ACACIA SPH^KOCEPIIALA AND CECBOPIA PELT AT A. 401 



3. Tincture of alkanet stains it bright pink* It is not 

 coloured by other staining fluids.] 



The green colour with sulphuric acid is the only one of the 

 above reactions which would show that the globules are oil and 

 not resin ; for Hanstein f mentions alkanet as staining .resin ; but 

 the fact that the ants search greedily after the food-bodies seems 

 to show that they are oleaginous. 



Relation between the Oil and the Protoplasm. — Sections cleared 

 with caustic potash show plainly that the oil is embedded as sepa- 

 rate drops in the protoplasm ; and similar sections from which 

 the oil has been dissolved with ether &c. exhibit the cavities in 

 which the oil is thus contained. The matrix may be stained 

 with alkanet ; and one then sees the crimson protoplasm pierced 

 in all directions by smooth-walled cavities corresponding with 

 the oil globules seen in the potash specimens, fig. 3. Sections 

 stained rather less brightly and from which the oil has not been 

 removed, show a pretty contrast between the bright crimson oil 

 drops and the less brilliant matrix. "Where the oil is in the form 

 of a number of very minute drops of equal size, the protoplasmic 

 matrix (as seen in thin sections cleared with potash and with 

 creasote) exhibits a delicate network, as shown in fig. 4. 



It may be of interest to note that the food-bodies of Acacia 

 partially resemble in structure some other forms of nutriment- 

 stores in the vegetable kingdom. It appears that the crater glands 

 and the food-bodies together supply nutriment sufficient to support 

 the ants • it is therefore evident that the latter must be con- 

 sidered in the light of protein stores as well as of stores of car- 

 bohydrates. "We may compare them analogically with the endo- 

 sperm of seeds, in which these substances are also stored. Now 

 m some seeds the oil is found as drops in the cavities of a proto- 

 plasmic matrix, just as in the food-bodies ; possibly also the (pro- 

 teinaceous ?) granules in the protoplasm of the food-bodies may 

 be compared to aleurone grains. 



Homologies and Development.— in the adult state the upper 

 leaflets terminate in minute points, '1 miliim. in length. They 

 appear to be embryonic structures ; for they are relatively well de- 

 veloped in a very early stage of growth, and as the leaf grows they 

 remain stationary and shrivel away. If we examine an extremely 

 young leaf, we find that all the leaflets, both upper and lower, termi- 



This is the test for oils mentioned by Sachs, 

 t Bot. Zeit. 1868, p. 708. 



