AXTENX.E OF J.EiM JJOPTKKA. 9 



compact ring suiTouiidiiig tlie |X)int of attachnient and rising up 

 around the l)ase of tlie In'i.-^tle. They vary greatly in size. Tliey are 

 usually shorter than the third type of sense-hair, but in some cases 

 they are longer. In Xotolojjhiis leucodujma, Fig. :-50, they measure 

 350 m. It is extremely difficult to obtain good sections of the soft 

 parts of these sense-organs, because of the toughness of the chitinous 

 parts, and the ease with which the cellular elements tear away from 

 them in the cutting. As shown in Fig. 14, we have here more than 

 one cell in connection with the sense-hair. I liave been unable to 

 determine whether all three are nerve-cells and receive branches from 

 the nerve-trunk. That at least one is supplied with a nerve is cer- 

 tain. At a later date, with fresh material, I hope to devote further 

 study to the histology of this form. Whether in all these cases the 

 cell is really a nerve-cell and has true nerve processes, is a matter 

 of <l(>uljt. AVhether the nerve is really a part of the cell, or sin)ply 

 intimately connected with it by the apposition of a brush end has 

 not been determined, and could j)erhaps be oatisfactorily demonstrated 

 only by the application of the ( Jolgi method. AVhichever condition 

 may be the true one, however, it is doubtless constant in all forms, 

 and so cannot affect the problem under consideration in this paper. 



The distribution of the sense-hair of the second ty})e is as wide as 

 that of the first type, but the number is limited to a very few on a 

 single segment. 



8. Long, rather stout and stiff hairs which, like those first de- 

 scribed, are hollow and closed at the free ends ; Fig. 23, 3, shows 

 these in relation to the rest of the antenna, and Fig. 12 represents a 

 section of one from a male Callosamia promethea, magnified about 

 500 diameters. It will be seen from the figure that the structure 

 differs but little from the first type. The nerve-cell, however, is 

 relatively smaller, and the hair is more firmly attached. The ex- 

 ternal portion of the hair usually measures from 80 to 100 )n. I 

 have found none less than 75 m., while some attain a length of 

 350 m. In distribution they are more limited than the other types. 

 They do not occur in the Jugatse, nor in the Hesperiina and Papili- 

 onina. 



4. Pits, or depressions, guarded at the opening l)y stiff i)rojections 

 of chitin, ami containing single rods or cones, which are connected 

 with nerves from the interior; Fig. 15 represents a section of one 

 from a male Callommi a promethea. In this species the pit is a de- 

 pression in the chitin from 8 to 10 m. deep, and of almut the same 



TEANS. AM. ENT. SOC. XX III. (2) JANUAKY, 1896. 



