Alternating Annual and Biennial Habit. 293 
cultivated and have nevertheless not been extirpated. 
The general opinion of botanists is that the represen- 
tatives of the main line of the evolution of plants have 
been for the most part perennials. From these the an- 
nual and biennial forms must have arisen independently 
in the various families and groups; and it is further 
natural to suppose that the biennials arose first and that 
the annuals arose from them. If this is true the pro- 
duction of a biennial from an annual or of a perennial 
by one of these two would have to be regarded as a 
phenomenon of reversion. 1 Instances of such atavism 
seem to occur very generally in the vegetable kingdom, 
but progressive transitions, that is to say, those that take 
place in the opposite direction, are also by no means 
rare. 2 
From the abundant literature on this point I select 
two cases which seem to me the most important. Pha- 
scolits imdtifloms (Ph. coccinens L.) is, with us, an 
annual plant, producing occasionally, however, a bulbous 
root which can be wintered and by means of which the 
plant can be perpetuated. VON WETTSTEIN, S to whom 
we owe our knowledge of this phenomenon, has obtained 
plants which lived four years, and in my own experi- 
mental garden I have succeeded in wintering several 
such Phaseohis tubers. VON WETTSTEIN'S view is that 
we are dealing here with the transformation of a peren- 
1 Many, however, hold the opposite view. See DARWIN, Varia- 
tions, II, p. 5 ; and RIMPAU, loc. tit. 
2 See the works relating to this subject by IRMISCH and WARMING. 
Also HILDEBRAND in ENGLER'S Botan. Jahrb., II, 1882, pp. 51-135; 
with regard to different sorts of beets : F. SCHINDLER in Bot. Ccntral- 
blatt, 1891, Nos. 14 and 15, and the literature cited there. 
3 R. VON WETTSTEIN, Die Innovationsverhaltnisse von Phaseohis 
coccincus L. (= Ph. multifiorus IVilld.}, Oesterr. bot. Zeitschrift, 
1897, No. 12, 1898, No. i. 
