September 15, 1911] 



SCIENCE 



347 



by many other workers whether they call at- 

 tention to the recapitulation shown or not. 



As a whole M. Cossmann's criticism shows 

 a total misconception of modern methods in 

 phylogenetic study and even the illustrations 

 which I have used to point out likeness or dif- 

 ference in descent are to him, judging as he 

 does by the standards of the older conchol- 

 ogists, only so many offences against the good 

 old fashioned rule of putting together species 

 that are alike in the adult and ignoring 

 " simples et legeres modifications dans I'orna- 

 mentaticn de la spire." A careful compari- 

 son of the detailed figures in my paper will, 

 however, show that the modifications are not 

 slight in cases where genera are separated. 



As might be expected, the choice of a geno- 

 type from the work of a pre-Linnean author 

 is questioned. This choice arose from the 

 difilculty of applying the established rules of 

 nomenclature in such a manner as to meet 

 the approval of all students of the subject. 

 Bruguiere, the first post-Linnsean author to 

 use the binomial nomenclature in connection 

 with Cerithium, did not select a genotype, 

 and Lamarck chose, at different times, two of 

 Bruguiere's species as illustrations of the 

 genus. At present one eminent authority 

 chooses as genotype the first of Lamarck's se- 

 lected species, while another chooses the second 

 and a third suggests a choice from Brug- 

 uiere's many species. If the general con- 

 sensus of opinion finally fixes upon Pseudo- 

 vertagus aluco or Cerithium? nodulosum in- 

 stead of C. tuberosum as the type of Cerith- 

 ium it would simply necessitate the choice 

 of a new name for the group represented by 

 C. tuherosum, C. adansoni, etc., which were 

 the first to be described, and which have long 

 been known by that name. However, the 

 name applied to a natural group is of minor 

 importance. The point of especial importance 

 is that such shells as Pseudoveriagus aluco 

 and Cerithium ec7uMaiMm,f or example, can not 

 be classed together in the same genus since 

 their ontogeny shows that they have an en- 

 tirely different ancestry. The main object 

 xii my paper is to trace the phylogeny of a 

 natural group and to show the methods by 



which relationship should be determined. 



Li summarizing I would emphasize the 

 following three points: 



1. A natural classification should be 

 based on community of descent. 



2. In tracing descent the whole ontogeny 

 is a more reliable guide than a single final 

 stage of it. 



3. There is suflieient evidence in favor of 

 the theory that ontogeny recapitulates 

 phylogeny to make it the only safe means of 

 determining relationship. 



Elvira Wood 

 Columbia IJNrvERsiTY, 

 May 31, 1911 



A NEW RACK FOR INDniDUAL TOWELS 



The " common towel " is a problem which 

 sanitarians so far have failed to solve. This 

 is especially true of the roller towel so often 

 found in public places. The fact that these 

 towels are dirty and unattractive, if not re- 

 pulsive, is comparatively unimportant, for if 

 the real facts were known it would be under- 

 stood that many cases of disease are trans- 

 mitted by this means. In the better places 

 where the wash room can have frequent at- 

 tention, small individual towels can be used. 

 In a good many places, however, it seems im- 

 possible on account of the expense, due to 

 those lost and stolen. One attempt to solve 

 the problem has been the substitution of 

 paper for cloth towels. In the minds of most 

 people, however, these paper towels are not 

 satisfactory, although, of course, they are 

 perfectly hygienic. 



An attempt has been made at the Univer- 

 sity of Wisconsin to solve this problem and 

 towel racks have been designed which make 

 it possible for each person to have an indi- 

 vidual towel. This rack is very simple, as is 

 seen in the accompanying illustration. It 

 consists essentially of a shelf on which are 

 placed a pile of the small, clean towels, and 

 just below this and at one side" is a receptacle 

 or basket into which the dirty towels are 

 placed. Connecting the shelf and the basket 

 is a rod with a goose neck. The towels are 

 provided with a perforation or loop through 



