A. B. Gurney.
Notes on Dilaridae and Berothidae, with Special Reference to the Immature Stages of the Nearctic Genera (Neuroptera).
Psyche 54(3):145-169, 1947.

This article at Hindawi Publishing: https://doi.org/10.1155/1947/78317
CEC's scan of this article: http://psyche.entclub.org/pdf/54/54-145.pdf, 2288K
This landing page: http://psyche.entclub.org/54/54-145.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. 54 SEPTEMBER, 1947 No. 3
NOTES ON DILARIDB AND BEROTHIDE, WITH
SPECIAL REFERENCE TO THE IMMATURE
STAGES OF THE NEARCTIC GENERA
(NEUROPTERA)
BY ASHLEY B. GURNEY
Bureau of Entomology and Plant Quarantine, Agricul- tural Research Administration, United States Department of Agriculture
To students of neuropteroid insects, as well as to ento. mologists interested in specialized types of holometabo- lous larvae, it will be significant that the immature stages of Nallachius and Lomamyia have been found. Nal- lachius is the only Nearctic genus of Dilaridae and con- tains two species in the United States, both of which were originally referred to the genus Dilar. Lomamyia is the sole Nearctic genus of Berothidse, represented in this country by 10 species, and, like Nallachius, they are all relatively rare insects. With the exception of the eggs and first-stage larvse of Spermophorella, an Australian berothid genus, the young stages of these two families have been entirely unknown. Though some details of the biologies of Nallachius and Lomamyia are still unknown, the main features may now be presented. Their larvae prove to be related, though perfectly distinct, predators which attack soft-bodied insects.
The discovery and recognition of the young of these insects may be largely attributed to the collecting zeal and generous cooperation of my colleagues, William H. Anderson and Herbert S. Barber. Dr. Anderson col- lected the male allotypel of Nallachiw arnericanus (McL.) in 1939, and has also collected four lots of larvae, with 1 Designated by Carpenter (1940) subsequent to the original description, 1881. This specimen was reared, but the cast skins were not recovered. 145




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

146 Psyche [ Sept.
one lot of which two pupae were associated. The pup% died before maturity, bat possess the characters of the adult sufficiently well to permit identification. A cast larval skin of one is identical with larvae which he col- lected.
This material of americanus was first correctly identified in 1945 by Henry K. Townes.
More than 50
years ago the late H. G. Hubbard collected a single nearly fully grown larva of Lomamyia which could not then be identified. A similar, though fully developed, larva taken in 1941 by R. J. Kowal likewise could not be named until the key to this identification puzzle was secured through study of first-stage larvae hatched by Mr. Barber in 1919 from eggs laid by a captive L~mamyia.~ F. M. Carpenter of Harvard University made some most help- ful suggestions, when he learned of the study in progress, and I am also grateful for the photograph of Nallachius wings (fig. 1) which he kindly offered to make. Finally, I would express my appreciation to George E. Wallace of the Carnegie Museum for the loan of adult specimens of Nallachius americanus and the privilege of retaining two of them for the United States National Museum. In this paper I have included a few notes on the distri- bution and variation of Nallachius americanus, though my primary object is to compare the immature stages of Nallachius and Lomamyia morphologically and to discuss the relationship of the Dilaridae and Berothidae to other families in thelight of present information. The genus Nallachius Navas
Nallachius Navas, 1909, Mem. Real Acad. Cienc. Artes Barcelona, vol 7, pp. 627, 666. Genotype: Dilar pres- toni McL., 1880, designated by Navas (1914). Carpenter's revision (1940) and subsequent additions (1942, 1947) should be consulted for information on the taxonomy and distribution of the Nearctic Dilaridas and 2 The specific identity of this female is uncertain. It was determined as
L. flavicornis (Walk.) by A. N. Caudell in 1919, but determinations current at that time are rendered untrustworthy by the description of several species then unrecognized. The specimen was utilized for morphological study and is not now available.




