1899.] NATURAL SCIENCES OF PHILADELPHIA. 103 



to account for a seemiug departure in regular law, wlieu accouut- 

 ing for its singular structure. 



In tracing the morphological relationship of the parts of a flower, 

 we are largely aided by examining a nuiiiber of species. In this 

 study of Asclepias tuherosa, A. verticillata, A. Syriaca and A, 

 incarnata also took part. 



From some vigorous shoots of Asclepkis incarmda, the lower 

 leaves showed that normally the leaves are alternate, and arranged 

 on a f plan. Whatever variation from this plan follows, must 

 be referred to some discordant occurrences. In A. incarnata,. 

 we see how this operates in forming the branching character. The 

 main stem is subtended, when it begins to branch, by a weak 

 branchlet on one side and by a leaf without an axillary bud on the 

 other. The exact fact is that what now appears the central axis 

 or leading stem was the axillary bud ! When the rhythmic growth- 

 wave resumed activity the chief energy was directed toward the 

 axillary bud. It pushed aside the former leader, and reduced it to 

 a mere branchlet. We can still trace this disturbing force through 

 many nodes, though Avith continued arrestations and developments 

 it is often difficult to trace the originals eventually. But we learn 

 by this to look for these aberrations of the various parts when con- 

 sidering A. tuhevom by itself. In the case of the latter species we 

 find in the earlier states of growth the axillary bud is not able to 

 dislodge the central shoot from leadership. In this case we have 

 the branchlet in the axil of the leaf, and not opposite, as in the 

 other species cited. Later on we may find the growth-energy 

 directed strongly in the axillary direction. But we have learned 

 the valuable lesson that variation in the period of rhythmic 

 growth -energy is sufficient to produce specific characters. 



With the ideal quinary plan in mind we find in Asckpias tuberoscc 

 five sepals, and the series of five petals is alternate Avith the sepals 

 as such a series should be. Then we have, alternate Avith these, a 

 series of five stamens, but the filaments have become united up 

 to the connective (represented by a minute dorsal gland), Avhab 

 should have been anthers are petaloid, and we have to call that a; 

 nectary Avhich should have been a more elaborate organ. The five 

 perfect stamens next appear, and properly alternate Avith the five 

 petaloid stamens (nectaries) beloAV. Noav Ave iJerceive Avhat, if 

 noted before, has not been recorded : fiv^e greenish scales, alternate 



