R. T. Holway.
Preliminary Note on the Structure of the Pretarsus and Its Possible Phylogenetic Significance.
Psyche 42(1):1-24, 1935.

This article at Hindawi Publishing: https://doi.org/10.1155/1935/91275
CEC's scan of this article: http://psyche.entclub.org/pdf/42/42-001.pdf, 2044K
This landing page: http://psyche.entclub.org/42/42-001.html

The following unprocessed text is extracted automatically from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

PSYCHE
--
VOL. XLII
MARCH 1935 No. 1
PRELIMINARY NOTE ON THE STRUCTURE OF THE PRETARSUS AND ITS POSSIBLE PHYLOGENETIC
SIGNIFICANCE
Biological Laboratories, Harvard University The uncertain reliability of morphological structures for determining ancestral relations is well known and the ex- treme care and attention to the vagaries of parallelism, spe- cialization, etc. so necessary for the development of valid theories have been rightly emphasized of recent years. How- ever, in spite of its late disrepute, there is yet much to be learned from comparative morphology if sufficiently long series are studied and checked with the trends of parallel structures and if a most judicious interpretation of the evi- denoe is demanded.
The primary purpose of this paper is to indicate the type of claw-segment found in the important insectan orders together with the common basic plan occurring throughout and also to outline possible homologies which further study and examination of selected and thoroughly representative series from each order may support. Although it is impos- sible at the present stage to offer any evidence substantial enough to warrant forming exact conclusions as to homolo- gies from order to order, to. say .nothing of attempting to solve phylogenetic problems, there are without doubt most significant agreements in many cases which provide definite possibilities to be tested by further study. Especially im- portant are those correlations which fall in with phylogen- etic concepts already accepted (or debated) from the evi- dence of the comparative morphology of other structures. Pacht 42:)-24 11935). http//psyche eniclub.mg/42/42-001 him)



================================================================================

2 Psyche [March
For example, the concurrence of Trichoptera and Lepidop- tera is not questioned and some investigators have suggested that these two groups might'even be conceived of as one valid order. The nature of the claw-segment of Astenophy- lax (Fig. 21) and Automeris (Fig. 22) shows at a glance the proximity of these structures even in such advanced re- presentatives of the respective orders.
The relative value of a structure like the pretarsus as a phylogenetic guide may be uncertain, but the above example indicates that such consistency of the claw-segment from order to order should make it of considerable utility in checking theoretical relationships. Although the basic plan of the pretarsus presents a certain conformity throughout the orders, there are sclerites which are variable enough to warrant their investigation from the standpoint of taxon- omy. For example, within the Hymenoptera the orbicula may be important in this respect. Hayes and Kearns (1934) have studied the empodium of Coleoptera and find that the Adephaga is the only group showing any consis- tancy in structure; they conclude, in general, that there is in some a tendency towards generic or family uniformity but that the super-families are decidedly heterogeneous. Arolia and pseudarolia have long been used as the basis of classification for the Miridae and for other Hemiptera the empodium may prove to be of taxonomic value. Acknowledgements
The writer is indebted to Professor C. T. Brues and Pro- fessor G. C. Crampton for their advice during the prepara- tion of this paper, and to them and to the Museum of Com- parative Zoology for much of the material herein figured. Terminology.
The general structure of the pretarsus has been figured many times for different orders: Snodgrass gives detailed descriptions of the condition in the honey-bee (1925) and figures several other types (1927) ; Crampton (1923) com- pares the claw-segment of Periplaneta with those of Leptid and Asilid Diptera; Hayes and Kearns (1934) have studied the pretarsus of Coleoptera and summarize past work on



================================================================================

193.51 Structure of the Pretarsus 3
this structure; de Meijere's investigations are more extlen- sive and cover all the important orders but his German terminology is not used at all in this country. A complete history of the terminology of the numerous parts of the claw-segment must bfe left for a later and more comprehen- sive study of this structure, therefore that system accepted by the more recent investigators is followed throughout. In some cases the same terms are used from order to order, even though true homology is not yet proved, instead of of- fering new names which in all. probability would have to be discarded aftler further study supplied evidence to substan- tiate suspected homologies.
However, all doubtful cases
are indicated as such.
The following is a list of the terms herein used with their definitions and equivalents.
Apodeme (Unguitractor tendon, Sehne) .-This is the tendon which runs from the unguitractor plate to the tibia1 musculature, the flexor of the claws? Supposedly the claws spring back into their normal position by natural resiliency when tension through the apodeme is released as there is no l'evator of the pretarsus in insects. (Crampton 1923) (Snodgrass 1929).
Arolium. In general the term arolium is restricted to a median paired or unpaired, pad-like structure which is ar- ticulated to the unguifer or arises from a position midway between the claws. Whether the complicated median struc- ture of Hymenoptera, Lepidoptera and Trichoptera, the paired ribbon-like parts known as arolia in certain Hemip- tera (Miridae), and the single median pad of Diptera and orthopteroid insects are all strictly homologous is prob- lematical. However, the term arolium is here used to des- ignatle these structures in every order in which they occur, according to the definition above.
Basipulvilli. (auxillise). This term was first applied by Crampton (1923) to the small lateral scleritlea at the bases of the pulvilli of Diptera and also to similar sclerites lat- erad of the unguitractor in Periplaneta although true pul- villi are not found in Orthopteroids. The auxillise (Hymen- optera), of MacGillivray, are evidently homologous with



