

1909] 



LEAVITT—HOMOEOSIS IN PLANTS 



51 



specimen 



N 



of the three main branches of the scape. 



of the florets in each umbel are replaced by long-pedicelled umbellules. 



That is, the form resident in the flower cluster as a whole has instantly 



passed into several of the parts (florets) . A similar variation I have 



seen in Hydrocotyle umbellata (specimen 



Jamaica, Ferry, no. 6168): 



and in Daucus Carota it was 



noted by Cramer,* s ra dial 



florets being replaced by 



umbels. 



Two false umbels of 

 Pelargonium in my posses- 

 sion are abnormal in a 

 similar way. In each case 

 one of the florets, eccentri- 

 cally situated, has been 



Fig. 15. — Normal leaf of jasmine (left) and 



transformed homoeoticallv a leaf With homoeotic segments (right). After 



• ■| - J 1/WTlLTtVW 



with the result that it 



Bonnet. 



rises 



as a pseudumbel of a secondary order from amid the flowers of 

 the first. The subordinate flower cluster is like the chief cluster as 

 regards centrifugal development, involucre, etc., though the flowers 

 are fewer (8 instead of 18) and shorter-pedicelled. The secondary 



peduncle 



off 

 flow 



is articulated to the expanded summit of the primary, as 

 nff C . e P edlcels of the flowers, and apparently would have been cast 

 n > m event of failure to fertilize, by an absciss-layer, as with the 

 m ers among which it stands. 



Homoeosis and reversion 



amon ^h ^ ^ homoeosis is often confused with reversion; indeed 

 ThT g M amsts the confusion may be said to have been habitual. 

 1 he old er writers havh 

 J^rsion in its ancient 



little rpfl<w,- 



homoeotic Dhenomena the word 



some contemporary 



in 



Cramer, 



modern 



864. 



Rhodo 



Aversion irl*a ** 7 \. I9 * I9 °5> I hay e discussed briefly the origin and use of the 



*ea as apphed to plants. 



