E. O. Wilson, T. Eisner, and B. D. Valentine.
The Beetle Genus Paralimulodes Bruch in North America, with Notes on Morphology and Behavior (Coleoptera: Limulodidae).
Psyche 61(4):154-161, 1954.

This article at Hindawi Publishing: https://doi.org/10.1155/1954/53278
CEC's scan of this article: http://psyche.entclub.org/pdf/61/61-154.pdf, 552K
This landing page: http://psyche.entclub.org/61/61-154.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.

THE BEETLE GENUS PARALIMULODES BRUCH
IN NORTH AMERICA, WITH NOTES ON
MORPHOLOGY AND BEHAVIOR
(COLEOPTERA : LIMUL0DIDAE)l
BY E. 0. WILSON, T. EISNER, and B. D. VALENTINE Biological Laboratories, Harvard University The family Limulodidae was erected by Seevers and Dybas in 1943 to receive the staphylinid subfamily Ceph- aloplectinae and the trichopterygid subfamily Limulodinae. It includes a small number of species with body form con- spicuously modified for a myrmecophilous existence : body tear-shaped or oblong, head covered by the pronotum, base of the antennae protected by deep fossae, labrum in re- pose contiguous with the mentum, etc. These specializa- tions are of the type generally characterizing "loricate synoeketes", ant guests more or less indifferently toler- ated by their hosts and protected from occasional molesta- tion by a shielded body form.
Of the five limulodid genera recognized by Seevers and Dybas, the most specialized and obscure is Paralimulodes Bruch. This genus has hitherto been known only from four specimens of the single species, P. wasmanni Bruch, taken with a colony of the army ant Neivamyrmex spegaz- zinii (Emery) at La Plata, Argentina in 1919. It was therefore a matter of some interest when the genus was recently rediscovered, this time in the southeastern United States. Living material was maintained with the host ant colonies in artificial nests long enough to record some as- pects of behavior, and enough preserved material was ob- tained to fill in most of the important detail inadequately covered by Bruch's original description. New information obtained in this study is presented in the sections below. Distribution and taxonomic status of the North Ameri- can population. Over two hundred specimens, all appar- ently belonging to the same species, were collected by Wil- son at the following three localities in central and northern 'Published with a grant from the Museum of Comparative Zoology at Harvard College.
154




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

19541 Wilson, Eisner, and Valentine - North American Paralimulodes 155 Alabama: Bryce Lake, Tuscaloosa, Tuscaloosa Co. [host Neivamyrmex carolinense (Emery) ] ; Hurricane Creek, near Peterson, Tuscaloosa Co. [host N. carolinensel ; King Cove, Bankhead National Forest, Lawrence Co. [host N. nigrescens (Creslson) 1. The species has been tentative- ly determined as P. wasmanni. Since syntypes of wasman- ni are not available at present, we have relied on Bruch's original description and detailed drawings. No difference could be found on this basis, and we feel confident in stat- ing that the two populations must be very close if not Text-figure 1. Paralimulodes riding on the head of a Neivamyrmex nigrescens major worker.
identical. Specimens have been deposited in the Museum of Comparative Zoology, Harvard University, and Chicago Natural History Museum.
Morphology. (See figures. Most of the specimens used in this study were cleared with 10% KOH and stained with Chlorazol Black E, acid fuchsin, or fast green. Because of the minute size of the material, most of the examination was made under oil immersion lens.)
Size extremely small, total length approximately 0.6 mm Antenna %segmented. Antenna1 fossa deeply excavated but nevertheless capable of accommodating only the scape. The antenna is otherwise well protected, being able to move freely between the prosternum and the concave ven- tral surface of the remainder of the prothorax. The ter- minal four segments are flattened dorsoventrally. A pair of what seem to be sensory pegs, one dorsal and one ven-



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

156 Psyche [December
tral, project beyond the tip of the terminal segment. These are visible only under the best conditions of lighting and were apparently missed by Bruch.
Two "eye-spots", apparently nothing more than tiny optically light areas, are present near the outer lateral margins of the head. These are essentially no different from other such areas which occur sp~rad~ically on several other parts of the body (see pi. 13, fig. d), but may be significant in that they are larger and uniform in size and are evident at the same location in every specimen examined.
Maxilla tripartite ; maxillary palp 4-segmented, the third segment greatly swollen and ovoid. The labium appears to be degenerate in comparison with that described by Seevers and Dybas for Cephaloplectus mus, although ad- mittedly this structure could be seen only in ventral view in our material.
Two segments are visible, the distal one subg globose and with no appendages other than the palps evident. Segmentation in the labial palps is apparently obsolescent.
The mandible consists of a well sclerotized piece, the distal portion of which is divided into a "molar" projection terminating in several sharp teeth, and a thin, nearly trans- parent "incisor area", edged by numerous seriate denticles barely within the limit of light resolution. The incisor area has never been described in other limulodid genera but may have been overlooked due to its almost invisible structure.
Paralimulodes does not differ much from Limulodes and Cephaloplectus in thoracic structure. Contrary to the claim of Bruch, the prosternum is not a distinct piece, but merges into the pronotum anteriorly by a pair of narrow arms. Posteriorly it is free and overlaps the mesosternum and anteromedian metasternal projection.
The tarsus is two-segmented, the basal segment sub- EXPLANATION OF PLATE 12
(Paralimulodes wasmanni Bruch; Lawrence Co., Ala.).-Fig. A, out- line, dorsal view, showing habitus. Fig. B, dorsal view, with abdomen distended by treatment with KOH. Fig. C, ventral view, abdomen also distended. (Sculptural detail not shown).



