BIOMETRIC DATA IN CRINUM LONGIFOLIUM. 83 
TABLE V. 
VARIABILITY IN NUMBER OF OVULES AND NUMBER 
OF SEEDS PER LOCULE (1) AND PER FRUIT (f). 
: Authority and Ovules Seeds 
ace O ication per Fruit per Frui 
Species Place of Publicati Frui Fruit 
Sanguinaria ipa Harris, Biometrika. 7 : 309-310.|30.28 — 30.66/32.66 — 39.26 
bina Pseud-acacia.. no 6 : 442, 18.69 45.45 
Hibiscus militaris (1).. “ = Si 55-646 Hh” Steere ene 20.70 
lasiocarpos (1) “ ae BB 508 occ en are 32.88 
ie Moscheutos (1) ia ie 8 66-645. Vleet mee. 18.93 
“ Trionum OE ; ss 8 : 55-64. 17.49 22.58 
a syriacus (CoP. a me 8 : 55-64. 8.79 — 10.21/27.65 — 46.78 
t ) oa “ o 8: 55-64. [11.15 — 13.80/34.91 — 54.35 
‘* Manihot (1) .. oe He 8 : 55-64. 13.27 35.55 
Celastrus scandens ....|Harris, Rep. M. B.G. = : ire or uae 31.42 
Cytisus scoparius......;Harris, Amer. Nat. 48:354. |........... 35.24 
E moe Rd aes |Pearson, Phil. Trans. i i8t: oT ee oes 36.78 
Cercis canadensis...... ‘Harris, Bot. Gaz. 19.26 26.71 
Nelumbium luteum... ./Pearl, Amer. Nat. 40 : 760. Pale ates oxevat 17.45 
Lotus corniculatus.... Pearson, Phil. Trans. A. 197 :367.| ........... 46.38 — 52.14 
Lathyrus odoratus..... s - “ “197 : 367.|12.65 — 13.39)27.67 — 34.96 
: sylvestris... . a . i 167 < 367. 6.79 32.91 
Victa Fabasf.c02.%. - a Be IGT. 2. 3G 7e | cecre cetates ans 33.50 
Vicia hirsuta.......... fe tr : ‘ 197 : 367 30.88 38.32 
In percentage variability of number of seeds per fruit, 
Crinum stands far above any of the considerable number of 
species given in this table. Only in Staphylea trifolia® have 
I found coefficients of variation at all comparable in magni- 
tude. These two species agree in having very skew distri- 
butions of number of seeds per fruit, but differ widely in 
range and mean, the mean number of seeds per fruit in 
Crinum falling in the neighborhood of the maximum num- 
ber in Staphylea. When a sufficiently wide series of these 
exceptional cases are gotten together, discussions of the 
underlying morphological and physiological factors to which 
their peculiarities are due can be profitably undertaken. 
3. Variation in Seed Weight. 
An investigation of the size of the exceedingly variable 
seeds of this form presents two serious difficulties. They 
are so irregular that weighing is the only feasible method 
5 For the data on which these statements are made see four 
papers: Biometrika 7 : 453-504. 1910.— Zeitschr. f. Ind. Abs.-u. 
Vererbungsl. 5 : 173-188. 1911.—Beih. Bot. Centralbl. 28:1-16. 1912. 
—Bot. Gaz. 53 : 204-218; 396-414. 1912. 
