BY ALEX. O. HAMILTON. 785 



Ilex spp. — Dr. Lundstrom describes backward curls of the 

 edge of the leaf near the base, forming a cylindrical room, and 

 found here the cast skins of mites. But so far as dried material 

 could show, there was not the peculiar structure found in domatia. 

 I have found in Eupomatia laiirina similar structures, but could 

 find no mites or traces of them. 



ScHiNUS spp. — These have a wing on the rhachis provided with a 

 small tooth on each side at the insertion of the leaf, which folds over 

 and forms a cavity. I am inclined to think that none of these 

 structures are true domatia, and would restrict that term to 

 cavities or depressions in the leaf surface showing the peculiar 

 appearances described under the types I have taken. But under 

 Dr. Lundstrom's definition of a domatium, viz., all those structures 

 of plants which act as dwellings or shelters for insects and receive 

 in turn some benefit from the latter, all these might be included. 

 Dr. Lundstrom classifies domatia into the following five groups 

 — (1) Hair tufts at axils; (2) bending back or folding of leaf or 

 edge of rachis ; (3) dimples with or Avithout hairs ; (4) small 

 pockets; (5) bags, kc His group 1 corresponds with my group 

 4; his 3rd with my 1st, and 4th with my 2nd. His 2nd and 5th 

 groups I have not taken to be domatia, and he does not 

 particularly notice my 3rd or 5th groups. 



I have arranged the groups of types as shown because it indi- 

 cates the order of development — beginningwith the highest. The 

 domatium usually begins either as a small hair-tuft or a depression. 

 Then an outgrowth from the veins begins extending right across 

 the angle. Later a ridge thickens up across the open angle and 

 runs round to the sidcjs, so that when all the parts are grown to 

 full height a circular orifice is formed. This is well seen at times 

 in Vitis Baudiniana, which usually has the triangular pouch, but 

 at times forms the circular cavity in this way. As the order of 

 types, beginning with the 5th, represents the development of the 

 domatia in a single plant, so also it probably brings before us the 

 order of evolution. 



So far as I have looked into the matter, it appears to me that 

 domatia are most common in plants of a southern origin. At any 



