Miscellaneous. .'573 



Bctwoou the rodleta in their course towards the arthrodial mumbrane 

 lie hero and there small bodies, which contain cliromatin ; these are 

 probably nuclei, but whether they belong to supporting cells or to 

 other cells of a nervous nature is a (juostion which for the present 

 may bo left undecided. On the outside of the antenna there is 

 neither a sensory seta nor any other appendage corresponding to the 

 ends of the rodlets, and the pores appear to be closed at their outer 

 ends. In no other joint of the antenna of the wasp have I been 

 able to find a trace of this structure or of one resembling it. T at 

 first thought tliat perhaps some relation might exist between this 

 organ and the rod-bearing sense-organs, or scolopophores, which 

 (according to the statements of Graber, von Leydig, and others) 

 occur in the antenna). A closer investigation, however, proved this 

 supposition to be erroneous. 



In the genera Melohmtha (Coleoptera), Epinephele (Lepidoptera), 

 Bomhus (Hymenoptera), Pachyrhina, Tabanus, Si/r^Jms, Jielojph litis, 

 Musca, Sarcophaga (Uiptcra), Sicdis, Panorpa, Phrycfanea (Neuro- 

 ptera), and Libellula (Pseudoneuroptera) the organ exhibits on the 

 whole the same structure as in the case of the wasp. 



Among the Hemiptera I have so fiir only examined the Homoptera. 

 Here the organ is but slightly developed. The ganglion-cells and 

 rodlets are present in small numbers, and are also situated at the 

 end of the second joint, which is not materially dificrent from the 

 other joints. 



Among the Orthoptera the genera PeHplaneta, Locusta, and 

 Stenobothrics have in the second joint of the antenna a structure 

 which is provided with ganglion-cells and long fibre-like rodlets. 

 I have hitherto had no opportunity of examining the Thysanura. 

 The organ undergoes a further development in the Culicidae and 

 Chironomidae(Diptera), in which, as is well known, the antennte are 

 different in the two sexes. At the base of the antenna is an almost 

 spherical joint, which is larger in the male than in the female. 

 Within this joint is a very complicated organ of a nervous nature, 

 in the structure of which, especially in the case of the female, an 

 unmistakable similarity to the organ above described is recognizable. 

 In the female the rodlets, instead of entering into pores, terminate 

 towards a plate of chitin, upon the centre of which stands the long 

 antenna. The very large anteunary nerve runs chiefly into the 

 ganglion-cells of the organ, but gives off two relatively small cords, 

 which run through the centre of the large joint into the other seg- 

 ments. The ganglion-cells pass without a sharp division into those 

 of the brain. The rodlets are delicate and provided with small 

 nuclei, which are rich in chromatin. In the male the structure 

 is very complicated, and, moreover, different in the Culicidae and 

 Chironomidae. It does not easUy lend itself to description in a few 

 words and without figures, but I would here remark that this form 

 is also readily traceable to the general type. 



The organogeny I have so far studied only in the Culicidse and 

 Chironomidae. For the general development of the antennae in these 

 insects I would refer the reader to the papers of Weismanu and 

 Hurst, already quoted. I will merely add that the entire organ is 

 formed from a fold at the base of the invaginated sac of hypodermis, 



