584 Burns. — Heterophylly in Proserpinaca pains tr is , L. 
Returning to the cuttings (p. 582) which were made in May from 
Proserpinaca plants producing entire leaves, it will be remembered that 
they continued to produce entire leaves, blossomed, and in many cases 
produced fruit (Fig. 1). Early in September it was noticed that the main 
stem did not grow so rapidly and that several side-shoots developed at the 
surface of the pot Here we meet a most interesting picture in leaf-forma- 
tion. The main stem is orthotropic and produces the ‘ land-type 9 of leaf, 
while the side-stems are plagiotropic and produce the ‘ water-type 9 of leaf. 
Both vegetation-points are under the same external conditions, as far as 
they could be controlled, and yet what a marked difference in the leaf 
formed ! Again a little later the main stem undergoes a change. It ceases 
growing erect and grows horizontally, and the vegetation-point, which since 
May has been producing the ‘ land-type 5 of leaf, begins to produce, in air, 
the ‘ water-type ’ of leaf (Fig. 5). This experiment finds its exact repro- 
duction in nature and is of the greatest biological importance. As has 
been pointed out, Proserpinaca , during the summer, sends its branches out 
of water, producing entire leaves, flowers, and fruit. In the fall the main 
stem becomes plagiotropic, produces divided leaves, and finally sinks into 
the water. Here it passes the winter protected by the water, and in the 
late fall and early spring multiplies rapidly by the growth of axillary buds. 
It must be noted that this return to the water-form is not caused by the 
water-stimulus. I have fastened many plants so that the vegetation-point 
could not possibly be closer than 30 cm. to the water, and yet these 
plants returned to the formation of the ‘ water-type 9 of leaf. 
In June of the present year the vegetation-point was removed from 
a number of plants producing entire leaves. This caused the lateral 
branches to develop. In many cases the leaves on these were more or less 
divided, although the plant grew in air under the same conditions as 
many other Proserpinaca plants, all of which were producing entire leaves 
(Fig- 3 )- 
Thus we see that the presence of water is not necessary for the 
development of the ‘ water-leaf. 9 
On the other hand, Proserpinaca palustris may produce entire leaves 
in spite of a water-environment. In the spring I always found a number of 
plants producing the { land-type ’ of leaf under 24 to 30 cm. of water. This 
I found only on rapidly growing branches. In June of the present year, in 
a large area at Dead Lake, I found that almost every plant was producing 
the ‘ land-type 9 of leaf, although dozens of them were growing under as 
much as 30 cm. of water (Fig. 4). This certainly points to other factors 
than water or checked transpiration as the cause of the division. 
In Fig. 4 attention is called to the two leaves a and b. This tip was 
growing under 26 cm. of water. These two leaves show a transition from 
the one type to the other ; they are * water-leaves 9 at the tip and ‘ land- 
