CHARACTERS EXAMINED. 6l 



Structural characters : 



(a) The form-index, which is determined by multiplying the greatest 

 anterior-posterior length by the greatest transverse diameter and the division 

 of this result by the greatest dorso- ventral diameter of the thorax. 



(d) Size ; determined by measuring the greatest anterior-posterior length. 



(c) Punctations on the pronotum and elytra. 



(d) Glandular openings on the elytra and pronotum. 



The nature of my material has made measurements almost impossible, espe- 

 cially in the color characters, hence I have in the main seriated it into classes 

 by inspection, and this method has on actual trial given exactly the same seri- 

 ations as were obtained by measurements. It seemed, therefore, inadvisable 



^ ^^ das. 



bas-prox Wz^^' ^I^SS^^k ^' ^^^' 



cu-\-an -■■^■if^BHB ' jy^ox. 



med. I 



prox-m.'-m, — I 



ni-dist,^ . 



dist-mar. .. * 



Fig. 3. — L. lacerata. To show areas of cuticala color in banded species ; 

 has., basal band ; proxr,, proximal band ; vi., middle band ; dist., distal 

 band ; mar., marginal band. Other abbreviations as in text-fig. i. 



to employ the complicated methods of statistical investigation elaborated by 

 Pearson. I have, moreover, proven to my own satisfaction by actual test 

 that the highly complicated methods of biometry are no more reliable in their 

 results as far as this material is concerned than other methods far less cumber- 

 some and slavish. It has been sufl&cient to determine the range of variation, 

 the empirical mode and the per cent of variates therein, and the mean. 



The color characters have been treated as though the material were all of 

 one sex. This I am warranted in doing, owing to the fact that there is no 

 dimorphism, nor even a distinguishable difference in the color pattern of the 

 two sexes. That no error was introduced by ignoring sex is shown by the 

 production in every case of clearly defined monomodal curves. As seriation 

 of the color characters of the two sexes gives identical results, this material 

 may be treated as though it were homogeneous, and as if sex were non- 

 existent. 



5— T 



