40 University of California Publications in Botany. [VOL. 4 



who have propagated them have had much difficulty because of 

 the readiness with which the young seedlings damped off. All 

 of these facts, together with the isolation of the genus from all 

 related forms, seem to indicate that we have here a decadent 

 race, marooned on these few islands where its last stand in the 

 fight for existence is being made. 



That this race of trees will become extinct unless protected 

 is a certainty. Steps should be taken, either by the State or 

 Federal Government, to acquire and protect at least a few of 

 the groves. An important factor in the weakening of this race 

 is doubtless that of inbreeding. There are apparently no special 

 devices for cross pollination and, judging from the botanical 

 characters of the two forms, there is almost certainly no transfer 

 of pollen from one island to another. If artificial cross polli- 

 nation were practiced between the typical form and the var. 

 asplenifolius, a good yield of fertile seed would probably result 

 and give rise to a stronger race. At least, the experiment should 

 be made. 



Propagation is best effected by seed. These are largely in- 

 fertile, so that only a small percentage can be expected to germ- 

 inate. In the past, this seed has all come from Santa Cruz 

 Island, but cultivated trees at Santa Barbara are now yielding 

 a more convenient supply. In propagating from seed, great 

 care must be exercised to see that the seedlings do not damp off 

 in the early stages. Stumps have been successfully transplanted 

 but their shipment from the islands is expensive. They may 

 be divided by splitting, but experience has shown this to be a 

 precarious proceeding. All attempts at layering and propa- 

 gation by means of cuttings have failed up to the present time. 



Eugenia myrtifolia Sims. 2 

 AUSTRALIAN BRUSH CHERRY. 



An evergreen glabrous tree, or shrub-like: leaves opposite, oval or 

 oblong, narrowed to each end, short-stalked, mostly acute, entire, 1^ to 

 3 in. long: flowers in loose clusters: calyx-tube top-shaped, the lobes 



2 Some confusion has arisen because of the various names applied to 

 this tree. Eugenia australis Wendl. is exactly the same, according to best 

 authorities. Jambos myrtifolia Nadz. is the name adopted in Engler and 

 Prantl 's NaturlichenpflanzenfamiUen. 



