thirty feet, like an anchor rope, to the surface, where it holds the large float and laminæ against the impact of the heavy surf. This stipe is often less than one centimeter in thickness for half its length, but of such surprising strength that the native fishermen tie their boats to these ready-made anchors and ride out a storm, as noted by MacMillan ('99). The stipe of Egregia, however, while slender and flexible, is not bare, but covered with very numerous short proliferations along its whole length giving it the appearance of a feather boa. Some of these are photosynthetic areas, some sporophylls, some floats filled with air. The presence of such organs as air vesicles so near the holdfast shows clearly the plant's adaptation to a shallow-water habitat. It grows attached to rocks which are never deeply submerged and are uncovered even by a moderately low tide, where its branches, buoyed up by their innumerable pneumatocysts, float with their whole lengths on the surface of the water. To the boatmen along that shore a thick bed of Nereocystis is a sure sign of deep water, but a bunch of Egregia as surely marks a rock to be avoided.

The youngest plants of Egregia are extremely difficult to separate from those of Hedophyllum. The juvenile forms of both these kelps are dark brown, distinguished from most others of their size by shorter stipes, together with a rather strong development of hapteres. The youngest plant of Egregia found (Fig. 28) was 25 mm. long, with a lamina about 20 mm. long and 10 mm. wide. The holdfast had already developed a circle of secondary hapteres, although the primitive holdfast could be made out beneath the secondary. The stipe was but 3 mm. long, cylindrical, and featureless except for a very slight thickening about a millimeter below the base of the blade. This appeared to be the beginning of the proliferations. which characterize the later stages of the plant.

The thickening of the stipe soon becomes more pronounced and develops into a pair of horns about a millimeter long just below the base of the lamina and lying in the same plane (Fig. 29). These are the only dis

Egregia. Five sixtos natural size.

Youngest plant, barely distinguishable from Hedophyllum at this

age, cf. Fig. 37, which is less eroded.

FIG. 29.
FIG. 30.

Plant showing the first pair of proliferations on the stipe.

Plant with the transition region roughened by many capillary proliferations, tuberculate ridges appearing in the base of the lamina.

FIG. 31. Base of a much older plant showing the differentiation of the first branch (b) made evident by the appearance of proliferations on its stipe, first pneumatocyst (p) just appearing, base of stipe remaining smooth.

FIGS. 32-34. Egregia. One half natural size.

FIG. 32. Whole of the plant shown in Fig. 31, proliferations on the lamina absent at the tip, but well developed below.

FIG. 33. Much younger plant than Fig. 32, capillary proliferations prominent in the transition region, laminar proliferations just beginning to appear on margin of both stipe and lamina, tip of lamina smooth, other portions covered with protective ridges.

FIG. 34. Older plant in which the lamina has reached its maximum development and the stipe has begun to grow, several well-developed pneumatocysts and young branches are evident among the outgrowths from the stipe.

tinguishing features of the plant until it reaches a length of about 40 mm. In plants of about this length a few round tubercles begin to appear at the base of the lamina, which has hitherto been smooth as in Hedophyllum. A specimen 75 mm. long (Fig. 30) showed numerous tubercles in the transition region, giving it a roughened appearance; and there were three instead of two horns below the zone of the tubercles. The basal portion of the stipe was still smooth as in the youngest specimen. In this plant the stipe had elongated scarcely at all and the growth had been restricted to the lamina, which extended through 70 of the 75 mm. of the plant's length. Tubercles similar to those of the transition region had also appeared and these were shown by transmitted light to be connected with streaks of denser tissue running lengthwise through the lamina.

