NATURAL SCIENCES OF PHILADELPHIA. 289 



tlii'ongh which passes the pistil, a complete cloud of pollen, en- 

 veloping the stigma upon every side. 



This operation can be performed artificially, by taking hold of 

 the under lip with the left thumb and fore-finger, and })ulling the 

 upper lip backward, by the right, and then releasing the hold of 

 the latter : the upper lip springs to its place, spirting the pollen 

 through the aperture upon the left hand. From the a1)Ove it is to 

 be seen, that the i)lant has two chances of being fertilized one by 

 its own pollen, and the other by that of another. Although the 

 flower seeds abundantly, yet I am disposed to think that it is 

 mainly through the pOllen of another that the seeds become per- 

 fect. I incline to this opinion because, in an examination of many 

 pods, I noticed that a few seeds were found in a rudimentary 

 condition, apparently manifesting a tendency to abort, while the 

 majority were in a vigorous condition ; the former, doubtless, being 

 the etfects of self-fertilization in part, which, as is well-known, is a 

 degenerating process. 



I desire also to call attention to an interesting discover}^ which 

 I was enabled to make recently, whilst engaged in an examination 

 of a double flower of Ranunculus fasciculai^is. In the genus 

 Ranunculus, the corolla of a normal flower is made up of five 

 petals, each of which on the inner side of its basal part is usually 

 provided with a scale. This scale from its position is denominated 

 the nectariferous scale. 



In the specimen under consideration three of these scales had 

 assumed the character of petals, agreeing with the flower's true 

 petals in every particular except size, being but three-fourths the 

 dimension of the latter. It very frequently happens that we find, 

 in examining flowers, parts which we can refer to no organ with 

 which we have become acquainted. They appear to be distinct 

 from any of the whorls which make up a perfect flower, although 

 located among them and attached perhaps to them. All such 

 parts are designated as appendages. Under this category are 

 placed the scales that are characteristic of some species of Crow- 

 foot. 



Prof. Lindley thinks that these small appendages are barren 

 stamens united to the bases of the petals. This opinion I think 

 is a just one. 



From the facts here indicated it is reasonable to conclude, tliat 

 the double flowers of the Ranunculus do not alwaj^s originate by 

 true staminal metamorphosis, but sometimes by scale transforma- 

 tion ; also that nectariferous scales when they exist are barren 

 stamens, whicli favorable conditions may develop into true petals. 



Whilst examining several specimens of PotentiUa canadensis 

 lately, I was struck with the variableness displayed in the number 

 of seo-ments which constituted their outer or calvcine whorls. This 

 series in Fotentilht., as is well known, consists of five sepals, with 

 as many intermediate bractlets. 



