1922] BAILEY—ANT-PLANTS 383 
their presence was discovered, and were hurled to the ground. 
How then may one account for the defoliation of trees which are 
inhabited by Aztecas? When a Cecropia is touched or shaken, 
the angry and aggressive Aztecas rush out of their nests and swarm 
over the whole plant, but under normal conditions only a limited 
number of workers are visible on the stems and bases of the petioles. 
Furthermore, there are periods during which the entire colony is 
quiescent, that is, has withdrawn into its nest. It is well known 
that these periods of quiescence are not the same in the case of 
different species of ants. Thus, certain ants are nocturnal, others 
are active at temperatures when other species are inactive, etc. 
In other words, as suggested by MtttER and MOLLER, a Cecropia 
may be defoliated by leaf cutting ants during periods when the 
Aztecas are quiescent. 
Prostomata 
Above the insertion of every leaf in Cecropia adenopus there is 
a shallow groove, which terminates just below the next node in a 
roundish depression (text fig. 8). As the external depression 
corresponds to an internal one, the wall of the fistulose stem is 
very thin at this place, and is a mere diaphragm ina tube. SCHIM- 
PER showed that this diaphragm is devoid of hard and tough ele- 
ments (fibrovascular bundles, collenchyma, lignified parenchyma, 
etc.), such as occur in the adjoining portions of the stem. He 
inferred from this that the diaphragm originated as an adaptation 
for facilitating the entrance of ants. At the phylogenetic com- 
mencement of symbiosis, the ants bored an entrance through the 
groove, evidently because the wall of the stem was somewhat 
thinner there. In accordance with a custom that is almost invari- 
ably followed and is connected with the domestic arrangements in 
their nests, they tended to penetrate into the internodal cavity at 
its upper extremity (at the apex of the groove). All features which 
facilitated their entrance at this place were retained in the struggle 
for existence, and were accentuated through natural selection. 
This led eventually to the differentiation of a thin, weak diaphragm 
or prostoma (von IHERING). In other words, although SCHIMPER 
admitted that the groove is due to the pressure of the axillary 
