60 THE ENTOMOLOGIST S RECORD. 



tibial spur as an organ varying a good deal in different species, and so 

 affording a specific character, Avhilst in the family of PsijcJddae it 

 affords a subfamily character. 



The Fumeid species may be divided into two groups : 



1. With short spurs (under -64 of tibia); median accessory cell; reticulate 

 wing markings (usually). 



Of these I have verified reticnlatella and comiteUa, and presume 

 from descriptions that raiblensis, rouasti, and norvc(jica belong to this 

 group. 



2. With spurs over -64 in length ; with median nervure simple ; without any 

 suggestion of wing reticulations. 



a. Species with spurs of length •65--75 : (1) Cmssiorella. (2) Affini'f. (3) ^Ht- 



fordella, n. sp. (4) Hlbernicella, n. sp. (-5) Suhjiavella. (6) Edirardsclla. 



/3. Species with spurs of length -TY-'Sl : (7) Casta with vars. nitidella, inter- 



mediella, boiverclla. (8) Scotica. 

 y. Spurs -85 : (9) Germanica, n. sp. 



All these Fumeas clothe their larval cases with straw-like material 

 placed lengthwise on the cases, and it is supposed that the nature and 

 size of this material affords specific characters of value. In the first 

 place, I think the material is more rarely straws of grass stems than is 

 generally believed, and many of the cases that look most white and 

 straw-like and are at once taken to be covered with grass stems are 

 really clothed with dead stalks and pedicels of various flowering plants. 

 Some cases that apparently belong to the same species as straw-covered 

 ones are encased in leaves of fir, Avhilst the size of the case and of the 

 materials covering it depend much more on the sex of the specimen 

 and on the materials available than on anything else. 



I doubt very much whether the larvfe will present any characters 

 of use to separate the most allied species ; even the most separated 

 species have larvfe so alike that it is difficult, if not impossible, to 

 define them by absolute description, and for the present, at any rate, I 

 have no material adequate to make any such attempt. 



The want of fully correlated material is perhaps to some extent the 

 reason I have to make much the same confession as to specific 

 characters in the $ s. The only definite distinction I know of is 

 between 2Iasonia crassiordla ? and Fumca capita ? , the former having 

 reduced tarsal joints, the latter the full five to each tarsus. 



The neuration gives a very definite division between the short- 

 spurred and long-spurred sections of the genus, the former possessing 

 the median accessory cell, in the latter the median nervure is simple. 

 This distinction is at least valid in the species I have examined. 



I hoped to find some useful characters in the antennae and have 

 not been altogether disappointed, but have been obliged to conclude 

 that there is considenable variation in the antenna3 within the limits 

 of a species. Whether 21. c rassiorell a v:ith 21 antennal joints is or is not 

 the same species as one with 24, may seem to be properly decided in 

 the latter sense, but when we find 21. cmsaiordla having 22 and 23 joints 

 to the antennfe amongst specimens that it is impossible to doubt are 

 all the same species, it is difficult on this ground to distinguish those 

 Avith 21 or 24 as distinct. The same considerations apply strongly to 

 the forms of F. casta. That the antennae are variable within the species 

 is confirmed by an examination of the antennal structure. The 

 antenna consists of a large basal joint, the second joint is also large 

 and globular, normally the third is the first of the clavola or flagellum 



