constantly changing both in position and in intensity. Across these streamers, pale green pulsating clouds drifted, in general from north to south, but occasionally assuming a spiral form around the zenith. They attained their maximum brightness near the zenith where they were especially conspicuous on account of their almost instantaneous changes in intensity.

Bright colors were not noticed during the evening, but after the moon set about midnight, pale reds and blues appeared on the edges of the streamers and clouds. The display continued at intervals throughout the night. It was not more conspicuous in the north than in other directions.

The aurora was undoubtedly due to the very large group of sun-spots which had just passed the center of the sun's disk.

MIDDLETOWN, CONN., May 15, 1921

FREDERICK SLOCUM

AGAINST a clear, moonlit sky, a brilliant auroral display was observed at Ames, Iowa, between 8:30 and 10:30 P.M. on May 14. The arch which was visible throughout this time except at short intervals, formed in our magnetic north and extended about 15 degrees above the horizon.

As the streamers, which were predominantly white, grew in number, in length and in extent along the horizon, they converged to a focus at a point somewhat variable in position but approximately 15° south and 5° west of the zenith, which point, the magnet zenith, became a center of radiation for the streamers. Abouth 15 minutes before the maximum development of the display, streamers of red were seen to rise from the horizon a few degreees south of east and to extend through the radiant center to the horizon about the same distance north of west, forming an arch along a magnetic parallel.

The maximum degree of brilliancy was attained at 9:27, when the streamers from a large coronal area formed about the magnetic zenith extended to the horizon in all directions, lighting the entire heavens. The radial

streamers were visible within a few degrees of the moon, which had just passed the first quarter. At this time a dark area a few degrees west of south on the horizon closely resembled an auroral arch, but a definite segment of a circle like that on the northern horizon could not be discerned.

The shades, tints and hues, changeable and increasing from the beginning of the observation, now became more distinct and all of the primary colors appeared in varying degrees of intensity. Reappearing intermittently, the colors gradually faded away during the remaining hour of the display.

DEPARTMENT OF GEOLOGY, IOWA STATE COLLEGE

JOHN E. SMITH

RUSSIAN GEOLOGISTS

THE sad fate that has befallen many of the leading Russian geologists and mineralogists constitutes a gloomy chapter in the history of these sciences. From particulars gathered by Professor Sederholm, of Sweden,1 and confirmed by a personal letter of March 30, 1921, received from Dr. Cornelius Doelter of Vienna, the following data have been secured.

Of some seventy Russian specialists in these fields eleven are dead. Of these, there died in Petrograd the well-known Professors Inostranzer, Fedorov (who died of hunger), Karakash, Derzhavin and Kasanski. Professor Sokolov died in Moscow. Professor Armasevski was shot in Kiev, as were Professors Samiatin and Mitkyevich in Petrograd. Stopnjevich died of smallpox and Snertkov of hunger-typhus. Baron Rebinder committed suicide, and it is reported that Faas is seriously ill.

The president of the Petrograd Academy of Sciences, and former director of the Geological Institute, Alexander Karpinsky, the Nestor of Russian geologists, who is now eighty years old, lives with his three daughters, a son-in-law, and his grandchildren, in a cold kitchen, and suffers great deprivation be

1 Given by Professor Mohr in Centralblatt für Mineralogie, Geologie und Paläontologie, 15 Jan., 1921, No. 2, p. 60, from the Svenska Dagbladet.

cause of the lack of necessaries of life, although his scholars, with touching zeal, bring everything they are able to secure.

Professor Andrussov and the Academician Vernadsky were fortunate enough to make their way to South Russia, and it is stated that the latter seems to be in good circumstances, as he has founded a new academy of sciences in Kiev, and also a new university in Simferopol. About ten of these scientists fled across the frontier, and escaped to Finland or Poland, or even to America or Japan, and perhaps as many more are scattered through Siberia. From fifteen to twenty are probably in the Russian provinces, but only about ten are managing to exist in Petrograd.

