( 565 ) 



as well as on the elytra. Applying this discovery to the immature and adult 

 specimens from China (ligs. 1 and 3 of PI. XII.), no dorsal combs can be expected 

 to put in an ajipearance in this species until the imago is being formed in 

 the last nymphal instar. 



Since in the insects with incomplete metamorphosis the elytra are in a state 

 of growth in the immature instars, while the head and thorax are already more 

 or less shaped as in the imago, one might deduce from this generally applicable 

 fact the a priori opinion that in the Polyctenids the comb of the elytra invariably 

 appears at a later instar than the comb of the pronotum. This is not the case. 

 Tiie s(iecimen fignred at PI. XIII. fig. 8, which is immature, has combs on the head 

 and elytra, lint not on the prothorax, and is, we think, a nymph belonging to fig. 5, 

 the specimens being obtained together. The specimen was unfortunately glued 

 on a card, and the new individual inside the old skin is not advanced enough to 

 show any bristles and spines. That is much to be regretted, since the appear- 

 ance of a pronotal comb on the specimen forming in the nymph wonld be the 

 best evidence that onr uymph fig. 8 actually belongs to the species fig. 5. It is 

 interesting to note that Dr. Speiser* draws the conclnsion that the elytral comb 

 ajipears before the thoracical one in the individual life, which does not seem 

 warranted by his specimen of a nymph, but is borne out by our fig. 8. It is 

 perhaps necessary to mention that this conclusion also does not apply to those 

 species of wliich the imago has a comb on the thorax but not on the elytra. 



We have further evidence of an irrefutable kind that the j'oung differ from the 

 •adult in the absence of certain combs. The proof that this is so was accorded to us 

 by the discovery of a young Polyctenid in the abdomen of its mother. The embryo 

 is well advanced, the combs, bristles, and claws being already strongly chitinised 

 (brown). While the mother (PI. XIII. fig. 5) has three dorsal combs, the offspring 

 jiossesses only one (on the head). Moreover, the first antenual segment bears only 

 one broad obtuse spine in place of the ro.v of spines found on this segment in the 

 mother. If fig. 8 really is the pupa of this same species, three stages are known : 

 the embryo, with one dorsal comb (ou the head) ; the pupa, with two (on the 

 head and elytra) ; and the imago, with three (on the head, pronotum, and 

 elytra). A less advanced embryo is contained in one of our females of another 

 species {talpa).^ 



Classification. 



Systematics are primarily based on a knowledge of the characters wliich 

 separate one species from the other. A classification cannot be built up merely 

 on the distinctions exhibited by the various specimens. We have to discriminate 

 between these difterences, as otherwise individuals which are different because 

 some are immature, for instance, and others adult, or because they are of opposite 

 sexes, might be placed widely apart from each other in the scheme of classification. 

 In our case a classification based, e.g., on the presence or absence of combs 

 would be entirely erroneous. An investigation into the taxonomic meaning of 

 the dirt'erences, therefore, is essential for the correctness of a classification as 

 opposed to a mere arbitrary grouping of the individuals. The difi'erences observed 

 may be those between 



(1) adult and immature individuals of the same species, 



(2) the sexes, 



♦ Rec. Ind. Mm. iii. p. 271 (11109). 



t These young I'olj-ctenKls will be described and figured in anutlier place. 



