May 27, 1898.] 



SCIENCE. 



751 



but we may confidently expect the beginnings 

 of such a genetic psychology in the future. At 

 any rate, in this field, as in most other fields, 

 progress and profit are increased by greater ex- 

 actness and care, by more accurate and conven- 

 ient apparatus and by shorter and more definite 

 methods. These elements are the ones which 

 experimental psychology is trying to introduce 

 into the exploration of mental life. The fact 

 that these methods are somewhat new in psy- 

 chological work gives us the right to call a 

 system of them a 'new psychology.' 



Professor Stanley's claim that biology is the 

 main standpoint of psychology is quite justified 

 — if 'psychology' means the science of mental de- 

 velopment. It must be remembered, however, 

 that there is a fundamental difference in aim 

 and method which marks off" experimental psy- 

 cbology from the other mental sciences. Its ob- 

 ject is to determine the fundamental laws of 

 mental activity in the adult human being under 

 ■ ordinary circumstances. The change of the prob- 

 lem to child-study, to the development of the 

 individual or of the race, or to abnormal circum- 

 stances, produces closely related sciences. All 

 these sciences are inter-dependent. In fact, 

 all these sciences — as Professor Stanley implies 

 — are needed for a concrete, practical under- 

 standing of mental life ; nevertheless conven- 

 ience and clearness sometimes require that at- 

 tention should be concentrated on one of them 

 at a time. 



E. "W. SCEIPTUKE. 



New Haven, Conn., May 20, 1898. 



rOSSIL FULGUE PEEVEESUM AT AVALON, N. J. 



On page 682 of Science the quotation from 

 Captain Swain, of the Avalon Life Saving 

 Station, N. J., with reference to the casting 

 ashore of Fulgur perversum is slightly inaccu- 

 rate. I now quote from his letter the passage 

 I read at the Academy that " the conchs in ques- 

 tion come ashore only during a strong north- 

 west (not northeast) wind that happens imme- 

 diately after a northeast or a southeast gale, a 

 northwest wind is the only kind that will bring 

 heavy substances ashore, it seems to make the 

 surface current offshore, and this creates an 

 under current on-shore." I have no doubt that 

 Fulgur perversum at the locality is raked out of 



a fossil bed a short distance offshore, and that 

 this off-shore wind after the on-shore gales favors 

 the tides and currents in doing so. 



Lewis Woolman. 



the definition of species. 



I HAVE Stated in this Jouenal (N. S. , VI, 329) 

 that I believe the quantitative study of varia- 

 tion to be the most pressing problem of biolog- 

 ical science. I have consequently read with 

 great interest the papers by Professor Daven- 

 port and Mr. Blankinship, on 'A Precise Cri- 

 terion of Species ' (page 685 above). It seems 

 evident that for the definition of species we 

 should not depend on a 'type specimen,' the 

 one first found, in the best state of preserva- 

 tion or the like, but should collate a consider- 

 able number of specimens taken at random, and 

 when the traits can be measured give the aver- 

 ages and the mean deviations. Then, as Mr. 

 Davenport explains, we have double-humped 

 curves showing a tendency for the type to 

 split up, and these are of the greatest possible 

 interest to the student of the causes of the evo- 

 lution of species. 



When, however, Mr. Davenport proposes to 

 use a given^relation between the height of the 

 smaller hump and the depression between the 

 humps* — namely 100 : 50 — as a precise criterion 



* This relation depends not only on the distance 

 between"^ the apices, but also on the relative number 

 of specimens of the two types, which, of course, has 

 nothing to do with the difference between the types. 

 There are other cases in Mr. Davenport's paper 

 where the statements seem scarcely to take account of 

 the complexity of the problems. It is meaningless to 

 say that ' in some cases fifty per cent, or even more of 

 the individuals will occur at the mode ' and that in 

 this case the curve is steep. The number of indi- 

 viduals at the mode depends on the unit of measure- 

 ment selected, and the steepness of the curve is arbi- 

 trary. The 'half range,' defined as three times the 

 'standard deviation' (error of mean square), is a 

 theoretically impossible point, and could only be de- 

 termined approximately from thousands of speci- 

 mens. Thus in Mr. Davenport's Fig. 9 the 'half 

 range ' of the right-hand curve is tripled by a single 

 specimen. In all these cases Mr. Davenport neglects' 

 the probable errors which when reckoned show that 

 his distinctions between species and varieties have 

 no validity whatever. The data of Fig. 9 can be ex- 

 pressed by a curve with a single apex. 