================================================================================

4 Psyche [ March
these structures but wherever such parts are found they are called basipulvilli in this paper.
Camera. "(Bugel, or bow of Arnhart). Camera was originally applied to the curved narrow sclerite supporting the paired lobes of the arolium in Hymenoptera (MacGil- livray). It is herein also used to designate the evident ho- mologue in Trichoptera, Lepidoptera and Mecoptera. Distitarszis. (onychium, digitus, ungula, etc.) . The dis- tal tarsal segment.
Empodium. (onychium). This term has been used for a variety of structures. Crarnpton (1923) finds that it should be restrictled to a process of the unguitractor as found in Asilid Diptera and other orders. Hayes and Kearns (1934) accept this view and apply the term in their work on Coleoptera.
Empodium is often used erroneously
to refer to a pulvillus.
Flexor membranes. These are the membranous areas be- neath the claws which apparently serve to transfer the ten- sion of the apodeme on the unguitractor to the claws thus forcing them downwards. They are the areas from which, it has been suggested, the puvilli may have developed. Gleitrinne or Gleitflache (de Meijer'e)'. A prolongation of the ventral wall of the distitarsus. It is especially devel- oped in Odonata and in some Coleoptera where it extlends forward between the claws.
Onychium.
This term has been used in such a variety of ways that it appears impossible to define it satisfactorily and therefore is not used in this paper. Orbicula. In Hymenoptera, a small dorsal sclerite at the base of the arolium and distad of the unguifer. The
orbicula is quite variable as to size and shape in this order and may be of taxonomic value.
Parempodia. (paronychia) . Bristle-like appendages of the empodium.
(Coleoptera and Hemiptera) .
Planta. As used by Snodgrass, Crampton and MacGil- livray planta applies only to the sole of the foot (typically in Hymenoptera and Orthoptera) . However, the examples



================================================================================

19351 Structure of the Pretarsus 5
of most of the various orders observed show a plate which is probably homologous with the hymenopterous planta and therefore this term is used wherever practicable. Coleop- tera, Diptera, Hemiptera, Dermaptera, Odonata and Eph- emeroptera do not exhibit the planta in its typical condi- tion although an empodium is present in these orders (ex- cept Ephemeroptera) which suggests the possibility that the empodium is merely a modified planta, closely associ- ated with the unguitractor. Further study is needed to de- termine the correct relationships of these structures. Snodgrass (1929) suggests that the planta may possibly be a subdivision of the unguitractor. However, the nature of the planta under high magnification is unlike that of the unguitractor which always gives a granulose appearance, is heavily sclerotized, and is sharply demarked in every case observed 'even to the primitive Odonata and Ephemer- optera. On the other hand the planta is variable in extent and degree of sclerotization and never shows the heavy, rugose aspect of the unguitractor. It seems more likely that the planta represents merely an area of the mem- branes distad of the unguitractor which has become sclero- tized in greater or less degree according to the stress to which it is subjected by the tension of the apodeme and un- guitractor. Of course the unguitractor itself is a sclerotized area of the membrane but it is always distinct in outline and not an indefinite structure as the planta often appears to be. (Note.-Planta has also been used to refer to the basal joint of the post tarsus in pollen-gathering Hymen- optera, to the soles of the post tarsal joints and to the anal clasping legs of catferpillars) .
PseudaroZia.-Paired structure which occurs beneath the claws of some Herniptera (Miridae) and which possibly will prove to bear some relation to pulvilli. Pulvillus.-Originally referred to the pad-like structure beneath each claw in many Diptera.
Other orders in which
undoubted homologous parts occur are Lepidoptera and Trichoptera. Membranous areas are found beneath the claws in most orders but it is impossible to consider them as homologous with the pulvilli. In this paper they are



