ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
309 
Cannabis sativa the normal pollen contains little fat or starch, and is 
highly vacuolate. Thus, it seems that the quantity of food-reserve is of 
greater importance than the quality. Experiments with Plantago lanceo- 
lata and other species show that germinating pollen-grains depend upon 
an external nourishing substratum for their growth. When pollen was 
germinated in a 5 p.c. glucose solution the tubes were longer and the 
absorption of starch w 7 as more complete than when germination took 
place in pure water. Under natural conditions the stigma is probably 
the source of this externa! supply of food-material. The water-content 
of the dry pollen of anemophilous species is very small, so that the 
materials stored in the pollen-grains apparently serve the purpose of 
maintaining a high osmotic pressure; in Juneus , Secale and Plantago 
this pressure attains a strength of 3 to 4 G.M. NaCl. S. G. 
General. 
New Theory of Mermecophily. — R. Chodat and Luis Carisso 
( Compt . Rend. Sc. Soc. Phys. et d'PHst. Natur. Geneva , 1920, 37, 
9-12). An account of the ant-chambers found in several South 
American species of Cordia and in Acacia Cavenia. In Cordia these 
chambers are the result, in the first instance, of a wound made by one 
of the Hymenoptera, allied to the genus Eurytoma. The insect pierces 
the plant and deposits one or more of its eggs in the opening. The 
larvas hatch in the cortex, and cause an irritating stimulus, which is 
followed by the appearance of a perimedullary meristem ; this forms a 
centrifugal ring of medullary cork, w T hich is gradually eliminated, and 
centripetal phelloderm, which ultimately hardens and lines the cavity in 
which the larvse develop. In Acacia it is possible to follow step by 
step the growth of the cavity in the stipules until the time when the 
fully grown insect makes its exit by a hole in the thorn. Thus it 
appears that the abnormal growth in both genera is a kind of gall 
w r hich the ants afterwards adapt for their ow 7 n purposes. Moreover, it 
is no longer possible to admit the theory that these ants are a protection 
against leaf-cutting ants, since within these cavities typical “ ant-gardens ” 
are found, made from portions of floral and vegetative organs torn by 
them from the host-plants. S. G. 
Acarodermatia in the Clethracese. — J. Briquet {Compt. Rend. 
Sc. Soc. Phys. et d'Hist. Natur. Geneva , 1920, 37, 12-15). An account 
of acarodermatia on the leaves of Clethra barbinervis. When young 
the leaves are covered with tw 7 o kinds of hairs — i.e. with simple, stiff 
unicellular hairs which are firmly embedded in the epidermis and attain 
a length of 0‘ 4 mm., and branched, stellate hairs at the bases of which 
are 3 to 6 trichomes grouped together to form a solid mass. The adult 
leaves are slightly hairy, the upper surface being almost smooth ; on the 
lower surface both kinds of hairs occur here and there in patches, but 
at the points where the median nerve gives off its branches, there are 
dense tufts of the stellate hairs, which serve as acarodermatia. The 
hairs forming each cluster are closely packed together, and the acari 
find an entrance between the trichomes. The dermatia are of the 
simple lophic (Lundstrom) type. There is at present no evidence to show 
what mutual benefit results from the symbiosis here recorded. S. G. 
