1895.] NATUKAL, SCIENCES OF PHILADEXjPHIA. 289 



constituent of every gelatinous secretion, is in this particular ease, in 

 all probability, derived directly from the water surrounding the plant 

 than that which is carried through the tissue of the plant to the 

 glandular organs. The secretion in the fruit of Peltandra undulata 

 Waf. is not in close contact with an aqueous medium, and it appears to 

 me a significant fact that the tissue upon which the glands of this plant 

 are supported is traversed by spiral vessels, being thereby well sup- 

 plied with water-conducting tissue. The water in this case must of 

 necessity be carried through the plant to the interior of the fruit. 



A few more observations deserve a passing notice: When I col- 

 lected my material for study October 7, 1894, I observed that 

 some of the fruits had left, the spadix while others still crowded 

 anmnd it surrounded by the tough spathe. About a dozen of the 

 seeds I threw into a jar containing water. Of these only three sur- 

 vived and germinated in March of this year. Figs. 8 and 9 

 are drawings of seedlings as they appeared on April 15, and 

 the plants continued to live until the beginning of May, although 

 they had no soil to root in. 



On April 18, of this year, Mr. MacElwee showed me a locality 

 near Gray's Ferry Station where Peltandra grows. Although the 

 seeds which I had collected last year had germinated a month ago, we 

 found that most of the seeds at this locality just barely showed the 

 protruding tip of the first leaf, and some showed no siinis whatever 

 of sprouting. It cannot be said that the appearance of the jelly is 

 invariably the first indication of activity in the seed, since I noticed 

 repeatedly that the bud may appear while the secretion is still 

 encased in the pericarp. The plantlet represented in fig. 8, did 

 not throw back the pericarp either during germination or after- 

 ward. 



Some of the fruits were placed on very wet sand and in a few days 

 there were signs of vigorous growth, fig. 10. Others were thrown on 

 water, and, although development took place, fig. 11, it did not 

 seem as healthy as when the fruits were allowed to rest upon a solid, 

 although thoroughly wet, substratum. The difference may be ap- 

 preciated on comparing figs. 10 and 11, and this difference 

 became more marked later on. 



So far I have not found any satisfactory explanation for the de- 

 velopment of this peculiar jelly-like secretion in the fruits of Pel- 

 tandra undulata Raf. , its persistence during germination, and its 