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

19471 Dilaridt~ and Berothidis 147
Berothi&. The only sizable collection of Nallachius adults yet reported is that of Steyskal (19441, who col- lected six females and 21 males of americanus flying around a single dead tree in Detroit, Mich., during the months of June and July. It is possible that larvse oc- Fig. 1. NdlaehWa bzertea.nw (McL.), female, dorsal view of right wings. Length of front wing, 7.5 mm. Brooklitte, Pittsburgh, Pa. Photograph by P. M. Carpenter.
curred in this tree, and also that oviposition was in progress.
Previously unreported adult material of N, americanus consists of four males and four females from the Pitts- burgh area of Pennsylvania, one male from Odenton, Md., and a male from Bainbridge, Ga. (Detailed data appear



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

148 Psyche [ Sept.
in the section headed ' ' Material examined. ' ') The wings (fig. 1) are specific for this insect, but male genitalia should be examined to confirm identifications whenever possible.
A photograph of the wings has not previously been published, though Costa Lima (1943, figs. 54,55) has given excellent photographs of the Brazilian N. prestoni. Size variation is indicated by the Pittsburgh series, a ' front wing of each of the males measuring (length in millimeters) 5.1, 4.8, 4.8, 3.6, respectively, and a front wing of each of the females from the same series measur- ing 7.5, 6.3, 6.1, 5.6, respectively. Forewings of the Odenton and Bainbridge males measure 4.2 and 4.3 mm. Male antenna usually include 9 pectinations (plus the apex of the central axis), though a few specimens have 10 pectinations.
Immature stages of Nallachius americanus Eggs.-One female examined (opposite Homewood Cemetery) has a cream-colored egg held at the apex of the ovipositor. It measures about 0.37 X 0.13 mm. and is rounded oblong.
Larva.-General body shape slender, elongate, convex (fig. 6); body set= sparse, inconspicuous, long, and slender; spiracles well developed on mesothorax and ab- dominal segments I-VIII, scarcely noticeable without compound microscope. Head with elongate, somewhat blunt jaws of usual Planipennia type (figs. 11, 14); no fracture line of jaws evident; mandible and maxilla sub- equal in size; apex of maxilla with sensory set~ and spiculelike teeth ; front triangularly produced between and anterior to antenna1 bases; a single dark pigmented, lateral eye spot ; a tiny curved spine immediately anterior to eye. Antenna with 2-5 ring segments following scape (total segments 5-8) ; penultimate segment enlarged, bearing postapical peglike organ (sensory?) and 2 small seta externolaterally ; apical segment with one long prin- cipal seta and several minor ones.
Prementum apically
incised; labial palpus with 2-5 ring segments following basal segment (total segments 5-8). Cervical region (microthorax of Tillyard, 1916) with well-developed dorsolateral lobes.




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

19471 Dilaridee and Berothidce 149
Thoracic segments without noticeable sclerotized plates; mesothoracic spiracle in fold created by small anterior subsegment. Front legs somewhat more elon- gate and robust than middle and hind legs. Leg (fig. 21) with femur and tibia moderately compressed ; tarsus dis- tinct but capable of little or no movement on tibia; tro- chanter weakly defined; front claws blunt with anterior claw decidedly shorter than posterior one (fig. 17) ; claws of middle and hind pairs much smaller than front ones, unequal but not conspicuously so; paired pulvilli (or similar pads) at base of claws; empodium slender, trumpet-shaped, segmented in apical third, a second line of segmentation indistinct. Abdomen uniformly, weakly sclerotized; segments I-VIII with lateral spiracle in anterior third ; each lateral margin of segments I-IX (on slide-mounted specimen) with two principal lobes, each lobe with a principal seta.
Coloration: Head gamboge yellow, somewhat darker above ; thorax and abdomen uniformly pale, almost color- less; front claws light brown, middle and hind claws paler.
The supposedly mature larvae of americanus range in length from 4.6 mm. to 12 mm., while the number of ring segments of the labial palpus and antenna vary as indi- cated above, and in several cases there is a difference of one segment in right and left appendages of the same specimen. The larva from College Park is the longest and has the largest number of antennal and palpal seg- ments, but the one from Jackson's Island is the shortest (4.6 mm.) and has one more segment in each appendage than two larvae from University Park examined in detail (4.8 mm. long, 5 segments in each appendage). The sex of these larvae is naturally unknown, aside from the cast skin from which a male pupa developed. Adult males average definitely smaller than females. From the evi- dence now at hand, it is accordingly uncertain whether all or only part of these larvae are fully developed, whether the number of palpal and antennal segments is variable from instar to instar, as well as in larvae of the same instar, and whether sex has a bearing on size of



