216 THE SCOLYTID BEETLES, 



STATISTICAL TAXONOMY. 



The subject of statistical taxonomy has received considerable 

 attention by the writer and an attempt has been made (Hopkins, 

 1911, pp. 28-30, PL II) to show its successful application in the genus 

 Pissodes. The detail and accuracy required for this method and the 

 difficulty of determining a specific and taxonomic formula are so great, 

 however, as to be rather trymg on the patience of the investigator. 

 Therefore it is the opinion of the writer that it should not be resorted 

 to except in cases in which other methods fail to give satisfactory 

 results and then only when the elements to be included are of such a 

 nature as to be readily available for the mathematical determination 

 of relative proportions. There is in this, as m many subjects relating 

 to taxonomy, a need of a more comprehensive investigation m order 

 that the most reliable basis for conclusions may be reached. 



THE SPECIES. 



Reference has been made (Part I, p. 66) to the writer's views on 

 specific distinction and the range and limits of specific variation. In 

 addition it may be said that no species can be established beyond dis- 

 pute without a knowledge of both the morphological and physiological 

 elements of distinction. Tliis does not, however, preclude the recog- 

 nition andnamingof imperfectly defined and poorly represented species 

 and of retaining them as long as they can be readily distinguished 

 from other allied forms. Such names and definitions serve as a basis 

 for study until it is more definiteh^ shown whether or not the forms are 

 worthy of permanent specific distinction. There is necessarily a wide 

 range in the relative prominence of the distinguishing characters as 

 between a species which is the only representative of a genus and sub- 

 family and one of a group of closely allied species. It is found, how- 

 ever, that for the purpose of systematic and economic mvestigation 

 the isolated species may be of no more value for study and may con- 

 tribute no more to the advancement of knowledge than the one which 

 is with difficulty separated from its congeners. 



Since we now insist upon a single specimen as the author's designa- 

 tion of the type of a described species it is deemed by the writer to be 

 perfectly proper to base the description of a new species on a single 

 specimen, provided the author of the description is sufficiently 

 familiar with the previously described allied species and genera to 

 enable him to recognize the characters and characteristics which are 

 of real specific importance. 



VARIETIES, RACES, ETC. 



There is in all species a greater or lesser range of variation from 

 the type, and often some groups of individuals may be so different as 

 to indicate a distinct varietv or race. If such variant forms are 



