298 PROCEEDINGS OF THE ACADEMY OF [1893. 



the lower lobe of the forked style — pressed into it, and covered with 

 pollen. A number of flowers were opened, all exhibiting the same 

 phenomenon. The branched filament, as is well known, bears an 

 anther cell on one fork only. At the time above noted the arms 

 diverge in nearly opposite directions ; but eventually the barren 

 one becomes nearly erect, which has led some authors to describe 

 the stamen bearing the anther on a lateral horn-like process. 



The opening of the flowers is the work of the lower lip alone. 

 In the unexpanded bud.it extends inward and under the upper lip, 

 and falls over, seeming to clasp the anthers. When the opening 

 time arrives this lower lip is withdrawn till it presents a right angle 

 with the upper lip, and finally falling back till it reaches the tube 

 of the corolla. Soon after mid-day the style branches close — an 

 indication that fertilization has been accomplished. The style, how- 

 ever, continues to lengthen. The corolla only continues about 24 

 hours, as early the day following they drop out of the calyx tube. 



A small species of bumble-bee, rather numerous, paid constant 

 visits to the flowers. Hypothetically one might reasonably regard 

 the flower as arranged for cross-fertilization by insect agency. The 

 branches of the style seeni favorably situated to receive pollen from 

 the back of au insect visitor. As a matter of fact, however, none 

 of these visiting bees were seen touching the style branches in any 

 way, though I watched them closely for a long time. As the lobes 

 close at noon, there could be no benefit if they did touch them after 

 that. Insect aid could be given between 8 A. M., and noon. If in 

 any other part of the world they are of service they must be larger 

 than these bees were ; but as the flower certainly fertilizes before 

 opening at this season, in this place, it is not conceivable that the 

 hypothetical arrangements are theoretically correct. 



The head of flowers is a contracted spike, and is composed of 

 numerous verticils in which there are three flowers to each axil. 

 The central flower of these three is always the first to open, and 

 hence the anthesis is centripital. Though this would thus seem to be 

 the general plan, strange to say, the first flower to bloom on a head 

 may be from any part — either near the apex or near the base. 

 They appear from any part without any regard to order or system. 

 No explanation of this disorderly blooming seemed satisfactory. 



It has been a point with the author of these " Contributions " that 

 a leaf does not originate at the node from which it seems to spring, 

 but at some indefinite point on the axis below. In Brunella the 