================================================================================

6 Psyche [ March
termed flexor membranes. Crampton (1923) suggests two possible derivations for the pulvilli: either they arose as portions of a divided arolium (see condition in Hymenop- tera, Fig. 26, Plecoptera, Fig. 7, and Mantispa, Fig. 19.) or they are detachled membranous areas from the under side of the claws (see condition in some Ephemeroptera, Fig. 2, Homoptera and Hemiptera, Figs. 14 and 15, and the relation of basipulvilli to such membranous areas in Blat- tids.)
Ungues.-The claws.
Unguif er.- (tubercula, and Gelenkhocker of de Mei jere) refers to a small dorsal sclerite by which the claws are ar- ticulated with the distitarsus.
Unguitractor- (tarsulus, calcanea) . This sclerite is the most constant structure of the pretarsus and is easily identified in all insects thus far observed because its gen- eral appearance is always the same though varying in size and shape. The surface of the unguitractor is sculptured to a greater or less degree with nodules which are some- times almost spinelike. These nodules appear to corres- pond with the polygonal cells of the hypodermis and these cells may be seen through the chitin with high magnifica- tion. Snodgrass (1927) shows that in Tibicen this plate may be divided into two sclerites and also he figures in the membrane distal to the unguitractor, two small plates which he says may represent the arolium. It seems more likely that thley are a divided planta but there is little evidence as yet for 'either supposition.
Discussion.
ORDER ODONATA
Anax junius Drury. (Fig. 1)
The claw-segment of Anax is simple and consists of the unguitractor plate (ut) with the expanded distal portion which is probably an empodium (em?) and of the lateral fleshy pads, flexor membranes (fm.) ,-which transmit tension to the ventral surfaces of the claws. The Gleit-
flache of de Meijere, or thickened ventral portion of the



================================================================================

19351 Structure of the Pretarsus 7
distitarsus over which the unguitractor slides, is well de- veloped in Odonata and a process of this structure extends beyond the bases of the ungues (un) . The only other orders in which this condition has been observed are Coleoptera and Hemiptera.
ORDER EPHEMEROPTERA
Siphlonurus sp. (Fig. 2)
Thle unguitractor is a sma1,l pear-shaped plate and the claws (un) are simple. The important feature in this case is the membranous pad-like structure (pv?) closely applied to the ventral surface of each of the ungues. It is this con- dition which has led Crampton to suggest that possibly the pulvilli (Diptera) were derived from some such situation, that is, a splitting off of a ventral membrane from the claw itself. Proximally these pad-like structures are closely as- sociated with the membranous areas transmitting tension from the unguitractor to the claws and in fact give evidence of being continuous with them.
Although these examples of Odonata and Ephemeroptera both show their ellementary position, (absence of arolium, planta, basipulvilli or other significant structures), they give no evidence of any close relationship for the presence of an empodium in Anax indicates a certain amount of spe- cialization.
Thus, if these cases prove to be typical, the con- clusions to be derived from the comparative morphology of the pretarsus will evidently be in agreement with the general conception of the relative positions of these orders. CURSORIAL ORTHOPTERA
Blattidae-Pmmiplanta mericana L. ( Fig. 3) The arolium (ar) is large in this family, although its structure is simple compared with the complicated types found in the higher orders. Vientrally it is membraneous and concave, so that it forms a most efficient adhering organ. The unguitractor is typically large and heavily sclerotized. Two other structures of importance appear in this group. The planta
(pi) is a median quadrate plate just distal to the unguitractor and seems to be lightly sclerotized; it bears a single large seta. On either side of the planta is a



================================================================================

8 Psyche [March
small sclderite, the basipulvillus (bp) of Crampton, from which the retractile membrane leads to the under side of the claw. The relation of these membranes to the basipulvilli (where they are present) is the same throughout the or- ders and with further study should present interesting evi- dence with respect to the hypothesis that the pulvilli are evolved from such membranes.
Mantidae-Tenodera sinensis Saussure. (Fig. 4) An a'rolium has not been found in Mantids thus far ob- served and such a structure is probably absent in this fam- ily. The general plan follows closely that of the roach al- though a seta is not present on the planta. The flexor mem- . branes are large and especially noteworthy are their lat- eral expansions which if further produced might 'easily take on the appearance of pulvilli !
Termopsis angusticollis Walker (Fig. 5)
Again, in this case the basic plan follows closely that of the Blattid; however, the planta is much smaller than in either Mantid or Blattid and the seta does not arise from the planta itself but from thfe membrane. This condition might be taken as evidence for the view that the planta is a variable sclerotized area of the membrane between the claws. Although there is no arolium we find between the bases of the claws in winged termites a small plate (ar?) which undoubtedly represents an abortive arolium for in winged specimens of Mastotermes, a definite arolium, even though small and rudimentary, is present. Unguitractor and basipulvilli, as in the Mantid, show no significant var- iations from thae Blattid type.
The proximity of these three groups has been noted many times by students of insect morphology so that such a cor- relation in type of pretarsus is not surprising. This com- bination appears even more natural when contrasted with the equally coherent union (from the standpoint of Pre- tarsus) of the Saltatorial Orthoptera.




