St R. .J. TILLtARll. 295 



series (Text-fig.5), we see that a distal forking of one of the 

 sectors may become interposed between two of the gradate \eins, 

 and thus allow of the occasional introduction of an extra cross- 

 vein. I propose, therefore, to lay down two rules for numbering 

 these cross- veins: — (a) only to count cross-veins from the most 

 anterior branch of Cu, upwards; (2) to give, as the normal num- 

 bei", the number counted in specimens in which no extra cross- vein 

 is interposed in the manner shown in Text-fig.5, c. 



Secondly, as regards the number of i-adial sectors in the fore- 

 wing. In all specimens of Drepanacra which I have examined, 

 the most distal branch of R, viz., Rj, gives off one or more 

 posterior branches, while the most basal branch gives off one or 

 more anterior branches. The branching of the former is nearly 

 constant, there being only fwo posterior branches (small forkings 

 distad from the outer gradate series are not taken into account). 

 But, for the most basal branch, we find two conditions almost 

 equally prevalent. Either this bi-anch gives off only a single, 

 anterior branch, and is followed distally along R by a fixed 

 number of simple sectors; or else it gives off two anterior branches, 

 and is followed by one less than this fixed number of simple 

 sectoi's. The explanation of this is, that what is really the 

 second sector from the base frequently be- 

 comes detached from R, and fuses on to the 

 most basal sector, giving it an extra branch 

 anteriorly. This is shown in Text-fig. 6. A 

 single specimen sometimes has the condi- 

 tion a on one side, and the condition b iext-ng. . 

 on the other. To deal with this Aariability, I propose the follow- 

 ing plan. An imaginary line, drawn approximately parallel to 

 and just inside the outer gradate series, will cut all the radial 

 sectors and their main branches. If we denote a single sector 

 by the figure 1, a sector with one branch by 2, and so on, and 

 reckon from the distal (anterior) end of our imaginaiy line 



* Variation in condition of branches of radius in forewing of Drepanep- 

 ttryx : a, archaic condition; h, the second bianch from the base becomes 

 hitched on to the most basal branch. 



