Certain phases of counter technique are reviewed. The conditions necessary to make and operate reliable proportional counters are discussed. It is shown that certain mixtures of gases, notably argon-carbon tetrachloride and argon-boron trifluoride, when used in counters give these the property of producing large pulses over considerable voltage ranges below the threshold. The size of the pulse (in volts) will be roughly proportional to the amount of ionization produced in the counters by the passage of the ionizing particle being studied. Because of this proportionality, these counters can, for example, distinguish between alpha-and beta-rays over considerable voltage ranges without the necessity of much vacuum-tube amplification. Neutron counters are made using B${\mathrm{F}}_3$ gas, the ${\mathrm{B}}^{10}(n,\alpha ){\mathrm{Li}}^7$ reaction producing the alpha-particle which is detected by the proportional counting procedure described. Flights have been made with instruments carried by small balloons and using single Geiger counters. The data were transmitted from the balloon to the ground station by automatic short wave radio. Such flights show no diurnal effect in the cosmic radiation at great elevations (½ meter of water) as large as four percent of the total cosmic-ray intensity. Further, the flights show that no increase in the ionization as great as two percent of that due to the cosmic-ray intensity takes place at an altitude of 20 km, which can be correlated with a solar flare, although this same flare produced abundant ionization below 80 km and caused a radio fade-out. The absorption coefficients for the flare radiation are calculated. Flights were made with neutron counters and the radio-balloon method, which yielded measurements of the neutron intensity as a function of elevation. It is found that the neutrons increase rapidly with elevation, the rate of increase being faster than that of the total intensity of the cosmic radiation, and approximately equal to that of the big bursts. The production of neutrons in the atmosphere by the cosmic radiation is discussed, and a production cross section of ${10}^{-35\mathrm{}}$ sq. cm is computed.

The first section of this report contains a brief summary of those experiments which have made substantial contributions to our knowledge of the nature and mode of absorption of cosmic-ray particles. Included is a discussion of the difficulties which inhere in the attempts to interpret the observed cosmic-ray phenomena in terms of the assumption that the cosmic-ray particles are practically all protons and positive and negative electrons. These difficulties are resolved by demonstrating that in the same momentum range there exist two types of particles, one of which is highly absorbed in a heavy material, mostly through radiation, while the other is relatively penetrating. Behavior of the first kind is typical of shower particles in general, and of all single negatron secondaries, which are presumably produced in elastic collisions with atomic electrons. This group is therefore to be identified with electrons. As it can be shown that particles of the penetrating group cannot be of protonic mass, it follows that they must be neither electrons nor protons. The simplest assumption is that their distinguishing property is a mass intermediate between the electron and proton. Confirmation of this view is found in the observation by other writers and by ourselves of particles whose range, ionization and curvature relations are such that they demand a mass in this region. The best range and ionization data seem to give mass determinations in the neighborhood of 200 electron masses. A discussion is given of certain difficulties which exist in the interpretation of cosmic-ray particles as electrons and mesotrons of a unique mass.

Chree's method of investigating recurrence phenomena in international magnetic character-numbers has been applied to the daily mean cosmic-ray intensities, as given by the Carnegie Institution's Model C cosmic-ray meter at Teoloyucan, Mexico, somewhat more fully than in the study reported previously by Mr. P. S. Gill. The five days of highest cosmic-ray intensity (high$I_0$) were selected from each of seventeen calendar months, and to each of these days was assigned the number $n=0\mathrm{}$. The $I_0$ for each of these 85 high $I_0$ days was tabulated, also the $I_0\text{'}\mathrm{s}$ for the consecutive days following (and preceding) each of these selected days, up to $n=+131\mathrm{}$ and $n=-132\mathrm{}$ The average of the 70 to 85 $I_0\text{'}\mathrm{s}$ tabulated for each $n$ value, when plotted against $n$ gives a curve analogous to Chree's "positive pulses." A similar "negative-pulse" curve was obtained by selecting the five lowest $I_0$ days per month to be the "zero" days. Chree's material (20 years' data) gave sharp pulses at $n=\pm 27,\pm 54,\pm 81,\pm 108$ with the curves relatively flat between pulses, and pulse-amplitudes decreasing with increasing $n$. The cosmic-ray curves have maxima (or minima) at intervals of 27 to 29 days and an amplitude of about 0.3 of one percent. Unlike Chree's curves, however, they are markedly sinusoidal, the maxima being much less sharp and showing very little damping, if any. Somewhat sharper maxima are shown by the difference-curve (positive-pulse minus negative-pulse), and these average between 27 and 28 days apart.

This high correlation, on the average, between the intensity of cosmic-ray ionization at one time and that about one synodic rotation of the sun later is, from this evidence, a recurrence or interval phenomenon, not necessarily a true periodic phenomenon. The possibility that the latter, or at least a quasi-periodicity, is involved is suggested by the persistence in pulse-amplitude.

Some results are given from the records of three different ionization-chambers, two under 110 cm Fe, one under 12 cm Fe. A daily influence generally parallel to temperature is found in the thin shielded instrument, not in the thick shielded one, and only in winter. It is found that the short periodic variations are not followed by the thick shielded instruments. The long periodic variation with temperature is generally antiparallel and moreover temperature and pressure variation is the same in all three instruments. A barometric coefficient of -6.4 percent per cm Hg is found and a temperature coefficient of -0.21 percent per degree C. The high value found for barometric variation is not explained, but seems not improbably to be related to the high values sometimes found at large angles from the vertical.

SOME sixty active investigators of cosmic rays, including representatives from most of the world's leading research groups engaged in this study, attended a congress held under the auspices of the University of Chicago from June 27 to 30, 1939. The attendance of those interested in the subject numbered about 300.

The papers presented at the symposium, either in full or in abstract, together with a part of the unusually active discussion, are published in this number of the Reviews of Modern Physics.

An editorial problem has arisen with regard to the designation of the particle of mass intermediate between the electron and the proton. In the original papers and discussion no less than six different names were used. A vote indicated about equal choice between meson and mesotron with no considerable support for mesoton, barytron, yukon or heavy electron. Except where the authors have indicated a distinct preference to the contrary, we have chosen to use the term mesotron.