December 2, 1904.J 



SCIENCE. 



765 



Dr. F. H. H. Calhoun, who discussed two 

 features of glaciation in the northwestern part 

 of Montana. The first was as to the relative 

 ages of the ice of the Kewatin ice sheet, and 

 the ice of the mountain glaciers. The second 

 dealt with the question of the Albertan drift 

 period. From data gathered in the three 

 river valleys, the conclusion was reached that 

 the mountain ice was the older of the two 

 sheets. In the valley of Birch Creek a valley 

 train from the mountain glacier was found 

 beneath lake deposit formed in front of the 

 Kewatin ice sheet. In the valley of the Two 

 Medicine River, the moraine of the mountain 

 ice was covered by lake deposit, and in the 

 same valley further east, the moraine of the 

 Kewatin ice sheet rested on a valley train 

 from the mountain glacier. In the valley of 

 the Belly River, the mountain drift was found 

 under the Kewatin drift. Dr. Calhoun also 

 arrived at the conclusion that the so-called 

 Albertan drift sheet of Dawson was nothing 

 more than old river gravels picked up by the ' 

 Kansan ice sheet, and now found at the bot- 

 tom of that drift sheet. This paper will be 

 published by the TJ. S. Geological Survey. 

 Haven Metcalf, 



Secretary. 



ONONDAGA ACADEMY OF SCIENCE. 



At the last meeting of the Academy, held in 

 Syracuse, N. T., October 21, Dr. H. C. Cooper 

 addressed the meeting on ' Physical Chem- 

 istry and its Service to the Other Sciences.' 



Physical chemistry was shown to have an 

 important bearing on physiology in virtue of 

 -the application of the modern theory of solu- 

 tion to vital processes and the discovery of 

 the similarity of behavior between enzymes 

 and inorganic contact agents. The work of 

 van't Hoff and his pupils on chemical equi- 

 libria in various geological formations, and 

 the determinations of the transition-points of 

 geologically important substances were com- 

 mended to the study of geologists. The 

 speaker also showed how, by physico-chemical 

 methods, various physiologically and miner- 

 alogically important substances have been 

 prepared in the laboratory. 



J. E. KlRKWOOD^ 



Corresponding Secretary. 



WELLESLEY COLLEGE SCIENCE CLUB. 



The first regular meeting of the club for 

 the year was held at the Whitin Observatory 

 on October 18. Professor Ellen Hayes pre- 

 sented a 'paper on ' Nature the Master of 

 Man.' 



The second meeting was held in the Physics 

 Lecture Room on November 8. Mr. Albert E. 

 Leach, of the Massachusetts State Board of 

 Health, read a' paper on 'The Purity of our 

 Food.' Grace E. Davis, 



Secretary. 



DISCUSSION AND CORRESPONDENCE. 

 Davenport's ' statistical methods.' 



To the Editor of Science: In the preface 

 to the second edition of Dr. C. B. Davenport's 

 useful ' Statistical Methods ' occur the follow- 

 ing sentences : " Especial attention is called 

 to Table IV., which is an extension of Table 

 IV. of the first edition that was calculated by 

 Professor Frederick H. SafEord, and appears 

 to have been the first published table of the 

 normal probability integrals based on the 

 standard deviation. More recently Mr. W. F. 

 Slieppard has published in Biometriha a sim- 

 ilar table. * * * " 



In justice to Mr. Sheppard I wish to point 

 out that his tables in Biometriha are only the 

 extension of the table which appears in his 

 memoir read before the Royal Society in 1897 

 and issued in 1898 (cited by Dr. Davenport, 

 second edition, pp. 84, 101). Dr. Davenport's 

 first edition of his ' Statistical Methods ' did 

 not appear until 1899. Thus there can be no 

 claim as to priority for Professor SafFord's 

 table of 1899. If we desire to be absolutely 

 accurate in the matter, we shall probably not 

 attribute the first published table even to Mr. 

 Sheppard, still less to Professor SafEord. On 

 the other hand, Mr. Sheppard's table at pres- 

 ent stands quite unsurpassed for its range, 

 accuracy and the number of decimal places 

 in the probability integral. 



Karl Pearson. 



Untversitt College, London, England, 

 November 5, 1904. 



CORALS. 



To the Editor of Science: I find myself 

 in the position of an author replying to a 