================================================================================

19351 Structure of the Pretarsus 9
ORDER EMBIOPTERA
Oligotoma sp. (Fig 6)
In this order we again find the Blattid type; the planta is more extended laterally and is in contact with the basi- pulvilli on either side. There is no arolium and the flexor membranes are large and slightly expanded. There seems to be nothing particularly outstanding about the claw-seg- ment of this insect but in general it approaches that of the Plecopteran to such a degree as would not prohibit the con- ception that these two orders are related. ORDER PLECOPTERA
Pteronarcys dorsata Say (Fig. 7)
Pteronarcys shows the lateral basipulvilli and median planta as found in cursorial Orthoptera and termites. The arolium is a large and complicated structure which is made up of three lobes : a median portion or true arolium, and two lateral smaller pads beneath the claws which may be sec- ondary divisions of the arolium or pulvilli which have be- come fused with the arolium. A more complete study of the Plecoptera may settle this question if intermediabe forms can be found. On the other hand, it is possible that this structure represents a new development with no traceable homologies to the parts referred to above. (This also ap- plies to similar lobies beneath the claws of many Homoptera and Hemiptera, to be discussed later). Dorsally, the mem- branous pads are supported by chitinous plates (indicated by broken lines) and the median lobe blears two clusters of small set= on its ventral surface. The presence of planta and basipulvilli make it unlikely that the Plecoptei-a should be considered with the Archipterygota,
SALTATORIAL ORTHOPTERA
Locustidae-Melanoplus bivittatus Say (Fig. 8) A large arolium has been found in all Locustids observed as contrasted to Tlettigoniids and Gryllids in which this structure probably does not occur.
It should be noted that
the presence or absence of an arolium is not a condition which is constant for an order, but further study should



================================================================================

10 Psyche [March
produce information concerning its consistency in groups below ordinal rank. The absence of basipulvilli is charac- teristic of all the leaping Orthoptera and it is this factor which gives such a clear-cut division between the two or- thopteroid groups. The planta of Melanoplus is more heav- ily sclerotized than the representatives of Tettigoniidae and Gryllidae which were examined.
Tettigonfidae-Ceuthophilus maculatm Say (Fig. 9) Except for the absence of arolium and basipulvilli, the condition of the planta seems to be the only important fea- ture of Ceuthophilus. This structure is not as heavily scler- otized as it is in Melanoplus and is quite different in shape. Often the planta is so lightly demarked, especially in small insects, that its boundaries are almost indistinguishable from the surrounding membrane.
Gryllidae-Gryllus pennsylvanicus Burm. (Fig. 10) Gryllus gives us the first radical variation in the ungui- tractor for here the form is quite different from anything else found among the Orthoptera. This sclerite is charac- terized by a longitudinal V-shaped ridge or protuberance, the significance of which must remain in doubt until serifes are studied in groups below the order. It is possible that the less notable variations (chiefly in outline) of thle ungui- tractor observed in these orthopteroid families will prove to be consistent to some daegree.
The planta may be easily
compared with that of Ceuthophilus and differs mainly in the length of the lateral arms and the nature of the pos- terior margin.
Gryllotalpidae-Gryllotalpa borealis Burm. (Fig. 11) Considerable modification of the planta is exhibited by Gryllotalpa and its condition in this insect is one of the factors suggesting a close relationship between planta and empodium. Note the numerous large setae and the manner in which the anterior portion of the unguitractor merges with the flexor membranes and planta. Comparison of the latter with the corresponding structure in Ceuthophilus and Gryllus indicates that it is a planta.
Empodium by definition refers to a process of the ungui- tractor, but this term must also apply to certain structures



================================================================================

19351 Structure of the Pretarszis 11
closely associated with the unguitractor which are not act- ual processes of it. Eg,, the setiform empodium of Asilid Diptera is articulated with the unguitractor rather than continuous with it; also the coleopterous empodium in many cases gives some evidence that it may be retracted within (or beneath?) the unguitractor as pointed out by Hayes and Kearns. At any rate the empodium is not necessarily a distinct part of the unguitractor, so that the condition in Gryllotalpa may be intermediate between the typical iso- lated planta (as of Melanoplus) and the empodium of many Coleoptera where there is no discernable break between this structure and the unguitractor. Ceuthophilus and Gryllus may represent a step beyond the Melanoplus type towards closer association of planta with unguitractor. About all that can be definitely said for the Saltatorial Orthoptera (sensu strictu) at the present time is that they are linked by a universal absence of basipulvilli and by the nature of other more variable features such as planta and flexor membranes.
In any case there appears to be a wealth of material to be investigated within this group. Phasmida+AnISomorpha buprestoides (Stoll) (Fig. 12) The absence of basipulvilli would place the Phasmids with Saltatorid rather than Cursorial Orthoptera and this is in agreement with conclusions presented by students of other structures. The unguitractor is narrower than in other Orthopteroids, while the planta is more heavily sclerotized and is triangular in outline, but is distinct from the ungui- tractor, a condition nearer to that of Locustids than to the other Salfcatorial Orthoptera. The arolium is large and is reinforced dorsally by thickened ridges. On the whole the condition of the claw segment of this insect may be con- sidered as supporting the general conception of the position of this family, i. e., related to the Saltatorial Orthoptera but developed along its own line of specialization. ORDER DERMAPTERA
Pscdis sp. (Fig. 13)
Apparently basipulvilli and planta are missing in Psalis. If this proves to be true for all Dermaptera, the order will



