular nature of that organ. It should be noted, however, that the intrinsic musculature of the pump is not nearly so well developed as in feeding species such as M, sexta 30 (Sphinginae) wherein histological examination revealed heavy muscle development. In Smerinthinae the pump takes up a great deal of room within the cranial capsule. The size of the pump of S, geminatus prompted Schmitt (1938) to comment, ...this development of the sucking pump has reached such a point that little space is left for the brain and the suboesophageal ganglion. While this might first appear to be the case I do not feel that the sucking pump in any of the Smerinthinae has reached the high point of development suggested by Schmitt. The apparent ”crowding" of other cranial components by the pump is evidently the result of the reduction of the size of the cranium which is, in all Smerinthinae studied, considerably smaller in proportion to the rest of the body than in other sphingid subfamilies. It is probably true that the size of the sucking pump in Smerinthinae is not reduced significantly over the size of the fully functional pumps of their probable feeding ancestors. In fact, if one were to consider feeding incapability on the basis of pump morphology alone, one would rapidly conclude that the pump of Smerinthinae could well be functional. Other morphological considerations render the conclusion that the whole subfamily is a non-feeding one. Schmitt (1938) makes no direct reference to the fact that _S. geminatus is a non—feeding species, although he makes a general statement that fully functional mouth parts must have at least two pairs of proboscis extensor muscles. 31 In the course of the present investigation I found it not impossible to misjudge the size of the sucking pump. Besides in- dividual variation, which seems to be slight, it is possible to observe the pump in a dilated or contracted state. If a specimen with the former preserved condition was under observation, the pump, of course, would appear larger than if the alternative were the case. 1, (Fig. 2) Always distinct and similar except for minor size variation in Smerinthinae. Muscles of the salivarium S, (Fig. 2) Small in all Smerinthinae with little specific variation. Proboscis extensor muscles 2, and 1S, Absent. 11, (Figs. 17, 19, 21, 23, 25) Rather small but present in all Smerinthinae. This muscle was observed by Schmitt (1938) in S, geminatus. He considered it to be the posterior proboscis ex- tensor muscle, based on judgement of its insertion. The present study lends validity to Schmitt's conclusion. The fact that none of the Smerinthinae examined in this report possessed a full compliment of proboscis extensor muscles renders substantial evidence that they are non—feeders. It is not conceivable that the proboscis extension mechanism could function without either one more set of extensor muscles, or at least heavier development of the single pair of muscles (11) thaQ is present. 32 Antennal muscles 12, (Figs. l7, 19, 21, 23, 25) Similar in all Smerinthinae. In S, juglandis the muscle diverged, somewhat, from the general smerinthin pattern in that it was smaller and its origin was mostly on the dorsal, rather than the dorsolateral surface of the anterior tentorial arm. 12, (Figs. l7, 19, 21, 23, 25) Only slight specific variation occurs. 1&, (Figs. l7, 19, 21, 23, 25) Similar in all Smerinthinae. 15. (Fig. 23) Present only in S, juglandis. H 6. (Figs. 17, 19, 21, 23, 25) Similar in all Smerinthinae. Muscles of the labial palps 17. (Fig. 5) Similar in all Smerinthinae. 1S, Absent in Smerinthinae examined. Head Muscles in Macroglossinae TWO species, Haemorrhagia thysbe (Fabricius) and Haemorrhagia diffinis (Boisduval) were selected for morphological studies. Field observation indicates that both species are active adult feeders. In the Macroglossinae the head muscles are so well developed that very little free space is present within the head capsule. The musculature agrees closely with that of feeding species in other sub— families. There is little significant difference between the two species with reference to head musculature. Schmitt (1938) examined the 33 proboscis extensor muscles in S, thysbe. Dilator muscles of the sucking_pump 1, (Figs. 26, 28) Virtually identical in both species. S, (Figs. 26, 28) Similarly divided into two parts (2a, 2b) in both species. S, (Figs. 26, 28) Highly developed in both species. In S, thysbe the muscle is