After this stage, as in Lessoniopsis there is some variation in the age at which the various structures appear. A specimen 18 cm. long (Fig. 33) will serve as an illustration of the next step. Here the streaks beneath the tubercles of the lamina had become prominent ridges, much larger than the small tubercles at their summits. The ridges stood out so strongly as to cause depressions on the opposite side of the lamina beneath them. This gave the lamina a wrinked appearance and added greatly to its strength. The margin was entire or slightly undulate, but at the base were a few short serrations which looked much like the tubercles of the stipe. The roughened region of the stipe was about 1 cm. in length and no longer terete like the lower smooth portion, but somewhat flattened. In place of the horns of the younger specimens were several outgrowths, the largest of which bore a small orbicular lamina. The holdfast had become nearly 2 cm. in diameter by the great elongation of a few hapteres.

In view of the proportions assumed by the adult plant the relation between the lamina and stipe in the juvenile forms is most interesting. In the smallest specimen the lamina is only about three times as long as the stipe.

But further growth is for a time almost restricted to the lamina until the ratio is increased to ten or fifteen to one. After this stage the stipe begins to grow and soon surpasses the lamina, which seldom exceeds half a meter in length, while the stipe sometimes becomes fifteen or twenty times as long.

A specimen 12 cm. in length (Fig. 32) though only two thirds as long as the one just described, was considerably more advanced. The uppermost quarter of the lamina was entire, as in the last plant, and below the tip were a few serrations like those at the extreme base of the former. Toward the growing point these outgrowths were larger and had become spatulate proliferations about a centimeter long, fringing the basal two thirds of the blade. The stipe had reached a length of 3 cm. Its numerous tubercles were much elongated and frequently dichotomously branched, once or even twice giving the peculiar roughened appearance characteristic of the adult. The proliferations along the lateral edges of the stipe were much more numerous than in the former specimen; some of them were simple laminar appendages; others were inflated into small globular pneumatocysts (Fig. 31, p); on others the stalks were roughened by small tubercles like those of the main stipe. Some of these last, if detached, might easily pass for young plants cut off just above the holdfast.

Though marked changes are yet to occur before the plant becomes mature, they may be understood by a comparison of the adult with this young plant (Fig. 32). The most conspicuous change is of course the great elongation. While this is especially noticeable in the stipe, the lamina likewise grows until it reaches a length of about 50 cm., but its width increases scarcely at all, seldom exceeding 4 cm. The proliferations from this narrow lamina become so numerous that they completely mask the distinction between it and the stipe, and it is only by close inspection that the lamina may be recognized. The growth of the stipe carries the lamina far away from the holdfast, where it is exposed to the severest action of the

waves, which lash the plant until the lamina together with the meristem is torn off and there remain simply the stipe and holdfast.

The stipe remains smooth for a few centimeters above the large branching holdfast, this being evidently a persistence of the smooth basal region of the young plant. Some of the lower tubercles, however, disappear, so that the smooth area now extends farther from the base than originally. This portion is terete, but at a length of about a decimeter the stipe becomes flat and strap-like about four times as wide as thick.

In the younger specimens the proliferations from the stipe and lamina are all small and not very numerous. In the adult they enlarge very greatly and increase in numbers so as to become by far the most conspicuous feature of the plant. The increase, both in number and size, is most marked toward the growing point, those at the base generally remaining small and scattered. Farther out along the stipe they are found of all lengths up to about 12 cm. and of various forms, as figured by Ramaley. They stand as thickly as possible along the stipe; in some places by actual count upwards of a hundred were found in a single centimeter of its length. Of these only a few were large and more than half less than a centimeter long. Crowded as they are along the edges of the stipe, they never arise from its faces, which are bare except for the tubercles described above. The air vesicles are formed at frequent intervals, providing sufficient buoyancy to keep the plant floating just beneath the surface with the tips of the proliferations emerging. When mature, they are about 30 mm. long, with an average capacity of about one cubic centimeter. Others of the outgrowths remain permanently small and become sporophylls. The outgrowths on the lamina also increase in size and number, but become neither so large nor so numerous as on the stipe. As noticed by Ramaley, no bladders nor sporophylls develop on the lamina.

Egregia becomes much branched before it is mature. Although Ramaley suggests that the branching may have