Whitby. — Variation in Hevea Brasili 'crisis. 321 
it may not infrequently be observed that the latex left on the tapping cut, 
after a tree has ceased to drip, darkens rapidly at the upper end of the cut, 
where there is a strong tendency for the tapping to fall short of the proper 
depth. But, quite apart from cases of rapid discoloration which could be 
avoided by deeper tapping, it was observed that in some cases rapid 
discoloration appeared to be characteristic of the tree. One tree, which 
showed this feature very strikingly, was kept under notice for five years, 
and was always seen to produce rapidly-discolouring latex, which gave an 
exceptionally strong reaction for peroxidase. 1 
(/;) Tendency to rapid coagulation. In the case of two trees in the 
population with which this communication more particularly deals, a 
tendency to exceptionally rapid natural coagulation was observed in the 
latex, and persisted over a considerable period. 
(c) A marked cream-straw colour appeared to be characteristic of the 
latex from a small percentage of trees. (This colour is to be distinguished 
from the transient yellow colour which latex from a new cut often has 
before wound response has established itself.) The latex in question has 
a noticeably ‘ rich ’ appearance. It was not, however, found in general to 
have a higher rubber content than the average. 
Postscript. — Since the above was in print, A. A. L. Rutgers 2 has 
given some interesting data which are in full accord with the observations 
(made in 1913), recorded in the present paper, as to the practical constancy 
of the yield from individual trees and as to the character of the distribution 
of the total yield over a group of rubber trees of equal age. Data for three 
areas are recorded by Rutgers. 3 As a result of comparing the classifica- 
tion of the trees made at the outset with that made ten to twenty months 
later, he concludes that * good trees remain good, poor trees remain poor \ 
From his data for a group of 1,467 trees (mean daily yield per tree, 9-1 g.) 
it may be calculated that, at one extreme, 9-0 per cent, of the trees (trees 
giving 17 g. or more) produced 23*9 per cent, of the total yield, and, at the 
other extreme, 27-3 per. cent, of the trees (trees giving 0-5 g.) produced 
only 7*2 per cent, of the total yield. 
1 Cf. Whitby : Roll. Zeit, 1913, vol. xii, p. 147. 
2 Selectie en Uitdunning. Archief voor de Rubbercultuiir, 1919, 3, 105-123. 
3 It may be remarked that in the case of two of these areas the volume of latex, not the weight 
of rubber, per tree was the quantity measured. Also, in the instructions with regard to the 
collection of yield data for the purpose of selecting trees for thinning-out operations, it is directed 
only that the volume of latex shall be determined. It would seem, however, from the results 
recorded in the present paper on the ‘ V ariation in Rubber Content of Samples of Latex ’, that this 
simpler procedure does not give exact results, and that it is necessary to weigh the rubber. 
