PARADISE APPLE STOCKS. 
3Sl 
Up to the present, except in a very few instances, we have had no 
blossom or fruit to aid us, and the remaining characters of leaf and 
bud just quoted in the footnote would not carry us far in an iden- 
tification of mixed types. Moreover, from the second description of 
vegetative characters the fact becomes apparent, read side by side 
with the present records, that certain characteristics may vary accord- 
ing to the district. The Doucin has proved at East Mailing the 
most " feathered " or " spiny " of all the types. Hence the necessity 
is proved for taking the widest possible survey of all characters, if 
this work is to give any common ground for comparison. 
XII. The Actual Value of the Present Figures and Measurements. 
The characters in Table IV. were compiled in an exactly similar 
manner to those in Table III., except that the observations were made 
during the growing season. Every stool has been repeatedly subjected 
to critical examination. The shoots described are the long shoots of 
the current year's growth, whilst the leaves are from similar wood. 
It is of course realized that the stools from which these characters 
were collected are very strong and healthy, being in their first full 
vigour and having been planted on good ground. However, it appears 
quite true to say that whilst actual measurements of strength and size 
may vary as the stools become older, the relative strength and vigour 
are likely to be maintained. Within a period of four years it has 
already become noticeable that as the stools grow older they send up 
more numerous shoots annually, whilst they tend to be slightly 
less strong individually, but the various types have maintained, 
with little exception, their relative number of suckers and degree of 
strength. This fact is very well illustrated in Table V., which gives 
the average measurements of the leaves of each type. First 
typical or medium-sized leaves — chosen from average shoots of known 
measurement— were selected from the middle of such shoots and 
measured. These afforded the ratio between length and breadth of 
leaf-blade on individual types and between one type and another. 
They also took into account the relative length of petiole and apex. 
The largest leaves were then chosen from the most vigorous shoots. 
They were similarly measured, and it was found that in nearly all 
cases the proportions remained closely similar. If there proved to 
be any deviation, it was generally in the form of an exaggeration of 
an already existing tendency. For example, in Type VI. the tendency 
in the medium leaves is towards length (6*2 cm.) considerably exceeding 
breadth (4*0 cm.) ; when the largest leaves were measured this tendency 
was exaggerated, the average length (9*3 cm.) greatly exceeding 
the breadth (6*4 cm.). In Type VII., where the leaf normally tends to 
approach the circular form, the roundness is almost complete in the 
largest leaves. Their average length is 77 cm., and breadth 7*2 cm., 
as against 5*9 cm. and 4*8 cm. in the medium leaves. Similarly in 
Types V. and VIII. shortness of leaf-stalk is accentuated. Finally, 
quite small leaves were taken off weak one-year bedded layers of the 
