suddenly stopping anywhere in open

ocean.

When I came to think of all the details of my new endeavor, the subconscious worry and fear of the whole expeditionary responsibility, always hanging over me, became more vivid, floating to the surface of my mind and unpityingly pointing out the situation. A ship is made to travel, its engine to throb; and although I was in complete command, yet the shadow of my old passenger subordination always lay heavy upon my decisions. There seemed, too, something against all the traditions of the sea in thus willfully turning a perfectly good vessel into a derelict of sorts even for a time. I pictured the weed and barnacles on the keel as sprouting forth in awful rapidity of growth during the period of inaction, the engines becoming rusty, the engineers and oilers falling asleep one by one indeed, before I knew it I had visualized another Flying Dutchman, only under a static, instead of a dynamic, spell; I seemed to be laying the foundation of a Pacific sea of dead ships.

I prepared for the experiment by the study of a wholly different type of fish fauna, the shore fishes of Cocos Island - that speck of land so beloved by the pirates of old, about five hundred miles off the coast of Panama. If preliminary success was augury of good luck, I should have been contented, for the finny inhabitants of Chatham Bay yielded up their secrets in wonderful fashion. The rainy season had been a jest at the Galapagos, but no season ever merited the epithet more than the one we spent at Cocos. As I passed most of my time in my diving-helmet beneath the surface, I hardly noticed the constant downpour, but it was a fact that the air was saturated most of the time. Dwight Franklin one day laid on my laboratory desk a water

color sketch marked 'Cocos,' a composition consisting of a wide expanse of sea with a small smudge of a rainstorm in the centre - a joke, but not an exaggeration.

When I had once halted my ship in mid-ocean I had no hesitation in knowing what to do. I wanted to learn all I could of what flew in the air, floated on the surface, dived in the depths, or burrowed into the substance of this tiny pin-point in the great Pacific. But now that I have finished and steamed away, and weeks have passed since the last dredge came up, I am confused as to the manner of telling about it. What I did, day and night, of dredging and trawling, was done so blindly, so gropingly, what came up was such a pitiful fraction of the great mass of life which must be below, that I feel like a deaf, dumb, and blind person attempting to interpret a wholly new and strange world.

With more usual islands one naturally begins with the life of the ground, then that of the trees; and finally, with net and gun and glasses, one collects and studies the beings of the free air. Here I shall reverse the process and begin with the top of the water column.

II

On Sunday, May 24, in the late afternoon, we pulled up anchor at Cocos Island and steamed westward out of Chatham Bay, slowly encircling the island. After skirting the southern headlands and passing the zone of uncharted shore, I gave orders to turn south; and in a swirl of wind and rain Cocos changed from dull green to gray, and finally was lost in the black mist of night. Under slow speed we crept southward, and at dawn, with the mountainous little island just visible on the northern horizon, Bill Merriam let go the sounding weight. Minute

"The centre of the island is four and a half degrees of latitude north of the equator, and eighty-seven degrees of longitude west of Greenwich. Its nearest terrestrial neighbor is Cocos Island, which is due north one degree, or sixty miles. To the southwest, three hundred and fifteen miles away, is Tower Island, the nearest of the Galapagos; the nearest point on the American continent is Llorena Point, Costa Rica, three hundred miles northeast. The inhabitants of Seventy-four are engaged chiefly in fishing, its exports being fish, sea cucumbers, jellyfish, and other marine products; while its imports consist of entangled dredges, coal ashes, and fresh-water rain. For ten days it was a colonial possession of the United States. It has now reverted to No Man's Land and the realm of memory and imagination.'

Cocos vanished from sight early in the evening of that damp Sunday, yet day and night thereafter we were constantly to feel her influence, even when sixty miles away. The rain steadied to a downpour, and as I looked out of my cabin door the deck was a maze of starred splashes, and the edge of the

blackness a thin screen of slanted pearlgray lines etched on the substance of night.

At midnight the unending warp of rain still threaded the invisible sky and sea. I lay in my bunk and listened to the unearthly cries of the confused sea-birds. The high, shrill, pitiful notes filtered through the murk, and then suddenly several ghostly forms would shape themselves, fluttering tremulously far out in the driving wind and rain, proving that the darkness was not darkness after all.

In the museum of Uyeno Park, Tokio, there was once an incomparable collection of kakemonos, the rarest work of the best old masters of Yeddo and China all taken now by the earthquake. Unknown to me there was hidden, deep within a forgotten cell of memory, a clear-cut vision of one showing sea gulls flying in the rain. And now, on this rainy midnight at sea, the picture flashed to consciousness, for there before me, framed in the long rectangle of my cabin door, Hokusai's kakemono lived again.