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

150 Psyche [ Sept.
larvae or number of palpal and antenna1 segments. Only one species of Dilaridae is now recorded from the eastern United States. Most but not all Neuroptera have 3 larval instars.
Pupa.-The pupa illustrated (fig. 16) is near maturity, as evidenced by well-developed genitalia and wing vena- tion, and dark color of the well-sclerotized portions of thorax and abdomen and on the apical two-thirds of tibia. The wings are practically black, except at the bases, and the Nallachius-type venation may be traced. Except as indicated by stippling on figure 16, the pupa is pale. Each antenna is a simple appendage, but the pectinations of the adult are visible within, wound about the central axis in spiral fashion. Body length is 3.26 mm. The other pupa studied is not so near maturity as the one illustrated, and, though it is clearly a male, the dark areas are scarcely developed and the wing venation can- not be readily traced.
Cocoon.-The cocoon (fig. 3) is a tightly woven white sack without a visible exterior sculpture of any kind. The outer surface is covered loosely, and without regard to pattern, with clay-yellow particles, which appear to be fragments of wood. Others appear to be the excretory pellets of wood-feeding insects and are darker. Frass
and various other particles frequently occur in the gal- leries of wood-boring insects, which explains their pres- ence on the cocoon of Nallachius.
Material of Nallachius americanus examined3 (previ- \-
ously unreported) :
Adults.-Pittsburgh, Pa., all collected by Hugo Kahl: Fern Hollow, July 15, 1911, near orchard (I<?) ; Fern Hollow, August 6, 1907, on tree trunk near hilltop (1 9) ; Southern Avenue Park, July 8, 1911, in coitu (8, 9) ; op- posite Homewood Cemetery, on an oak trunk, July 9,1910 (1 9) ; Brookline, on tree trunk, July 17, 1910 (1 9). Ingram, Pa., June 28, 1931, W. D. McIlroy, Jr. (2 3). Odenton, Md., August 14,1918, on Robinia pseudacacia, W. L. McAtee (1 3).
3 All the Pennsylvania specimens except the two from Fern Hollow, July 15, and from Brookline, are deposited in the Carnegie Museum, Pittsburgh, Pa., all other material of both Nallachius and Lomamvia in the United States National Museum.




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

19471 Dilaridca and Berothidae 151
Bainbridge, Ga., April 6, 1943, at light, H. R. Dodge (13-
Larvae and pupae.-Jackson's Isl., Md. (also known as Scott's Isl. or Turkey Isl., located in Potomac River 11 miles above Washington, D. C.), found May 19,1913, near a larva of Micrornalthus, while breaking up wood frag- ments taken from a decaying log, May 5 ; rearing unsuc- -
cessful ; dead June 9 ; many other kinds of larvae in same log; E. A. Schwarz and H. S. Barber, 1 larva. Greenbelt, Md., April 13, 1938, beneath bark of moist rotten log, W. H. Anderson, 1 larva.
College Park, Md., October 18, 1942, under bark of dead oak, W. H. Anderson, 1 larva.
University Park, Md., January 28, 1945, in hard dead tulip-tree wood in galleries of living larvae of Pentar- firinus and Phloeophagus weevils, W. H. Anderson, 4 larvae. Two pupae in cocoons recovered from same wood sample, but not in same gallery with the 4 larvae. Langley, Va., November 1, 1939, with bark from tree, W. H. Anderson, 1 larva.
Biological notes: While the larvae of Nallachius have not been observed feeding, there is no doubt that they attack other insects, as indicated by the piercing-sucking jaws and their presence within the galleries of wood-in- habiting insects. They themselves are not equipped to bore in wood, though the rather heavy claws, particularly those of the front legs, are presumably well adapted to locomotion in galleries often somewhat choked with the frass of their would-be victims. The integument of most of the larval body is pale and weakly sclerotized, appar- ently an adaptation to a minimum exposure to light and the open air. Most adults evidently emerge during spring and early summer. Apparently there is a single generation per year and a 1-year life cycle. A large percentage of adults recorded were collected on or about tree trunks, and eggs are probably inserted into the cracks of dead wood or the crevices of or beneath dead bark. Judging from the long ovipositor, it is extremely unlikely that the eggs are stalked.




