lepidopterous lame and piqHS. 159 



of defence, and that the latter have saved the larva from 

 extermination. A relation grows up between a larva 

 and its parasites similar to that between plants and 

 insects (well exemplitied by Mr. A. B. Wallace in his 

 'Essays on Natural Selection,' 1875, p. 271, &c.), 

 although the benefits are rather one-sided in the former 

 case. Any improvement in the means of larval defence 

 may be met by greater ingenuity or boldness in the 

 attack, and so it comes about that many of the best 

 protected larvae are often those which die in the largest 

 numbers from the attacks of insect-parasites. The 

 exceptional standard of defence was only attained by an 

 exceptional need. Under such circumstances a very 

 sudden advance in the means of defence miglit cause 

 the entire extermination of the parasitic organism, and 

 if this were achieved it is likely that the special defensive 

 measures would be slowly relaxed. It is thus possible 

 to imagine the conditions under which a larva would 

 preserve a functionless rudiment of a formerly efficient 

 flagellate organ, such as the caudal horn of Sphingidce 

 has been shown with a great degree of probability to be 

 by Professor Meldola (Appendix to his translation of 

 Weismann's Essay, "On the markings of Caterpillars," 

 p. 527), and myself (Trans. Ent. Soc. Lond., 1885, 

 Part II., August, p. 302, etscq.). But the withdrawal of 

 the means of defence at once leaves the way open for 

 renewed attacks from other parasites, and may render 

 the larva very helpless. In fact the comparative im- 

 munity of the large, helpless, and often imperfectly 

 concealed larva of Sphingidce is often a source of wonder 

 to me, and it would seem that the only explanation can 

 be that the peculiar parasitic enemies are not flourishing 

 and abundant, being themselves overweighted in the 

 struggle with organisms which prey upon them. It 

 would be extremely interesting to compare the pro- 

 portions which succumb to the attacks of insect- 

 parasites out of large numbers of many species of larvae 

 protected in various ways. 



9. An eveesible "gland" in the larva of Org\'ia 

 PUDIBUNDA. — I do not kuow whether this structure has 

 been previously noticed ; it is not mentioned in the well- 

 known text-books. The single gland to which I refer is 

 situated in the median dorsal line of the seventh abdo- 



