786 



SCIENCE. 



[N. S. Vol. IV. No. 100. 



jacket lias been decreased as might be ex- 

 pected ; but this decrease is quite irregular, 

 reaching a maximum of 0.064 inch about 

 two feet from the upper, or muzzle, end of 

 the jacket, while at the other end of the 

 jacket it is 0.017 inch. The diminution of 

 bore extends to the extremity of the breech 

 beyond the region of compression due to 

 the jacket. 



The only hypothesis upon which these 

 irregularities can be even tentatively ex- 

 plained is lack of homogeneity in open- 

 hearth steel. The better the steel the more 

 nearly perfectly does it recover from strain 

 after the removal of stress. Irregularity in 

 heating during the first attempt would have 

 produced warping in any metal whatever. 

 Perfectly homogeneous steel would have re- 

 covered completely when the temperature 

 became uniform, but any lack of homo- 

 geneity implies a permanent set. Assuming 

 such lack, the larger the scale of experi- 

 ment the more difficult it is to secure uni- 

 formity of temperature. The steel may 

 satisfy completely the demands of prelimi- 

 nary test experiments on elastic limit and 

 tensile strength, yet it may fail to meet the 

 requirements of accurate fitting and com- 

 plete recovery after wide variation of tem- 

 perature. Krupp, with his admirably 

 homogeneous but high priced crucible steel, 

 has already been successful in assembling 

 the parts of guns twice as massive as the 

 recent subject of experiment at Watervliet. 

 It remains to be seen whether equal success 

 will be possible by the use of open- hearth 

 steel in connection with the 16-inch gun 

 yet to be constructed. In the light of re- 

 cent difficulties the approaching work will 

 be watched with interest. 



The publication of these observations 

 in the present form would not have been 

 possible but for the friendly courtesy of 

 Major Isaac Arnold, Jr., the commandant 

 of the Watervliet Arsenal, who has granted 

 me the utmost freedom of access to the gun 



shops and who kindly invited my inter- 

 est and cooperation in the attempt to 

 solve the problem of the unmanageable 



jacket. 



W. LeConte Stevens. 



Eensselaeb Polytechnic Institute, Teoy, N. Y. 



ONTOGENIC AND PHYLOGENIO VARIATION. 



In an article published in 1894, in 

 Merckel u. Bonnet's Ergebnisse der Anatomie 

 u. Lntwicklungsgeschichte, the writer proposed 

 the distinction which forms the title of this 

 article. This subj ect was further elaborated 

 in three papers before the Biological Section 

 of the New York Academy, in March, April 

 and May, 1896. As Prof. C. Lloyd Morgan 

 and Prof. J. Mark Baldwin have quite inde- 

 pendently reached somewhat similar conclu- 

 sions, it seems of interest to publish the 

 second and third papers, above referred tOr 

 in their original form as they were mailed 

 to the Secretary of the Academy. These 

 papers, by an unfortunate oversight, were 

 never sent to the printers. The first paper 

 was published in the Transactions and ab- 

 stracted in Science. 



The title of the paper of April 13th was 

 'A Mode of Evolution requiring neither 

 ]S"atural Selection nor the Inheritance of 

 Acquired Characters.' It was discussed by 

 Prof. Baldwin and Prof Cattell. "I pre- 

 sent a continuation of the subject of Onto- 

 genic and Phylogenic Variation, discussed 

 at the last meeting of the Biological Section, 

 The latest papers upon selection are signifi- 

 cant because they show that the hypothesis 

 of evolution purely by the selection of fortu- 

 itous variations is losing ground. Defi- 

 nite or determinate variation is now ad- 

 mitted by nearly all writers except Wal- 

 lace. If we assume the transmission of 

 acquired characters the explanation of defi- 

 nite variation becomes simple enough, but 

 in this contribution I propose a view of 

 the facts which does not assume the trans- 

 mission of acquired characters nor the im- 