usually divided into two distinct parts (3a, 3b), but the division may not always be complete. In S, diffinis the division of the muscle is not complete. &, (Figs. 26, 28) Mbderately well developed and similar in both species. S, (Fig. 28) Apparently absent in S, thysbe, although one specimen showed an extra pharyngeal dilator muscle which appeared to be a subdivision of 3, as the division from £_was not complete. In _S. diffinis muscle S_is distinct. Muscles of the wall of the sucking pump 6. and 1, (Fig. 2) Virtually identical in both species. Muscles of the salivarium S, (Fig. 2) Similar in both species. Proboscis extensor muscles 9., 19,, 11, (Figs. 27, 29) All very well developed and similar in both species. 34 Antennal muscles 12, (Figs. 27, 29) Arising on the dorsal surface, in S, diffinis, or the dorsolateral surface, in S, thysbe, of the anterior tentorial arm. Similar in other respects in both species. 13., 14., 1S,, 1S, (Figs. 27, 29) A11 similar in both species. ———— Muscles of the labial palps 17., 1S, (Fig. 4) Similar in both species. 1Z_is the largest —— of the muscles. Head Muscles in Philampelinae Species examined were Pholus satellitia pandorus (Hubner), Ampeloeca myron (Cramer), and Amphion nessus (Cramer). Field study shows that these species, as well as other members of the subfamily, feed as adults. Head musculature in the Philampelinae is quite similar to that found in feeding members of other subfamilies. Muscles associated with the feeding structures are among the most highly developed of any sphinx moth group. Some of the head muscles of Darapsa pholus are figures by Schmitt (1938). Dilator muscles of the sucking pump 1, (Figs. 30, 32, 34) Similar in all Philampelinae examined. 2. (Figs. 30, 32, 34) Not divided into two parts in S, myron and S, nessus, but divided into two parts (S3,_SS) in_S. satellitia. 35 S, (Figs. 30, 32, 34) An extremely well developed muscle in all three species and divided into two parts in each (SS, SS) but the division in S, gyggg_is quite different from the other two species. In S, Eyggg_muscle SS_is rather small while SS_is quite extensive and partly divided, that part originating on the frontal area of the cranium probably homologous with SS_in the other species, while that part originating on the antennal ridge is probably homologous with S3_ in S, satellitia and S, nessus. Muscle SS_is more highly developed in_S. satellitia than in S, nessus. S, (Figs. 30, 32, 34) Similar in Philampelinae examined. 5. (Figs. 30, 32, 34) Paired and similar in the three species. Muscles of the wall of the sucking pump S,, 2, (Fig. 2) Similar and well developed in Philampelinae. Muscles of the salivarium S, (Fig. 2) Similar in all three species. Proboscis extensor muscles 2,, 19,, 11, (Figs. 31, 33, 35) All strongly developed and similar in Philampelinae. Antennal muscles 13,, 1S,, 14., 1S,, 1S, (Figs. 31, 33, 35) The five muscles, two levators and three depressors, form a similar and well balanced system in Philampelinae. 36 Muscles of the labial palps ,11,, 1S, (Fig. 4) These muscles are similar in Philampelinae examined. 1S is smaller than 12, Head Muscles in Choerocampinae Celerio lineata (Fabricius) is the only member of the Choero— campinae that the author has had an opportunity to examine. There— fore specific assumptions cannot be made with reference to how typical the muscle arrangement of S, lineata is for its subfamily. Still it can be assumed that the head musculature of the species is represen— tative of a large portion of the possibilities within the subfamily. The head musculature is extensive and all systems are well developed. This species feeds extensively as an adult and the capability in this respect is well substantiated by anatomical fact. Dilator muscles of the sucking pump 1, (Fig. 36) Rather small when compared to that of other feeding species. S, (Fig. 36) Divided into two parts (S3,_SS). 22.13 dis- tinctly compressed laterally, not rounded as in most sphinx moths. S, (Fig. 36) The largest and most powerful pump muscle in this species as in the others. There is only a hint of subdivision into two parts. 4., S, (Fig. 36) Muscle S_is smaller than S_and paired. 