PEIMARY AND SECONDARY SEXUAL CHAKACTERS IN LEPIDOPTERA. 201 



between two correlated parts, he would substitute a "food-bond," or, 

 in other v.'ords, a chemical one, and he adds: "To revert to the 

 illustration already borrowed from the stag, if the influence be purely 

 nervous, as is commonly believed, the path by which it is conveyed all 

 the way from the testes to the horns over the intricate and interlacing 

 lines of the sympathetic system is hard to conceive, but substitute a 

 food-bond, and the connection is at once obvious and easy of com- 

 prehension." 



After showing that in plants structure largely depends upon food, 

 and that the profound modification of the plant protoplasm in the 

 formation of galls, is possibly due to the fact that the peculiar 

 substance injected into the tissues of the plant plays the part of food, 

 " not however in the ordinary sense of nourishing, but rather in that 

 of combining and uniting with the protoplasm or one of its con- 

 stituents, after the manner of a chemical agent, and so altering its 

 molecular constitution and affinities as to change it into something 

 altogether different from what it Avas before." He then concludes : 

 " of such a kind is the nature of the ' food-bond,' which I would 

 suggest may underlie some of the phenomena of correlation, that is, 

 certain substances are secreted by the controlling organ, Avhich 

 combine chemically with the protoplasm of the correlated part and 

 endow it with new capabilities." 



If we are unable to accept this view, it is not because we do not 

 fully appreciate the exceedingly clever suggestion made by Dr. Wood 

 to explain a difficult subject, but because Ave are unable to apply his 

 reasoning to definite cases that have come under our notice. 



Talve for example, first, the case of those bilaterally gynandro- 

 morphous examples of lepidoptera — Trichinra crataef/i, Lasiocampa 

 qnercm, Malacosoma castrensis, M. ncnstna, &c. — that are knoAvn to all 

 entomologists, and scA^eral of which we have described at length in our 

 recently-published second volume of The Natural History of the British 

 Lepidoptera. In these the external portions of the sexual organs are 

 on one side male, on the other female, presumably the internal 

 portions of the organs are so also, and as a result the side AA'hich 

 possesses the seminal glands presents all the secondary sexual charac- 

 ters of the male — antenna, legs, shape of wings, markings, colour 

 (both of head, Avings, and thorax)— whilst the side AAdiich possesses 

 the ovaries presents all the secondary sexual characters that dis- 

 tinguish the female. So complete is the division that a central line 

 diAades the insects into tAvo distinct halves, male on one side female 

 on the other. I cannot understand a food-bond that is not diffused 

 throughout the Avhole system. Ave Ave to understand that the internal 

 secretions from the ovaries of an indiAddual such as I have described 

 can be absorbed only by one side of the insect, and similarly that the 

 internal secretions of the seminal glands can only be absorbed by the 

 opposite side ? 



A second class of gynandromorphous individuals differs Avidelyfrom 

 the first. These appear to have the OA'aries developed in part on either 

 side of the insect and similarly the seminal glands. Such individuals 

 Avill present the general characters of a male or of a female, according 

 as the male or female genital organs are best de\'eloped, with sundry 

 areas and structures on either side of the insect showing characters of 

 the opposite sex to that Avhich is generally exhibited by the insect. 