================================================================================

12 Psyche [March
be set off from the B'lattid group on the one hand, and the remainder of the orthopteroid insects on the other. (Speak- ing from the standpoint of pretarsus only, of course). The unguitractor of Psalis is interesting because of its anterior projection (em?) which could be considered either empo- dium or planta ; superficially, this structure appears to be continuous with the unguitractor and therefore more like a coleopterous empodium except that its margins are vague and it is not heavily sclerotized, i. e., more like a planta. It should be noted and may prove to be significant that all examples of Coleoptera thus far observed show neither basipulvilli nor planta but do have an empodium, whereas all specimens of most higher orders studied are fundamen- tally of the Blattid type with numerous secondary special- izations.
ORDER HOMOPTERA
Agallia, constricts Van Duzee
(Fig. 14)
The small unguitractor bears a triangular empodium (em) which in turn is provided with a single heavy seta or parempodium (pem).
As there seems to be considerable
variation in the number of setae and in the shape of the empodium among the Hornoptera, this group must be more fully studied to determine the extent and constancy of such variations. Attached to the under surface of the claws and extending beyond their tips are large membranous pads (pv?) which are reinforced meso-dorsally by small sclero- tized areas.
There are no basipulvilli.
The pretarsi of Cicadlidae are not of this type nor do they show any features characteristic of the more advanced Hemiptera. As figured by Snodgrass (1927) the ungui- tractor of T. septendecim is divided into two platces; there is no empodium, but two small setae-bearing sclerites occur in the membrane between the claws, and ~nod~rass sug- gests that these may represent the arolium but there seems to be morle reason to consider them as a divided planta. In
T. cinera these plates are very faint and difficult to distin- guish from the surrounding membrane. There are no pads beneath the claws.




================================================================================

193.51 Structure of the Pretarsus 13
ORDER HEMIPTERA
Brochymem arborea (Say) (Fig. 15)
The large triangular unguitractor is extended anteriorly to form a quadrate empodium which bears a parempodium on each of two lateral protuberances. Beneath the claws are pad-like structures (pv?) reinforced by chitinous plates but not attached to "the claws beyond their bases in con- trast to the condition in Membracidae, Cicadellidae, etc. Although it seems very probable that.these pads will prove to be identical with those of the Hornoptera, but separated from the claws, it is more questionable whether they can be considered as homologous to the pulvilli of higher orders especially because of the fact that typical basipulvilli do not appear in the Hemiptera-Hornoptera. The condition in these groups, however, possibly indicates a way in which sub-ungual pads may be developed and split off from the claws. Further reduction of the membranes might result in the typical pseudarolia of the Minds! It is hoped that
an extensive study of the Hemiptera-Hornoptera will illum- inate the relations of these various structures. ORDER COLEOPTERA
Phy11ophaga fusca, Froe. (Fig. 16)
As mentioned above, the empodium, rather than basipul- villi and planta, is the characteristic feature of Coleoptera. In this specimen the unguitractor is narrow and the em- podium is expanded distally into a broad plate which bears two parempodia or large setae. Each claw is provided with a large median tooth.
Hayes and Kearns have figured many specimens from this order and although they find the empodium absent in many cases, there seems to be a typical fundamental plan for the coleopterous pretarsus.
That is: First, a well de-
veloped unguitractor with or without some form of em- podium which in turn may or may not bear parempodia; Second, total absence of any membranous structures which could represent arolia, pulvilli, etc.
Hayes and Kearns find
that "some cleared specimens show what appears to be a basal part of the empodium withdrawn into the body of the