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

Psyche
[Sept.
The genus Lomamyia Banks
Lomarnyia Banks, 1904,4 Proc. Ent. Soc. Wash., vol. 6, p. 209.
Genotype : Hemerobius flavicornis Walk., by monotypy. Two species, fiavieornis and bawksi Carp., are recorded from the vicinity of Washington, D. C., and it is possible that still others occur here, so the specific identity of the larvae described below cannot be determined. Immature stages of Lomamyia
Eggs.-While confined in a glass tube, a female col- lected at light on Plummer's Isl., Md., June 16, 1919, laid clustered stalked eggs similar to those of many Chrysopidae. There was one cluster of 12 white eggs, each about 0.7 mm. long, attached to the end of a thread- like stalk 6 mm. long, the terminal end of the stalk being more slender than the base. There were several smaller and less perfect clusters. After 24 hours the eggs had transverse dark bars due to the development of banded embryos. Smith (1923, p. 139) has described a single infertile unstalked egg laid by a captive female of Lomxwqia. The unstalked condition may have been ab- normal.-
First-stage larva.-General body shape (fig. 2) much as in Nallachius, head more elongate; jaws subequal in length to palpi and antenna; 3 pairs of equal legs ; max- illa dominant, with scalelike dorsal sculpture (fig. 8), the apical tenth specialized and weakly serrate (fig. 7) ; man- dible reduced, slender, lateral ; labial palpus and antenna 3-segmented, second and third segments each with numer- ous annulations. Front broadly produced ; two lateral eye spots ; cervical region and legs well developed ; latero- dorsal thoracic sclerites absent or inconspicuous ;5 thorax and abdomen with conspicuous reddish purple transverse bands, as shown by stippling (fig. 2).
4 The genus Lornarnyia was first proposed in a list of species; the formal description appeared late the following year in Banks' revision of the Nearc- tic Hemerobiidee.
5 See Killington (1936, fig. 43, p. 97).



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

19471 Dilarida and BerotMdae 153
Fully developed larva (possibly more correctly de- scribed as a prepupa, though no cocoon is associated).- Abdomen much more enlarged than head, in contrast to first instar (fig. 5) ; legs reduced; jaws truncate, much shorter than palpi (loss of apices by fracture possible but uncertain) ; maxilla with conspicuous sculpture (fig. 12). Front more produced than in the first instar, broadly rounded; anterior half of median dorsal surface of head with broadly convex area poorly demarked; posterior half with pattern of pale sutures and 3 lateral setae. Pre- mentum small, irregular, anterior margin entire ; mentum narrowed apically, broadly rounded basally, flanked by elongate sclerites (undifferentiated cardo and stipes ?) ; cervical region with conspicuous, laterally concave apo- demes, each with a detached, V-shaped base. Front and middle legs subequal (hind legs lost) ; tarsus rather freely articulated with tibia (fig. 22) ; trochanter distinct; tarsal claws equal; no observed pulvilli at base of claws; em- podium apparently broadly incised apically, otherwise much as in Nallachiuq a definite " sole " ventrad of claws (fig. 18). Pronotum with large paired laterodorsal sclerites on anterior two-thirds, a weak transverse suture on posterior third ; meso- and metanotum each with small, narrow, semilunar sclerites ; anal sucker apparently well developed.
Coloration : Head, cervical apodemes, thoracic sclerites, and cox= brown ; claws and " sole" pale brown ; abdomen with poorly demarked dorsal pattern of light purple transverse bands ; remainder pale. The unpublished notes of H. G. Hubbard contain the following description, made from life, of the Washington larva : "General color is purple-brown, darker on thoracic segments and almost black on the head; lighter beneath; variegated with pure white. The legs and mouth organs are translucent white. Beneath, each abdominal segment has a large triangular spot of yellowish white, which looks like the luminous glands of Lampyridae, but was found not luminous above or below. The dorsal surface is gaily decorated with snow-white markings-viz. : On prothorax, on each side, a narrow white line beginning behind the



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

154 Psyche 1 Sept.
middle ; on meso- and metathorax, on each side,, just above the legs a quadrate white spot; on second, third, fourth and fifth abdominal segments, also a narrow transverse band of white extending entirely across the segment near the hind margin, the ends of the bands enlarged, rounded and slightly curved forwards; the same white band on sixth and seventh segments but interrupted or less con- spicuous (obsolete) ; the last two abdominal segments are pale and the terminal one is a suckerlike organ, not used in progression, but assisting the insect to cling to smooth surfaces, as in Hemerobius."
The slide-mounted Washington larva shows that the prothoracic laterodorsal sclerites bear several setae, each -
arising from a well-developed pit. The sclerites of meso- and metanotum each bear a single posterior seta arising from a large pit. Other thoracic and abdominal setae are sparse, inconspicuous, mainly arranged in transverse rows.
Both the Washington and Beltsville larvae have the jaws proportionally much shorter than those of first- stage larvae, but there is no definite fracture line, as described in Osmylus by Killington (1936, pp. 99, 224) and Withycombe (1923, p. 516), and the maxillae of the Washington larva are not broken off evenly with the mandibles. Study of more material will be necessary to determine whether there is a normal shortening of the jaws subsequent to the first instar, whether there is a regular fracture prior to pupation, or whether the jaws of the two larvae here studied were accidentally broken. Pupa.-Unknown.
Material of Lomamyia examined :
Larvae.-Washington, D. C., July 19,1895. "Found in the fungus garden (nest) of Atta occidentalis6 in wooded knoll, on Brentwood Road, near Soldiers ' Home. " H. G. Hubbard, 1 larva.
Beltsville, Md., October 28,1941, in fallen log, in pocket containing living and dead termites, R. J. Kowal, 1 larva. Plummer's Isl., Md. (located in Potomac River 8 miles 6 My friend, M. R. Smith, has informed me that the ant in question was almost surely Trmhymyrmex septentrionalis (McCook). Specimens are not
known to have been preserved.




