and on the Structure of the Seed in the Rhamnaceje. 91 



a thm, dark-coloured, brittle, submembranaceous, loose enve- 

 lope, open at the top and sometimes also along the dorsal face ; 

 but m all cases the margins of this opening overlap each other^ 

 and thus entirely conceal the seed. This was first described by 

 ienzl*, who calls it "a brittle testa, covering the smooth carti- 

 laginous endopleura/' Endlicherf, on the other hand, desig- 

 nates It as an "arillus, covering a smooth corneous testa." 

 Dr. Asa Gray J, speaking of this tunic, "doubts if this mem- 

 brane be a true arillus, as our materials are not sufficient for the 

 complete nivestigation of its nature. Although marked with 

 what seems like a raphe, it cannot be the testa or any proper 

 integument of the seed, as Fenzl (in PI. Hugel) took it to be, 

 tor It has no connexion with the corneous seed-coat at the cha- 

 laza but only at the hilum ; the chalazal end appears to be 

 slightly open, as would be the case with an arillus : perhaps this 

 membrane is a separable lining of the coccus." 



I have not been able to detect the presence of any kind of 

 vessels in this tunic, particularly in the prominent line that 

 "seems like a raphe," and which Endlicher positively calls a 

 raphe : this prominence appears to me nothing more than a mere 

 impression of the sutural line of the coccus, and is a common 

 feature seen upon the corneous coating of the seed in the Col- 

 letiecB and other Ilhamnaceous genera. The tunic certainly pre- 

 sents all the characters of an arillus; and, as Dr. Asa' Gray 

 justly remarks, it cannot be one of the proper integuments of 

 the seed. The hard seed of Alphitonia is perfectly sessile within 

 It, and is attached to it by a small open transverse fissure in its 

 base, as in Zizyphus. 



Other circumstances, however, present themselves in the or- 

 ganization which I have found in the seed of Alphitonia excelsa, 

 that are worthy of record, as they completely confirm all the 

 details of structure previously described, and which seem to 

 prevail thoughout the RhamnacecB. The second hard tunic, 

 which lies within the outer brittle coating last described, becomes 

 softened by maceration, when it is easily detached ; it is then 

 very lax, is of a pale- brown or yellowish colour, quite homo'^e- 

 neous in texture, void of any vessels, and encloses an oval body 

 covered by a third tunic, which is notably smaller in size, and 

 therefore leaves a considerable vacant space all round its 'sides 

 and its summit, and is connected with the second tunic only at 

 a small point in the base, which corresponds with the basal slit 

 in the outer tunic of Colletia and Zizi/phus. The summit of this 

 third tunic is marked by a large tumid spot, evidently a chalaza, 

 which exhibits no indication of any previous connexion with the 

 * Plant. Hugel, p. 20, sub Colubrina excelsa. 

 t Gen. Plant, no. 5712. + Bot. Explor. Exped. Un. St. p. 278. 



