124 THE entomologist's record. 



however, defined as expanding 15mm., and doubtless it also as well as 

 B. reticnlatella has the " cellula intrusa," which 2L. edwardsella is 

 without. Its coloration and darkening towards the hind margin dif- 

 ferentiate it from M. mitfordella and associate it with M. subflavella, 

 from which it differs in size and number of antennal joints {vide, pi. 

 iv., fig. 37, anterior tibia). 



The great ^^ casta group," with a spur length of -TT-'Sl, is, perhaps, 

 the most puzzling and difficult of all. I incline to consider them all 

 as one species with a number of local races and varieties. This cer- 

 tainly gets over the great difficulty of saying what many intermediate 

 forms should be called. On the other hand, it is probable from the 

 specimens I have examined that each colony of the species is fairly 

 well-defined, and does not present, except in rare aberrations, specimens 

 agreeing with other varieties, and it is evident that if this is so it must 

 be largely a matter of individual predilection rather than anything else 

 whether each such race is to be called a species or not. The only 

 ground for doubting my conclusion to call them all one species is 

 that there is also some ground for believing that the nitidella 

 and inter mediella forms do occur together in some localities, and yet 

 maintain themselves as separate races. This certainly requires 

 further investigation. I may say that I have put names to these forms 

 without sufficient certainty that I am right as to the use of the names 

 nitidella and intermediella. If I am wrong this can be put right, but I 

 have assumed nitidella to be the small form, which Bruand thought he 

 was dealing with under the name roboricolella, but with which he 

 unfortunately mixed some Proutia betulina, and so gave an account 

 which is incorrect whichever species we apply it to. 



Fumea casta. — The name casta appears to cover the great mass of 

 the Fumeas in British collections, called often, with much apparent 

 capriciousness, roboricolella, nitidella, inte)')iiediella, and even crassiorella 

 and crassicolella. The definite character that unites all the forms is a 

 length of tibial spur of from -77 to -81. The antennal joints differ from 

 16 to 18 and 20, and the expanse from 9mm. to 15mm. There is a good 

 deal of variation in wing form, generally there is a good breadth basally 

 by the inner margin commencing at the base with a deep rounded lappet, 

 and the inner and costal margin making some approach to parallelism. 

 But not a few show a considerable approach to the form of M. mit- 

 fordella. It is very possible that by measuring the spurs of a number 

 of such specimens some would prove to be really M. mitfordella. The 

 several forms difl'er chiefly in size, but also in the number of antennal 

 joints : 



a. ab. minor. — Exp. al. 9mm. -10mm., antennal joints 18, wings often more 

 diaphanous than type. I have called this an aberration rather than a variety, as it 

 occurs in odd specimens in different collections, and is usually probably a starveling 

 form rather than a distinct race. 



(3. var. nitidella. — Exp. al. llmm.-12mm., antennal joints 18. This is a very 

 definite race, and appears to be the most common and widespread form. I am by 

 no means prepared to assert that this form is not a true species and distinct from 

 the next form, only, if so, I cannot divide them with even approximate confidence 

 (anterior tibia, pi. iv., figs. 46-47). 



y. var. intermediella. — Exp. al. 13mm. -14mm., antennal joints 18-20. This is 

 rarer than the last species, and is, in places, apparently a distinct race, in others 

 merely an aberration of nitidella. Some of the largest specimens have only 18 

 antennal joints and some of the smallest 20, so that I feel unable to divide the 

 forms nitidella and intermediella into two distinct species, defined as nitidella, 18 