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

19471 DiZarida and Bero thidce
above Washington, D. C.), hatched from
female collected at light June 16, 1919, H. eggs laid by
S. Barber, 6
first-stage larvae and several eggs.
Biological notes: The first-stage larvae reared by Mr. Barber hatched from 2-day-old eggs and were kept in a vial. His notes and those of A. N. Caudell describe these larvae as extremely active, running very rapidly on the smooth glass surface without use of the anal sucker. A crushedfly offered as food was not touched, and they at- tempted unsuccessfully to pierce larvae of ants. Living larvae and pupae of Scolytidae and immature termites proved very attractive. A larva would rush up to a victim, stop, then lunge forward and pierce the prey with the jaw of one side. After the jaw was withdrawn, the head would be turned to direct the jaw of the other side toward the prey, whereupon another lunge would be made. With a lens Mr. Caudell observed the jaws in- serted into scolytid larvae. It was seen that some of the Lomanqia larvae increased in size, and the scolytid larvae were dead the following day. The Lomamyia larvae un- fortunately died within a few days, perhaps due to a low atmospheric humidity.
There is no indication that the Beltsville larva (fig. 5) - ,
was removed from a cocoon; its robust abdomen and very small legs may point to a recent engorgement or a seden- tary life, rather than preparation for pupation. Killing- ton (1936, p. 128) should be consulted for a discussion of neuropterous prepupse. The Washington larva (4.9 mm. long) is apparently in an earlier stage than the Beltsville specimen (9.36 mm.), and an enlarged abdomen is not evident, though its condition is somewhat unsatis- factory. It is suggestive of nest predatism7 that one larva was taken among ants, the other among termites. I am inclined to believe that the later larval instars of Lomawnja are less active than those of NaZlac7%ius. Sny-
der (1920, p. 120) reported a strange neuropterous larva which he found in a termite colony at Falls Church, Va., in 1918. His notes lead one to suspect that the larva may 7 Wheeler, Ants, 1926, p. 382, has discussed enemies living in ant nests, calling them syn~chthrans.
True commvsals are of a different nature, living at the expense of but without harm to the host.



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

156 Psyche 1 Sept.
have been Lomamyia, but the specimen is not now avail- able.
Although the larval feeding mechanism of Lomamyia is specialized to function almost identically like that of Nallachius, the smaller legs of the mature Lamamyia larva, and the fact that the front leg and claws equal the middle ones, suggest somewhat different habits. Dr. Anderson consistently sought wood-inhabiting beetle larvae in his collecting, and five times he secured Nal- lachius larvae or pupae, never Lomamyia. This fact sug- gests that Lomamyia larvae occupy a different habitat. The eggs of the Australian Spermophorella are borne on separate stalks, according to Tillyard (1916), and the first instars are much like those of Lomamyia. He re- ported that the larvae walked with the combined use of legs and anal sucker, moving with a looping action similar to that of Geometridae.
Principal larval characters distinguishing Lomamyia and Nallachius Lomamyia
Penultimate antenna1 segment
without subapical peg.
Antennae and palpi 3-segmented,
no supernumerary segments.
Maxilla much larger than man-
dible, dominant both ventrally
and dorsally.
Antennae borne at lateral ex-
tremities of head.
Anterior margin of front broadly
rounded.
A single simple eye at each side
of head.
Several ventral sclerites in region
of mentum.
Prementum apically entire.
Laterodorsal thoracic sclerites
well developed.
Legs much reduced in proportion
to abdomen (mature larva).
Front legs equal to middle (and
hind?) legs.
Claws equal.
Tarsal claws with a ventral
( ( sole. ' ?
Tarso-tibia1 joint mobile; tro-
chanter distinct.
Nallachius
1. Subapical peg present.
2: More than 3-segmented, super-
numerary segments present.
3. Mandible and maxilla subequal ;
mandible conspicuous dorsally,
maxilla ventrally.
4. Antenna1 bases much less re-
mote.
5. Anterior margin acute.
6. Two simple eyes at each side.
7. Ventral sclerites of head reduced
in number.
8. Prementum apically divided.
9. Absent or indistinct.
10. Legs only moderately reduced in
proportion to abdomen.
11. Front legs larger than middle
and hind legs.
12. Claws unequal, especially those
of front legs.
13. No " sole ' present.
14. Tarso-tibia1 joint apparently not
mobile ; trochanter poorly de-
veloped.




