1893.] NATURAL SCIENCES OF PHILADELPHIA. 291 



flower branch remaining in the dichotomous fork. In Boragina- 

 cece with its alternate leaved system, there is but a single bud to 

 push into growth and become the main axis, which it does, push- 

 ing the arrested flowering branch wholly to one side, aud which, in 

 technical language becomes extra-axillary. 



Missing Virticel in Glaux maritima. 



Examining on July 14th at Seal Harbor some specimens of 

 Glaux maritima, commenting on the well known and remarkable 

 alternation of the stamens with the lobes of the calyx, a friend 

 remarked that this arrangement would necessitate the lobes of the 

 corolla being opposite the lobes of the calyx if the flowers had not 

 been apetalous, which would be an anomaly in floral structure. 

 But as the stamens in Primulacece are opposite the lobes of the 

 corolla, we might naturally look for the same phenomenon in Glaux 

 had the corolla been developed. Yet there is one point in this con- 

 nection worth noting. In Samolus and some allied genera there are 

 imperfect anthers in between the lobes of the corolla, and the 

 opposite stamens are therefore easily accounted for when we under- 

 stand that a series between them aud the corolla has been sup- 

 pressed. The position of the stamens alternate with the sepals in 

 Glaux, again proves the existence of two series, as the suppression 

 of both would bring the stamens just where they are between the lobes 

 of the calyx. Another point is that the ease with which the outer 

 series of stamens, united with the corolla in Primulacece generally, 

 has been suppressed, causes no surprise when we see the corolla 

 itself wholly suppressed in Glaux. 



The tendency to suppression of parts and general irregularity is 

 particularly striking in this genus. Though I examined many 

 flowers of this plant on the north Pacific coast, I did not note any 

 that were not 5-lobed and 5-stamened. Fully one-fourth of the 

 flowers at Seal Harbor were tetramerous. 



The irregularity of the leaves is well known — opposite until the 

 flowering stage is reached, when they become alternate and fre- 

 quently three-whorled. 



Every flower is fertile ; but all the flowers I could examine were 

 expanded, and afforded no opportunity to observe the condition of 

 affairs before opening. 



Morphology is much aided by a study of the acceleration and 

 arrestation of parts in erratic plants like these, and I regretted 



