Chapter IV. Individual Variation in Size and Growth of Hylocomium spiendens
30 CARL OLOF TAMM
·;.
Fig. 5
%
Figs. 5 and 6. Distribution of weight classes of segment 2 of Hylocomium splendens in samples from Os prestgårdsskog, collected 15. VI. 1950. Fig. 5· Sample 672, moderately exposed. Fig. 6. Sample 677, shaded.
tions of the individual moss plants must be still higher. Evidently the biologiear variation inherent in the Hylocomium community is much greaterthan the errors in the measurements.
Unfortunately we must also pay attention to more or less systematic errors du e to possible changes in the segment useq as a base for the calculations .. While these errors can hardly invalidate our conclusions as disenssed on p. 23, they prevent more accurate growth determinations by the method used. Sampling from a large number of randomly assigned places on each occasion can diminish the error in absolute figures, thus improving the curve in Fig. 3· This very large in-crease in labour would, however, only earrespond to a small inin-crease in accuracy.
If a more detailed curve for the seasonal growth of Hylocomium splendens is desired, some method invalving direct growth measurements is preferable.
Distribution in size-classes
We may now consicler the main problem of this chapter: the individual variation in size and growth. At first we shall study the distribution in size-classes of the Hylocomium segments from small sample plots.
Figs. 5 and 6 present the distribution in weight classes of segment 2 (seg-ment 49) from two samples collected in
J
une, 1950, in Os near Bergen, western Norway. Similar diagrams have also been drawn for several other samples, especially from Sites I and II (Grenholmen), bu t these samples have not contained as many individuals as the samples from Os. Apart from this these other diagrams agree with Figs. 5 and 6. Fig. 5 is based upon all segments 2of Hylocomium splendens from sample 672 (207 in number), which have been weighed individually. Sample 672 is the entire m oss earpet from a plot 25 X 25 cm beneath a spruce, but with a good light supply (at the margin of a clear felling). Fig. 6 represents the 127 segments 2 from sample 677, which is a similar moss carpet, collected beneath a dense canopy of spruce.
3I We see at first in the diagrams that the average weight of a segment is much higher in sample 672 (exposed) than in sample 677 (shaded); the same is true of the maximum weight. The most striking feature in Fig. 5 is, however, the low frequency of small segments. The same would be the case in Fig. 6, had not the low average weight (in comparison with the dass width) concealed the fact. This will be shown more clearly in Fig. 8.
In communities of perennial plants a high degree of replacement of individ-uals by seeds (or spores) is indicated by the presence of a large number of small individuals, of different age but including seedlings (cf. LINKOLA 1935, PERTTULA 1941). Where the small size-classes are under-represented, the re-newal of the plants by seedlings must be slow, if occurring at all (TAMM 1948). Consequently such an underrepresentation of young and small plants is a characteristic feature of many closed plant communities, where propaga-tion by seedlings is possible only when some disturbing influence decreases competition.
The under-representation of moss individuals with small segments, which recurs in Hylocomium samples from very different habitats (see Figs. 14 to r8), should probably be interpreted in the same way as the corresponding phenom-enon in higher plant communities: propagation by spores is rare in the estab-lished community. A high frequency of germinating spores or protonemata is no proof that propagation occurs, as seedlings are also common in many communities where they never attain maturity. Indeed, no cases where young Hylocomium plants grow out from protonemata have been observed in the communities studied.
Variation in growth-rate. Renewal of the moss individuals Before we can accept definitely the conclusion that propagation by spores is rare in the Hylocomium splendens community, we need to know more about the growth rate of the individuals of different size. Another pertinent question is the origin of the small individuals, if there are no "seedlings".
The growth-rates of the Hylocomium individuals in samples 672 and 677 are depicted in Figs. 7 and 8, where the weight of segment 2 (49) is plotted against the weight of its parental segment (48). All individuals with the seg-ments in question are weighed except those damaged during preparation.
The following conclusions can be drawn from the diagrams: r) Segment 49 is much larger than segment 48, an observatiqn also valid for other samples, cf. Table IX. z) There is a very large growth variation, slightly greater in the exposed sample (Fig. 7) than in the shaded on e (Fig. 8). 3) There is no clear difference in growth-rate between small and big individuals, i.e., the regression line follows fairly well a straight line from the origin. 4) The dots representing sympodia branching from 1948 to 1949 lie well within the range
32
segment 49
50
4
30
20
10
.
"
00
. . .
•
... . ·' .
CARL OLOF TAMM
o
•
• •
•
o
•
•
•
.
• o•••••••
•.
•
• ••.. , ...
' . .. . ·-· ·:· ...
•••
• • 'b. ' . .
•... . . .
• • ••.. . ...
• o. . .
• ~
•
\o
•
··~
.... : ...
•• . ....
. . '
•• • •
.
_...
. .
•.. •
•
10 20 30
o
•
o
.
• o
o o
40 50 mg
Segment 48
Fig. 7· Weight of segment 49 of Hylocomium splendens plotted versus the weight of seg-ment 48. Sample 672, moderately exposed.
e "Unbranched" individuals O "Branching" individuals
of variation of the unbranched ones. When large samples have been studied, we have as a rule found higher average weights and faster growth of "branching"
individuals (cf. Table II), but the individual growth variation is apparently big enough to prevent this difference from showing up when merely a few individuals are compared.
According to 3) individuals of all size-classes have approximately the same average growth rate, and the growth of the small individuals is not sufficiently rapid to suggest a continuous renewal of the m oss individuals by "seedlings".
The occurrence of a few "branching" but rather small segments in Figs. 7
segment 49 mg 40
30
2 '!
• • • • l
. . ·.
l' J~·:·
• • • J~
10
. .. ,,
··.:·~··,
• · - · . ~.... ·· ..
, ... : .
• -
•
• o•
:! . .
0~---~---~---~---,
O 10 20 30 40mg
Segmenl 48
33
Fig. 8. Weight of segment 49 of Hylocomium splendens plotted versus the weight of seg-ment 48. Sample 677, shaded.
e "Unbranched" individuals O "Branching" individuals
and 8 may give us a clue to the origin of the very small individuals. (Small individuals .. are of course those with small segments; unbranched sympodial chains containing both very small and large segments are exceptions, in spite ofthe great growth variation.) The daughter segments of such small "branching"
segments, or at least one of each pair, are usually very weak and slender.
Such slender segments may also, as mentioned on p. 26, grow out from old segments, isolated from their daughter segments by the death of stem parts.
It is very difficult to decide whether all very small individuals have this origin or not, as they are extremelyfragile and often break off during prepara-tion; but i t is clear that a large percentage of the small individuals have started as such side-sympodia. Of those small individuals which cannot be traced to a paren t segment of more normal size, a large number have maintainedapproxi-mately the same size for several years, as long as they can be followed back.
I t has not been possible to decide whether the remaining group of defective small individuals contain some "seedlings" or not.
A renewal by vegetative propagation, branching, is continuously going on in the moss community. Though most sympodia seem to be long-lived, in-juries always occur, e.g. by animal excrement or trampling, and the places
3· Meddel. från Statens skogsforskniagsinstitut. Band 43: r.