1892.] NATURAL SCIENCES OF PHILADELPHIA. 163 



thrips can gain entrance t(> the nectary. The mouth of the tube is so 

 densely matted with hair, that Faux clausa is the term used in the 

 description of the species by Latin authors. If a pollen- clothed 

 tongue were thrust through the mass, it would be thoroughly 

 cleaned, and in like manner the flower's own pollen would be brushed 

 back, when the insect withdrew its tongue. But a greater difficulty 

 presents itself The capitate stigma with its surrounding rim, com- 

 pletely fills the upper portion of the tube. There is no space for 

 an insect's tongue to get past the stigma. But even could this 

 rubicon be passed, a dense mass of hair presses close against the 

 style, and the most powerful insect known to the writer, could 

 hardly force a passage. The entrance of insects is completely 

 blocked. To provide for pollination, the anthers curve over and 

 rest on the stigma, and the pollen on ejection from the anthers, can 

 do no more than cover the stigma with their own pollen. 



In many plants which have flowers that are generally fertilized 

 by their own pollen, the arrangements will often permit of pollina- 

 tion from some other; but in the case of this ^msom'o, nothing but 

 self-pollination is possible. 



To those who may not have flowers for comparison, the figure of 

 this plant in " Botanical Register," Plate 151, will aid in making some 

 of the above noted points clear. 



On a special form of Cleistogamy in Polygonum acre. 



In a paper published in the Proceedings of the Academy (1889, 

 p. 59,) I pointed out that in almost all — probably in all cases, the fer- 

 tile flowers were pollinized in the bud in all the species of Polygonum 

 that I had had the opportunity of examining: that they were really 

 cleistogamous. There are two classes of flowers in the inflorescence. 

 Many expand and are to all appearance hermaphrodite, w^th all 

 their sexual organs perfect, but infertile ; another class never opens, 

 but are invariably fertile. 



In May, 1890, I noticed a quantity of P. acre in a swamp in 

 Chester County, Pennsylvania, with a short and close habit. The 

 leaves were shorter and broader, and the ochrea shallower than 

 usual. Small wliite flowers Avere protruding above the sheaths, 

 and I suspected I had found a new species in the section with 

 axillary flowers. But on examining P. acre in other localities, 

 I found, in every case, flowers hidden under the ochrea from even 

 the lowest axil on the branch. It was the shallowness of the ochrea 



