OF WASHINGTON. Ill 



the parasitic families, is the longest, and that of the larva the 

 shortest. 



In the Cynipidse or Gall-flies we have again every variation, 

 though the larval life is usually long compared with the life 

 of the adult, which, on the average, is brief . This may be said 

 to be the rule with species producing summer and autumn 

 galls on the firmer textures, whereas the opposite is true of 

 spring forms on the more succulent parts of plants, the same 

 species, in alternate generations, often representimg both condi 

 tions. I have known of many remarkable instances of the pro 

 longation of the life of the individual in hard woody galls under 

 unnatural conditions of dryness, in a number of cases the larva 

 being carried over two or three years and then ultimately trans 

 forming, while in other cases the adult retained life for two 

 years, unable to issue from its bonds. In one case, that of an 

 undescribed species of Callirhytis ( C. fruticola Riley MS.), 

 where the gall, occurring in the acorn, is as hard as a stone, the 

 larva retained life for a period of six years ; yet in all these cases, 

 under normal circumstances with the softening and rotting of the 

 woody tissue on the moist ground, the individual life would 

 hardly have been extended beyond the year. 



In the parasitic families it is difficult to say whether more 

 species hibernate as adults or as larva? or as pupa? ; but in what 

 ever state the winter is passed, the other states are all of relatively 

 short duration, and we have here again, frequently in the same 

 genus, species which produce one and others which produce more 

 than one annual generation, though it is very generally true that 

 the longevity of the parasite is dependent on the character of its 

 host. 



In the Terebrantine Hymenoptera we have again a great 

 majority of the forms limited in life duration by the recurring 

 year. In the Uroceridae for the most part wood-borers one 

 annual generation is produced, and the winter is passed in the 

 larva state, the life duration of the adult not exceeding three 

 months. In all cases of this kind the larva lives longer than the 

 adult. In the Tenthredinidas we find great variation, the individual 

 life varying according as there are one or more generations annually. 

 Yet in no case, so far as known, does any species normally ex 

 ceed a year in individual life duration, while the great majority 

 have an adult existence of but a few weeks, with a larval exist 

 ence of a much longer period. Most of the species hibernate in 

 the cocoon, either above or below ground, and in the ultimate 

 larva state, transforming to the true pupa in early summer, only 

 a few weeks prior to the issuing of the adult. 



COLEOPTERA. In this Order, viewed as a whole, the larval 



