May 31, 1912] 
quent in the southeast quadrant of a depres- 
sion. Professor Waldo (“ Elementary Meteor- 
ology,” p. 221) says: 
In the eastern and northern parts of the United 
States the area of maximum rainfall lies southeast 
of the center of the cyclone, and usually at a dis- 
tance of about 300 miles from it; but the distance 
varies greatly in individual instances. 
In New England, however, the rainfall seems 
to be heaviest in the northeast quadrant of a 
cyclone, the precipitation accompanying a 
“northeaster,” in which the center of the de- 
pression remains south of the observer during 
its eastward movement, is usually heavier than 
that of a storm whose center passes down the 
St. Lawrence Valley. Dr. Shaw (“ Forecast- 
ing Weather,” p. 206) seems to be of the 
opinion that the precipitation is most abun- 
dant, or at least is most frequent, in the “ left 
front of the depression.” In the latest dis- 
cussion of the problem, Mr. F. J. Wardale 
(Symonds’s Meteorological Magazine, Febru- 
ary, 1912, p. 8) concludes that when a depres- 
sion erosses England the bulk of the rain 
falls in subsidiary eddies on its northern side, 
the region of heaviest precipitation as the 
storm advances being “a broken band parallel 
to or gradually diverging from the central 
track on its left side.” He believes that these 
eddies, too shallow to be evidenced on the 
meteorological charts, have a counter-clock- 
wise orbital movement around their primary, 
at the same time sharing its forward move- 
ment. These eddies, in which the heavier 
downpours occur, pass quickly over and hence 
give slight precipitation to a place south of 
the storm track, for there they are accelerated 
by the general forward movement. For a 
place to the north of the center, the orbital 
velocity of the secondary is subtracted from 
the general forward movement, consequently 
the eddy passes slowly and in some cases might 
remain stationary, resulting in prolonged and 
therefore heavy rain. These eddies, he be- 
lieves, are formed at intervals during the prog- 
tess of the eyclone, thus accounting for the 
patchiness of the band of high rainfall. Mr. 
Wardale’s suggestions are well worth the seri- 
ous consideration of forecasters, since unno- 
SCIENCE 
873 
ticed secondaries and trough-like isobars have 
often resulted in heavy precipitation when 
fair weather was expected. His conclusions 
agree closely with those previously reached by 
Mr. W. G. Reed in a study of the cyclonic 
distribution of rainfall in the United States 
(Monthly Weather Review, October, 1911, p. 
1609). 
NEW BOOKS 
Amone the books which have recently ap- 
peared are: (1) “ Meteorology,” by W. I. Mil- 
ham. New York, The Macmillan Co. §8vo. 
549 pp. $4.50 net. (2) “ Weather Signs and 
How to Read Them, for Use at Sea,” by W. 
Allingham. Glasgow, J. Brown & Son. 117 
pp. 2s. net. (8) “The Sun,” by C. G. 
Abbot. New York, D. Appleton & Co. 448 
pp. (4) “ Uber die Helligkeit des Himmels 
in der Nahe der Sonne,” by H. Diercks. Kiel, 
Liidtke & Martens. 48 pp. (5) “Uber die 
Gesetze der Wirmestrahlung,” by W. Wien. 
Leipzig, J. A. Barth. 21 pp. 1 Mark. 
Among the books soon to appear are: (1) 
“The Meteorology of the Globe,” by W. N. 
Shaw. (2) “Clouds,” by C. T. R. Wilson. 
(38) “Structure of the Atmosphere,” by OC. J. 
P. Cave. (4) “ Weather-science,” by G. F. K. 
Lempfert. (5) “ Radiation,” by P. Phillips. 
Professor R. DeC. Ward has begun the prepa- 
ration of a “Climatology of the United 
States.” 
ANDREW H. PALMER 
BLuE HILL OBSERVATORY, 
May 1, 1912 
SPECIAL ARTICLES 
THE ORIGIN OF ERYTHROCYTES BY A PROCESS OF 
CONSTRICTION OR BUDDING 
It is gradually becoming apparent that in 
the solution of some of the more intricate 
morphological problems, involving phenomena 
of development and histogenesis, recourse will 
be necessary, in part at least, to data other 
than those obtained from fixed and stained 
preparations alone. Sabin, 705, in a discussion 
of evidence from fixed histological material 
bearing on certain questions regarding lymph- 
