360 SECTIONAL TRANSACTIONS.—K. 
the first two leaves are directly continuous with the intercotyledonary poles of the 
tetrarch root, but those of the succeeding leaves are not connected with root poles. 
The occurrence of a ‘double strand’ in epicotyledonary leaves, and more 
especially in those unrelated to root poles, is of interest in connection with the cases 
quoted by Dr. Thomas and Prof. Compton, and lends support to the suggestion put 
forward by Dr. Thomas that ‘ doubleness may be a primitive foliar arrangement.’ 
15. Discussion on Adaptive Characters. 
(a) Prof. F. O. Bower, F.R.S. 
At the opening of the discussion it will be well to see clearly what is meant by the 
words used in its title. The expression ‘adaptive ' is often applied loosely for any 
character to which a reasonably probable use can be ascribed. Causality in relation 
to that use is then liable to be assumed, without any evidence being adduced to show 
that the character actually originated in relation to the conditions which it may 
effectively meet. To speak thus of a character as adaptive is merely to apply to it 
a question-begging epithet. Few of those who lightly use the word have ever adduced 
evidence that the character really is adaptive in the evolutionary sense in any specific 
case: that is, that the development, individual or racial, originated in accommodation 
to circumstances. On the other hand, if the character were actually adaptive in this 
sense, it might be expected that, in the absence of the causal condition, the character 
should be modified, or even disappear. But it cannot always be presumed that a 
feature held as adaptive must necessarily disappear: provided that it be not harmful 
it may persist, even though its primary cause is absent, for it may have passed into 
the category of inherited characters. Thus we may hold it as possible that fluctuating 
characters, arising first as consequences of immediate accommodation, may become 
permanently fixed. The discussion will naturally lead to the question of the in- 
heritance of characters thus acquired. ‘ 
It is desirable that the discussion should not be conducted merely along general 
lines ; but that it should relate first to specific instances, upon which alone any sound 
conclusions can be based. In the comparative study of ferns, checked by reference 
to the related fossils, it is believed that evidence has been obtained of the secular 
inheritance of characters in the first instance acquired and adaptive; and that in the 
course of evolution, extending from the palsozoic period to the present day, those 
characters have become permanently fixed. The best instance is in the adoption of a 
protective superficial position of the sorus, though comparative evidence of ferns, 
fossil and living, indicates that the distal or marginal position was the original one. 
Other examples are seen in the adjustment of the vascular tissues in relation to 
increasing size and in the absence of secondary thickening, so as to maintain a suitable 
proportion of surface to bulk. Such adjustments have frequently become hereditary. 
(6) Prof. J. Prirstiry. 
The fact that the existence of adaptive characters receives a partial explanation 
upon the hypothesis of natural selection, has led to a much too facile interpretation 
of structural features of the plant as adaptive. Such an interpretation should always 
rest upon definite experimental evidence that the structural features concerned per- 
form the functions assigned to them. The result of such experimental examination 
has recently been illustrated in the study of various epidermal structures which are 
interpreted on theoretical grounds as devices for controlling transpiration. 
Interpretation as an adaptive character still leaves the etructural feature in question 
open to elucidation upon the lines of causal anatomy. This alternative method of 
approach is illustrated by a brief examination of the main features of stelar anatomy 
in the ferns. 
(c) Dr. D. H. Scorr, F.R.S. 
The modern reaction against adaptation appears to be due to several causes :— 
1. The weakening of the Darwinian position, involving less reliance on Natural — 
Selection. 
2. Distaste for the too facile assumption of hypothetical functions on the part of i 
some zealous Darwinians. 
3. The influence of Mendelism, which has no use for adaptations. 
-. 