III

I lay back in the bunk, writing on my drawn-up knees, my posture recalling Stevenson or Mark Twain in everything except the value of what I wrote. A half-hour passed, and the rain was Monday's rain, when I heard a gentle whipping of wings - the sharper tone which is given out when wings are very wet.

In mid-air in my cabin, beating a little cross-current to my electric fan, was one of the fairy terns of Cocos. As I looked, the immaculate little beauty fluttered upward and poised close to the wall light, then sank slowly and came to rest on my knee. I finished my sentence and began to write a description of the dainty bird while it ruffled

and shook and settled its plumage into place, showering me with drops. I felt no envy for Stevenson or Mark Twain now. For the space of several minutes we looked at each other, the tern much the more composed and less breathless of the two of us. Then, light as thistledown, it rose, fluttered over to my desk, and alighted in the middle of a large map of Cocos Island which happened to be lying there. (Go ahead, reader, say it yourself; I won't bother to write it.)

For a long time the bird preened its white plumage, looking about with its dark, quick eyes and burying the slender beak deep in the feathers, fluffing them out. The chicory-blue of the beak was just the touch needed to set off the snow-white plumage. As it preened, it walked slowly about on the paper Cocos, the violet-blue webs between the toes pattering softly. Then the long, angled, capable wings were stretched, high, high up, and a halfdozen quick beats lifted the whole little being, making palpable the thin air. Without haste, yet without hesitation, the fairy tern drifted out of the door, glimmered like a painted kakemono ghost for a moment, and vanished.

I watched the same slanting lines, listened in vain for any last call it might have sent back, and wondered whether I had not dreamed a dream. But the map of Cocos Island showed a cluster of little swollen blisters where the damp drops had raised the paper, and to the paper-flat slopes of Mount Harrison there clung a tiny feather- not a soft and downy feather from the body, but a little tertiary from the wing itself.

Again I looked out and marveled how such a pinch of white fluff, scarcely a foot in length, weighing less than five ounces, could have the courage on such a night to leave light and shelter

and safety for it had showed not the. slightest fear of me and launch out into the driving rain, with the nearest tree forty miles away.

During this first night of rain and wind, boobies by the dozen also sought haven on the lighted steamer, in a manner far different from that of the white tern. They heralded their coming with squawks, sounding muffled through the distance and rain, and then flopped to the decks or against the cabins with a bang. Thereupon they raised their voices to the highest pitch of raucous outcry, launching awful protests, screaming curses of anger and fright, until the steamer rang with the noise. Toward morning a great red-footed booby bludgeoned into my room, missed my face by a narrow margin, and thrashed his way out again.

I snapped on the light and envisaged a mill of devil birds. At my threshold my visitor encountered another of his kind, a hated rival of long standing, it appeared to me. Each immediately credited the other with all the blame for the storm, the confusion, and an intense dislike for this newfound sanctuary. A battle ensued, and with beaks gripped on one another's bodies the combatants remained locked, lying on their sides, squawking full steam through half-closed beaks until I went out and hurled them both over the rail. After the voluntary leavetaking of the white tern I had no fear for the safety of these great birds, provided the plunge cooled their frenzy of hate.

The rain ceased just before dawn and gave place to a strange, hard sunrise a scarlet slit in the ashgray of the east, and an unreal, pallid, greenish expanse in the north. In this eerie light, at five-thirty, we made the first sounding, which I have described.

IV

In the ten days during which I floated over my island I had rather remarkable luck in recording birds. I observed seventy-four altogether, comprising thirteen species. Six of these were sea birds from Cocos, which had come this great distance to some favorite feeding-ground, or, in a few instances, had perhaps been blown farther than they had intended to fly. Of those which came on board on nights of stress and storm, some were obviously exhausted, but most were apparently strong on the wing, and only confused and distracted from their true course by the sudden vision of the ship's lights.

Five species, three petrels and two shearwaters, were true pelagic birds, feeding as they flew and paying no attention to the vessel. Then there were two strays, probably storm-driven a gull and a warbler.