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




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

168 Psych6 [December
triangular in outline and partly overlapping the cylindrical distal segment. The pretarsw contains some detail not resolvable by the light microscope, but there is visible a single (?) irregularly shaped, nearly transparent distal pad flanked by at least two bristles.
GIawa in the conven-
tional sense are lacking.
Seven abdominal segments are visible in ventral view (see pi. 12, fig. B). Division in the anterior tergites is indistinct, so that correspondence with individual sternites is uncertain. Perhaps the most unusual feature of the ab- domen is the shape of the first six visible sternites. These are very wide and are produced dorsally to overlap in part the adjacent tergites. The dorsal sternital tips are pre- sumably what Seevers and Dybas refer to as "paratergiteaY' in their general description of the limulodid abdomen. This peculiar modification in abdominal structure seema best interpreted as providing a sort of shingle-armor pro- tection f or the lateral intersegmental membranes. Behavior. As noted previously, Paralimulodes occurs naturally with the army ants Neivamyrmex carolheme and N. nigrescens in Alabama. Both of these species are abundant within the presently known range of the beetle. That Padimulodes is truly hosbspecific for Nebamjrmex is suggested by the following- observations. A colony of N. nigrescens from Hurricane Creek, found within fifty feet of 'the infested carolinerne colony, was maintained for sev- eral days with the queen in an artificial nest. No PmraIimu- lodes could be found with the thousands of corkers, des- pite the earlier proximity of the two ant colonies. But å´whe beetles were isolated from the caroZinense and placed in the nigrescens nest, they proceeded with little hesita- tion or opposition into the mass of resting workers. Twen- ty-four hours later they were still ensconced with their new hosts. When other Pa~dimdodes were transferred to E~~~~AMATION or PLATE 13
(Paralå´tmå´alod vsaamantii Brueh; Lawrence Co., Ah.).- Figs. a-d, mouthparta in ventral view.
Fig, e, ventral view of left antenna. Pig. f, aedeagus tube, ventral view. Fig. g, spermathem, ventral view. Fig. h, tarsus, including only structures clearly visible witit the light micro- scope; terminal tareal segment rotated to show pretarsus in dorsal view.



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




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

160 Psyche [December
nests containing colonies of the non-ecitonine species Mono- morium minimum ( Buckley ) , Crematogaster minutissima Mayr, and Camponotus pennsylvanicus (DeGeer) , they completely ignored these ants and wandered aimlessly about the nest chambers.
The Paralimulodes present a very bizarre appearance in life. When resting they superficially resemble mites. They are usually to be seen riding on the heads and ab- domens of the workers and only rarely run along the ground in a foraging column of the ants. On several oc- casions two were observed riding on the same worker and in this circumstance were positioned symmetrically on each side of the head or abdomen, a phenomenon already noted in some myrmecophilous mites (Wheeler, 1926) and in Limulodes (Park, 1933).
When the ants are moving in columns the beetles re- main more or less stationary on single individuals. Only when the ants crowd together to rest do the beetles show much activity of their own. Then they begin to travel over the bodies of their hosts in light, rapid, jerky move- ments, climbing appendages, scurrying around curved body surfaces, and skipping freely from one individual to an- other. The ants do not seem to be bothered by this activity at all, and in fact show no sign that they are aware of their tiny guests7 existence. Once a beetle was seen to cross the extended antennae of two workers which had met and were palpating one another, and even this imperti- nence failed to evoke a response.
Like Limulodes par& (see Park, 1933), Paralimulodes is a strigilator, obtaining its food by scraping organic matter from the cuticle of its hosts. On several occasions individuals were seen moving their mouthparts along the anterior gular rim and coxal insertions of resting workers. No ant brood was present in the nests at the time of ob- servation, so that it could not be determined whether the beetles confine their attention to the adults. Acknowledgement. The authors are indebted to Dr. H. S. Dybas for examining part of the Lawrence County series and making certain valuable corrections and sug- gestions relative to the foregoing section on morphology.



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

19541 Wilson, Eisner, and Valentine-North American Paralimulodes 161 LITERATURE CITED
PARK, 0.
1933. Ecological study of the ptiliid myrmecocole, Limulodes para- doxus Matthews. Ann. Ent. Soc. Amer., 26: 255-261, fig. SEEVERS, C. H. and H. S. DYBAS
1943. A synopsis of the Limulodidae (Coleoptera) : a new family proposed for myrmecophiles of the subfamilies Limulodinae (Ptiliidae) and Cephaloplectinae (Staphylinidae). Ann. Ent. Soc. Amer., 36: 546-586, 81 figs.
WHEELER, W. M.
1926. Ants, their structure, development, and behavior. Columbia Univ. Press. 663 pp.




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