FISHER: SEED DEVELOPMENT IN THE GENUS PEPEROMIA 147 
In the abnormal, lobed ovules, it is frequently more than two cells 
thick, is usually irregular in shape,Jand is often more or less 
incomplete. The fertile lobe develops into a ripe seed, which is 
more or less deformed by the presence of the sterile lobe. The 
latter may cause a flattening of one side of the seed, or there may 
be a depression in one side of the seed due to its presence (Fic. 
27). The sterile lobe is always more or less cut up into smaller 
lobes, and this subdivision is extremely irregular and often very 
complex. The sterile lobes often contain large oil-cells, which are 
regularly found in the carpels, bracts, and rachis, but are not 
present in the normal ovules, nor in the fertile lobes of these lobed 
ovules. 
Frequently, in young ovules, one or more of these sterile lobes 
appear potentially able to form an embryo sac, as indicated by 
their shape, their affinity for certain stains, and sometimes by the 
presence of cells which appear to be sporogenous in character. 
In fact a few cases were found in which the two lobes of the ovule 
were fertile, each having an embryo sac and a separate integument. 
But no ovule was found in which more than one embryo had been 
initiated. In this respect, the phenomenon resembles the similar 
abnormality in Morus alba (Hofmeister, 29) in which, in all the 
investigated cases, the ovules aborted after the egg apparatus was 
formed 
The flower-spikes which showed this abnormality were all 
obtained from the two plants in the greenhouse at the Botanical 
Garden of The Johns Hopkins University. No such cases were 
found in the Jamaican material. It may be concluded that the 
greenhouse environment was responsible for the phenomenon, 
but this seems improbable when we take into consideration the 
fact that almost nothing of this kind was seen in the half dozen 
other greenhouse species which were studied in detail. On one 
plant of P. verticillata, which grew in the greenhouse, nearly one 
half of the ovules were of this lobed sort. In fact, in some spikes 
of this plant every ovule was lobed or branched. On the other 
plant, a very much smaller proportion of the ovules showed this 
abnormality. : 
Whether these abnormal structures have more than a terato- 
logical significance is difficult to determine. Their relative abun- 