================================================================================

14 Psyche [March
unguitractor." This condition is evidently the same as that here shown in Phyllophaga fusca, (Fig. 16) in which the broken lines indicate the basal part referred to. It is questionable whether this is really within the unguitractor ; this portion of the empodium might be beneath the ungui- tractor, or, merely a degree of sclerotization. There is considerable similarity between the claw-seg- ments of Dermaptera and Coleoptera. Although the em- podium of Coleoptera is of course much more varied than that of Dermaptera, these orders agree in the absence of other significant structures. Some of the figures from Hayes and Kearns in which the empodium is absent are strongly reminiscent of thte condition in Psalis, as in the latter, the area distad of the unguitractor is rather indefin- ite so that it is not difficult to imagine the sclerotization of this area to form the kind of empodium characteristic of Coleoptera-or its total lack of sclerotization with a result- ing absence of empodium which is also a condition commonly found in Colheoptera. This must remain pure speculation, however, until checked by further study. ORDER NEUROPTERA
Sialidae-Corydalis cornuta L. (Fig. 17)
There is no arolium but the planta is large and the basi- pulvilli are contiguous with its lateral margins. The claws are simple and the unguitractor is triangular like that of Ithone. In general the pretarsus of Corydalis is surpris- ingly different from that of Mantispa, but except for the absence of arolium and basipulvilli setae, it is easily com- parable to the claw-segment of Ithone.
Ithonidae-Ithone sp. (Fig. 18)
The arolium is a median pad somewhat like that of P. americana, but much smaller and without the ventral con- cavity of the latter.
The relations of planta and basipul-
villi are of the Blattid type, but the planta is very lightly sclerotized and is continuous with the arolium, so that it might almost be considered as a part of this structure. The basipulvilli occupy the normal position and are provided with three setae, incidentally, the only case observed where



================================================================================

19 3 5 1 Structure of the Pretarsus 15
the basipulvilli bear setae. The unguitractor and claws are not distinctive.
The similarity of the claw-segment of Ithone with that of the Blattids is evident and perhaps significant, for this in- sect is considered to be the most primitive of the Plannipen- nia. Comparison of Figs. 17, 18 and 19 illustrates why Ithone may be considered as annectant between Corydalis and Mantispa.
Mantispidae-Mantispa brunnea Say (Fig. 19) The pretarsus of Mantispa is provided with a complicated type of arolium which has probably developed from a simple pad such as found in Ithone. Two fleshy lobes are sup- ported by lightly sclerotized areas, indicated by broken lines, and at the base, posteriorly, are two ovoid sclerites also serving as supports. There is evidently no planta, a condition consistent with the light sclerotization of the planta in Ithone. At present it is impossible to intaerpret the significance of these various parts in Mantispa but it is hoped that other specimens of Neuroptera will make it possible to find homologies with the structures of higher insects. The basipulvilli are large and extend laterally a short distance along the membranes. The claws are broad and pectinate; the unguitractor differs considaerably in out- line from that of Corydalis or Ithone and bears a median ridge. It will be noted that the claw-segment of Mantispa is not comparable to that of Corydalis in any of its features and, in fact, the divergency of these two types is greater than any variations found in other orders, a condition which emphasizes the fact that the Neuroptbera are known to be a highly diversified group.
ORDER MECOPTERA
Panorpa ruf escens Rambur (Fig. 20)
The unguitractor is oval and bears a conspicuous median ridge such as is found in Mantispa; the claws are pectinate. A rod-like planta extends between the bases of the claws and the basipulvilli. The arolium is simple and is supported by a curved structure (em) which is probably the same as the camera of Hymenoptera.




================================================================================

16 Psyche [March
ORDER TRICHOPTERA
Astenophylax argus Harris (Fig. 21)
The arolium is much like that of Panorpa and differs only in the curvature of the camera which is U-shaped in Asten- ophylax. The planta is also similar, that is, a narrow rod- like structure distad of the unguitractor. The basipulvilli are large and are located at the bases of the lleathery pul- villi (pv) which are heavily fringed with long narrow scales. The claws are simple instlead of pectinate and the median ridge is absent from the unguitractor,-two fea- tures in contrast to the condition in Panorpa. ORDER LEPIDOPTERA
Automeris io Fab. (Fig. 22)
The claw-segment of Automeris is very similar to that of Astenophylax and varies only in the narrower, curved pul- villi and the nature of the arolium which gives evidence of developing a secondary lobe. The condition in many Lepid- optera where the arolium is lacking has not been investi- gated.
This agreement between Lepidoptera and Trichoptera was to be expected, although such a complete resemblance is perhaps surprising. There are also certain features trace- able to Panorpa.
(see Figs. 20 and 22). It is conceivable that some Mecopteran will be found with traces of pul- villi which appear to be common structures in the orders Lepidoptera, Trichoptera and Diptera. There can be no question but what the condition in Panorpa approaches that of the Trichoptera and Lepidoptera more closely than it does the Neuropteran types, except for the absence of villi.
ORDER DIPTERA
Tabanidae-Tabanus atratm Fab. (Fig. 23)
The unguitractor is broadly triangular and the pul and basipulvilli are typical. The arolium is articulated with the unguifer and on the ventral surface its membrane is continuous with that surrounding the basipulvilli and ungui- tractor. This is the structure erroneously referred to as pulvilliform empodium; it is really an arolium. (Cramp- pul-
villi




