Decembee 7, 1917] 



SCIENCE 



567 



is locally applied. In the case of any curved 

 surface, e. g., of a suspended drop of fluid, the 

 tangentially acting force due to surface-ten- 

 sion must similarly tend to draw the surface- 

 fluid away from any area where the tension is 

 locally lowered; for geometrical reasons this 

 lateral traction is necessarily greater than the 

 externally directed force acting on the surface- 

 fluid at the same area — due to the radial com- 

 ponent of surface-tension which compresses 

 the drop and tends to cause outflow at that 

 area; hence in this case also the surface-layer 

 of fluid will tend to be withdrawn from regions 

 of lower and heaped up at regions of higher 

 surface-tension. If the drop is in contact with 

 a solid, such displacements may by reaction 

 cause movements of the drop as a whole. The 

 author's account of the mechanics of amoeboid 

 movement and cell-division needs to be recon- 

 sidered, since he assumes throughout that pro- 

 trusion or outflow always takes place at re- 

 gions of lowered surface-tension. 



The whole subject, however, is full of de- 

 batable questions, and in his preface the author 

 expressly defers judgment upon most of these, 

 urging that the present need is for further in- 

 vestigation rather than for theoretical discus- 

 sion. Most of us will readily grant this, and 

 it is as an aid to investigation that the present 

 manual will find its chief usefulness. 



The reviewer feels bound to point out that 

 the book suffers greatly from carelessness in 

 composition and proofreading. The responsi- 

 bility for this is not the author's alone. A 

 University Press should be careful to maintain 

 high standards in such matters. 



Ealph S. Lillie 



THE PROCEEDINGS OF THE NATIONAL 

 ACADEMY OF SCIENCES 



The seventh number of Volume 3 of the 

 Proceedings of the National Academy of Sci- 

 ences contains the following articles: 



The Cayleyan Curve of the Quartic: Teresa 

 Cohen, Johns Hopkins University. 



A Search for an Einstein Relativity-Gravi- 

 tational Effect in the Sun: Charles E. St. 

 John, ilount Wilson Solar Observatory, Car- 

 negie Institution of ^Vashington. A series of 



observations stretching over several years 

 indicates that the Einstein effect does not ex- 

 ist. 



Triads of Transformations of Conjugate 

 Systems of Curves: Luther Pf abler Eisen- 

 hart, department of mathematics, Princeton 

 University. 



The Molecular Weights of the Triaryl- 

 methyls: M. Gomberg and C. S. Schoepfle, 

 Chemical Laboratory, University of Michigan. 

 After discussing factors influencing dissocia- 

 tion and the relation between dissociation and 

 the nature of the aryl groups, seven triphenyl- 

 methyls are investigated in detail and various 

 inferences are drawn from the graphs of their 

 dissociations against their concentrations. 



Sex-Determination and Sex-Differentiation 

 in Mammals: Frank E. Lillie, department of 

 zoology. University of Chicago. Discussion 

 of the results of studies of the anatomy of 

 twenty-two fetal free-martins ranging in size 

 from 7.5 to 28 cm. Sex determination in 

 mammals is not irreversible predestination; 

 with known methods and principles of physiol- 

 ogy we can investigate the possible range of 

 reversibility. 



The Crystal Structure of Magnesium: A. 

 "W. Hull, Research Laboratory, General Elec- 

 tric Company, Schenectady. The structure 

 is analyzed by means of X-ray. 



The Structure of High-Standing Atolls: 

 "W. M. Davis, department of geology. Harvard 

 University. Attention is drawn to the rela- 

 tion of atoll limestones to their supposed 

 foundation of volcanic rocks. The relative 

 merits of the glacial-control theory and of 

 Darwin's theory are discussed. 



Studies of Magnitude i7i Star Clusters, VII. 

 A Method for the Determination of the Rela- 

 tive Distances of Globular Clusters: Harlow 

 Shapley, Mount Wilson Solar Observatory, 

 Carnegie Institution of Washington. The 

 median magnitude of short period variables is 

 constant in each cluster and may be used to 

 determine the distance of the cluster which, 

 with one or two exceptions, is found to be 

 greater than 30,000 light-years. 



The Principal Axes of Stellar Motion: H. 

 Raymond, Dudley Observatory, Albany, Xew 



