Miscellanea 
107 
There seems therefore considerable danger both in drawing conclusions from selected shoots 
of a single tree in a problem of growth and further in citing such results as an argument as to 
honiotyposis. Probably, but we have no certainty for it, Miss Tammes in selecting shoots 
with 9 leaves, really made a selection of 'sun' branches as distinguished from 'shade' branches, 
but whether her results are universally true for the leaves on all such branches of all beech 
trees, it is quite impossible to say until a large number have been examined. In the case 
of the big trees such as we have dealt with from Hampden and Highmore, the task of 
collecting year's shoots from 'sun' branches will not be an easy one, for in the majority 
of cases they are very inaccessible. Whether the homotyposis of 'sun' branches difi'ers from 
that of 'shade' branches, it is impossible to say ; the larger number of leaves to the shoot on 
the former, however, would make them better material for homotypic investigation. 
(v) It may perhaps not be out of place to note here that since the first memoir on homo- 
typosis was written we have been able to get a short series of 150 woodruff sprays from a lane 
at Horsham, Sussex. The woodruff was in flower and the whorls were counted from the flower 
downwards. As compared with the Great Hampden* series we found: 
Mean number of branches Standard Deviation 
Gi-eat Hampden ... 6 9 -93 
Horsham 6-7 -84 
The correlation between position and number of branches was "26 and the homotyposis 
allowing for this diS'erentiation was ^2. This is a very sensible increase on the "17 of the 
original Hamjtden investigation, but while the number of whorls to the spray at Hampden was 
on an average 7 and might reach 10, the mean number at Horsham was 4-5 and never exceeded 
5. Hence the material did not provide a good determination of the relationshijj between number 
of branches and position of the whorl. What it does suffice to show is that the Hampden series 
was justly described as having a value much reduced by neglect of differentiation. We should 
be very glad if any reader of Biometrika would provide us with 400 to 500 healthy sprays of 
woodruft' after the plant has ceased to grow in the autumn of next year. 
Description of Plates. 
Plate I. Illustrations of the so-called 'sun' and 'shade' branches of the beech. 
Plate II. Backs of typical beech leaf sprays. I and II. Usual forms of spray, with four 
leaves and indicating the marked difierentiation between first and second pairs of leaves. 
II and III. Seldom forms with five leaves. V and VI. Less seldom forms with three leaves 
only. The letters L and R indicate the left and right-hand side of the back of the leaf and 
the subscript figures the number of the veins. The larger number gives the order of the leaf 
on the spray stem. 
III. Albinism in Sicily and Mendel's Laws. 
It has been suggested by Castle (1) that albinism in Man behaves in accordance with 
Mendel's Laws of Inheritance, and the suggestion is considered probable by Bateson (2). In a 
recent paper Garrod (3) has endorsed the same view, and has cited the work of Arcoleo in support 
of it. Arcoleo's work (4) records 62 cases of albinism observed in Sicily, the greater number in 
the province of Palermo. Three of the albinos recorded were produced by albino parents, the 
rest had normal parents. In ten cases it is known that the parents were pigmented, but it is 
not known whether they produced pigmented children in addition to the albinos. 
The forty-nine remaining cases were distrilmted among twenty families, which contained alto- 
gether 133 children. Now, on the Mendelian hypothesis, that albinism and possession of pigment 
* The Great Hampden Lane gave no woodruff when examined this autumn, but the hedges had 
been recently cut down and the banks trimmed. 
14—2 
