38 THE entomologist's record. 



tions. It results from this that, while ostensibly a certain portion of 

 ourlarvje are "known", a part of these are insufficiently described, 

 opening again a field for the student of the twentieth century. 



An important advance which has been made is the growing 

 practice of describing separately each stage of larval life. This often 

 records interesting and important characters which appear and 

 disappear during ontogeny, and it is of course the only proper way to 

 describe a larva. 



Descriptions of larvfe should serve other purposes than that of 

 mere recognition. We need an accurate and detailed record of these 

 organisms for the use of students of the general subject and for future 

 comparisons. Much advance has been made in this direction, the 

 increased detail of descriptions, but unfortunately it has not become 

 the general rule. Under this view it is often important to record 

 many characters which are of no value from the point of view of 

 recognition, as they are common to large groups. It is, however, 

 sufficient if a standard of reference has been established, to do this 

 briefly by referring to the character in question as " normal" to the 

 group. A most commendable advance is the custom of giving enlarged 

 figures of the larva or of single segments ; yet this needs more care 

 than is usually given, as most of the published figures, especially of 

 newly-hatched larvae, are somewhat unreliable in detail. I refer 

 more particularly to the exact number and position of the hairs. 



As to the value in classification of the various larval characters 

 that have so far been brought out, it cannot be doubted that some of 

 them are considerable. The head and its appendages are so similar 

 throughout the order that but little of general value has been drawn 

 from it. The number of the body somites is practically invariable, 

 only a slight difference in degree of consolidation of the last two being 

 apparent. Yet where this is manifest, it indicates the relative degree 

 of specialisation as shown by other characters. The shape of 

 individual somites is occasionally subject to modification, yet these 

 specialisations are sporadic and do not indicate any broad lines of 

 phyletic differentiation. The number and modification of the larval 

 hairs or sette, though often shoAving striking characters, are likewise 

 of an adaptive nature and are not conclusive. Still they yield certain 

 indications in line with the general scheme. 



The number of the abdominal prolegs is not a character of wide 

 importance. It appears that the primitive lepidopteron must have 

 acquired the normal proleg formula. Perhaps this was established 

 even before the true lepidoptera arose. At any rate, the character is 

 practically fixed throughout the order, except in the Tincidae where 

 it is subject to a certain flexibility. Above this group, there is no 

 modification, save a partial disappearance of the prolegs, and that is 

 not even generally of family value, though occasionally it seems to be 

 so. The Mi'iialopiiiiidac are superficially defined by the presence of 

 extra abdominal prolegs, yet I have shown that these are secondary, 

 are superposed upon the normal pedal structures, and, therefore, a 

 character of specialisation, not indicating a low position for this group, 

 but the contrary. 



The arrangement of the adhesive hooks or crochets of the 

 abdominal prolegs shows a good character, as has long been known. 

 Kecently Karsch (^Ent. XacJiric/iten) has carried to an extreme a 



