INTERCELlwUIvAR CANALS 257 



matic origin, filled with a dark resinous material are arranged in a 

 compact peripheral row. They occupy the entire space between the 

 rays but no interruption of the rays was noted. 



Burseracccc. — In specimen (Yale No. 3677) of Elaphrhim siuiaruba 

 from the Guatemala-Honduras boundary, radial canals are common. 

 They are visible with lens. 



In two Brazilian woods labeled respectively "imburana" (Yale No. 

 3410) and "umburana" (Yale No. 3415), believed to be Bur sera sp., 

 ducts are present in some of the rays. They are barely visible under 

 lens on tangential surface as dark specks in the rays. 



In a specimen of Santiria nitida (Yale No'. 2224) a few scattered 

 radial ducts are barely visible without lens. One of these canals was 

 figured in the original paper but no reference to it was made in the text. 



Scattered radial canals, barely visible to the unaided eye, were 

 observed in the Philippine "bogo" (Yale No. 2223), Garuga sp. 



Dr. Henry H. Dixon of Trinity College, Dublin, states in a letter of 

 January 20, 1919, that he has observed intercellular canals in Protiiim 

 altissimiim. It is presumed that he refers to vertical canals resulting 

 from injury. 



Radial canals have been observed by the writer in representatives 

 of four genera, namely, Bursera ( ?), Blaphriuui, Garuga, and Santiria. 

 Traumatic vertical canals have been reported by others' in Canarimn 

 and Protiiim. 



Combrctacccc. — Vertical canals, of the gummosis type, have been 

 observed by the writer in various specimens of Terminalia spp. In 

 r. bialata (Yale No. 3878) from India and T. sp. (Yale No. 2238) 

 from the Philippine Islands, there are direct evidences of wounding. 

 Other specimens in which this feature was observed are T. bclcrica 

 (Yale No. 3875) from India, and the "araga de agua," T. afif. januar- 

 ensis (Yale No. 3940), from the region of Bahia, Brazil. A market 

 specimen of "araga" (Yale No. 3105) from Rio de Janeiro shows two 

 compact peripheral arcs of ducts, in one instance running the full 

 length of the piece, namely, eight inches. This wood is unquestionably 

 Terminalia though probably not of the identical species of the other 

 Brazilian wood mentioned. In all of the above, the ducts are small 

 and, though distinct, are not conspicuous. The contents are gummy 

 and mostly of a dark reddish color. 