37 Muscles of the wall of the suckigg pump S,, 1, (Fig. 2) Well developed in S, lineata. Muscles of the salivarium S, (Fig. 2) Rather small in this species. Proboscis extensor muscles 2,, 19,, 11, (Fig. 37) All these muscles are strongly developed in this species. S_arises mostly on the inner surface of the well developed laterofacial inflection. Antennal muscles 12,, 1S,, 1S,, 1S,, 1S, (Fig. 37) Three depressors and two levators are present. This well balanced system is similar to other sphinx moths with five antennal muscles. Muscles of the labial palps 12,, 1S, (Fig. 4) ‘1Z_is comparatively small in S, lineata. 1S_is extremely small and easily overlooked. COMPARATIVE SUMMARY OF HEAD MUSCLES (Tables I, II) 1. Anterior cibarial dilator muscle: Because so little varia— tion occurs in this muscle from species to species, no characteristics are assigned to it. Generally, it is more developed in feeding species than in non-feeding ones. It is never paired. 2. Medial cibarial dilator muscle: Characteristics: A: Distinctly subdivided into two separated parts. B: Subdivided into two parts but parts contiguous. C: No subdivision. 3. Posterior cibarial dilator muscle: Characteristics: A: Not subdivided. A.1: Origin on laterofacial inflection and antennal ridge. A.2: Origin on antennal ridge entirely. A.3: Origin on antennal ridge and frontal area of cranium. A.4: Origin on laterofacial inflection, antennal ridge and frontal area of cranium. B: Subdivided into two distinct parts. B.1: Origin of first (anterior) division on laterofacial inflection only. Second division on antennal ridge only. B.2: Origin of first (anterior) division on laterofacial inflection. Second division on antennal ridge and frontal part of cranium. B.3: Origin of first (anterior) division on laterofacial inflection and antennal ridge. Second division on frontal region of cranium. 38 C: B.4: 39 Origin of first (anterior) division entirely on antennal ridge. Second division on frontal region of cranium. Subdivided into three distinct parts. C.1: 0.2: 4. Origin of first (anterior)division entirely on laterofacial inflection. Second division on laterofacial inflection to antennal ridge. Third division entirely on antennal ridge. Origin of first (anterior) division on latero- facial inflection and antennal ridge. Second division on antennal ridge. Third division on frontal region of cranium. Lateral pharyngeal dilator muscle: This single, paired muscle varied only in size from species to species. It was present in all and its points of origin and insertion were virtually identical in all. B: 5. Medial phagyngeal dilator muscle: Characteristics: Muscle present. A.1: A.2: Paired. Unpaired. Muscle absent. 40 Table I PUMP MUSCLE CHARACTERISTICS Sphinginae: Manduca sexta Ceratomia undulosa Ceratomia catalpae Sphinx eremitus Smerinthinae: Smerinthus geminatus Paonias excaecata Paonias myops Cressonia juglandis Pachysphinx modesta Macroglossinae: Haemorrhagia thysbe Haemorrhagia diffinis Philampelinae: Pholus satellitia Ampeloeca myron Amphion neasus Chaerocampinae: Celerio lineata *Considerable variation Muscle_1 Muscle_S A.3 B.4 B.4 B.4 B.2 B.4 A.2 41 Table II NUMBER OF ANTENNAL, LABIAL PALP, AND PROBOSCIS EXTENSOR Sphinginae: Manduca sexta Ceratomia undulosa Ceratomia catalpae Sphinx eremitus Smerinthinae: Smerinthus geminatus Paonias excaecata Paonias myops Cressonia juglandis Pachysphinx modesta Macroglossinae: Haemorrhagia thysbe Haemorrhagia diffinis Philampelinae: Pholus satellitia Ampeloeca myron Amphion nessus Chaerocampinae: Celerio lineata MUSCLES Antennal Muscles Palp Muscles Proboscis Ext. Muscles 42 Table III PROBOSCIS LENGTH Number after specific name represents number of specimens examined. (*) indicates moths not selected for internal dissection. Lengths are in millimeters. Range Average Sphinginae: Manduca sexta (5) 66—93 80.0 Manduca gginquemaculata* (2) 87-93 90.0 Ceratomia amyntor* (2) 11—13 12.0 Ceratomia undulosa (5) 9-11 9.8 Ceratomia catalpae (5) 4—5 4.4 Sphinx eremitus (3) 38—40 39.0 Sphinx chersis* (2) 41-51 46.0 Sphinx kal’é‘miae* (1) 40.0 Sphinx drupiferarum* (1) 44.0 Smerinthinae: Smerinthus geminatus (5) 2-3 2.9 Paonias excaecata (5) 3-4 3.4 Paonias myops (5) 2—3 2.6 Cressonia juglandis (4) 2—3 2.5 Pachysphinx modesta (5) 3-5 4.0 Macroglossinae: Haemorrhagia thysbe (5) 18—20 19.4 Haemorrhagia diffinis (2) 17-17 17.0 Philampelinae: Pholus satellitia (2) 33—36 34.5 Ampeloeca gyron (3) 14-15 14.7 Deidamia inscriptum* (3) 13-14 13.3 Amphion nessus (5) 15—17 16.0 Choerocampinae: Celerio lineata (5) 34-41 37.0 DISCUSSION It is clear from the morphological and field evidence presented that many species of sphinx moths have lost their ability to feed as adult insects. It is apparent that none of the