both man and horses and common in France and occasionally in America, is identically the same species and should be called by the name first used by Hektoen in this country. A butter having only a few yeasts and molds, when other conditions are favorable is a safer hazard for shipments and storage is the claim of F. W. Bouska and J. C. Brown of Chicago in their paper on 'Yeasts and oidia in pasteurized butter." Creameries which have the best commercial reputation for their butter also have the lowest yeast and mold counts. These two men give methods for sampling and counting butter which they have recently devised. The late Dr. Edw. Birge presented his study on the activities of certain bacteria in sewage. He believed that some bacterial forms can be found which will play an important rôle in the treatment of sewage, and that the time will come when septic tanks will be seeded as alfalfa fields and cream vats are seeded now. A method for the detection of pasteurized milks is described in detail by Dr. W. D. Frost, of the University of Wisconsin. The addition of a special dye stains the blood cells, always present in pasteurized milks. In raw milks the cells will not be stained. A strong plea for the thorough investigation of all waters whose potability is questioned, and for thoroughly trained investigators experienced in laboratory and field work, is put forth by H. A. Whittaker, of the University of Minnesota, in a paper on the "Investigation of drinking water supplies." A. L. Amott, a commercial milk expert in Chicago, has given much time, energy and thought to "The milk supply of Chicago," and discusses the source of supply, amount, production, transportation, city distribution, prices, farmers' organizations, and milk inspection. He calls attention to the improvement of the milk supply and the lowered baby death rate in recent years in Chicago. B. W. Hammer, of the Iowa Agricultural College, in a paper on "The bacteriology of ice cream," summarizes the knowledge of such points as number and kinds of bacteria, sources of materials, effect on the bacteria during freezing, hardening and holding, softening and rehardening. He also treats of the manufacture of ice cream with a low bacterial count, and the relation of ice cream to the public health, and bacterial standards. SPECIAL ARTICLES THE QUANTITATIVE BASIS OF THE POLAR CHARACTER OF REGENERATION IN BRYOPHYLLUM WHEN the defoliated stem of a plant of Bryophyllum calycinum is cut into as many pieces as it possesses nodes, each piece will produce shoots from the two dormant buds of its node and roots at its basal end. When a long piece of stem possessing 6 or more nodes is cut out from such a plant only the most apical node will produce shoots from its two buds while the other nodes will show no or only inconsiderable growth. The question is, Why do all the nodes except the most apical fail to produce shoots when they are part of a long piece of stem, while they would each produce shoots when isolated? This is the problem of polarity in regeneration in its simplest form. Earlier biologists, especially Sachs, have suggested that this polarity is due to the fact that the ascending sap carries the substances needed for shoot regeneration and that if a piece of stem is cut out from a plant the sap must collect at the apex and thus give rise to the shoots at the most apical node. This explanation is only satisfactory if the assumption is added that in the case of the stem of Bryophyllum practically none of these substances reach the dormant buds in the nodes below the most apical one. The problem is how to furnish a scientific proof for this suggestion. This can be done by treating this problem from the viewpoint of chemical mass action. The formation of new shoots in an isolated node of a defoliated stem of Bryophyllum can only be the result of synthetical processes the velocity of which depends for a given temperature and degree of moisture upon the relative mass of the material reaching the dormant buds of the node in the unit of time. The material required for growth will be taken from the sap reaching the node. The disappearance of this material from the sap will cause similar material to leave the cells of the stem and to diffuse into the sap. If this purely chemical reasoning is sound it would follow that the larger the mass of the stem the greater the mass of chemical substances available for the growth of shoots per unit of time. On this basis we should expect that the mass of shoots formed on the node of an isolated piece of stem would be in proportion with the mass of the piece of stem. That this is correct can be shown by cutting a defoliated stem of Bryophyllum into as many pieces as it possesses nodes. In this case, each node will produce shoots but their mass will be unequal in the different pieces, and will be greatest where the mass of stem is greatest. If it is true that in a long defoliated piece of stem only the two shoots of the apical node grow out because practically all the material available in the stem flows to the apex; and that the shoots in the nodes below do not grow out because practically none of the material reaches them, then we should expect that the mass of the two shoots formed at the apex of a long piece of stem should approximately equal the mass of all the shoots which would have been formed if the stem had been cut into as many pieces as it contained nodes. A large number of experiments have been made which have shown that this is correct. The following example may suffice: Four large stems of Bryophyllum were defoliated and a piece containing 9 nodes was cut from each defoliated stem. From each piece of stem the three uppermost nodes were cut off and cut into three pieces containing one node each. These 12 one-node pieces produced 23 shoots. The 4 stems, with together 8 shoots. 