With regard to moult staging, Stevenson (1985: 21) found it unfortunate that investigators had often attempted to establish “their own arbitrary criteria for each group studied” rather than using criteria suggested by Drach (1939, 1944), Drach & Tchernigovtzeff (1967) and Stevenson (1968). Whilst these suggested criteria are appropriate for the larger species to which they have been applied, development of refined techniques has allowed study of moulting in animals that are too small or frail for accurate estimates of cuticle hardening or visually determined staging of setal changes to be applied. The only attempt to stage moulting in copepods to date has been by Dexter (1981) who studied setal changes in Calanus marshallae copepodids. The use of preserved material in this latter study and the attendant likelihood of artefactual changes particularly relating to shrinkage has been pointed out by Buchholz (1982).
Without recourse to assessments of cuticle hardness and setal development the present study has sought to tie changes seen to those recorded in other species. In particular the study by Buchholz & Buchholz (1989) on Euphausia superba Dana has been used, since the cuticle of this species appears to be most closely allied with that found in L. salmonis. In order to facilitate comparison with other studies, the exo- and endocuticle referred to elsewhere will here be referred to as p1 and p2 respectively.
Moult stages A-C are very difficult to determine precisely in L. salmonis since the criteria generally used to determine them in the other species have relied principally upon assessing cuticular rigidity and setal development. Whilst the A1 stage should theoretically be easily diagnosable, as it is the stage entered by the animal immediately post-ecdysis, in practice this is not straightforward in L. salmonis. Difficulties arise firstly from the fact that the chalimus larva is attached to its salmon host at the time of moulting and therefore can not be observed to moult directly and secondly from the likelihood of a very short duration of this stage in the rapid moult cycle of the chalimus larva. Because sampling was only undertaken every 24 h it is likely that this stage will have been missed and was therefore unlikely to be represented in the present study. Stage A2 has been suggested to begin with the initiation of post-ecdysial production of the p2 layer (Stevenson, 1968). Observation of this event classically requires recognition of external changes, which is again impossible under the normal conditions of chalimus development, although recognition of production of the p2 ultrastructurally should be possible where the p2 layer may be differentiated from the p1 layer. Stevenson (1968) also expressed concerns over the possibility for recognition of such a stage in cases where cuticle production did not stop during ecdysis. Stages B1 and B2 are similarly externally assessed and rely on criteria associated with the rigidity of the cuticle. Again, this cannot be easily recognised in the present species, although stages A/B are indicated by a reduced complement of procuticular laminae with respect to the completed cuticle. Stage C1 has been proposed to be reached when the chemical changes in the p2 layer have been completed (Stevenson, 1968) and again, it has been suggested that this event be linked where possible to externally recognisable events. Stage C4 comprises completion of the cuticle and hence the initiation of the “intermoult” phase of the cycle. In Orconectes this stage might be recognised by the presence of a completed membranous layer within the cuticle. Because such a layer does not occur in L. salmonis, this criterion clearly can not be used. As indicated above there is, furthermore, no indication of reduced epidermal activity in L. salmonis that might suggest true anecdysis. Instead, this stage can only be recognised by attainment of the full complement of procuticular laminae.
The separation of the old cuticle from the epidermis, “apolysis”, is well defined in this study. Because such a change is universal and clearly recognised ultrastructurally, it presents a good point to tie the cycle seen in L. salmonis with those recorded for other groups. This stage corresponds to stage D0-D1 of Drach & Tchernigovtzeff (1967) according to whom, secretion of new epicuticle begins at the start of the D2 stage and secretion of the new procuticle (p1) towards the end of it. The D3 stage corresponds to the resorption of most of the old p2 layer and the D4 stage corresponds to the completion of resorption of the p2 layer and the splitting of the exuvium prior to moulting. Stage E correspond to ecdysis of the exuvium.
In contrast to most studies of crustacean moult cycles, the epidermis of L. salmonis did not display a clear cycle of structural organisation or cellular activity such as has been described by Green & Neff (1972) for Uca pugnax (S.I.Smith) and instead gave indications of continuous activity with no clear intermoult period (C4) being displayed. This may result from the rapid transition between moults and from the fact that there is a constant food supply available from the host which obviates the need for gradual accumulation of materials needed for growth, metamorphosis and cuticle construction. A similar lack of decline in structure or activity of the epidermis was also noted in Daphnia magna by Halcrow (1976) and in Euphausia superba by Buchholz & Buchholz (1989) and in the former paper this was also suggested to result from a rapid transition time.
As opposed to the findings of Buchholz & Buchholz (1989), there was no evidence in the present study of the extensive pleating of the apical plasma membrane although some signs of small microvilli were observed. There was, however, considerable evidence of large-scale folding (macro-folding) of the new cuticle which is suggested to allow expansion of the cuticle following ecdysis. Extensive cuticular folding of the trunk cuticle has been noted by Smith & Whitfield (1988) for moulting females of Lernaeocera branchialis where it appeared to be restricted to the p1 and epicuticle layers. This folding was suggested to result from epidermal folding in the moulting chalimus IV stage and was said to facilitate elongation during metamorphosis of fertilised females. Larger second-order folds were also described in L. branchialis and were suggested to perform a similar function. Such second-order folds appear closer in morphology to those seen in the present study. Extensive cuticular folding prior to ecdysis has also been reported for the lobster Homarus americanus Milne-Edwards by Cheng & Chang (1994).
In the present study the new epicuticle / tp1 layer was demonstrated to be elaborated even after initial deposition of the p1 layer. Such elaboration has also been reported by Smith & Whitfield (1988) although whether it results from material passing from the epidermis via the p1 layer or through the addition of materials resulting from resorption of the old cuticle is unclear. The rugose surface of the new epicuticle / tp1 layer may be allied to the first order folding of Smith & Whitfield (1988) although it lacks the regularity of the folds demonstrated in that study and so may not be concerned with expansion of cuticle surface area following ecdysis.
As with the completed cuticle described elsewhere in this volume, there was no sign of intracuticular pore canals in the present study. Whilst these were suggested to be important for transport of material from the epidermis into the cuticle and for transport of material resorbed from the old cuticle in E. superba (Buchholz & Buchholz 1989), it seems likely that the restricted cuticle thickness of L. salmonis chalimus larvae allows movement of material without such elaborations. Whilst Buchholz & Buchholz describe the continued attachment of tonofibrils to the old cuticle despite the intercalation of the new p1 and epicuticle and suggest that this may facilitate continued swimming in premoult euphausids, it seems likely that chalimus larvae are relatively inactive prior to the moult and therefore such an attachment is not necessary.