THE AUSTRALIAN NATURALIST. 14] 
LAMBERTIA FORMOSA. 
A PARTIAL STUDY IN VARIATION. 
By H. Tasman Lovell. 
The character examined in this study was the number of 
component flowers. In all 354 examples were dealt with in 
one day, the locality being French’s forest and the time the 
early part of February, 1912, at the close of a dry spell ex- 
tending over seven or eight weeks. The results may be 
arranged in a frequency table as follows— 
No.of Component Flowers|6 |> 7 | 8 | 9 | 10 | lL 
UO gee ¢ iy ae a 
Tt will at once be noticed that 7 is the normal. Tt will 
also be remarked —and this would not easily be revealed 
to the casual observer—that there is a fairly wide range 
of variation in this character, especially above 7. 
The exact extent of the variation within the limits 
of the number of cages examined may now be figured out 
thus: yer 
No. of 
Component Flowers. Frequency. 
6 xX SS 6 
7 x 346 A, 
8 x 1 = 8 
9 x 3 = 27 
ll x 1 — ll 
354 2494, 
Therefore the arithmetic mean = #7045 Compt. Flowers. 
The deviations of 6, 7; 8, 9, 10, 11 from 7.045 is now 
taken : 
Deviations. Frequency. 
aan a l = 1.045 (neglecting the 
— 045 x 346 = 15.070 negative sign) 
955 x 1 == 955 
1.955 x 3 = 5.865 
2.955 x 2 = 5.910 
3.955 x i — 3.955 
354 33.800 
Thus the mean deviation == ‘095 roughly ‘1 compt. flowers, . 
Therefore while 7.045 may be viewed as the type, .095 
represents the average extent of deviation from the type; 
i.e., of variation. This will scarcely be regarded as a con- 
siderable variability, yet it may be of interest to those wha 
haye not previously observed its presence at all. 