To be more specific, one day a frigate bird flew past, with its marvelously slow wing-beats, headed for Cocos. It may have been out for days without tiring, and in the case of such low-lying storms as those hereabout could easily rise above the level of the rain. The two Cocos boobies, the redfooted and the white-breasted, came in numbers to our lights. These birds travel thirty and forty miles to and from certain fishing-grounds, but are not capable of nearly so prolonged flight as the frigates. The boobies of Tower Island feed, for the most part, forty miles away from home, in the direction of Indefatigable, although fish seem quite as abundant near at hand; and at Cocos the same inexplicable habit would seem to hold. We caught several boobies on the decks and caged them for exhibition in the Zoological Park. When first caught they were fiends incarnate, dashing

themselves against the wire, screaming, and striking fiercely with their powerful beaks. Within three days they had become quiet, almost gentle, making no attempt to injure the hand that provided them with fish. A hint of the wonderful sight and balance that they use in diving after their prey is shown in the way they catch pieces of fish, for no matter how swiftly it is thrown, or at what awkward angle, with a slight twist of the neck the fish is caught.

Shearwaters were in sight almost every day - the dusky, and the larger white-fronted species. One day while I was watching a school of tunnies leaping high in air, a dusky shearwater wheeled into sight directly in front of the bow. I watched it with the glasses for a time and, as I had paper and pencil, followed its flight. I know of no bird better named than this. First on one side of the bow, then on the other, the bird described loops, doubling almost into figure-of-eights. Once it put on full brakes with wings and feet, spattered for a few feet through the water, with quick-paddling webs, snatched a small fish, swallowed it, and left.

When I had it in the field of my glasses I saw what to me was a wholly new observation — the dipping of the under wing-tip well into the water at almost each outer edge of the turns; and not only this, but a very apparent throbbing or successive fluttering of that wing alone (the other being held quite still), as if to increase the braking power or the fulcrum value of the heavier medium. It reminded me somewhat of my old days of polevaulting, when, running at full speed, I struck the tip of the pole into the ground. Time after time I watched the little furrow that the wing made, and saw the tremulous pressing against the slight hold of the water. After

forty or fifty observed repetitions I have not the slightest doubt of the material assistance that this habit gives to the ease of swift pivoting and steep banking.

Mother Carey's chickens, or stormy petrels, were present on most days, regardless of waves and winds, flickering cheerfully about their business of finding small prey. Leach's and darkrumped petrels I expected to find, but when a white-faced petrel (Pelagodroma marina) flew on board late one evening I knew I had a prize. This bird has its centre of distribution near Australia and New Zealand, but here was a straggler thousands of miles away, and yet strong on the wing and in good health. It became confused by the ship's lights, flew on board, and was not able to rise from the flat deck. It is accidents such as this that keep scientists from becoming conceited, realizing, as they must, how much of their knowledge depends on chance or accident.

The stray gull was peculiar to the Galapagos, and it flew around the ship wing-wearily one morning, like the one I had seen the week before at Cocos. Storm or wind or some strange wandering instinct must have brought them over more than three hundred miles of ocean. The white tern and the two species of noddies were all Cocos birds, out fishing when the drenching rain and high wind forced them to come aboard for rest. Numbers of birds must perish in every severe storm; for although these sea birds have welloiled plumage and webbed feet, yet a strange fear of the water obsesses them, and they alight on its surface only as a last resort, dreading some danger unknown to me, whether of shark or other fish, or of the fatal water-soaking of already drenched wings. Of my island avifauna there remains to be described only the most

unexpected visitor- a dainty Cocos Island yellow warbler, which appeared one morning in the rigging. The wind of the preceding night had blown from the east; it was not overstrong, and the night, although dark, was without rain; the arrival of this land bird was therefore wholly unexpected. For an hour it preened its plumage, then halfheartedly sang a single phrase of its simple ditty. It next flew down to the deck, where, with the skill of a professional flycatcher far transcending that of an ordinary warbler, it caught two flies which were humming about a dead fish. A moment later it rose, in a steep ascending spiral, and after gaining an elevation of about two hundred feet it darted along the compass line for Cocos, fifty-eight miles away.

V

As to completeness of representation of vertebrate classes on my island, I announce failure at once. No amphibian, whether frog, toad, or polliwog, existed nearer than the American mainland, but this was the only group missing. I lay flat in my bow pulpit one day, while we were slowly steaming in a great circle, drawing a half-dozen large tow-nets, when I saw two rocks ahead, just awash. Before the first impression could crystallize into actual belief, I detected the rounded, upturned heads, and knew that the class of reptiles could be included in my island fauna. They were big green seaturtles, although one belied its name, for its shell was a warm brick-red in color, dotted here and there with large white barnacles. They drifted slowly past me, one on each side of the Arcturus, merely turning their big heads, but not otherwise moving until they were tumbled by the bow waves, when they immediately dived.

Two species of sea mammals paid