276 



The original weight of the thousand seeds given above is 4,035 

 grams; another thousand from the same trees weighed when fresh 

 4,122.5 grams; and a third thousand 4,223 grams. This gives an 

 average of 4,126'83 grams; or, in other words, 1,000 fresh seeds from 

 untapped trees weigh 91 lb. This agrees exactly with the result obtain- 

 ed by Mr. Carruthers five years ago from fresh seeds from untapped 

 trees at Peradeniya. 



The kernels and shells of 100 fresh seeds were weighed separately, 

 and this was repeated with 100 seeds which had been drying for six 

 weeks, with the following results : — 



Fresh seed. Dried seed. 

 Original weight ... 454 grams ... 408*5 grams 

 Weight when dried ... ... ••• 341 grams 



"Weight of shells ... 1597 grams = 



35*2 per cent ... 154'5 = 45'3 per cent. 

 Weight of kernels ... 294*3 grams — 



64'8 per cent ... 186"5 = 54'7 per cent. 

 The loss of weight, therefore, takes place almost entirely from the 

 kernel. The kernel is enclosed in a brittle shell, but there is a circular 

 patch, about 3 mm. in diameter, at one end of the seed, over which 

 the shell is not developed, and water can evaporate quite easily through 

 this patch. 



The thousand seeds from group B weighed when fresh 3,511*5 

 grams ; another thousand from the same trees weighed 3,500 grams ; 

 and a third thousand 3,611 grams. The average weight is 3,540'8 

 grams; or 1,000 fresh seeds from tapped trees weigh 7*8 lbs. But 

 these are the trees from which Mr. Carruthers obtained his. seed in 

 1902. We have, therefore, the following result for group B : — 



1902 ... Untapped trees ... Weight of 1,000 seeds — 9*1 lb. 



1905 . . . Trees tapped 



1907 ... ... ... Weight of 1,000 seeds == 7*8 lb. 



If this had been the only group experimented with in 1907, it 

 would have been doubtful whether the diminution in weight was 

 not merely the effect of a bad season ; but, fortunately, we know 

 that the weight of 1,000 seeds from untapped trees (group A) in 1907 

 was 9*1 lb., exactly the same as that found by Mr. Carruthers for 

 group B before they were tapped. We are forced to conclude, there- 

 fore, that the general opinion that tapping affects the seed is correct. 

 Kough results obtained by planters show that the loss of Weight is not 

 due to the increased age of the trees. One planter states that 1,000 

 seeds from his trees (15 years old) formerly weighed 10 lbs., but now 

 they average 7 lbs., having diminished 1 lb., each year during tapping. 



The loss of weight after drying for six weeks was 19*8 per cent 

 in lot 7,20*5 per cent in lot 8, 21*8 per cent in lot 9, and 18*3 per 

 cent in lot 10, or an average loss of 20*1 per cent. Thus, the loss of 

 weight is greater than that in the case of seeds from untapped trees. 



100 seeds — original weight 339 grams — weighed 277 grams after 

 drying for six weeks; of this, the shells weighed 117 grams, or 42*2 

 per cent, and the kernels 159*5 grams, or 57*8 per cent. 



It was quite obvious that the seeds of the tapped trees were smaller 

 than those from the untapped. The latter averaged 2*5 cm. in length, 



