March 11, 1898.] 



SCIENCE. 



337 



The other case is that of the Report on 

 the Mollusks of the deep-sea dredging ex. 

 peditions sent out by France, 1880-83, in 

 the Travailleur and Talisman. The first 

 volume of this report by Arnould Locard,* 

 on the Testaceous Mollusks, includes the 

 Cephalopods, Pteropods and Gastropods as 

 far as Litiopidse. It is illustrated by excel- 

 lent lithographic plates and is chiefly de- 

 scriptive. A superficial examination gives 

 the impression that the abyssal fauna of the 

 eastern Atlantic does not materially differ 

 in character from that of the American 

 border of the same ocean, but that, so far as 

 it does differ, it confirms the impression that 

 the abyssal mollusk fauna of any coast is 

 strongly tinctured with the faunal charac- 

 teristics of the shallow waters of that coast • 

 so that, while there are some ubiquitous or 

 almost ubiquitous species and many ubi. 

 quitous genera, the deep-sea fauna will 

 eventually be divisible into almost as many 

 provinces as there are recognizable among 

 the different faunas of the sea margin. 



We congratulate the author on the ap- 

 pearance of this weighty instalment of his 

 work, and desire to assure him that we also 

 know what it is to publish through a gov- 

 ernment printing office. 



"Wm. H. Dall. 



ON THE LAW OF ANCESTRAL HEBEDITY.f 

 The Darwinian theory has for its main 

 factor the perpetuation of favorable varia- 

 tions by natural selection under the law 

 of heredity. Hence any complete quanti- 

 tative treatment of evolution must deal : 

 (1) with the nature and distribution of vari- 

 ation ; (2) with the nature and influence 

 of selection, and this not only upon the 



* 4°, pp. iv + 516, PI. I.-XXII. ; Paris, Masson et 

 Cie., 1897. 



t ' Mathematical Contributions to the Theory of 

 Evolution.' Abstract of a paper read before the 

 Eoyal Society by Professor Karl Pearson, F.R.S., 

 University College, London, January 27, 1898. 



selected but upon all the correlated charac- 

 ters or organs; and (3) with the law of he- 

 redity. Earlier published and other written 

 but unpublished papers of the present writer 

 cover to some extent the ground of (1) and 

 (2). Although the mathematical theory of 

 variation and selection is yet very far from 

 completion, the general lines on which it 

 will proceed seem, to the present writer at 

 any rate, fairly clear. With the law of 

 heredity, however, the case has hitherto 

 been different. Much has been written on 

 the subject, much has been attributed to 

 inheritance, but the quantitative measure- 

 ments and facts have formed such a small 

 and slender proportion of the whole that it 

 has been extremely difficult to base a 

 rounded mathematical theory on what is 

 really known. It was with a view to the 

 collection of further facts that the writer 

 started his collection of Family Measure- 

 ments, which would now have reached com- 

 pletion were it not that certain collateral 

 relationships are still numerically some- 

 what deficient. Such facts are so all-impor- 

 tant for real progress in our. knowledge of 

 heredity that the writer is convinced that 

 there ought to be a comprehensive and sys- 

 tematic collection of them by some public 

 body ; the labor is beyond the powers of 

 any unaided individual. 



When the writer of the present paper 

 wrote his memoir on Heredity, in 1895,* 

 the only available material was contained 

 in Mr. Francis Galton's Natural Inheritance, 

 and in the data and measurements in Mr. 

 Galton's hands, which he at once placed, 

 with his usual generosity, at the writer's 

 disposal. The very suggestive theory of 

 heredity developed in the Natural Inherit- 

 ance has two main features : (a) a theory 

 of regression, which states the average pro- 

 portion of any character which will be in- 

 herited under any degree of relationship. 

 This theory was very simple; if the aver- 



* Phil. Trans., Vol. 187, A, p. 253. 



