BIOLOGIC RELATIONS BETWEEN PLANTS AND ANTS. 430 



apparently is not necessary to induce the primary thickening- of the 

 tubercle nor the formation of the internal cavities. An important 

 observation of Beccari is opposed, however, to these conclusions. He 

 noticed that only those plantlets increased in volume that were provided 

 with a small cavity at the base of the enlargement. Those that did not 

 have this rapidly perished. He supposes that the cavity in question i-; 

 hollowed out by ants. The irritation produced by these insects would 

 then be absolutely indispensable for the proper development of the 

 plautlet. 



A demonstrative experiment still remains to be made: to sprout the 

 seed and obtain hollow tubercles without the intervention of ants. This 

 has not been done up to this time, for in tropical countries the abun- 

 dance of ants is such that under the conditions proper for the develop- 

 ment of the plantlets we can never be sure that no ant has approached 

 the germinating seed. But there is a very direct and very conclusive 

 experiment due to Treub, which is as follows: Transported from their 

 normal habitat into tropical botanical gardens, the tubercles are aban- 

 doned by the red ants that inhabit them in the forest. Very often 

 these are replaced by little black ants, in spite of this change of 

 inhabitants the plant continues to flourish, its tubercle enlarges, new 

 galleries are formed in it, new leaves are produced, and fructification 

 ensues. It appears, then, that if the presence of ants is necessary to 

 the plant, it is not a particular species of ant that is required. But 

 this is not all. The majority of the tubercles commence to rot at the 

 time of their transportation, and this rotting causes the flight of the 

 ants. At the end of a certain time the tubercles again become swollen 

 and healthy, again commence to grow, to thicken, and to form new 

 galleries, and all this in the absence of ants. 



The ants, then, are not at all indispensable for the renewal of develop- 

 ment in the adult plant. From the experiments of Treub we can, 

 indeed, deduce but one fact : that in the adult state certain Myrmecodias 

 can dispense with the presence of ants. We do not yet know whether 

 this is the case with young plants, whether in the absence of ants a 

 Myrmecodia can pass through all stages of development from the germ 

 up to the adult. In certain species of this genus the myrmecophilism 

 may be facultative, in others obligative. 



Beccari justly remarks that the tissue which lines the galleries has 

 all the characters of a young tissue likely to actively proliferate and to 

 react energetically to any irritation that might be caused by the pres- 

 ence of ants. Its reaction to irritation might take the form of a pro- 

 liferation of its constituent elements, as occurs in the tissues of a newly 

 formed leaf when it is subjected to the irritation caused by a gall- 

 insect. The irritation produced by ants upon an adult tissue would 

 not cause the proliferation of its elements, though it might easily do 

 this upon young tissue. 



Though direct proofs are wanting, we are yet authorized in some 