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

1947 I Dilaridee and Berothid& 157
The foregoing generic characters are subject to modi- fication as Iarvse of other species become known. In sort- ing material, the subapical peg of the penultimate anten- nal segment of NallmJ~s and the sculptured dorsal sur- face of the maxilla of Lomamyiu have been found particu- lady useful characters. The dominance of the maxilla in Lomamyia is one of the most fundamental differences be- tween the two genera. Withyeombe (1925, p. 328) has attached considerable significance to the relative domi- nance of mandible and maxilla in neuropterous families. Tillyard (1916) stated that the mandible of 8permoplw- rella is dominant, but his figures indicate that the first instar larva (figs. 4, 10) is very much like that of Lomamyia (figs. 2, 8), and it is possible that he misin- terpreted the mouthparts. If that be true, the close similarity of SpermophoreIh and Lomamyia larvae strongly suggests the stability of larval family charac- ters in the Berothidse, even when represented by genera occurring in distant parts of the world, FAMILY RELATIONSHIPS
The larvae of neither the Berothidse nor the Dilarid~e, as represented by the genera studied, may readily be con- fused with those of any other family. The terrestrial neuropterous larvae most familiar to entomologists are Chrysopidse and Hemerobiidfe. The latter are superfi- cialy suggestive of Berothidse and Dilaricte, but in each case, as in the hemerobiid genus Wesmcslius (fig. 91, the jaws are strongly incurved, not extending straight for- ward. In fact, the berothids and dilarids clearly belong to what Withyeombe (1925) has termed the straight-jawed families of Neuroptera, in contrast to the many families whose larvse consistently have strongly incurved jaws. First-instar Hemerobiidse have a trumpet-shaped empo- drum, but this becomes a broad pad in later instars (fig. 20). The Chrysopidse is one of the very few families to retain a trumpet-shaped empodium in all larval instars (fis. 19). The well-illustrated key by Townsend (1935) will enable students to recognize the families of most Nearctic neuropterous larvae,




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

158 Psyche isept.
Withyeombe placed the Dilarids~ and Berothi&, to- gether with the Mantispidse, Osmylidse, Sisyridse, and Myiodactylidse, in a superfamily called the Osmyloidea. With the exception of the Myiodactylidre, I believe that this grouping is entirely correct.
Larvae of the latter
have been illustrated and described by Tillyard (1926, fig. 11171, and, on the basis of larvae, it seems clear that the family is related to the Ascalaphidse and associated families. So far as I am aware, there was no description of mylodactylid larvse prior to 1926, though Tillyard (1917, p. 543) mentioned the larva of Mifiodactylv-s and indicated it would later be
described. It is probable,
therefore, that no information concerning these larvae was available to Withyeombe, and that the resemblance between Oamylidse and Myiodactylidse with respect to wing venation led him to associate closely the two families.
Klingstedt (1937) has suggested that the Dilaridse may be closely related to the Raphidiodea, basing this view on chromosome structure.
Tjeder (EM'), who has studied
the external and internal anatomy of adult Dilaridse and Eaphidiodea, points out both similarities and dissimi- larities, and concludes that sufficiently close relationship is shown to justify transferrhg the Dilaridse from the Newoptera (strict sense, Planipennia of authors) to the Raphidiodea.' Larval characters of the dilarids, previ- ously rmlmown, impress me as more trustworthy indica- tors of relationship than the adult characters thus far studied, and consequently I believe the primary affinities of the Dilaridse are with the Osmyloidea. The Berothidse are close relatives.
The presence of an ovipositor in both raphidiids and dilarids has led Tjeder and others to suspect close rela- tionship, A comparison of the adult female ovipositor of Nallachius (fig. 15) with those of mantispids of the genera Symphrasis and Plegu, and with those of the raphidiid genera Agulla and Inocellw shows that there is an abrupt bend or elbow at the base of the ovipositor in the first two families which is absent from the Raphi- diodea. In several other ways the Mantispidae suggest