6 nodes each, produced all After 20 days the dry weight of the shoots and of stems was determined. It was found that the 12 small pieces of 1 node each had produced 23.2 mg. dry weight of shoots per gram of dry weight of stems, while the 4 large pieces with 6 nodes each had produced 26.3 mg. dry weight of shoots per gram of dry weight of stems. This shows that the mass of the two shoots produced at the apex of a long piece of stem equals approximately the mass of shoots which would have been produced in the same stem in the same time under the same conditions if the shoots could have grown out in all the nodes. This leaves no doubt that the polar character of the regeneration of shoots is due to the fact that all the material available for growth reaches the apical and none of the other nodes of a long piece of stem. The average growth of shoots in small pieces is slightly less than in large pieces in the experiment mentioned (23.2 mg. instead of 26.3 mg.), probably because the extreme ends of each piece die or cease to participate in the supply of material for growth. As a consequence the mass of a stem which supplies material for growth is less when the stem is cut into smaller pieces than when it is left intact. It had been shown in previous papers that the mass of shoots and roots produced by a leaf of Bryophyllum is also in proportion to the mass of the leaf.1 A fuller description of the results will be given in the Journal of General Physiology. JACQUES LOEB THE ROCKEFELLER INSTITUTE THE SCATTERING OF ELECTRONS BY NICKEL A STUDY of the electron emission from a nickel target under electron bombardment has revealed certain features of this emission which appear to be of considerable interest on account of their probable bearing on the structure of the nickel atom. Besides the emission of slow-moving secondary electrons characteristic of all metals the emission from nickel contains an appreciable fraction of electrons of higher speed which appear to be scattered directly from the incident beam of primaries by the atoms of the target. The fastest of these scattered electrons have speeds almost if not quite equal to the speed of the primaries. It would appear that the sharp deflections experienced by these scattered electrons must result from their penetrating into the atom structure and being 1 Loeb, J., J. Gen. Physiol., 1918-19, I., 81; 1919-20, II., 297, 651. swung about by the strong field there encountered. For an electron to thus enter and emerge from an atom without appreciable loss of energy would seem to require that it do so without having a near encounter with any of the electrons of the atom structure. On the other hand the structural electrons may be so anchored in position that no energy is transferred to them in any except very close encounters. The fraction of the primary electrons scattered from a nickel target without appreciable loss of energy is small, not more than one in a thousand being turned back with a loss not to exceed one per cent. of its initial energy. The distribution of these high-speed scattered electrons in the region in front of the target is particularly interesting. Our observations suggest that it is entirely symmetrical with respect to the incident beam and independent of the inclination of the target to the incident beam except as this affects the region into which the scattered electrons are free to emerge. With a target inclined at an angle of 45 degrees to the incident beam the intensity of scattering as a function of angle has been studied in the plane including the incident beam and the normal to the target. The range of 135 degrees on one side of the incident beam has been explored with the exception of 25 degrees adjacent to the beam. The Faraday box collector used for picking up the scattered electrons can not be brought nearer the primary beam in our present apparatus. The principal features of the angular distribution are two maxima of emission, one back along the path of the bombarding electrons (v=0) and another lateral to the primary beam whose position depends upon the bombarding voltage. The relative importance of these two maxima also depends upon the speed of the primaries. Fig. 1 shows such a distribution curve for a bombarding potential of 150 volts. The intensity is measured as the ratio of the current entering the Faraday box collector to the total current reaching the target. The opening in the Faraday box subtends about .03 of unit solid angle to the spot under bombardment. The retarding potential between box and target for the curve, Fig. 1, is 135 volts, so that only electrons that have lost not more than 10 per cent. of their initial energy are caught. The effect of bringing the retarding voltage nearer the bombarding volt age is to reduce the number of electrons caught and to increase the sharpness of the pattern. On decreasing the bombarding potential without altering the ratio of retarding to bombarding potential the lateral maximum moves away from the primary beam toward the plane of the target, and the ratio of the intensity of this maximum to that of the other becomes greater. In attempting to interpret these results we have been led to consider the scattering of electrons by a positive nucleus of limited field, one for which the central force on an electron is Ee/r2 for values of r less than and zero for all values of r greater than p. ρο Such a field would exist for a concentrated positive charge E surrounded by a spherical shell of uniformly distributed charge -E and of radius p. The field of a system comprising a central positive nucleus of n electronic charges surrounded by n electrons uniformly distributed over the surface of a sphere of radius P will also be roughly of this nature, provided n is not too small. Neglecting the change of mass of the bombarding electrons while traversing the field within the shell it turns out that when such a system is under random bombardment by electrons approaching on parallel lines, the number of these emerging per unit solid angle in a