FISHERY BULLETIN: VOL. 80, NO. 3 



Nair 1979; Dawirs 1979, 1980; Anger and 

 Dawirs 1981). 



MATERIALS AND METHODS 



Ovigerous Hyas araneus were dredged from a 

 deep channel near the island of Helgoland 

 (North Sea) during early winter in 1978-79 and 

 1979-80. After hatching, the zoeae were isolated 

 in vials and maintained individually at 12°C. 

 Food (a mixture of freshly hatched Australian 

 Artemia sp. nauplii and the rotifer Brachionus 

 plicatilis) and filtered seawater were changed 

 every second day. The methods of obtaining and 

 rearing the larvae have been described in detail 

 by Anger and Dawirs (1981). 



For determination of wet weight, larvae were 

 caught individually with pen-steel forceps, 

 briefly rinsed in water from an ion exchanger, 

 blotted for about 10 s on filter paper, and trans- 

 ferred to preweighed silver cartridges. All 

 weight measurements were carried out on an 

 Autobalance AD-2 (Perkin-Elmer) 3 to the 

 nearest 0.1 m£- The techniques and equipment 

 used for obtaining dry weight (DW), carbon (C), 

 nitrogen (N), and hydrogen (H) content of larvae 

 and young crabs were the same as described by 

 Anger and Nair (1979) and Dawirs (1980): deep 

 freezing, vacuum drying, weighing, and com- 

 bustion in aC-H-N analyzer (Model 1106, Carlo 

 Erba Science). Only rinsing of the material (see 

 above) was added as an initial step. This 

 standard procedure was adopted to remove pos- 

 sible adherent salt and thus to increase the 

 accuracy of the measurements. Comparison of 

 test measurements, however, did not show signif- 

 icant differences (Anger and Nair 1979). 



Energy estimates (J) were obtained from 

 carbon values by applying the N-corrected 

 regression equation given by Salonen et al. 

 (1976). Statistical procedures were the same as 

 referred to in detail by Anger and Dawirs (1981). 

 In regression equations, intercept (6) and slope 

 (m) are given; in addition, correlation coeffi- 

 cients (r) and their level of significance (P) for 

 deviation from zero are provided. For logarith- 

 mic transformations, In (= log e ) was applied. All 

 statistical tests were two-tailed. 



In May 1979, a first series of 46 analyses com- 

 prising 123 individually reared zoea-1 larvae (Z- 

 1) of Hyas araneus was carried out to compare 



3 Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



their growth patterns with those previously ob- 

 served by Anger and Nair (1979) in commonly 

 reared zoeae. This set of data showed unsatisfac- 

 torily high variation, and high mortality pre- 

 vented a larger number of analyses. For this 

 reason, in February 1980 another set of 92 

 analyses comprising 110 prezoeae, 274 Z-l, and 

 30 early Z-2 was obtained (Table 1). The data for 

 later stages given in Table 2 had been obtained in 

 March and April 1979 (112 samples, 149 

 individuals). 



RESULTS 



Larval Growth 



Fresh weight (FW) values fluctuated around 

 constant levels in all larval stages without clear 

 increase in a single stage (Tables 1, 2). This 

 steplike growth pattern did not allow any 

 analysis of actual body growth during larval in- 

 stars. 



The gain in total live weight (FW) from the 

 prezoea to the freshly metamorphosed crab was 

 ca. 770%. It was 640% in DW, and only ca. 470% in 

 C, N, and H. The absolute increase during larval 

 development is shown in Figure 1. During the 

 extremely short, nonfeeding prezoea stage there 

 was no gain in C, N, H, and energy. Molting to the 

 Z-l resulted in a minor loss of organic constit- 

 uents (cast cuticle) and in some uptake of water 

 and salt (Table 1; Fig. 1). During the following 

 instars there was an appreciably absolute in- 

 crease in all parameters considered. It was 

 generally strongest in the second zoeal stage and, 

 surprisingly, weakest in the megalopa. 



The values shown in Figure 1 for Z-l, Z-2, and 

 magalopa form a straight line when arranged in 

 a semilogarithmic scale. This indicates that 

 growth from stage to stage followed an exponen- 

 tial pattern during this period. 



The different growth patterns in wet weight 

 (steplike) and DW (gradual) were caused by a 

 combination of these two patterns in the water 

 content of the larvae; during each molt, it 

 suddenly increased, and then it gradually de- 

 creased during the molt cycle. This decrease 

 could be expressed as a power function in all 

 larval instars: In (% H 2 0) = b + m In (t + 1), where 

 b is approximately the logarithm of the initial 

 water content, m is the slope, and t is the time 

 (days from the beginning of a particular stage). 

 All r's for these fitted curves were significantly 

 different from zero (P<0.001). The rate of de- 



420 