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

19471 Dilarida and Berothida 159
affinities with Nallachius and Lomamyia that appear stronger than any shown by the raphidiids.' Trichosors or marginal dots and dashes of the wings, discussed by Killington (1936, p. 34) and Comstock (1918, p. 167), occur in Dilaridse, Berothidse, and some Mantispidse, as well as certain other families, but not in the Raphidiidas that I have examinel.
The swollen body and reduced
legs of mature Lomamyia larvae suggest an approach to the parasitic habit, with even more swollen body, of Mantispidae. The claws and empodium, as well as cer- tain other structures discussed by Withycombe (1925), of first-stage larvae of ~antis~idi are very similar to those of Lomamyia. I have examined larva of an un- identified mantispid found by 3. C. Bridwell in 1938 in alcohol in which spiders had been collected. In addition to the Nymphidas and Myiodactylidae, the Berothidae, Mantispidae, and Chrysopidae are known to lay stalked eggs. While raptorial front legs give the Mantispida a quite different superficial appearance from either Dilaridae or Berothida, they are regarded by Withycombe (1925, p. 329) as a specialization rather than as evidence of fundamental lackof relationship.
LITERATURE CITED
Carpenter, F. M.
1940. A revision of the Nearctic Hemerobiidse, Berothidse, Sisyridse, Polystcechotidse and Dilaridse (Neuroptera) . Proc. Amer. Acad. Arts and Sci., vol. 74, pp. 193-280, illus. 1942. Notes on Nearctic Neuroptera.
Psyche, vol. 49, pp. 49-51, illus.
1947. Taxonomic notes on the Dilaridse (Neuroptera). Psyche, vol. 54, pp. 100-109, illus.
Comstock, J. H.
1918. The wings of insects. 430 pp., illus. Ithaca, N. Y. Costa Lima, A. da
1943. Insetos do Brasil. Vol. 4, 141 pp., illus. Rio de Janeiro. Ferris, G. F.
1940. The morphology of Plega signata (Hagen) (Neuroptera : Man- tispidse) . Microentomology, vol. 5, pp. 33-56, illus. Ferris, G. F., and Pennebaker, Phyllis
1939. The morphology of Agulla adnixa (Hagen) (Neuroptera: Raphi- diidse) . , Microentomology, vol. 4, pp. 121-142, illus. Killington, F. J.
1936. A monograph of the British Neuroptera. Vol. 1, 269 pp., illus.
London.
8 Students are referred to two papers on the adult morphology of Raphi- diidsa and Mantispidse, by Ferris and Pennebaker (1939) and Ferris (1940).



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

Psyche
[ Sept.
Klingstedt, Holger
1937. Chromosome behavior and phylogeny in the Neuroptera. Nature, vol. 139, pp. 468-469, illus.
Navas, Longinos
1914. Fam. Dilaridas. Genera Insectorum, Fasc. 156, pp. 1-14, illus. Smith, Roger C.
1923. The life Histories and stages of some hemerobiids and allied species (Neuroptera). Ann. Ent. Soc. Amer., vol. 16, pp. 129-151, illus.
Snyder, Thomas E.
1920. Biology, in Banks, Nathan and Snyder, Thomas E., A revision of the Nearctic termites, with notes on biology and geographic distribution. U. S. Natl. Mus. Bul. 108, pp. 1-228, illus. Steyskal, George C.
1944. Notes on Nallachius americanus (McL.) (Dilaridse, Neuroptera) . Psyche, vol. 51, pp. 183-184.
Tillyard, R. J.
1916. Studies in Australian Neuroptera. No. iv. The families Ithonidas, Hemerobiidse, Sisyridee, Berothidse, and the new family Tri- chomatidse; with a discussion of their characters and rela- tionships, and descriptions of new and little known genera and species. Proc. Linn. Soc. N. S. Wales, vol. 41, pp. 269- 332, illus.
1917. ~donata, Planipennia, and Trichoptera from Lord Howe and Nor- folk Islands. Proc. Linn. Soc. N. S. Wales, vol. 42, pp. 529- 544, illus.
1926. The insects of Australia and New Zealand. 560 pp., illus. Sydney.
" "
Tjeder, Bo
1937. A contribution to the phylogeny of the Dilaridse and the Raphidi- idee (Neutroptera) . Opuscula Ent., Bd. 2, pp. 138-148, illus. Townsend. Lee H.
1935. Key to larvas of certain families and genera of Nearctic Neurop- tera.
Proc. Ent. Soc. Wash., vol. 37, pp. 25-30, illus. Withycombe, C. L.
1923. Notes on the biology of some British Neuroptera. Trans. Ent. Soc. London, 1922, pp. 501-594, illus.
1925. Some aspects of the biology and morphology of the Neuroptera. With special reference to the immature stages and their pos- sible phylogenetic significance. Trans. Ent. Soc. London, 1924, pp. 303-411, illus.