================================================================================

19 3 5 1 Structure of the Pretarsus 17
ton 1923). The arolium of Tabanus is simple and lacks the supporting sclerite, or camera.
For this reason there may
be some question as to its strict homology with the arolium of Mecoptera, etc. described above. The absence of a planta may be significant as a similar absence was noted in Mantispid Neuroptera.
On the other hand, the arolium
is like that of Ithone in nature i. e., its ventral surface is continuous with the membrane distad of the unguitractor) and the planta of Ithone is so lightly sclerotized that it is conceivable that in many related forms the planta would be entirely absent.
Asilidae-Diogmites umbrinus Loew. (Fig. 24) This insect has no arolium but is characterized by the presence of a setiform empodium, a long chitinous setiform structure arising from the distal margin of the unguitrac- tor. The pulvilli are also long and narrow and similar to those of Trichoptera and Lepidoptera, but the basipulvilli are small.
A study of more primitive Diptera may bring out some annectant features which will be significant in regard to the origins of this order. Thus far there seems to be more common features in the Neuroptera than in the Mecoptera. ORDER HYMENOPTERA
Siricidae-Tremex colurnba L. (Fig. 25)
The unguitractor is long and grooved and the basipul- villi are very small; pulvilli probably do not occur in the Hymenoptera.
Tremex lacks an arolium which is a com-
plicated structure when present in this order. The planta bears four setae and the claws are provided with a large median tooth or spur. The orbicula (rb) is a dorsal scler- ite which is here shown only in part.
Sphecidae-Chalybion csereuleum L. (Fig 26) The arolium, here shown extended, is normally folded back against the planta when not in use. Except for the
paired lobes this structure is much like that of Panorpa especially in reference to the camera. The planta is pro- vided with many small setae and the unguitractor differs



================================================================================

18 Psyche [March
from that of Tremex in outline and absence of dorsal grooves. The claws are simple and the basipulvilli small. The orbicula, a small dorsal scllerite, is not shown in this view but this plate is variable throughout the Hymenop- tera and together with the planta and unguitractor should prove valuable in studying the inter-relations within the order. As with the Diptera, it seems probable that further study of primitive types will be of phylogenetic value for the variety of structures does provide a wealth of material to work on.
Summary.
Without losing sight of the fact that only a few speci- mens selected at random from each order are studied, the evidence thus secured does seem to point out certain ten- tative hypotheses to be further investigated. 1. That the planta occurs in most of the insectan orders as a sclerotized area of the membrane just distal to the un- guitractor and that this structure is homologous throughout by position, ie., its relation to the unguitractor, basipulvilli and claws. The presence of one or more setae may be a distinguishing feature ; note the condition in cursorial Or- thoptera and Hymenoptera. The degree of sclerotization is quite variable for in some cases the planta is almost en- tirely membranous while in others it forms a heavy plate. 2. That the 'empodium is a highly modified planta. No
planta was found (in specimens so far observed) in the orders Diptera, Hemiptera, Colleoptera and Dermaptera but an empodium is present in at least some representatives of all these orders. On the other hand, wherever a planta oc- curs an empodium has not yet been observed. The condi-
tion of the planta in Gryllotalpa and Psalis, where it is closely associated with the unguitractor, illustrates how difficult an attempt to distinguish between these two struc- tures may be. Comparison of Gryllotalpa with other Or- thopteroids leaves but little doubt but what the part re- ferred to is a planta even though numerous set~ are pres- ent, for such appendages also occur on the plank of cur- sorial Orthoptera and Hymenoptera. The occurrence of pa- rempodia, which are really setae, on the empodia of Coleop-



================================================================================

19351 Structure of the Pretarsus 19
tera and Hlemiptera may also point to a close relation be- tween these structures. These considerations, if supported by further investigation, indicate that the empodium is really a specialized planta, i. e., a planta which has become closely associated with the unguitractor and in 'some cases lengthened into a setiform process.
3. That the arolium or median pad-like structure occur- ring in many insects, although varying in cornpllexity, is homologous throughout the orders. This structure is not of ordinal significance for it is present in some groups and absent from others within most orders.
4.
That the small lateral sclerites (basipulvilli) are ho- mologous in all orders; they form a support at the bases of the pulvilli when the latter are present and in such cases there is no question as to their homology. Many insects show these scleritles without the pulvilli but all other rela- tionships are the same.
5. The problem of the origin of the pulvilli remains in doubt. Very possibly further study will support the sug- gestion of Crampton (1923) that they are membranous areas detached from beneath the claws. Thus far what evidence there is remains inconclusive, that is, briefly: the membranous claws of Ephemeroptera and the similar con- dition in certain Homoptera (Fig. 15) ; the pseudarolia of Miridse which may be true pulvilli; the association of pul- villi and basipulvilli in Diptera, Trichoptera and Lepidop- tera and the fact that the basipulvilli (if present) are al- ways located at the bases of the flexor membranes in orders where typical pulvilli do not occur.
6. An interesting condition is that found in the orthop- teroid insects. The similarity of certain anatomical fea- tures of the Blattidse, Mantidse and Isoptera is well known so that the agreement found in the claw-segments of these three groups is not surprising. The saltatorial Orthoptera, including the Phasmidse, are also set apart as a coherent group by the nature of the pretarsus and the Dermaptlera would seem to present a third Orthopteran type although less distinct and nearer to the Tlettigoniids and Gryllids- plus coleopterous affinities.




