^^ 



MTSrETJ.ANEOUR FOREST INSECTS. 



Tlu' length of iho Ix'ak also vaiios; thus, in soiuo of the foiiialos it 

 is shorter than in some males of the same species. The elytra are 

 more or less variable in form, but ap])ear to be more constant than 

 the other parts, and show little or no sexual tliiierence. 



The chai'acter and i)osition of the spots of densely placed scales 

 appear to be of special value in the classification of the genus, but 

 these are sometimes rendered obscure in old, rubbed, and dirty speci- 

 mens. The scales are so firmly attached, however, that they are 

 often sufficiently retained in old specimens to be of value. Dirty 

 specimens can be easily cleaned with chloroform or xylol, the latter 

 being especiall,y valuable for the removal of resin. 



STATISTICAL TAXONOMY. 



In a comparative study of the characters which distinguish the 

 major and minor divisions and species of a genus or a larger group of 



organisms, a progressive 

 modification of certain 

 parts of the body struc- 

 ture is usually found to 

 serve (together with other 

 characters) as an index to 

 a natural classification. 

 Therefore the importance 

 of having some clear and 

 definite method of indi- 

 /nc/,^<c/u^/ /nc/ej! }> 9** catlug thc raiigc and 



Fig. 9.— P/ssodfs fraseri: Lateral aspect, showing method of de- limit of SUch liuCS of modi 

 termining 

 prothorax 



termining individual index, a, Length of teak; b, length of fl„„f,-^„ ^„ cinpcifllizfltion 



c;c, length of elytra. (Original.) iicaiion or Specialization 



IS apparent. 



The writer's experience with the statistical method in comparative 

 studies of such modifications has convinced him that when it is 

 accurately applied a mathematical formula may be determined to 

 express the limit and relative taxonomic importance of a given 

 modification in one or more structural characters, to indicate specific 

 differences, and to show the relative position and rank the species 

 occupy in a natural classification. Thus we may adopt for certain 

 groups of insects a statistical taxonomy as a guide toward the 

 classification of the species into natural divisions. 



In the bark-weevils of the genus Pissodes we have a good example 

 for the application of this method. One of the principal lines of 

 modification available for statistical study is the progressive elonga- 

 tion of the beak. Therefore when we compare the average ratio or 

 mode of the length of the beak to both the length of the prothorax 

 and length of the elytra (fig. 9) in a number of individuals of one 

 species with that of an equal number of individuals of another species, 



