24 THE REPORT OF THE [19 



abundant material we can kill a specimen or two occasionally in the cyanide jar and then 

 describe them at our leisure, but often we cannot spare any for this purpose, especially in 

 the case of species which are difficult to rear. 



Occasionally we come across a species, the larva of which is easy to describe, but as a 

 rule it is very difficult to make a satisfactory description. The markings are often 

 puzzling and the colours are frequently indescribable, it being almost impossible to give 

 them names. This chiefly arises from the more or less translucent nature of their bodies 

 and from the way one shade melts, as it were, into another. In many cases in which 

 there is not much change after the first moult, we Btill notice, as the larva grows, certain 

 things which we had not noticed in the earlier stages, and then we are haunted by a 

 horrible uncertainty as to whether these points have really developed in the later stages, 

 or were only overlooked when the creature was much smaller. 



But, doubtless, the chief difficulty in the way of the general preparation of useful 

 descriptions of preparatory stages is that probably at least nineteen men out of twenty do 

 not know what are the chief points to be observed and described, and hence we find some 

 men describing, with conscientious and laborious exactness, organs or appendages which 

 are common to at least all the larvte of the particular group to which the subject of their 

 observations belong, or on the other hand, contenting themselves with making the slightly 

 vague statement that there are " a number of warts with radiating bristles on each seg- 

 ment." 



To overcome these difficulties and to bring the work of as many observers as possible 

 into harmony, it seems to me that we require a simple but comprehensive " Manual " for 

 the study of preparatory stages which, while avoiding diffuseness and unnecessary techni- 

 calities, should still be precise and explicit, taking only the most elementary knowledge 

 for granted, and as fully illustrated as possible ; not only showing a figure of a typical 

 caterpillar, but having outline figures of all types of larvae, showing how organs and 

 appendages are modified and how they can be homologized, and with the fullest parti- 

 culars of the important points to be observed in each type of larva, and with typical 

 descriptions drawn up from common and well-known larvae as models. 



After a careful study of such a work with as many specimens as we could conveni- 

 ently lay our hands on, we would be in a position to make thoroughly useful descriptions 

 because knowing what we should look for, we would not overlook important points, but 

 would find them, if present. But I shall probably be asked who is the man for this task, 

 and in reply would say that, in my humble opinion, Dr. Harrison G. Dyar, who has made 

 such extensive studies upon many different groups of larvae, is well qualified to undertake 

 it, and it could probably be issued as an official bulletin by the Department of Agriculture 

 at Washington. 



Another work which is also much needed is a supplement to the "Bibliographical 

 Catalogue of the Described Transformations of North American Lepidoptera," prepared 

 by the late Henry Edwards, and issued as Bulletin No. 35 of the United States National 

 Museum in 1889 ; or what would be still better, a revised edition brought up to date of 

 the same work. Such a work is most important in order that people may be informed 

 as to what work has already been done, and what is lacking to fill up gaps. 



I have said that a knowledge of the preparatory stages is necessary for a satisfactory 

 classification of species, but I believe it to be also necessary in some cases even for the 

 discrimination of species. An extremely interesting case of this kind is that of the two 

 forms, or as I believe them to be, the two species of Halisidota, viz. Tessellata, A and S., 

 and Harrisii, Walsh. You are doubtless aware that as long ago as 1864 the late Benjamin 

 D. Walsh called attention to the fact that there were two kinds of larvae of Halisidota 

 producing imagos which were indistinguishable, one being the species named Phalaena 

 Tessellaris, by Abbott & Smith, but now known as Halisidota Tessellata ; and the other 

 an undescribed form or species which he named Halisidota Harrisii. These two forms of 

 larvae differ remarkably, and would never be taken for anything but species of the same 

 genus. The larva of Tessellata, as you know, varies remarkably, but only within well- 

 known limits. The body is usually black or blackish ; the head generally black, but 



