492 
NATURE 
[March 22, 1883 
fixed to the end of a small indiarubber tube in order to 
be laid near the insect whose notice it may be desired to 
attract. FRANCIS GALTON 
PRELIMINARY NOTE ON THE BACILLUS OF | 
TUBERCULOSIS (KOCH) 
if5 ops absorption and consequent retention of certain 
stains by this bacillus does not appear to be 
effected by the hydrates of potassium, sodium, and am- 
monium and by aniline alone. Sodic phosphate, potassic 
acetate, vegetable alkaloids, &c., appear to exert a similar 
action. Further experiments are in progress. I have 
some very good preparations which were rapidly stained 
with a very faintly coloured stain containing sodic phos- 
phate (sod. phos. cryst. B.P.). 
II. The sections of tissue shown (by the kind arrange- 
ment of Mr. Blaker) at the Brighton meeting of the British 
Medical Association, in which the bacilli were very dis- 
tinct, were stained, &c., then floated on to the glass slides, 
dried over concentrated sulphuric acid (or fused CaCl,), 
and mounted in balsam. Hitherto my attempts to fix 
the colour of the bacilli, by means of a mordant, in such | 
a way that it might remain unaffected by alcohol, and by 
oil of cloves, have not proved successful. 
III. Treatment with a solution of potassic acetate will 
probably prove well adapted to free preparations from 
those last traces of nitric acid which so often cause their 
ultimate destruction. 
From (I1.) I should omit a very beautiful and remark- 
able preparation showing the spores of this bacillus in 
the lymphatics of the lung. This slide was prepared by 
Dr. Barron, of University College, Liverpool, and for his 
kindness in lending it to me and for much invaluable 
advice I am very grateful. 
To Mr. Blaker, M.R.C.S., of Brighton, and to Mr. 
Black, M.R.C.S., of the Sussex County Hospital, I am | 
under many obligations for their kindly interest and 
assistance. J. W. CLARK 
THE SHAPES OF LEAVES* 
IIll.—Origin of Types 
bee two most general and distinctive types of foliage 
among angiosperms are those characteristic of 
monocotyledons and dicotyledons respectively. 
owe their principal traits of shape and venation to the 
manner in which these two great fundamental classes 
have been separately evolved from lower ancestors. 
Mr. Herbert Spencer has shown that there are two 
‘chief ways in which a central axis or caulome may con- 
ceivably be developed from an integrated series of primi- 
tive stalkless creeping fronds. The /ist way is by the 
in-rolling or folding of the fronds so as to form a com- 
plete tube, often with adnate edges, as represented in the 
accompanying diagram (Fig. 20), modified by Mr. 
Spencer’s kind permission from the “ Principles of 
Biology.” For details of the explanation, the reader 
must be referred to that work (vol. ii. part iv. chap. iii); 
it must suffice here to note that as in such case each 
frond must envelop the younger fronds within it, the | 
process is there shown to eventuate in an endogenous 
stem and a monocotyledonous seed—two characteristics 
found as a matter of fact constantly to accompany one 
another in actual nature. The second way is by the 
thickening and hardening of a fixed series of midribs, as 
shown in the next diagram (Fig. 21), also modified after 
Mr. Spencer; and this method must necessarily result in an 
exogenous stem anda dicotyledonous seed. The diagrams 
in Figs. 22 and 23, which represent according to Mr. 
Spencer (slightly altered) the development of the monoco- 
* Continued from p. 466. 
They | 
| parallel. 
| in those cases among dicotyledons where the lamina is 
tyledonous and dicotyledonous seedling respectively, will 
help further to illustrate the primitive characteristics of 
the two types. 
The monocotyledonous type of foliage is for the most 
part extremely uniform and consistent, in temperate 
climates at least, for in the tropics the presence of large 
arborescent forms, such as palms and screw-pines, as 
well as of gigantic lilies, amaryllids, and grasses, such as 
the bananas, yuccas, agaves, and bamboos, gives a very 
distinctive aspect to the ezsemdble of the class. Being in 
principle a more or less in-rolled and folded frond, every 
part of which equally aids in forming the caulome or 
stem, the monocotyledonous leaf tends as a rule to show 
little distinction between blade and leaf-stalk, lamina 
and petiole. For the same reason, the free end also tends 
to assume a lanceolate or linear shape, while the lower 
part usually becomes more or less tubular or sheathing in 
arrangement. Again, for two reasons, it generally has a 
parallel venation. In the first place, since the leaves or 
terminal expansions are mere prolongations or tips to the 
stem-forming portion, it will follow that the vascular 
tissues will tend to run on continuously over every part, 
instead of radiating from a centre which must in such a 
case be purely artificial. In the second place it is clear 
that parallel venation is the most convenient type for long 
narrow leaves, as is plainly shown even among dicotyle- 
dons by such foliage as that of the plantains, descended 
from netted-veined ancestors, but with chief ribs now 
Still better are both these principles illustrated 
suppressed altogether, and the flattened petiole assumes 
foliar functions, as in Oxalis bupleurifolia and Acacia 
melanoxylon (Fig. 24). These phyllodes thus resembling 
in their mode of development the monocotyledonous type, 
and continuous throughout with the caulome-portion of 
the primitive leaf, exhibit both in shape and venation the 
chief monocotyledonous characteristics. A typical mono- 
cotyledon in shape and venation is represented in Fig. 25. 
The dicotyledonous type, though far more varied, is 
| equally due in its shape and venation to the original 
characteristics implied by its origin. Only the midrib 
instead of the whole leaf being here concerned in the 
production of the stem, there is a far greater tendency to 
distinctness between petiole and lamina, and a marked 
preference for the netted venation. The foliar expansion 
is not here a mere tip ; it becomes a more separate and 
decided element in the entire leaf. And as the petiole 
joins the lamina at a distinct and noticeable point, there 
is a natural tendency for the vascular bundles to diverge 
there, making the venation palmate or radiating, so as to 
distribute it equally to all parts of the expanded surface. 
| Fig. 26 shows the resulting characteristic form of dicoty- 
ledonous leaf. Its variations of pinnate or other vena- 
tion will be considered a little later on. 
Among monocotyledons, the central type is perhaps 
best found in the mainly tuberous or bulbous orders, such 
as the orchids, lilies, and amaryllids. These orders, 
having rich reservoirs of food laid by underground, send 
up relatively thick and sturdy leaves ; but their shape is 
decided by the ancestral type, and by their strict sub- 
ordination to the central axis. Hence they are usually 
long, narrow, and rather fleshy. Familiar examples are 
the tulips, hyacinths, snowdrops, daffodils, crocuses, &c. 
Those which have small bulbs, or none, or grow much 
among grass, like Sisyrinchium, are nearly or quite 
linear ; those which raise their heads higher into the 
open, like Listera, are often quite ovate. Exotic forms 
(bromelias, yuccas, agaves) frequently have the points 
sharp and piercing, as a protection against herbivores. 
In the grasses there is generally no large reservoir of 
food, and their leaves accordingly show the central type 
in a stringy drawn-up condition. So also in sedges, 
woodrushes, and many others. But where the general 
monocotyledonous habit has been more lost,and something 
