[The factors determining temporal variation in Brachionus calyciflorus pallas (rotatoria)].

1. In the rotifer Brachionus calyciflorus a pair of posterolateral spines (PL) is subjected to extensive temporal variation. The length of these spines may vary from complete absence to more than lorica length. Production of spines is induced by starvation, low temperature, and a substance produced by the predatory rotifer Asplanchna. The aim of this study was, by comparing the effectiveness of these factors, to gain insight into the physiological mechanisms involved, and to assess their relative significance in natural populations. 2. The other spines of the lorica vary in accordance with PL (Fig. 1), however, they never disappear. In the absence of PL, they continue to vary in the same sense (Table 2). 3. In the laboratory starvation and low temperature induce only short PL. To elicit long spines the Asplanchna-substance is always necessary (Fig. 2). The effects of the three factors are additive. High temperature seems to inhibit spine production. The type of the inducing factor has no recognizable influence on the variability of the reaction (Figs. 3 and 4); even under conditions where the reaction is maximal no reduction of the standard deviation is found. 4. There is no clearly recognizable short-term inheritance of spine length: almost the full reaction to each of the inducing factors is realized in the first generation following induction (Fig. 5). 5. Low temperature increases body size. Analysis shows that this temperature effect consists of two components, one acting on egg size, the other on the animals growth after hatching (Figs. 6 and 7). 6. The final spine length is always determined before hatching (Fig. 5). The inducing factors determine the body proportions of the neonates; after hatching the spine length grows allometrically. The average growth rate of the PL relative to that of the body was k=0.54, i.e. the growth rate of the spines is smaller than that of the body (negative allometry). The value of k appears to be a constant, at least it is not influenced by the type of spine induction, the initial spine length, and the conditions after hatching (Table 4 and Fig. 9). 7. Using the allometric relation, an index of spine length (I ) was defined, which is independent of body size. ID permits the quantitative comparison of populations with different age structures, and of age (size) classes within the same sample. Differences between the I -values of the age classes of a sample indicate fluctuations of spine inducing factors (Fig. 10). It is thus possible to estimate the strength of spine inducing factors during previous periods by studying a single plankton sample (Figs. 11 and 12). 8. A reproducible bio-assay was developed to test the spine inducing activity of the medium. A relative unit of Asplanchna-substance (as) was defined. The quantity of substance produced per unit volume and per unit time is proportional to Asplanchna density and is influenced by temperature (within 10 to 25\dg C: Q =4). The amount of substance produced by 1 Asplanchna in 25 sec at 20\dg C in 1 ml medium will induce just recognizable spines in the experimental clone of Brachionus. Further data of substance production at 10, 15, and 25\dg C are given in Table 6. The substance decomposes exponentially (Fig. 14a). The velocity of this decomposition is influenced by temperature. Within the temperature range studied the half-life varies from 1 to 3 days (Fig. 14b). Theoretically, a specific equilibrium concentration of active substance corresponds to each population density of Asplanchna. This concentration can be computed using the data of production and decomposition (Table 6). However, in experimental populations as well as in the field, the actual concentrations were always considerably lower than expected on the basis of the Asplanchna density. This fact suggests that Brachionus incorporates the substance. Therefore, a pragmatic correction of the theoretical balance of the Asplanchna-substance was introduced by assuming that there is no accumulation of substance for more than 24 hours. With this correction, there is good agreement between the expected and the observed spine-indices in laboratory and field populations. 9. A quantitative comparison of the effectiveness of the three spine inducing factors shows that under natural conditions, long spines (mean I >0.3) cannot be induced by temperature and starvation, but must be due to Asplanchna-substance (Fig. 17). 10. In the field the Asplanchna-substance is the most important agent. Bioassays show that water of all ponds containing Asplanchna possesses spine inducing activity (Table 7). Both Asplanchna sieboldi (s. 1.) and A. priodonta are effective. The spine-indices of neonate Brachionus show a good correlation with the prevailing Asplanchna density (Figs. 20\2-22). The correlations of the indices of adult Brachionus show time-lags which correspond to the age of the animals. After the disappearance of Asplanchna, the spine-indices decrease with age-dependent time-lag. The spine-inducing influence of temperature is lower in the field than in the laboratory (Fig. 23). The possible effects of unknown factors in the field are discussed. 11. The reaction norms of two stocks from different ponds were tested. They exhibited remarkable and significantly different I -values at identical conditions of induction (Fig. 2), which occured at all three inducing factors. 12. Hypothetical mechanisms of spine production are discussed; possible differences in the physiological mechanisms of the three spine-inducing factors are indicated. Some aspects of the adaptive value of spine length variation are discussed.