Smerinthinae examined could possibly feed. Their morphological equipment is simply not adequate. The lack of two pairs of proboscis extensor muscles, and the reduction of the proboscis are noteworthy in this respect. It is surprising to find only one or two statements in the literature that indicate the lack of ability of moths of this subfamily to use the proboscis. Rothschild and Jordan (1907) state: Russel nie uber den Hinterleib hinausragend, zuweilen zu zwieiganz kurzen Lappen verkummert, bei den meisten Arten nicht mehr als ein Saugorgan brauchbar. Others, including Forbes (1948) and Holland (1941) indicate the reduced proboscis in this group, but say nothing about its function. It is clear, also, that certain members of a typically feeding subfamily, Sphingidae, are non-feeders. Members of the genus Ceratomia considered in this investigation are certainly not capable of being successful at taking food, but these insects are non—feeders for different morphological reasons, in part, than is the case in representatives of the Smerinthinae. While Ceratomia shares the re— duced mouth parts of the latter, there is only reduction in the size, not loss, of the proboscis extensor muscles in S, undulosa, all three 43 44 typical pairs of muscles being present, and reduction in size as well as loss of one set of muscles in S, catalpae. The possibility of three muscles still being useable cannot be ruled out. However, in Ceratomia the sucking pump and its associated muscles have become so reduced that it is apparently incapable of functioning. S, catalpae deserves some special attention for in that species, where the tendency to lose functional feeding apparatus is more ad— vanced than in S, undulosa, there is, more than in any other sphinx moth used in this study, considerably more individual variation of muscles associated with the sucking pump. There is a possibility that this insect is presently in a state of losing these muscles since some individuals have fewer and/or smaller muscles than others. In this respect it has not yet reached a fixed genetic state as have, apparently, other non—feeding species. Thus far adaptation has been mentioned only briefly and that matter, of course, is pertinent to this discussion. Classical bio— logical thought puts great store in structural and functional relation- ships and there is nothing really wrong with this line of thought. A structural feature of an organism frequently does have adaptive significance. A paper by Ferris (1943a) that deserves more attention than it probably will ever get, discusses the problem of adaptation and its meaning for the morphologist. He states: It is the contention here maintained that morphologists in general, whether aware of it or not, have kept hidden away in their thinking their own comforting dryad, their own animated spirit, to which they have appealed whenever they have gotten into difficulties. If a morphological situation is encountered which does not yield readily to analysis, recourse is immediately had to the dryad. 45 The name Dr. Ferris assigned to his dryad is adaptation, some sort of outward power which imposes itself upon the inward mechanisms of creatures and bends them to its will. Appeal is made to this dryad when the morphologist remarks that morphology must be allied to function, since it must see forms as plastic physical adaptations to the work that has to be done. But morphologists are bothered by the dryad who inserts a muscle here, reduces it there, puts in a suture, or a bend, or reduces a proboscis over there, so that they may be dismissed as ”merely adaptations." Dr. Ferris offers a way to remove the need for having a dryad to watch over morphological situations and recognize adaptation for what it really is by suggesting that all evolution depends upon the capacity of genetic materials to bring about change, that those changes can be only those that have been prepared for by the preceding steps in the evolution of genetic materials, that no evolutionary step is a phenomenon independent of other steps, that change may be in every direction within the genetic capacity, without regard to the environment, the latter acting only as a sorting device determining which changes shall survive, that adaptation is progressive sorting of continued changes, each change being dependent on the last set of changes, and that the evolutionary process does not consist only of endless change, but also of the setting