blattid.e peeserved in ambee, feom prussia. 353 



Larval Forms. 

 A large part o£ the collection is made up of immature forms and moults. 

 The latter were doubtless left adhering to the bark of the trees whence the 

 resin exuded, and became enclosed in it. All but three of the species 

 enumerated below belong- to the subfamily Phyllodromiinse, and I find it 

 impossible to allocate any of them with certainty to any of the adult forms 

 that I have described. Germar and Bereudt figure four distinct forms, which 

 they merely label A, B, 0, D. The second of these is the only one that I can 

 identify with certainty. 



? Phyllodeomia sp. (Larva B.) 



Germar and Berendt, op. cit. pi. 4. fig. 3 b. 



Nos. 5426, 5430, 5438, 5442, 5448, 5453, 5459, 5464, 5466, 5467, 5471, 

 5478, 5488, 5492, 5510, 5512, 5524, 5532, 5546, 5555, 6707, 6720, 6728, 

 6739, 6741. 



I hoped to be able to identify this very abundant species with the equally 

 abundant P. lorenz-meyeri, but the front femora are completely spined on 

 the anterior margin beneath, and therefore the larvse cannot belong to that 

 species. The thoracic tergites are bordered laterally and in some examples 

 posteriorly also with rufous. The third joint of the antennae is very lono-, 

 equal in length to about six of the succeeding joints. The body is provided 

 with a sparse erect pubescence. The supra-anal lamina is triangular, and 

 the sub-genital lamina of the older examples of the male sex is slightlv 

 asymmetrical and furnished with 2 styles. Superficially this larva bears a 

 close resemblance to larvpe of the recent species Loboj^tera nitida, Germ. 



? Phyllodromia sp. 



Folyzosteria tricuspidata, Germar & Berendt, op. cit. p. 35, pi. 4. fig. 1 (1856). 

 Nos. 5431, 5443, 5517, 6718. 



These larvse certainly do not belong to the genus Polyzosteria as now 

 defined. 



The following are very young larvae, which it is not possible to refer to any 

 of the species described above : — 



Nos. 5485, 5499, 5506, 5514, 5516, 5518, 5526, 5536, 5541, 5550, 5552, 

 5553, 5559, 6725, 6727, 6729, 6733, 6736, 6737, 6738, 6742, a 3. 



All of the above belong to the section Blattse armatse, but the following are 

 Blattge muticse : — 



No. 5435. Apparently identical with Polyzosteria parvula, Berendt- (Ann. 

 Soc. Ent. France, vol. v. p. 542, pi. 16. fig. 1 (1836) ; Germar & Berendt, op. 

 cit. p. 35, pi. iv. fig. 2 (1856)). The species, of course, is not a Polyzosteria. 



No. 6716. A larval moult. The supra-anal lamina is triangular, the sub- 



28-^ 



