Mabch 5, 1909] 



SCIENCE 



387 



associated with two of these in metabolic in- 

 sects. The pupa of the Holometabola is re- 

 garded as being to a certain extent a phylo- 

 genetic stage, analogous to the subimago of 

 the Ephemeridea, but as having developed its 

 peculiarities (quiescence, unchanged external 

 form and profound internal changes) in cor- 

 relation with the structural differences that 

 separate the larva from the imago. These 

 differences are described as follows : 



In the Hemimetabola the whole development 

 appears as at first progressively imaginipetal 

 (total habitus), later as temporarily and pro- 

 gressively imaginifugal (provisional organs), with 

 ontogenetic adaptations, and finally as regressively 

 imaginipetal (involution of the provisional or- 

 gans). In the Holometabola, on the contrary, 

 development is at first regressively imaginifugal 

 (total habitus and imaginal organs), then pro- 

 gressively imaginifugal (development of provi- 

 sional organs of first and second order) and 

 finally (in the pupa) progressively (total habitus 

 and imaginal discs) and regressively imaginipetal 

 ( involution of provisional organs ) . Hence the 

 Holometabola are characterized in the metembry- 

 onic portion of their life cycle by a regressively 

 imaginifugal type of development, which changes 

 to the progressively imaginipetal type in the pupa. 

 In other words: Whereas the continuously pro- 

 gressive development of the Hemimetabola is not 

 interrupted and is only slightly affected by the 

 formation of provisional organs, the progressive 

 development of the Holometabola up to the im- 

 aginal stage sufi'ers a long interruption (during 

 the larval stage) and is not resumed till the 

 transition to the first imaginal stage (the pupa), 

 in order to attain, by passing through this, the 

 definitive imaginal form. 



Deegener, like many other students of in- 

 sect metamorphosis, regards the pupa as a 

 teleological development which enables the 

 organism greatly to lengthen its larval life, 

 and through the magnitude and intensity of 

 the changes which it undergoes, to drop out 

 or fail to recapitulate, a great number of 

 phylogenetic stages and thus to pass directly 

 into the adult condition. The development of 

 such a pupal stage, he believes, has been facili- 

 tated by the ability, so frequently observed in 

 insects, to fast for long periods of time. In 

 this connection he might also have called at- 

 tention to the adaptation of the pupal stage 



to tiding over unfavorable seasons (cold win- 

 ters in temperate and boreal, dry seasons in 

 tropical regions), as has been pointed out by 

 Lubbock, Haacke, Handlirsch and others. 



William Morton Wheeler 



The Systematic BelationsMps of the Coccacece, 

 with a Discussion of the Principles of Bac- 

 terial Classification. By Charles-Edward 

 Amory Winslow and Aune Eogers Wins- 

 low. New York, John Wiley & Sons. 

 1908. 



The book before us is the completed results 

 of work by these authors of which we have 

 had preliminary information through articles 

 in Science^ and the Journal of Infectious 

 Diseases' 



This work is by far the most important 

 contribution to the purely scientific side of 

 bacteriology which has appeared in a long 

 time. It marks the beginning of a new era 

 in bacteriological classification and nomen- 

 clature. 



The systematic classification of the bacteria 

 has always been extremely artificial and arbi- 

 trary. Outside of the three large morpho- 

 logical groups, the cocci, bacilli and spirilla, 

 classification has probably never expressed 

 natural relationships. However useful for 

 purposes of identifying species artificial classi- 

 fication may be, it never reaches its highest 

 function until it tells us more than whether 

 a species has been previously described in the 

 literature. It can never be really useful until 

 it expresses for us the real position of the 

 species in question in relation to other forms, 

 and to some extent, at least, tells us the prob- 

 able line of descent which the species has 

 followed in its development from other forms. 

 This is the ultimate goal which the classifica- 

 tion of all living forms should seek. 



A few attempts have been made to recog- 

 nize certain " groups " among the bacteria, 

 and undoubtedly some of these groups repre- 



'" A Revision of the Coccaeeae," Science, N. S., 

 XXI., 1905, 669. 



^ " A Statistical Study of Generic Characters 

 in the Coccaeeae," Biological Studies by the Pupils 

 of William Thompson Sedgwick, Boston, 1906; 

 also Journal of Infectious Diseases, III., 1906, 485. 



