December 22, 1905.] 



SCIENCE. 



835 



by direct evaporation from depths of over two 

 inches in the rock must in general be insig- 

 nificant from an agricultural point of view, 

 hence if the capillary flow of water upward 

 can be nearly stopped an inch or two below 

 the surface by the formation of a dry surface 

 layer, the wastefiil loss of water by evapora- 

 tion will be much decreased. 



Laboratory experiments by Mr. J. O. Belz 

 in the physical laboratory of the Bureau of 

 Soils, in which arid and humid climatic con- 

 ditions were simulated, showed that such a 

 dry layer may be formed naturally under very 

 arid conditions. A very rapid initial evapora- 

 tion forms a dry surface layer, and the rate of 

 evaporation then falls off very greatly, the 

 result being that in the long run the total loss 

 of water from the soil is less under very arid 

 conditions than under humid conditions. 

 Under arid conditions, a soil has thus an au- 

 tomatic tendency to protect itself from the 

 great loss of water which would at first sight 

 be expected to occur under such conditions. 



Mr. Briggs applied- these principles to the 

 conditions of desert plants. 



Mr. J. C. Blake then read a paper on ' The 

 Electrical Behavior of Colloidal Mixtures.' 



The early work of Quincke and others on 

 the migration of visible particles was shown 

 to be in accord with the recent work on ' Col- 

 loids.' The rate of migration of visible par- 

 ticles as well as of colloids is almost identical 

 with that of the common ions in electrolytic 

 solution. It was shown that each colloidal 

 particle is probably accompanied by an ion, 

 which causes the migration of the particle, 

 the ion and colloidal particles being sur- 

 rounded by an electric double layer. The 

 absolute conductivity of the colloidal material 

 in a colloidal mixture agrees with the idea 

 that each colloidal particle carries the ionic 

 charge. Hence colloidal mixtures are to be 

 regarded as true solutions largely ionized, one 

 of the ions being associated with the colloid. 

 Charles K. Wead, 



Secretary. 



THE NEW YORK ACADEMY OF SCIENCES. SECTION 

 OF ANTHROPOLOGY AND PSYCHOLOGY. 



At a meeting held ISTovember 27, 1905, in 



conjunction with the New York Section of 

 the American Psychological Association, Pro- 

 fessor Woodbridge occupying the chair, Pro- 

 fessor Robert MacDougall was elected chair- 

 man for the coming year, and Professor R. S. 

 Woodworth secretary. The following papers 

 were presented: 



Smell Discrimination of Students: Will S. 



Monroe. 



A statistical inquiry into the ability of 255 

 girls to recognize odors showed that, on the 

 average, 6.72 out of a set of 20 common odors, 

 chiefly essential oils, were correctly named. 

 Those most often identified were wintergreen, 

 camphor, peppermint, vanilla and cloves ; those 

 least often, hemlock, bergamot, asafoetida, 

 wormwood and lavender. A census of odor 

 names showed that some of those least often 

 recognized, as lavender, were believed to be 

 familiar. The fact is simply that people do 

 not know as many odors as they suppose. 



Linguistic Standards : Frederic Lyman Wells. 

 The speaker contended that the current 

 standards of ' good use,' based as they are on 

 the individual introspection of rhetoricians, 

 or on a reactionary adherence to selected his- 

 torical models, were not adequate to the 

 changing needs of a language. By the appli- 

 cation of statistical methods, it is possible to 

 obtain standards that are free from the arbi- 

 trariness of one-man introspection, and also 

 furnish, in the ' probable error,' a measure of 

 their own validity. A statistical study of the 

 relative ' force ' of synonyms yielded results 

 having a very small probable error, which 

 nevertheless did not agree with any of the 

 definitions of force which the introspective 

 grammarians have laid down. 



A Threshold Study of the Beading Pause: 



P. M. Hamilton. 



The author showed how the tachistoscopic 

 method could be adapted to the reading not 

 only of isolated words, but also of sentences 

 and paragraphs; to the analysis of processes 

 at the threshold of word recognition; and to 

 the study of the marginal field of perceptual 

 regard. Light is thrown by these experiments 

 upon the questions of literal reading, reading 

 cues, value of context, etc. 