of limits which confirm future changes. Evolution, then, becomes in part a process of "sequential stabilizations of genetic patterns" which involves both the loss and limiting of capabilities and the attainment of them. 46 Certainly the present morphological study presents good cases for studies in adaptation. One can notice the refinement of the design of feeding mechanisms in any of our feeding sphinx moth species. Few insects are better equipped to draw nectar from deep throated flowers. The mechanism reaches the ultimate development in some sphinxes, as in the genus Manduca. It could be said that the morpho— logical modifications associated with these structures are highly adaptive, and it must be admitted that these moths may, indeed, have feeding advantages over others with less highly developed feeding features. But what of the members of the genus Ceratomia and the sub- family Smerinthinae? The question of adaptation, Dr. Ferris's dryad, comes up. The casual observer might say quickly that the non~feeders are at an adaptive disadvantage because they can't feed. The flaw in this reasoning is obvious. Biological success is based on repro- ductive success. If a creature can mate, lay eggs, and give rise to future generations without having had a bite to eat or a drop to drink it has been a biological success. In this respect one might consider the ability to survive without having to trouble oneself with feeding an adaptive advantage. There is no doubt as to the success of the members of the Smerinthinae and the genus Ceratomia considered in this paper. Members of these groups are among our most abundant species. Ardrey (1966) contributes the following statement that de— serves the attention of morphology. He says, ”A bird does not fly because it has wings; it has wings because it flies." He suggests 47 that such a statement, at first, seems to be a triumph of obviousness, of absurdity, or of unimportance but that upon a little reflection something in the statement begins to nag at you. The statement really asks one to think about the relationship of body to behavior. It is not very logical, of course, to think of a bird that could fly without wings. But is it any more logical to think of a creature with wings that does not have a behavioral pattern compelling it to use them? It is fitting to apply this reasoning to feeding, or lack of it, in sphinx moths. Do sphinx moths feed because they have the mor- phological equipment to do so, or do they have the morphological equipment because they feed? According to Mr. Ardrey a positive answer to the latter part of the question would get closer to the truth. Putting this in terms of generality this reasoning states that behavioral acquisitions precede structural ones. One could possibly take issue with which came first, but there is little denying that behavioral traits have a great deal to do with whether or not struc- tural acquisitions will gain a selective advantage or even manifest themselves at all. It might be assumed that the ancestors of the Smerinthinae, who were capable feeders, gradually stopped feeding, their behavioral acquisitions not demanding that their functional mouth parts be used. Thus as reduction of the functional mouth parts took place, and tended to become non—functional, the moths were not put at any selective disadvantage. The morphological literature clearly shows that morphologists have, until recently, practically ignored behavior all together as 48 they applied their skills and logic to answering questions of adapta— tion, phylogeny, and speciation. I should like to put in a plea to morphologists and other biologists concerned with these matters to start taking a careful and long look at behavior. It is felt that the Ardrey proposition is best resolved by suggesting that birds fly because they have wings and have wings because they fly, and that sphinx moths feed because they have func- tional feeding structures, and have functional feeding structures be- cause they feed. Behavior and structure must be combined in our thinking. The two have to be considered together. They must go hand in hand. One without the other results in morphological chaos. Little is known about sphingid ancestory and nothing was un— covered in the literature concerning it. It must be pointed out that any conclusions in this respect, in the present paper, are somewhat hypothetical. Any morphologist reaches his evolutionary conclusions with various