444 



THE GRAMMAR OF SCIENCE 



differential fertility, I merely laid aside selected repre- 

 sentatives of each kind of capsule. My series practically 

 included all the fives to eights, and seventeens to nineteens. 

 To my great surprise, however, my receptacles for twelves 

 and thirteens were filled up with the contents of very few 

 capsules, those for 1 1 and 1 4 more tardily, those for i o 

 and 1 5 only with emptying a very great number of 

 capsules, while I could hardly get any seed at all from 

 those capsules with very many or with very few bands ; 

 they were practically sterile. The type capsules were 

 enormously fertile, those with even a moderate deviation 

 from it relatively or even absolutely infertile. I then 

 repeated the experiment in another way, taking individual 

 plants. For plants with a low or high average of stig- 

 matic bands there was little seed, but where the average 

 was near the mode the seed was ample. 



The same observations were repeated on the wild 

 poppy, — capsules of 9, 10, and 11 bands were far more 

 frequently fertile than those with a greater deviation from 

 the mode ; arid when these latter had been fertilised, they 

 contained relatively few seeds. 



Lastly, I investigated the seed capsules on a number 

 of plants of Nigella Hispanica. The following is the 

 distribution of the segmentation on 3212 capsules: — 



Here the 8 capsules were highly fertile, but for capsules 

 with 10, I I, or 12 segments I could find hardly any seed 

 at all. Sixes and sevens were only moderately fertile, 

 and those with five and fewer bands practically sterile. 

 Lastly, in the cases of both poppy and nigella deformed 

 and abnormal capsules were absolutely or relatively sterile. 

 These experiments require, of course, repetition on a 

 wider number of species and on a larger scale ; ^ they are 



^ I should be very glad to hear from my readers of experiments or 

 observations on this point. The stages are quite easy ; observe the frequency 

 distribution of some character either of the leaves, flowers, or seed vessels of 