direction making an angle with the path of the incident beam is given by 28-1 2 one half the scattered electrons become more and more concentrated in and near the direction = 0, the intensity in this direction being infinite for ẞ=1/2. For values of ẞ less than 1/2 the distribution curves for the range 1 > B > 1/2 are identically repeated, the distribution approaching uniformity in all directions as ẞ approaches zero. For a neutral system of two or more concentric shells the distribution will be broken up into various beams or lobes corresponding to groups of electrons whose trajectories pass through one, two or more of the shells. In particular a system comprising two shells will give, in an appropriate range of bombarding potentials, distribution curves similar to that shown in Fig. 1. All of the main features of the distribution curves so far observed for the scattering from nickel seem reasonably accounted for on the supposition that a small fraction of the bombarding electrons actually do penetrate one or more of the shells of electrons which are supposed to constitute the outer structure of the nickel atom and, after executing simple orbits in a discontinuous field, emerge without appreciable loss of energy. If the theory of the scattering here proposed proves to be the correct one, there seems no reason why the careful study of such distribution curves as shown in Fig. 1 may not reveal much of interest concerning the disposition of electrons within the atom. It is (( 28 − 1)2 ( 1 + cos x) + (1 — cos sv)) hoped to report more extensively on this work and van Haagen's3 recent determination of the atomic weight of boron, 10.900, indicates the proportions of these isotopes as nearly 1 to 9, the relative intensities of the positive-ray spectra point to a considerably larger proportion of the lighter isotope. Since we have redetermined the atomic weight of boron by analysis of the chloride and bromide, and have obtained a result more nearly in accord with Aston's experiments than with those of Smith and van Haagen, it seems advisable to state the outcome of our preliminary experiments, without waiting for the completion of the investigation. Boron was obtained by reduction of boric oxide with an excess of magnesium and extraction with either hydrochloric or hydrobromic acid. To prepare the chloride, dry chlorine was passed over the boron at about 700°. To prepare the bromide, helium saturated with bromine nearly at the boiling point of the latter substance was passed over boron at 700°. After removal of the excess of halogen with mercury both halides were repeatedly distilled with the use of Hempel fractionating columns in sealed all-glass vessels, with complete exclusion of air. Quantitative testing even before the completion of the fractionation showed the absence of silicon halides which constituted the worst impurity. Material was collected for analysis in sealed glass bulbs. Analysis was effected by comparison with silver in the usual way. The results of the analysis of the chloride agree with those of the bromide in yielding the value 10.83 0.01 for the atomic weight of boron. On the assumption that constant boiling mixtures with the halogen acids were not formed and that no separation of the eight possible combinations of two isotopes of both boron and chlorine took place, this new value for the atomic weight of boron indicates the proportion of the heavier isotope to be about five times that of the lighter. G. P. BAXTER, A. F. SCOTT HARVARD UNIVERSITY 3 Car. Inst. Pub., No. 267 (1918). 4 Aston, loc. cit. THE AMERICAN CHEMICAL SOCIETY (Continued) Isomeric alkyl-pyrimidines and color phenomena: ARTHUR W. Dox AND LESTER YODER. A series of alkyl-diketo-pyrimidines was prepared by condensing alkyl-malonic esters with amidines. In this series four types of isomerism occur, of which the following derivatives are examples: (a) 5-butyl and 5,5-diethyl; (b) 5-phenyl-2-methyl and 5-methyl-2-phenyl; (e) 5-isoamyl-2-phenyl and 5,5-diethyl-2-p-tolyl; (d) 5-allyl and cyclobutane1,5-spiro. Some of these derivatives are white, others are bright yellow. Color is dependent upon the presence of an aromatic group on the 2-carbon and a labile hydrogen on the 5-carbon. The latter makes possible a rearrangement into a tautomeric enolic form with three double linkages in the ring. The only exception to the color rule is the spiro derivative, which is yellow. Spectroscopic examination of a typical yellow derivative showed an absorption band in the violet between 260 and 330 μμ. An octet formula for benzene: ERNEST C. CROCKER. Proposed formula is ring of six carbon atoms acting as single complex atom. Individual carbons bonded together by sharing single pairs of electrons (single bonds), with hydrogens associated with pairs of electrons, as usual. The six excess electrons of system are " aromatic "'electrons, and vibrate between the carbons, in unison. "Aromatic '' electrons cause two distinct patterns, o.p., and m., according to the influence of substituents in the ring. The theory accounts well for mono, di and tri substitution products of benzene. It accounts for aromatic structure in general; particularly thiophene, furane, pyrrol, naphthalanene, and anthracene. Diisopropylhydrazine. J. R. BAILEY, W. A. NOYES AND H. L. LOCHTE. Diisopropylhydrazine can be easily prepared by treating a solution containing acetone, hydrazine chloride, gum arabic and colloidal platinum with hydrogen under pressure. Dimethylketazine (CH,),C: N-N:C(CH,), is at first formed and this is reduced to diisopropylhydrazine, (CH3)2CHNHNHCH(CH,). The latter is a monacid base, which forms stable salts. The free base is very easily oxidized, even by exposure to the air, probably forming an azo compound. The investigation of this and other relations will be continued. The chlorination products of formanilide: W. LEE LEWIS AND R. S. BLY. When formanilide is chlorinated in the presence of chlorides of sulfur |