degrees of validity. It may be stated, for instance, that such and such a structure was derived from a more primitive type that may have had certain features. The conclusions made in this respect are apt to reach a higher degree of validity in direct pro— portion to the amount of observation the worker has made on the modi— fications of the structure as it occurs in a long series of species that possess the structure. The attempt has been made to show muscle homologies in sphinx moths and while it would be most presumptuous to base a phylogeny of sphingid subfamilies just on the basis of head muscles and cranial structure, it is possible, I feel, to point out some valid relationships. 49 It is an accepted assumption among morphologists that muscles may split, drop out, and change origins. It can be assumed that the sphingid ancestor was a feeding creature, with a full complement of head muscles associated with feeding. From that ancestory there branched one group which are all non—feeders; the Smerinthinae, whose head muscle complement is no longer complete. These moths are dis— tinctly set apart from other sphinx moths on this basis and other aspects of the head. From a feeding line, more recently in time probably, there branched from another subfamily of typical feeders, a small group of non-feeders; the genus Ceratomia of the Sphinginae (other genera, in- cluding Lapara which has a much reduced proboscis, might fit in this group also). In Ceratomia the tendency to lose feeding abilities is less advanced than in the Smerinthinae, and, in part, for different morphological reasons as has been already noted. The plastic evolu— tionary state, as evidenced by S, catalpae, in regard to loss of feeding ability and modification of head muscles concerned, is note— worthy. It demonstrates, at least, what steps, might have been taken by other sphinx moths as they ”advanced” from a feeding to a non— feeding state. Namely loss of muscles, reduction in size of muscles, reduction in the size of the sucking pump (not in Smerinthinae) and proboscis reduction. The Macroglossinae, Philampelinae, and Choerocampinae would, on the basis of head muscle configuration seem rather closer to one another than to the other subfamilies, and closer to the Sphinginae 50 than the Smerinthinae. More investigation is needed in this respect, however, before decisive conclusions can be drawn. It may be noted that specific differences, when all the head muscles and their characteristics are considered, are rather clear cut. While species, within the Sphingidae, are not a taxonomic problem in most cases, it is important to notice differences of muscle patterns. In other groups of Lepidoptera, where the determination of the species is more difficult, internal examination may be helpful in settling taxonomic problems. It is the contention maintained here that too long have many taxonomists ignored internal morphology. Lepidopterists seem to be especially guilty. DuPorte (1965) states that taxonomic categories are largely segregated on morphological characters and therefore the taxonomist should be a competent morphologist, able to recognize homologies and to use the same names in all groups of homologous structures, thus bringing about uniformity in taxonomic terminology. DuPorte (1965) with good reason, recognizes the problem in— volved with convergence or parallelism influencing two species with remote common ancestory to resemble each other quite closely. He contends that they are grouped together because of community of morphological characters and not primarily because of phylogenetic relationship. It is urged therefore that in segregating taxa as many characters as practicable should be used. None the less I feel that it is possible to suggest that the most generalized condition within the sphinx moths would be a situation wherein the sucking pump was moderately to well developed, the proboscis 51 was moderately developed, one undivided, and two paired sets of cibarial dilator muscles were present, two pairs of pharyngeal dilator muscles were present, three pairs of proboscis extensor muscles were present, two pairs of labial palp muscles and five pairs of antennal muscles were present. Divergent plans which may be considered progressive rather than conservative, include a reduction of the length of the proboscis as in Ceratomia and the Smerinthinae, or a more extensive development of that organ as in Manduca, a reduced sucking