The famous mineralogist Fedorov, whose death from hunger we have noted, was the first to proclaim, at a meeting in St. Petersburg, in 1889, the great advantages that would result from the application of the principle of the theodolite to goniometrical researches. Four years later, in 1893, he published his classic work, "The theodolite method in mineralogy and petrography." 2

G. F. K. AND E. T. W.

SCIENTIFIC BOOKS

The Coccida. Tables for the Identification of the Sub-families and Some of the More Important Genera and Species, together with Discussions of their Anatomy and Life History. By ALEX. D. MACGILLIVRAY. Scarab Company, Urbana, Ill., 1921. Pp. viii+502. $6.00.

Entomologists who have been acquainted with Dr. MacGillivray's thoroughgoing studies of the scale-insects have long awaited the appearance of this volume. The material was originally collected for the use of students in the identification of Coccids. Prepared in its first draft about fifteen years ago, it has been greatly extended, modified and revised as it was being tested out in laboratory and class work.

2 W. W. Nikitin, "La Méthode universelle de Fedorov," French transl. by Louis Dupare and Véra de Dervies, 2 vols. Geneva, Paris and Liège, 1914, Vol. I., p. 6.

In no group of insects of equal importance is so much reliance in systematic work placed upon minute structural details. Many a would-be student of the group has been deterred by difficulties of preparation of material and by lack of a comprehensive discussion, in English, of the morphology. To such the volume will prove a veritable boon.

A chapter is devoted to details of technique. In this are considered necessary equipment, tools, clarifying and the various stages in the making of permanent preparations. This is followed by a chapter on the external anatomy of the Coccidæ. The "great number of species and the dearth of usable characters, because of the simplification of their external form and structure, makes it necessary to employ every available structure." In spite of the lack of illustrations, the discussion and definition of these structures is clear-cut.

Figures were omitted for pedagogical

reasons.

The tables were prepared primarily for the use of students. Those who have had any experience in teaching know that most students will not undertake anything they are not forced to do. The omission of figures makes it necessary for them to study their specimens rather than figures.

The author's detailed studies on the phylogeny of the different subfamilies, genera and species have led him to the establishment of a considerable number of new genera, which are here defined for the first time. The group as a whole he divides into seventeen subfamilies, which have been treated in ascending order. A tabular arrangement indicates what the author believes to be the relation of these subfamilies, and the scientific and vernacular names that have been applied to them.

an

Dr. MacGillivray has done a real service in making the materials of his course available to a wider audience. The book will prove indispensable to future students of the Coccidæ. WM. A. RILEY

The Soils and Agriculture of the Southern States. By HUGH HAMMOND BENNETT, of

the Bureau of Soils, United States Department of Agriculture. The Macmillan Company, New York. 1920. Pp. xviii+ 399. Illustrations: 56 plates, general soil map of the Southern States (frontispiece), and four additional maps.

This book departs from the usual trend of books on soils in that instead of dealing with the properties and nature of soils in general the author describes the origin, geographic distribution, physical characteristics, agricultural adaptations and management of all the important soils occuring in the area under discussion. The states included in the work are those lying south of the north boundaries of Delaware, Maryland and West Virginia, south of the Ohio River, and south and east of and including Missouri, Kansas and Texas.

In the introduction the author explains the division of the country under consideration into soil provinces and subordinate soil regions, and describes the United States Bureau of Soils system of classification and nomenclature of soil series and types. The introduction further takes up the geographical distribution and in general the adaptation to different soils of the various crops grown in the South; and the influence of climate on soils and crops.

farmers and especially to those contemplating
settling in the south.
WM. B. COBB

DEPARTMENT OF AGRONOMY,
LOUISIANA STATE UNIVERSITY

SPECIAL ARTICLES

AN AGE-COMPUTING DEVICE

1. In a recent issue of SCIENCE (1920, No. 1336, pp. 134-135), Dr. Slonaker describes a device for the simultaneous determination of the ages of two individuals at different times in their lives, involving the use of a calendar in which the days are numbered consecutively throughout the year. The present device obviates the need of the calendar and the need for resetting for dates in different years. used with reference to human beings, two accessory scales aid in determining in years the age of an individual at different episodes in his life, when his present age and the years in which the episodes occurred are known,

The general geography, topography, geology and agriculture of each soil province and its subordinate soil regions are discussed, followed by detailed descriptions of the individual soils. These descriptions include the location, physical and frequently chemical characteristics, topography, drainage and crop adaptation of each soil, and methods of soil management and fertilization which actual farm practise and experimentation have proven to be most effective.