================================================================================

20 Psyche [March
7. That the planta and basipulvilli found in the Plecop- tera would place this order with the Neopterygota rather than with such types as Odonata and Ephemeroptera in which parts as advanced as the above have not been found although the presence of an empodium in Odonata must be considered.
8. At this stage it is impossible to attempt any valid cor- relations between the types of pretarsi found in the higher orders other than such obvious relations as that of Trichop- tlera to Lepidoptera. However, it is evident that the varia- tions of the several parts of the pretarsus found among the Holometabola, even in such a random selection of cases, warrant a thorough investigation of the condition in these orders with a view to possible clarification of the phyloge- netic implications concerned.
Arnhart, L. (1923). Das Krallenglied der Honigbiene. Archiv f. Bienenkunde, 5 :37-86, 1 pi.
Crampton, G. C. (1923). Preliminary note on the terminol- ogy applied to thfe parts of an insect's leg. Canadian Ent,., 55 :126-132.
Hayes & Kearns. (1934). The pretarsus (articularis) in Coleoptera. Ann. Ent. Soc. American, 27 : 21-33, 41 figs. MacGillivray, A. D. (1923). External insect-anatomy. Ur- bana, 111.
deMeijere, J. C. H. (1901). Ueber das letzte Glied de*r Beine bei den Arthropoden. Zool. Jahrb., Anat. 14 :417-476, pis. 30-37.
Snodgrass, R. E. (1925). Anatomy and physiology of the honey-bcee. McGraw-Hill Co., New York.
- (1927). Morphology and mechanism of the insect thorax. Smithsonian Misc. Coll., 80, No. 1, 108 pp., 44 figs.
(1929). The thoracic mechanism of a grass- hopper and its antecedents. Smithsonian Misc. Coll., 82, No. 2, 111 pp., 54 figs.




================================================================================

19351 Structure of the Pretarsus 21
Psyche, 1935 VOL. 42, PLATE I.
FIG. 1. Anax junius Drury.
FIG. 5. Termopsis ungusticollis
FIG. 2. Siphlonurus sp. Walker.
FIG. 6. Oligotoma sp.
FIG. 3. Periplaneta americana L.
FIG. 7. Pteronarcys dorsata Say.
FIG. 4. Tenodera sinensis Sauss.
FIG. 8. Melanoplus tiwittatus Say.




================================================================================

22
Psyche, 1935
Psyche [March
VOL. 42, PLATE 11.
FIG. 9. CeuthopMIus maculatus FIG. 13. Psalis sp. Say.
FIG. 14. Agallia constricta Van
FIG. 10. Gryllus pennsylvanicus
Duzee, ventral and lateral
Burm. views.
FIG. 11. Gryllotalpa borealis
FIG. 15. Brochymena arborea Say.
Burm. FIG. 16. Phyllophaga fusca Froe-
FIG. 12. Anisomorpha buprestoi- lich.
des Stoll.
FIG. 17. Corydalis cornuta L.




================================================================================

19351 Structure of the Pretarsus 23
Psyche, 1935
VOL. 42, PLATE 111.
FIG. 18. Ithone sp.
FIG. 22. Automeris io Fabr.
FIG. 19. Mantispa brunnea Say.
FIG. 23. Tabanus atratus Fabr.
PIG. 20. Panorpa rufescens Ram- FIG. 24. Diogmites umbrinus Loew. bur. FIG. 25. Tremex columba L.
FIG. 21. Astenophylax argus
FIG. 26. Chalybion caeruleum L.
Harris.




================================================================================

Psyche
[March
ap -apodeme
ar - arolium
ar ? - arolium (doubtful)
bp - basipulvillus
cm -camera
dt - distitarsus
em - empodium
em? - empodium (doubtful)
fm - flexor membrane
pem - parempodium
pi - planta
pv - pulvillus
pv ? - pulvillus (doubtful)
rb - orbicula
uf - unguifer
un -ungu1es
ut - unguitractor
Except where otherwise noted the pretarsi figured were taken from the right meta-tarsus and are ventral views.



================================================================================