1264 
COSSIDAE. General Topics. By Dr. A. Seitz. — Cossidae. By H. G Dyar 
outside. The number of eggs produced by some $$ of the Cossidae is therefore amazingly great, and if in certain 
years favourable conditions of life are presented to the Cossidae owing to the non-appearance of certain enemies 
or to climatic constellations, the number of certain species may sometimes be so tremendous that they almost 
resemble a swarm of locusts. 
The American Cossidae exhibit numerous schemes of colouring that are well known from other faunae. 
The Zeuzera pyrina which we already mentioned to have been imported, has not yet changed in America, while 
in North Africa it invariably shows a different thoracal marking compared with the European species. In America, 
especially in the nearctic region, we also find the approximate exterior of the palaearctic Cossus cossus in quite 
a number of species; some of them have even been described as Cossus, as for instance Acossus ore Strecker; 
A. populi from the Hudson Bay and undosus from the Rocky Mountains are extremely similar to North Asia¬ 
tic forms of cossus, except that the shade of the grey groundcolour has slightly changed according to the bark 
of the food-trees. Miacore difjidens Dyar from Mexico exhibits a colouring and marking very similar to that 
of Cossus terebra, and also other species occur in America, which show a great resemblance to the smaller 
Egyptian Oos^s-species and some Central Asiatic species, such as the recently discovered C. sheljuzhkoi Zuk. 
from the Thian-shan. 
The marking of the Cossid wing being adapted to rimous bark as well as the imitation of broken off 
little twigs represents an assimilation of nearctic Cossidae to palaearctic Heterocera, which is merely due to 
convergency and therefore accidental. The image of the resting Phalera bucephala most strikingly copying a 
piece of twig by the way it holds its wings and by showing the broken part, reoccurs in American Cossidae, as 
in Cossula magna Schs. from Guiana and the likewise similarly marked Cossula gaudeator Schs. from Costa Rica. 
As we mentioned above, it is extremely risky to conclude affinities from the colouring, marking and even the 
exterior habitus, yet it seems that in classifying some very similar species in distantly separated genera, the 
characteristic structure has sometimes been considered too exclusively, as we must not ignore an extraordinary 
polyformism, for instance of the antennae in the Cossidae. It was chiefly the shape of the antennae, which started 
the idea of dividing the Cossidae into two entirely separated families — the Cossidae and Zeuzeridae. In nearly 
all the other Heterocera families it is advisable to concede the shape of the antennae an essential influence 
upon the construction of the system, but in single cases the structure of the antennae is most probably caused 
by certain conditions of life, which being sometimes due to the landscape change the function of certain organs 
and thereby also their structure; we merely remind our readers of the differences of the antennae in certain spe¬ 
cies from other groups, which are otherwise extremely similar and evidently very closely allied, as for instance 
Brephos parthenias L. and notha Film. 
Family: Cossidae. 
By H. G. Dyar f. (Revised by Dr. W. Schaus.) 
Already in 1894 I divided, together with B. Neumoegen, the Cossidae for the first time into two sub¬ 
ordinate families according to the presence or absence of a cross-vein between the veins 7 and 8 of the hinclwing. 
This classification was kept up by Schaus for the genera of the South American Cossidae, considering at the 
same time the formation of the antennae. In 1911, Barnes A Me Dunnough supplied an excellent key to the 
North American forms of this family, in which the aforementioned characters were ascertained to be subject 
to a certain variability; they therefore proposed to give up this division. Instead of it they proposed as a basis 
for a better classification to use as distinguishing characters the variations of the anal veins of the forewing, 
according to these veins being united near the margin or extending separately. 
The collection of the National Museum at Washington contains now the collections of both Schaus and 
Dognin, being the main describers of the American species of this family, beside Herbert Druce, so that 
most of all the species known are present there. In trying to range these species among the corresponding genera 
I came across a species (Lentagena Ophelia Schs.) which renders also the division of Barnes A Me Dunnough 
impossible. But as this species corresponds well with Carohamilia Dyar in all the other characters, it shows 
that this cannot be used for diagnoses either. 
Consequently, I have been looking for another principle of classification for the American species and 
have likewise used for it the neuration of the forewing. Vein 11 arises from the areole in Xyleutes, Cossula etc., 
from the centre of the discal cell in the others, for which reason I choose this character as a division. Hampson 
proposed to divide the Cossidae into two families, the Hypoptidae without a frenulum and the Cossidae with a 
frenulum, but I found this character not to be suitable for the classification of the genera. 
Judging from the material at hand, no species of the family of Lepidarbelidae occur in America. This 
family — formerly named Arbelidae (according to Arbela Air., a homonym of Arbela Stal) — is closely allied 
with the Cossidae. The latest catalogue by Dalla Torre and Strand enumerates 9 American “Lepidarbelidae”. 
Only 7 of these are known as genuine Cossidae, quoted here as Cossula and Givira. The two others will probably 
also belong to this place. 
