364 
NATURE 
[ Fed. 15, 1883 
physical surroundings.” } But such admissions would make no 
change in the logical aspect of the case ; for, however many sup- 
plementary causes of this kind we may choose to imagine as 
possible, the evolutionist is bound to regard them as all alike in 
this—that they are of a physical or natural kind, 
And this leads me to the core of the whole subject. 
Gray says :— 
‘What is probably meant is, that natural selection is a rival 
hypothesis to design, that it accounts for all adaptations in the 
organic world on physical principles, and so renders . . . the 
evidence of design from these adaptations of no other or bet-er 
value than that from anything else in Nature.” He then proceeds 
to object to this view, and says :—‘‘ If means and ends are prac- 
ticable in inorganic nature at all, it is only by remote and 
indirect implication ; while in organic nature the inference is 
direct and unavoidable. With what propriety, then, can it be 
affirmed that organic nature furnishes no other and no better 
evidence of underlying intelligence than inorganic nature? The 
evidence is certainly o¢/er, and to our thinking detter.” 
This, I say, is the core of the whole subject. If once it is 
fully admitted and understood that organic nature is one with all 
.the rest of the universe in the matter of physical causation, so 
that all the wonderful adaptations which we there encounter are 
the results of natural causes—survival of the fittest p/wes any 
number of other natural causes—then it appears to me, as I 
have said in the essay already alluded to, that all such cases of 
adaptation must fall into the same logical category, with reference 
to the question of design, as all or any other series of facts in 
the physical universe. For the only element of difference arises 
from the greater intricacy of the physical causation in the cases 
contemplated, rendering it more difficult to perceive the opera- 
tion of the causes, at work, and therefore, as Prof. Gray truly 
asserts, rendering their operation more suggestive of design. 
But this element of difference does not really affect the question, 
For, ex hypothes?, the law of causation is everywhere and equally 
uniform, and for this reason the evidence of design in organic 
nature is certainly #o¢ other than it is in inorganic nature, nor, 
in view of the same reason, is it, to our thinking, better. 
Florence, February 3 GEORGE J. ROMANES 
Prof, 
Tue letter of Prof, Asa Gray (NATURE, vol. xxvii. p. 291) 
‘contains a sentence which seems to me to contain the essence of 
the difference between the views of organic life, as held by the 
supporters of Natural Selection and Natural Theology. He 
says: ‘‘ How is this presumption negatived or impaired by the 
supposition of Darwin’s theory, that the ancestors were not 
always like the offspring, but differed from time to time in small 
particulars, 3e¢ so as always to be in compatible relations to the 
environment?” The italicised portion is just such a statement 
as ‘‘ Design” would require, but cannot be held by scientific 
evolutionists, otherwise why are there so many extinct species ? 
With ‘‘ Design” there ought to be a perfecting of all species ; 
whereas we know of so many which have been ruthlessly 
swept aside, owing to their having ‘‘ differed (or owing to their 
not having sufficiently differed) from time to time in small par- 
ticulars, yet’? zzo¢ ‘‘so as to be in compatible relations to the 
environment.” Change is the evolutionist’s view of life—change 
sometimes caused by the environment, sometimes beneficial, 
sometimes eventually detrimental : where beneficial, the species 
increases ; where detrimental, other changes or extinction must 
ensue. Design would never have supplied us with a ‘* Nature 
red in tooth and claw with ravine,” nce would it have built up a 
system by the expensive and cruel mode of trial and error, 
Cove Castle, Loch Long, N.B. J. B. HANNAY 
Two Kinds of Stamens with Different Functions in the 
Same Flower 
To the Melastomacee and Commelynacee mentioned in 
NATURE (vol. xxiv. p. 307, vol. xxvi. p. 386, and vol. xxvii. p. 30), 
may be added the genera Mollia (Tiliaceze), Lagerstremia (Lythra- 
eex), and Heteranthera (Pontederacex), for having differently 
coloured anthers. In several species of A/o//ia, according to Dar- 
win (‘‘ Forms of Flowers,” p. 168, footnote), the longer stamens 
of the five outer cohorts have green pollen, whilst the shorter :ta- 
_ *Inmy “‘ Nature Series” essay I expressly stated that natural selection 
is probably not the only cause of organic evolution, and therefore I think it 
might have been well if my critic had taken the trouble to refer to this essay 
before indulging in the general proposition at the close of his letter with 
reference to exactitude. 
mens of the five inner cohorts have yellow pollen ; the stigma 
stands close beneath the uppermost anthers. In a Lagerstremua 
in my garden the six outer stamens have green pollen, and are 
much longer than the numerous inner ones, which have bright 
yellow pollen ; the stigma stands on a level with the outer anthers. 
I have repeatedly seen bees alighting on, and gathering the 
pollen of, the inner anthers without noticing the outer ones. 
In Heteranthera reniformis there is one long stamen (be- 
longing to the outer whorl) having pale bluish pollen, and two 
short stamens (of the inner whorl) with bright yellow pollen. 
The stigma stands generally on a level with the anther of the 
long stamen. When the white flower opens, pistil and long 
stamen diverge, the pistil bending (almost without exception) to 
the right, and the stamen to the left; at the withering of the 
flower, they again approach each other, so that the stigma may 
be fertilised by the pollen of the long stamen. Visiting insects 
are attracted yet more to the yellow anthers of the two short 
stamens by their being placed close to a yellow spot, surrounded 
by a violet border, at the base of the upper petal. 
Thus it may be safely assumed that in all these flowers, as 
well as in the above-mentioned Melastomaceze and Commelyn- 
acez, fertilisation is almost exclusively effected by th2 pollen 
of the longer stamens, whilst the shorter stamens serve only to 
attract pollen-gathering or pollen-eating insects. It is far from 
surprising that the pollen of these latter stamens, though often 
Fic. 1.—Flower-spike of Heteranthera rentformis (natural size). Fic. 2.— 
Upper end of the flower-tube, seen from behind. a’, the one anther-of 
the outer whorl, with pale bluish pollen; a, the two anthers of the inner 
whorl, with bright yellow pollen ; s¢, stigma. 
produced in large quantity, should tend to degeneration. Dar- 
win long ago came to this conclusion with respect to some 
Melastomacez with differently-coloured anthers, of which he had 
raised seedlings from pollen both of the longe* and shorter 
stamens (‘* There is reason to believe that the shorter stamens 
are tending to abortion.”—‘‘ Cross- and Self-Fertilisation,” 
p. 298, footnote). The Lagerstremia in my garden being self- 
sterile, I fertilised some flowers with green, and others with 
yellow pollen of a different variety (or species ?) growing in 
other gardens ; both produced fruits with apparently good seeds, 
but only some of those from the green pollen have germinated. 
As in all the flowers above-named, with differently-coloured 
anthers, the dull colour of those of the longer stamens evidently 
serves to make them less visible to insects, may not the green 
colour of the anthers of the long stamens of the mid-styled and 
short-styled flowers of Zythrum salicaria also protect them 
against the attacks of pollinivorous insects, to which, from pro- 
truding far from the corolla, they would be more exposed than 
those of the shorter stamens? 
Even without being differently coloured, the stamens of the 
same flower may be divided into different sets with different 
functions. Thus in a species of Cassia the visiting humble-bees 
gather the pollen of the four intermediate stamens (the three 
upper ones being pollenless), which are short and straight, 
whilst the three lower ones are very long and curved in such a 
way that their pollen is deposited on the back of the humble- 
bees. The pistil is of the same length and curved in the same 
way as the longer stamens. Another very striking instance has 
been carefully described by Prof. J. E. Todd of Tabor (Iowa) in 
