The age of the trees on Wizard Island in Crater Lake has been used to suggest the least possible time lapse since the last eruption of that volcanic cone (Waesche, 1934; Williams, 1942). Evidence currently available suggests further that Crater Lake has at some time engulfed a higher portion of the Wizard Island cone. The cooling patterns found in the lava on Wizard Island indicate that neither the Wizard Island cone nor the block-type lavas of the western side of the island, which are the more recent, were formed under water (Williams, 1942).

Hence, the last eruptions must have preceded the period during which the island was more extensively engulfed. Establishment of woody vegetation in the area covered by water would have been dependent upon a decrease in the lake level relative to the island rather than the cessation of volcanic activity. Therefore, the trees on the lower portion of Wizard Island should serve as a measure rather of the minimum time which has elapsed since this greater engulfment than of the minimum time which has passed since the last eruption.

Diatoms are members of a primitive plant group, the Algae. Freshwater sponges belong to a primitive group of animals called the Porifera, or "pore-bearers." Both the valves and frustules of diatoms and the spicules of fresh-water sponges are composed of hard, and very resistant, siliceous material. For this reason, these hard parts of diatoms and freshwater sponges are frequently well preserved in the fossil record.

Skeletons of diatoms (valves or frustules) and the skeletal framework (spicules) of fresh-water sponges have been found on Wizard Island at considerably higher levels than the present surface of Crater Lake, which averages 6,176 feet above sea level. Hegeness, for example, has found diatoms some forty feet above the lake level (Williams, 1942). The writers collected soil samples in areas protected from erosion, at intervals of ten feet, extending from the margin of the lake to 110 feet above the present lake level. Diatom valves and frustules were present in all the samples. Most of the valves were corroded or broken and gave every indication of age and weathering. In some of the samples, the diatoms were mixed with the spicules of fresh-water sponges.

Some of the diatoms obtained represent genera known to occur only in standing or flowing waters -- never in soils. Included here are specimens of Melosira and large forms of Pinnularia (Sovereign, 1955).

Freshwater sponges do not live in snow seeps or moist earth, but only in standing or slowly-flowing water. Melosira valves and the spicules of fresh-water sponges were both found in samples of soil collected at 100 feet and 110 feet above the present level of the lake (Sovereign, 1955). This means either that the level of Crater Lake has at some time been at least 110 feet above its present elevation or that Wizard Island has been raised in toto -- or possibly that a combination of both activities has been in operation to produce this result. In 1954, the summit of Wizard Island was 764 feet above the average lake level for that year.

Counts of seasonal growth rings made for trees on Wizard Island indicate that trees have been growing there throughout the past 800 years (Waesche, 1934). Waesche's report has been essentially confirmed by similar investigations which the writers carried out during the summer of 1954.

The increment borer used to obtain cores from trees on Wizard Island was twelve inches in length. The writers are indebted to the Union Creek District Ranger Station, Rogue River National Forest, for the use of this instrument. With this borer it was not possible to obtain a complete radial sample of growth rings from the larger trees. It was necessary, therefore, to estimate the age of trees whose diameters were greater than sixteen to eighteen inches.

The increment borer inuse From Kodachrome by C. Warren Fairbanks

In attempting to approximate the age of these trees by means of partial samples of the growth rings, there are several important sources of error. One of these is the general decrease in the thickness of growth rings with age -- from the center of the tree outward. As is typical, the young trees on the island had thicker rings, especially toward the pith (center), and the old trees had thinner rings in the outer layers, these being the only ones accessible with the borer. Thick rings are associated with rapid growth, thin ones with slow growth.

A second important source of inaccuracy, so-called false (or extra) rings and absent rings, can affect the results even in an actual count of the growth rings from pith to bark. A "false ring" may be formed during a year of unusually irregular climatic conditions, two rings then being produced for the one year. A year of relatively constant climate may result in the failure to form a readily detectable growth ring, thus an "absent ring."

Errors from this latter source would be exaggerated in an estimate based on the rings in a short, partial boring (core). In order to minimize such errors, it is important to base a calculated estimate of age on as many growth rings as possible. In a boring representing a period of many years, these deviations from the expected pattern of growth would tend to cancel out. Therefore, a seven-inch unit of core length was used as a basis for our calculations, this being the greatest core length that could be obtained with the twelve-inch increment borer from trees having a thick layer of bark.

