FLOWERS OF APPLES AS AIDS IN IDENTIFYING VARIETIES. 237 



ineasuremenb of fifty flowers in which the average length of column 

 before division was 6^ mm., gave 5^ mm. as the shortest and 7 mm. 

 the longest. 



The second type has the styles divided to the base as in fig. 90, and 

 the constancy of the form is even better maintained. The amount of 

 pubescence is also a guide, though not so constant. In some varieties 

 the whole pistil is quite glabrous, as in ' White Transparent,' in others 

 small hairs are present at the junction, as in fig. 89. A third case is 

 in fig. 91 ; the styles are hairy almost to their summits. 



So far in the form in which they are divided to the base, we have 

 not yet found a glabrous variety. It will be seen that taking the two 

 types, columnar and separate, we have two very clear-cut differentiating 

 characters. Examples of their utihty are shown in the following cases, 

 in which two similar fruits are compared: — 



Styles divided to base. Styles forming column, 



* Cox's Orange Pippin.' * The Houblon.' 



* Forfar Pippin. ' * Norman's Pippin. ' 



Correlation. — The fascinating but dangerous task of searching for 

 some broad and general correlation between these varying characters of 

 apples is one at which we can only permit ourselves a hasty and hesi- 

 tating glance. That large flowers indicate large fruit is frequently 

 true, but exceptions exist. The largest flower we have noted is that of 

 ' Nelson Oodlin,' an apple of medium size. Colour in flower and fruit 

 seems to go more by contrary rule, as the finest coloured flowers are 

 generally those of pale yellow or green varieties, as in ' Lord Derby,' 

 ' Forfar Pippin, ' and ' Golden Spire. ' 



Lateness in flowering has on the whole a correspondence with late- 

 ness in the fruit, but by no means in all varieties. 



The writer hopes that any departures from the facts stated above 

 may be brought to his notice, as he has not had the opportunity of 

 studying the possible effect of any markedly different environments 

 upon them. 



VOL. XXXVIII. 