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

Dilaridce and Berot/~itIa
cr = cervical region
ob = base of ovipositor
ex = coxa
pe = sensory (?) peg of antenna
ds =: dorsolateral thoracic sclerites
pm = prementum
of pronotum
so = ' ' sole ' ' of tarsal claws
em = empodium
87 = seventh sternum
f e = femur
ta = tarsus
m = mentum
ti = tibia
md = mandible
tr = trochanter
mx = maxilla
(Figure 3 drawn by Arthur D. Cushman, other drawings by the author.)



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

Psyche
Fig. 2. Lomamyia sp., first-instar larva, dorsal view, right middle leg missing. Length, 2.04 mm.
Fig. 3. Nallachius arnericanus, cocoon. Length, 3.26 mm. Fig. 4. Spermophorella disseminata Till., first-instar larva, dorsal view. Length, about 2.5 mm.
(Adapted from Tillyard, Proc. Linn.
Soc. N. S. Wales, vol. 41, pi. 18, fig. 32, 1916.) Fig. 5. Lomamyia sp., mature larva, dorsal view. Beltsville, Md. Length, 9.36 mm.
Fig. 6. Nallachius americanus, mature larva, dorsal view. College Park, Md. Length, 12 mm.




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

19471 Dilaridce and Berothide 163
PSYCHE, 1947
VOL. 54, PLATE 11
2.
Lomamyia
Nall ac
Spermophorella L~marnyia Nall achius




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

Psyche
[ Sept.
'Fig.
7. Lomamyia sp., first-instar larva, apical fourth of left maxilla, dorsal view.
Fig. 8.
Same specimen as fig. 7, dorsal view of head. Fig. 9. Wesmaelius quadrifasciatus (Reut.), mature larva, dorsal view of head.
(Adapted from Killington, A Monograph of the Brit- ish Neuroptera, vol. 1, fig. 44, A, 1936.) Fig. 10.
Spermophorella disseminata, first-instar larva, dorsal view of head and neck. (Adapted from Tillyard, 1. c., fig. 33.) Fig. 11. Nallachius americanus, mature larva, dorsal view of head and neck. Same specimen as in fig. 6.
Fig. 12. Lomamyia sp., mature larva, dorsal view of head and neck. Same
specimen as fig. 5.




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

Spermophorella
IMI
19471 Dilaridce- and Berothido& 165
PSYCHE, 1947
VOL. 54, PLATE 12
Nallachius
GURNEY-DILARIDB AND BEROTHIDB




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

Psyche
[ Sept.
Pig.
Fig.
Fig.
Fig.
Same specimen as fig. 12, ventral view of head and neck, antennae omitted.
Nallachhs americanus, mature larva, ventral view of head and neck, antennas omitted.
Same specimen as fig. 6.
Same, adult female, right lateral view of apex of abdomen, only base of ovipositor shown. Brookline, Pittsburgh, Pa. Same, pupa, left lateral view. Length, 3.26 mm.



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

19471 Dilarida and BerotMdcs 167
PSYCHF, 1947
VOL. 54, PLATE 13
achius k V




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

Psyche
[ Sept.
Fig. 17. Same, apex of right front leg, ventral view. University Park, Md. Fig. 18. Lomamyia sp., apex of left middle leg, anteriolateral view. Same
specimen as fig. 5.
Fig. 19. Chrysopa sp., apex of right middle leg, ventrolateral view. Chipley, Fla.
Fig. 20.
Unidentified hemerobiid, apex of right middle leg, ventrolateral view. Mexico.
Fig. 21. Nallachius arnericanus, mature or nearly mature larva, right front leg, anterior surface, femur turned into ventrolateral view, only coxal attachments shown. Greenbelt, Md. Fig. 22. Lomamyia sp., mature larva, right front leg, anteriolateral view. Same specimen as fig. 5.




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

1947 I Dilarida and Berothidce 169
PSYCHE, 1947
VOL. 54, PLATE 14
8. Lo'mamyia
\^< ., , .,,,:. .,.. -L . - , .;a >.
2 0.
Hemerobiid
Nallachius Lomamyi a




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