An attempt was made to minimize inaccuracies resulting from the former source -- decrease in the thickness of growth rings with age -- by the following procedure: For each species, growth rings were counted in the outermost seven inches of several larger (older) trees and in the innermost seven inches of several smaller (younger) trees. An average number of growth rings in each of these two seven-inch units was then calculated for each of the three species. By the additional averaging of these two averages for each species, an average age per seven-inch sample was established for each of the three species.

It was intended that this method should produce three values (constants) whose use would result in the most accurate estimates possible under the circumstances for the true ages of trees in the respective species.

As a check upon this method of calculation, a cross-section was cut from a Shasta red fir which had fallen during the 1953-1954 winter. This tree had a diameter of 25.5 inches, and its total age was counted, not estimated, to be 406 years. Applying the above method of calculation, e. g 189 years per seven-inch sample, a Shasta red fir 25.5 inches in diameter (considered as a 12.7-inch radius) would be an estimated 340 years old. This single check indicates that the ages calculated for Shasta red firs in the table may be on the order of fifteen percent too low. If this should actually be the case, then the age of the Shasta red fir here estimated to be the oldest (701 years) would slightly surpass Waesche's report of 800 years.

With due appreciation of these difficulties, together with such others as reliable diameter measurements and differences in the thickness of growth rings on various radii, we have estimated ages for fifteen of the largest trees, representing the three dominant species, on the lower portions of Wizard Island. The results are presented in the following table.

Data concerning variation in the width of seasonal growth rings, although they may have meteorological significance, have not been included here because of their fragmentary nature at present. These data, along with the increment borings, are on file in the Crater Lake National Park Library.

If some of the trees now on Wizard Island were living when the lake level was relatively higher, extensive sampling and precise dating of trees at all levels should reveal that the oldest trees are located at some distance, presumably more than 110 feet, above the present water level. Since none of the tree species listed here will tolerate submersion, those on the lower parts of the island would have to have become established after a decrease in water level relative to the island.

At present, the largest trees on the island are located at the lower levels, but it is not necessary that trees of the greatest diameters be the oldest. To the contrary, examination of trees at higher elevations indicates that their growth has been slower than those near water level. Therefore, the oldest trees may well be of lesser diameters. Growth of trees on Wizard Island, it should be noted, is suppressed in general and is not comparable with the growth of the same species under more favorably habitat conditions.

It is, of course, quite possible that none of the trees now growing on Wizard Island represent the first generation since either the exposure of the lowermost portions of the island or the cessation of volcanic activity which would have effectively deterred the establishment of trees above the all-time-high water level. However, the soil contains little organic material, which suggests that trees, other plants, and animals have been sparse in the past, if not essentially lacking prior to the time of establishment of the oldest trees now growing on the island.

Although we did not locate any tree having an estimated age equaling the 800 years reported by Waesche, the estimated ages of several of the trees sampled approached it fairly closely. On the basis of the evidence at hand, it is concluded that an interval of not less than 750 years has elapsed since the water of Crater Lake covered a substantially greater portion of Wizard Island than at present. It has been at least this long, then, since the diatoms and sponge spicules were deposited where these trees now grow.

It must be emphasized that only the Shasta red fir referred to above has been dated accurately and that many of the preceding suggestions are speculative, especially those predicated on the estimates. It is hoped, however, that these speculations approach the facts at least reasonably well.

The writers wish to express their gratitude to Mr. H. E. Sovereign for determinations of the diatoms and sponge spicules, and for his invaluable suggestions.

References

Sovereign, H. E. 1955. Personal communication.

Waesche, H. H. 1934. Geology of the boat trip around Crater Lake. In: Crater Lake National Park, Ranger- Naturalists Temporary Manual of Operation. Berkeley, California, The Field Division of Education (Mimeographed). 109 pp.

Williams, H. 1942. The Geology of Crater Lake National Park, Oregon. Carnegie Institution of Washington Publication 540. vi, 162 pp, Washington, D. C., Carnegie Institution of Washington.