pump develop- ment, as in Ceratomia, division of the cibarial dilator muscles, re- duction in the size of the latter, dropping out of one or more sets of muscles, as in the proboscis extensor muscles of Smerinthinae, loss of one set of labial palp muscles, and loss of one set of antennal muscles. If these criteria are born in mind, the moths of the subfamily Smerinthinae must be considered most divergent from the basic plan and, with reference to their head morphology at least, the least primitive. SUMMARY The morphology of the head musculature of 15 species of sphinx moths, representing five subfamilies was studied. Special emphasis was placed of the muscles associated with feeding mechanisms. Endo- and exoskeletal structure of the cranium was considered wherever necessary. Four points are to be stressed by way of conclusion: 1. In the generalized sphinx moth head a maximum of 18 muscles may be found exclusive of intrinsic proboscis and antennal muscles. Most of the muscles are paired. The muscles have been classified into the following six groups: 1) the dilator muscles of the sucking pump, 2) the muscles of the wall of the sucking pump, 3) muscles of the salivarium, 4) proboscis extensor muscles, 5) antennal muscles, and 6) muscles of the labial palps. The dilator muscles of the sucking pump are always present, but may be divided or reduced in size or number. The most anterior of these muscles is never paired but the others always are except for the most posterior one which is sometimes unpaired or absent. The latter is the only muscle of this group to ever drop out. Muscles of this group dilate the sucking pump. Muscles of the wall of the sucking pump are always present but may be reduced as in the moths of the genus Ceratomia. These muscles constrict the pump. 52 53 Muscles of the Salivarium are always present but may be reduced, as in all Smerinthinae. These muscles exert forces on the salivarium which presumably control its secretions. The proboscis extensor muscles may be reduced to one set, as in all Smerinthinae, or two sets, as in S, catalpae. All other sphinx moths examined had three sets but these may be reduced as in S, undulosa. These muscles exert forces on the stipes which press may blood into the outer lumen of the proboscis, thereby extending it. The antennal muscles are well developed in all species and consist of four or five sets. Two sets elevate the antennae and three sets depress them. Four sets were found in most Smerinthinae and the genus Ceratomia. The muscles of the labial palps occurred in two sets in most species, but only one in S, catalpae and the Smerinthinae. These muscles depress, elevate, and move the palps outward. 2. Loss of the necessary feeding musculature, reduction of the proboscis, and correlated field observations indicate, most clearly, that members of the subfamily Smerinthinae do not feed as adults. 3. Reduction of the size of the sucking pump, reduction of the size or loss of the necessary feeding muscles, and reduction of the proboscis, correlated with field observations, clearly show that members of the genus Ceratomia do not feed as adults. 4. The sphinx moth ancestor was a feeding species. The most removed from the primitive type are members of the Smerinthinae. On the basis of internal and external cranial morphology Macroglossinae, Philampelinae, and Chaerocampinae show more affinities with one 54 another than with the Sphinginae, but the Sphinginae are more similar to them than to the Smerinthinae. If head morphology is considered, it may be assumed that Macroglossinae, Philampelinae, and Choero— campinae are closer to the primitive sphinx moth type than are the Sphinginae and Smerinthinae. LITERATURE CITED Ardrey, R. 1966. The territorial imparative. Atheneum Press: New York. 390 pp. Berlesse, A. 1910. Gli insetti. V01. 1. Milan. 1004 pp. Borror, D. J. & D. M. DeLong. 1964. An introduction to the study of insects. Holt, Reinhart, and Winston: New York. 819 pp. Burgess, E. 1880a. The structure and action of a butterflie's trunk. Amer. Nat. 14: 313—319. 1880b. Contribution to the anatomy of the milkweed butter- fly. Anniv. Mem. Boston Soc. Nat. Hist. 16 pp. Comstock, J. H. 1924. An introduction to entomology. The Comstock Publishing Co.: Ithaca, N.Y. 1044 pp. Cook, E. F. 1944. The morphology and musculature of the labrum and clypeus of insects. Microent. 9: 1-35. DuPorte, E. M. 1946. Observations on the morphology of the face in insects. J. Murph. 