CONTRIBUTIONS FROM THE WISLEY LABORATORY. 367 



not a single one. It never occurred to my observation, before the 

 year 1820, when I was much occupied in the artificial impregnation 

 of different kinds of fruit, that, out of from nine or fewer, to fifteen 

 or more florets, of which the cluster (botanically corymbus) of the 

 pear-tree consists, only the three lower ones (generally speaking) 

 set, or, in other words, are effectually impregnated, for fruiting. 

 Recollecting the practice of the best gardeners, of topping their early 

 beans, i.e. of pinching off with the forefinger and thumb the upper- 

 most blossoms, some apparent, and others in embryo, of the general 

 spike, for the purpose of setting the lowest and earliest ones, which 

 would otherwise in most cases prove abortive, I conceived, that 

 removing the upper and central blossoms of the corymbus of the 

 pear, as soon as it could conveniently be done, would have a similar 

 good effect in invigorating the remaining ones, and causing them to 

 set with greater certainty. With this in view, in the spring of 182 1, 

 as soon as the three lower blossoms of the corymbi began to show 

 their white faces, I set to work with my sharp-pointed scissors on 

 two pear-trees, the one, the ' Gansell's Bergamot ' above mentioned 

 and the other a ' Brown Beurre,' and in as short time as I could have 

 properly thinned two dozen bunches of grapes I divested both these 

 trees of at least three-fourths of all their budding honours. On the 

 ' Beurre,' this operation subsequently appeared to have the best 

 effect. For there was scarcely an instance in which the three remaining 

 blossoms did not set, which afterwards produced the finest crop of 

 pears I have yet gathered from that tree. But on the intractable 

 * Gansell,' although the blossoms at first seemed to set, and many of 

 them did not fall off till midsummer, when they were nearly as large 

 as common gooseberries, yet not a single pear arrived at maturity. 

 By dissecting many of the largest of those which fell off last, and 

 comparing them with some of the 1 Beurres ' of the same age and 

 size, it was plain that the kernels of the former had not been 

 impregnated. This circumstance induced me to think that there 

 must be some imperfections in the essential parts of the blossoms. 



" In the following spring of 1822, on attending to the blossoms of this 

 tree, which blooms earlier than any other pear-tree which I have, they 

 appeared to me to remain much longer in a globular state, without 

 expanding, than any other variety of pear which I have had an oppor- 

 tunity of noticing. I fancied likewise that the pointal was fit for 

 impregnation before the anthers were ripe, and even before the petals 

 expanded ; and from the peculiarly slender and delicate make of the 

 latter, as it struck me, I supposed that it ceased to be in a proper state 

 as soon as it became exposed to the sun and air ; I therefore concluded, 

 that there might possibly be a chance of obtaining fruit, by depriving 

 the blossoms of their petals before they expanded, and inclosing with 

 each floret in this state, within a paper envelope (as is my mode of effect- 

 ing artificial impregnation), a riper blossom, viz. one that had just begun 

 to diffuse its farina, either one of its own or, preferably, of some other 

 variety of pear. Accordingly on March 27, 1822, 1 began this operation, 