Four appendixes include discussions of the meanings of terms used in soil classification, chemical analyses of representative southern soils, a bibliography of important publications on soils and related subjects, and statistics bearing on some of the important farm products of the southern states.

The book is valuable not only to students and agricultural investigators but also to

able pointer (P), bearing a median hair line throughout its length, is placed between the two disks and consists of a strip of transparent celluloid pivoted at one end on the eyelet and extending with a sharply pointed tip to the periphery of the larger disk.

3. Scale A, marked on the acircular sheet exetrnal to the circumference of the larger disk, is comprised of 365 equidistant radial lines numbered by tens from 0 (index a) to 360, representing successive days. Every seventh line if accented by a dot and numbered serially from 0 (index a) to 52, indicating weeks. An outer scale divides the circumference into twelve equal parts, each in turn divided into fourths, and is numbered from 0 to 11 to indicate months.

4. A coinciding scale (B) extends along the circumference of the larger disk but is gradu

Scale B.

(1) Set P at March 14 (March 13 + 1). (b) To find the age on September 17, 1920, of an individual born March 23, 1867 (if the entire

1 Although the reading in total days is invariably accurate (not considering leap year), readings in months and days may deviate from the customary calculation by not more than 3 days (because of the inequality of the calendar months). After the number of whole months is read, the number of days in the remainder may be determined accuScale A.

Scale B. (1) Set P at November 3.

ated by months, the last line of each being prolonged centrally and the intervening spaces labelled with the names of the corresponding months. The index (b) of this scale is an accentuated line representing December 31.

5. An inner scale (C) on the same disk lies around the circumference of the smaller disk and, with 0 (index c) on the same radius as index b, divides the circumference into 100 equal parts. The lines are numbered by fives from 0 to 95 and represent successive years.

6. Individual years are represented by scale D on the smaller disk. This scale coincides with scale C and is numbered from '05 to 1900 (equivalent to any even hundred year). Attention is called to leap years by a dot on every fourth line from that of '04 to that of '98, inclusive. An erasable mark accentuates the present year.

OF THE CIRCULAR SLIDE RULE

steps are taken is indicated by the numbers in the schematic diagrams.)

(3) At P, read the number of days (weeks or months). (2) At index a, set earlier date. November 3.

(3) At P, read 166 days (23 weeks, 5 days, or 5 mos., 14 days).1

(2) At index a, set May 21.

follows:

Example: For the interval from July 5, 1919, to March 13, 1920.

(3) At P, read 252 days (or 8 mos., 9 days). (2) At index a, set July 5. process were to be carried out):

rately by rotating scale B, carrying P (already set at the later date) in the clockwise direction past index a until the first day represented by the same figure as the earlier date is at index a. Then, at P on scale 4, read the number of days in the remainder. Thus, in the same example (from May 21 to November 3):

(5) At P, read 13 days.

(4) At index a, set October 21.

In step (3) it is seen that P lies outside the sector extending clockwise from index a to index b and accordingly 1 is subtracted from the reading in step (5). In step (2), correction is made for the occurrence of February 29 in the interval, and by steps (7) and (8) is obtained the correct reading of days in the remainder of the reading in months (step 6). The age, then, is found to be 46 years, 7 months, 16 days.

(c) To determine the period of liability in industrial insurance:

A new index (a') may be marked on scale A at a point representing, in the counter-clockwise diScale A.

(5) Read 47-1, or 46

years.

(4) Opposite '20.

rection from index a, the number of days in the legal interval between the date of injury and the first day of compensation. For example, in Massachusetts, where compensation starts on the 15th day of incapacity, index a' is marked at 351. With the day of injury (scale B) set at index a', the number of days of compensation may be read directly on scale A at the day on scale B representing the date of termination of disability.

Example: For what number of days shall compensation be paid to an employee injured on April 9 and incapacitated until August 51

(3) At P, read 104 days (or 14 weeks, 6 days).

(2) At index a' (351), set April 9.

Example: What per cent. of a year is an interval of 211 days?

(2) Set P at 211.

(3) At P, read 58 per cent.

Example: What is the equivalent in days of .36

(3) At P, read 131 days.

year?

(2) Set P at 36.