79: 371-417. 1956. The Hedian and facial sclerite in larval and adult lepidoptera. Proc. R. Ent. Soc. Lond. (Series A) 31: 109-116. 1965. The role of morphology in taxonomy. Ann. Ent. Soc. Quebec. 1965: 91—97. Eastham, L. E. S. and Y. E. E. Eassa. 1955. The feeding mechanism of the butterfly Pieris brassicae L. Philos. Trans. R. Soc. Lond. B 239: 1-43. Ehrlich, P. R. 1958. The integumental anatomy of the monarch butterfly, Danaus plexippus L. (Lepidoptera: Danaiidae). Univ. of Kansas Sci. Bul. 38: 1315—1349. . 1960. The integumental anatomy of the silver-spotted skipper, Epargyreus clarus Cramer (Lepidoptera: Hesperiidae). Nficroent. 24: 1—23. , & S. E. Davidson. 1961. The internal anatomy of the monarch butterfly, Danaus plexippus L. (Lepidoptera: Nymphalidae). Microent. 24: 85-133. 55 56 , & A. H. Ehrlich. 1962. The head musculature of the butterflies (Lepidoptera: Papilionoidea). Microent. 25: 1-89. , & . 1963. The thoracic and basal abdominal muscula— ture of the butterflies. (Lepidoptera: Papilionoidea) Microent. 25: 91-126. Ferris, G. F. 1943a. Some fundamental concepts in insect morphology. Microent. 8: 2—7. . 1943b. The basic materials of the insect cranium. Microent. 8: 8-24. Forbes, W. T. M. 1948. Lepidoptera of New York and neighboring states. Part II. Cornell Univ. Agric. Exp. Sta. Mem. 274. 263 pp. Holland, W. J. 1941. The moth book. Doubleday, Doran & Co.: New York. 479 pp. Kellogg, V. L. 1893. The sclerites of the head of Danaus archippus. Kansas Univ. Quart. 2: 51—59. Kirbach, P. 1883. Uber die Mundwerkzeuge der Schmetterlinge. Zool. Anz. 6: 553—558. Madden, A. H. 1944. The external morphology of the adult tobacco hornworm (Lepidoptera: Sphingidae). Ann. Ent. Soc. Amer. 37: 145-160. Matheson, R. 1951. Entomology for introductory courses. Comstock Publishing Co.: Ithaca, N.Y. 629 pp. Matsuda, R. 1965. Morphology and evolution of the insect head. Mem. Amer. Ent. Inst. No. 4. 334 pp. Michener, C. D. 1952. The Saturniidae (Lepidoptera) of the Western Hemisphere. Murphology, phylogeny, and classification. Bul. Amer. Mus. Nat. Hist. 98: 341—501. Ross, H. H. 1965. A textbook of entomology. John Wiley & Sons: _New York. 539 pp. Rothschild, W. & K. Jordan. 1907. Sphingidae—Gen. insectorum. Fasc. 57. Wytsman Co.: Brussels. 157 pp. Short, J. R. T. 1951. Some aspects of the morphology of the insect head as seen in the Lepidoptera. Proc. R. Ent. Soc. Lond. (Series A) 26: 77—88. Schmitt, J. B. 1938. The feeding mechanism of adult Lepidoptera. Smithsonian Misc. Coll. 97(4): 1—28. 57 Snodgrass, R. E. 1935. Principles of insect morphology. MtGraw- Hill Book Co.: New York. 667 pp. . 1947. The insect cranium and the "epicranial suture." Smithsonian Misc. Coll. 107(7): 1—52. . 1960. Facts and theories concerning the insect head. Smithsonian Misc. Coll. 142(1): 1-61. Tillyard, R. J. 1923. On the mouth-parts of the Micropterygoidea. Trans. Ent. Soc. Lond. 71: 181-206. 58 Generalized Cranium, Frontal aspect with portion of frontal Paonias myops, sclerite removed. , Sagittal aspect. , Lateral aspect with left eye removed. H , Ventral aspect. Ventral aspect. ANTENNAL SOCKET TRANSFRWTAL SUTURE , , , * r-ANTENNAL RIDGE FRONTAL SANGLION , 7 LATEROFACIAL SUTURE FRONTAL AREA ' ~LATEROFACIAL IMUECTION FRONT OCLYPEAL AREA _ ANTERIOR //TENTORIAL PIT TRANSCLYPEAL BAND ~ Var/PILIFER PROBOSCIS ANTERIOR ARM 0F TENTORIUM I SALIVARIUM HYPOSTOMAL BRIDGE POSTERIOR ARM LABIAL SCLERITE OF TENTORIUM TENTORIAL BRIDGE LABIAL PALP MAXILLARY PALP PILIFER 60 6. 7. Manduca sexta, Frontal and lateral aspects. 8. 9. Ceratomia 10. Ceratomia 11.12. Ceratomia 13. Ceratomia undulosa, Frontal and lateral aspects. catalpae, Frontal aspect. catalpae, Frontal aspect showing variations in muscle S, catalpae, Lateral aspect. 62 14. 15. Sphinx eremitus, Frontal and lateral aspects. 16. 17. Smerinthus geminatus, Frontal and lateral aspects. 18. 19. Paonias excaecata, Frontal and lateral aspects. 64 20. 21. Paonias myops, Frontal and lateral aspects. 22. 23. Cressonia juglandis, Frontal and lateral aspects. 24. 25. Pachsphinx modesta, Frontal and lateral aspects. 66 26. 27. Haemorrhagia thysbe, Frontal and lateral aspects. 28. 29. Haemorrhagia diffinis, Frontal and lateral aspects. 30. 31. Pholus satellitia pandorus, Frontal and lateral aspects. 68 32. 33. Ampeloeca myron, Frontal and lateral aspects. 34. 35. Amphion nessus, Frontal and lateral aspects. 36. 37. Celerio lineata, Frontal and lateral aspects. “'IIIIIIIIIIIIIIIinIIIIIIII“