1899] VARIATION-STATISTICS IN ZOOLOGY 335 



objects, because they can easily be experimented on in regard to 

 heredity and local variation. 



If I have succeeded in showing the statistical method of 

 investigating variation to be based on logical foundations, and to be 

 capable, by its special nature, of yielding new and valuable results 

 which cannot be acquired by any other method of investigation, it may 

 lead, I hope, to the increased use of the method by zoologists. The 

 mathematical training its application requires, does not exceed the 

 standard of final high-school examinations. The exact and unambiguous 

 results of the analytical method have a charm of their own, and it 

 ought not to be forgotten that a precise terminology helps in every 

 scientific work. The merely statistical research, whether directed 

 towards aetiological or towards morphological problems, is only, I believe, 

 the introduction to a more important kind of work, by which statistical- 

 analytical and therefore critical results will be established with the 

 help of quantitative experiments. To carry out this purpose we 

 would need a separate institute, distinguished from the ordinary 

 biological laboratory by larger accommodation for breeding experi- 

 ments, and from the taxonomic museum by facilities for storing in an 

 accessible fashion large quantities of homogeneous individuals which 

 may readily be investigated at any time either for controlling or for 

 completing former researches. To the eminently practical value of 

 such an institution for agriculture, forestry, horticulture, as well as for 

 fishery and cattle-breeding, I can now only refer ; its chief aim of 

 course would be scientific investigation, of which the results are always 

 either directly or indirectly valuable to practical life. The first step, 

 however, is that the statistical-analytical method be duly recognised 

 as a new and important organon in the advancement of biology. 



LIST OF PAPERS CITED. 



1. Brewster, E. T., "A Measure of Variability and the Relation of Individual Varia- 



tions to Specific Differences," Proc. Amer. Acad. Arts Sci. vol. xxxii. No. 15, pp. 

 268-2S0, 1897. 



2. Bumpus, H. C, "The Variations and Mutations of the Introduced Sparrow," Biol. 



Lectures IVoods Holl (1896), pp. 1-15, 1897. 



3. " The Variations and Mutations of the Introduced Littorina," Zool. Bull. vol. i. 



No. 5, pp. 247-259, 1898. 



4. " On the Identification of Fish Artificially Hatched," Amer. Natural, vol. xxxii. 



No. 378, pp. 407-412, 1898. 



5. Czernay, A., " Beobachtungen liber das Variieren der Artkennzeichen der Susswasser- 



fische in der Umgegend von Charkow," 1857. Bull. Soc. Imp. Nat. Moscoiv, vol. 

 xxx. No. 1, pp. 227-249, 1857. 



6. Davenport, C. B., and Bullard, C, "Studies in Morphogenesis; VI. A Contribu- 



tion to the Quantitative Study of Correlated Variation and the Comparative Variability 

 of the Sexes," Proc. Amer. Acad. Arts Sci. vol. xxxii. No. 4, pp. 85-97, 1896. 



7. Duncker, G., " Die. Methode der Variationsstatistik," Arch. Entivickclungsmcch., vol. 



viii. No. 1, pp. 112-187. Separately: Leipzig, Engelmann, 1899. 



8. Field, W. L. W., "A Contribution to the Study of Individual Variation in the Wings 



of Lepidoptera," Proc. Amer. Acad. Arts Sci. vol. xxxiii. No. 21, pp. 389-395, 1898. 



