31 



The results of the tappings here show a much more marked 

 variation than those on Block A. The average of five days continuous 

 tappings with suitable intervals gave the following : 



Ist fiA^e days average from 49 trees, 146 units. 

 2nd „ ., „ 164 „ 



3rd „ „ „ 120 „ 



During the third experiment the trees wei*e in the middle of 

 wintering. 



The avei-age yield of the first two experiments gives 155 units as 

 in Block A, the average over 10 days of the best yielder was 771 

 units, and of the worst yielder only 31 units. Twenty-three trees 

 gave yields above the avei-age and 26 below the average. The daily 

 yields of the best milker were as follows : 501, 858, 810, 901, 849, 

 754, 669, 830, 905 and 634. Daily yields of the lowest yielder were : 

 11, 26, 25, 7, 17, 10, 30, 52, 36, 89. 



It will be seen from these figures that the yields from both best 

 and worst milkers are fairly steady. Such variations as are shown 

 being easily accounted for by climatic conditions. In no instance 

 has it yet been found that a tree yields heavily one day and poorly 

 on another. Assuming that the seed had been properly selected from 

 the best milkers in Block A and that each tree gave a yield equal to 

 the average yield of the best 23 ti'ees — i.e., 252 units per tree, the 

 output per acre would be inci*eased by over 60 per cent. 



Even allowing that trees vary in their yields very considerably 

 during the course of a year, it is difficult to suppose that the 

 variation is as great as the differences between the best and worst 

 milkei's. 



It will be noticed that the avei-ages in Block A of the best and 

 worst milkers are 281 and 75, respectively, against 252 and 73 of 

 those on Block B, or a slight reduction in the following generation. 



While I am well aware that the figai*es quoted cannot by any 

 means be called conclusive, as the experiments have only been carried 

 on during a short period of time, I do think they are convincing 

 enough to call for a thorough investigation under strict scientific 

 control. I may also add that 1 am going to continue to take records 

 of these trees for at least twelve months, when 1 hope to be able to 

 publish the results. 



If a case can be made out to support the suggested degeneration 

 of our stock what is to be done in the matter ? Two alternatives 

 seem to suggest themselves. Firstly, seed can be taken from trees 

 that have been proved good milkers through a sufficiently long 

 period of testing, and planted out in a clearing completely isolated 

 from any chance of cross-fertilization fi-om outside trees. When 

 these trees are tappable. careful records of their milking capacity, 

 resistance to disease, etc., should be kept over another sufficiently 



