The evaporation of droplets of 1 to 2 mm. diameter into still air has been studied with nine liquids forming a series of graded volatility from toluene to aniline and including water. The experimental results agreed with calculations based on diffusion and heat transfer theory. The surface temperature of the more volatile droplets was found to be significantly lower than the bulk temperature, in contrast to Johnson's results for water. Convection effects were found to be present with 43 mm. diameter evaporation vessels in confirmation of Gilliland's suggestion, but were absent in the smaller vessels used in the main investigation.

I n an extension of earlier \\-orl;, sti~tlies have been macle of the ti111e variations in the particle nuntbcr ; I I I~ size i111c1 111as~ c011ce11tratio11 of 'still' ~I I I I~I O I I~~I~I chloride sn1ol;es having ;I \\.itlc range of initial mass concentration. Direct measurement of lo5ie.i 1 0 chan~ber aurfaccs was an important f e a t~~r c of the cxperitner~ts since ir pro\,itletl a n atltlitional q u a n~i t y to be saLilietl by ~h e o r y ancl pernlitLet1 a nlore critic;~l c\.aluation of the coagulation and loss constants than was prcviousl!. possillle. Sa tisfac~ory agreement \\,as found betweer1 theory ancl observation. I t has been slloxrn pre\~iousl!. (2) that time variations in the particte number of ammonium chloricte smokcs in still ancl moving air are described by ecluations decli~ced on the assu~llption that the rates of coagulatio~z and loss to surfaces are proportional respecti\.el>. to the square and to the first pol\-er ot the particle number. Application of the ecli~ations to the data permitted evali~ation of the loss and coag~~lation constants. The experiments, however, \Irere limited to an approximately constant initial mass concentration of smoke (40 mgm. per m 3 ) . Furthermore, as \\-as pointed out, the ~)~-o c e d~~r e employecl in the analysis of data dicl not lead to a \-cr>-critical evaluation of the constants for still or slightly moving air, although it \\-as satisfactory for the data obtained with the more rapid air n~otions.New cxperirncnts ha1.e been pertornled with ammonium chloride s~nolce in 'still' air over a \ride range of ~nitial mass concentration (36 to 120 nlgm. per 111.9. In man>. ot them the rate of loss of particles to surfaces was determined directly in aclclition to nleasurenlcnts of particle number and size and mass concentration. These detcl-minations \\lere an important feature since (a) they provided data for an adclitional q~tantity to be satisfied by theory, and (b) they permitted use of a more critical procedure for calculation of the coagulation and loss constants. A special set of experiments was performed with a series of chambers of graded size ranging from 11.8 to 0.0033 m.3 in order to test the predicted dependence of the loss constant on the ratio of the wall surface area to the chamber volume.All the data, including those for the rate of particle loss to surfaces and its dependence on chamber size, were in accord with the theory. In the light of the newer results some revision was desirable in the constants for 'still' and slightly moving air as deduced from earlier data (2). On the basis of this revision it appears that the coagulation constant is little affected by turbulent air motion

The anim als were p h o to g rap h ed ag ain st a sq u ared background, an d p rin ts m ade w ith th e squares enlarged to th e sam e size. E ach anim al w as th e n c u t o u t from th e p rin ts an d all m o u n ted on squ ared p a p e r a n d rep h otographed. R elativ e sizes an d body proportions can be com pared. M ethod of p h o to g rap h y a n d rep ro d u ctio n as in P la te 14. 612.313The secretion of protein m aterial in the parasym pathetic subm axillary saliva This paper deals with a study of the secretion of protein material by the submaxillary gland of the cat in response to electrical stimulation of the chorda tympani nerve. In the attempt to make a more generous use of certain advantages offered by physical methods in research on the digestive glands, the submaxillary gland was chosen for the initial investigation because of the relative ease of physiological experimentation. The study is characterized by (1) the use of quantitative spectroscopic methods of analysis which permit an extension in the range of the investigation over that possible with a purely chemical procedure, and (2) the application of mathematical analysis in the interpretation of the results.

Smoke chamber studies have been made of the ageing of ammonium chloride smokes under various controlled degrees of turbulent air motion as well as in still air. The changes in mass concentration and particle number were followed for 5.5 hr. under each set of conditions, and some data on particle size distribution were obtained. The logarithm of the mass concentration was found to vary linearly with time under all conditions. None of the particle number data lent themselves to interpretation on the basis adopted by previous workers. They were described quantitatively by equations developed from general postulates which take into account loss to various surfaces. The equations permitted analytical separation of coagulation and surface effects, and the constant descriptive of the latter was closely related to that associated with mass loss. The coagulation constant for still air was found to be only slightly greater than the ideal Smoluchowski value. A description was obtained of the manner in which the various constants increased with the degree of air motion.

Studies have been made of the aging of magnesium oxide, zinc oxide, carbon, copal resin, and silica powder aerosols, under essentially uniform experimental conditions. The changes with time in tlie particulate ~~clinber and in the n~~m b e r of particles lost from the system by deposition on charnber surfaces are well described in terms of the coagulatio~~ c o~~s t a~l t k and the deposition col~stalitp ~~4 t h the aid of eq~lations known to be applicable to r u m r n o~~i~~m chloride aerosols over a wide range of conditio~~s. Marked differences \irere found in tlie i-alues of k for different aerosols and lesser differences i l l the \ d u e s of p. AII electron microscope study of the shape and structure of the particles led to the conclusion that differences in these characteristics are not predominantly responsible for the observecl differences in behavior, although they may be a contributir~g factor. . A few prelirninary experi~iients in which silica powder was dispersed through an electrically charged screen indicated that the coagulatio~~ and deposition processes are influenced by changes in the electrical charge d i s t r i b~~t i o~~.