TRANSPARENCY IN WIN(rS OF LEPIDOPTERA. 115 



any, to avoid getting them laid upon cotton wool as they would the 

 plague. 



It is unnecessary for me to go into any detail as to methods of 

 developing the negative after it has been taken, as these will vary with 

 the make of plate and the particular developer used, and are easily 

 obtained from the manufacturers ; but it is helpful to keep a full note 

 with each exposure of at least the name of species, magnification, 

 lighting, make of plate, exposure given, developer used, and colour 

 and appearance of the ovum. I write these particulars upon the out- 

 side of a paper envelope, into which I afterwards slip the finished 

 negative, before storing it away, and I ahrai/s pencil the name of the 

 species upon one corner of the plate before development, so that if it 

 gets separated from its proper envelope at any future time it can be 

 identified with certainty. The envelopes alluded to above should each 

 bear a consecutive number, and if an alphabetical index to these is 

 compiled as they are made, any particular species required in future 

 can be found without loss of time. 



The exact amount of magnification is important, and once a con- 

 venient size has been fixed on, that, or multiples of it, should be 

 strictly adhered to, as otherwise any comparison will be difficult. An 

 easy method of ascertaining the magnification is to photograph a 

 finely ruled scale in place of the ova, and then compare your negative 

 with the oriainal scale. 



Transparency in wings of Lepidoptera {irit/i plate). 



By WILLIAM J. KAYE, F.E.S. 

 (Concluded from p. 86.) 

 As it is ejuite impossible to do justice to any but a very small 

 portion of the sulDJect, I have confined myself almost wholly to diag- 

 nosing, in detail, two distinct groups of transparent butterflies and 

 moths from British Gviiana, all the individuals of each having occurred 

 on a single forest track. The first group consists of two Syntomid 

 moths, Aiiijrta iiiirilia and Eiiaj/ra coelestina, one Geometrid moth, 

 Pseud arhessa dccDrata, one Hypsid moth, lostola dirisa, and an Erycinid 

 butterfly, J^sthetnopsis sen'cina. In examining such a mimetic group 

 as this the results are doubly interesting, (1) from the variety of 

 methods adopted to attain one end, (2) the almost conclusive proof that 

 these different species are mimics, from the very fact that the methods 

 of obtaining transparency are so different, whereas an universal method 

 would indicate that a common cause was at work, and would be merely 

 accidental or fortuitous if species of dift'erent genera looked alike- 

 Most unquestionably, the Syntomid, A(iyrta micilia, is now the most 

 abundant, and one assumes that it has been the model ; but why this 

 species should have developed transparency is difficult to say. If one 

 judged wholly by the scales modified into hairs, one might argue that 

 KucKjra nidcstiiia, in being more complete in this respect, had, so to 

 speak, set the pattern. In either case, however, both species are 

 brilliant on the wing, and are not specially rapid in flight. When 

 one considers the extreme complexity of a Miillerian association, it 

 becomes exceedingly difficult to decide which species originated, or 

 originally possessed, the type pattern. One recognises that a species once 

 brought into the group, may perhaps develop certain tendencies more 



