1907] CURRENT LITERATURE 349 
strongly monomodal curves, the former with the mode at 13, the latter with the 
mode at 5, thus supporting Lupwic’s view as to the prevalence of the terms of 
the Fibonacci series as modal numbers in the variation curves for ray-flowers. 
(c) Individual variation is shown in several tables, including about 70 individuals 
of Anthemis and 4o individuals of Achillea. These show that considerable 
differences occur, but so far as can be seen each habitat studied shows about the 
same series of individual variabilities. (d) Collections of the same species from 
different localities show little place-variation except in one instance. The number 
of rays in the heads of Achillea are only four in a much larger percentage of the 
heads at Ames, Iowa, than at three other localities from which material was 
secured. (e) Collections of Anthemis in a single habitat near Ames, in August 
and again in September, show a somewhat higher average number of rays at the 
time of the earlier collection, in this agreeing with results reached by the reviewer 
in the genus Aster. e modal number was 13 in both collections. It is stated 
that greater variability was observed in the ray-flowers of Anthemis growing in 
nich soil than elsewhere, but the data upon which this statement rests do not appear 
in the tables. 
The number of seeds of the lotus, Nelumbium luteum, as it occurred in the 
western part of Lake Erie in 1902, has been studied statistically by PEARL,® who 
finds that the distribution of the variates gives a good example of the normal 
curve with the following constants: average 24.874+0.078, standard deviation 
4.33940.055, coefficient of variation 17.44540.162. A complete analysis of 
the curve is made and its departures from the theoretical are accounted for as 
due to random sampling. An interesting numerical relation bearing upon the 
distribution of fecundity is pointed out. It appears that capsules bearing fewer 
than the median number of seeds produce only 42.96 per cent. of the total number 
of seeds, while the remaining 57.04 per cent. are borne by plants which severally 
produce more than the median number. This fact is conditioned by the normal 
form of the curve. The maintaining of the normal curve would seem to require 
either that the inheritance of seed-number is inappreciable or that there is unequal 
elimination of the seeds coming from large and from small heads.—G. H. SHULL 
Development of the seed of Alsinoideae.—Miss Grsss,*° in working over the 
development and germination of the seed in the subfamily Alsinoideae, of the 
Caryophyllaceae, has shown that in these features there is a marked uniformity 
in this group, with evidences of a progressive specialization showing a trend from 
the Sperguleae to the Alsineae. The latter is indicated especially by an increase 
in size, and presumably in efficiency as an absorbing organ, of the suspensor; and 
also by the increased specialization of the inner layer of the outer integument, 
which is believed to serve as a water jacket in the developing seed. The structure 
9 Peart, R., ate in the number of the seeds of lotus. Amer. Nat. 40:757- 
768. figs. 4. 19 
to Gis, L. . Notes on the development and structure of the seed in the Alsinoi- 
deae. Annals of Botany 21:25-55. pls. 5-0. figs. 4. 1907- 
