All bivalves were counted monthly and dead individuals were
recorded. The shell lengths of at least 200 individuals were
measured monthly using a Vernier calliper accurate to 0.1 mm. In
clams, the length corresponds to the anterior/posterior axis and is
measured perpendicularly from the height line, matching the
dorsal/ventral axis. Conversely, in mussels, length is defined as the
distance from the umbo to the opposite shell margin.
To avoid stress to the animals and minimise errors during
weighing, the bivalves were dried as much as possible using blotting
paper and their total live weight was estimated in pools of 40
specimens using an analytical balance (0.01 g). The number of pools
that were weighed monthly in each tank varied according to the
abundance of the surviving juveniles, ranging from 10 to 8 for
mussels and from 14 to 8 for clams.
After 98 (3rd month) and 202 (6th month) days in the first
experiment and after 31 (1st month) and 98 days (3rd month) in
the second experiment, further measurements were made. Shell
lengthwas measured in 30 mussels and 30 clams per tank and shell
alterations were evaluated both qualitatively, using a specific
damage index, and quantitatively, by measuring the percentage of
damaged area, mostly occurring at the umbo level. In the same
individuals, the dry weights (after 48 h at 60 C) of the shell and soft
tissues and shell thickness were also evaluated, but only in the first
experiment.
In both species, the shell damage index (DI) was graded according
to the type of alterations found, regardless of the extent of
the damaged area. For M. galloprovincialis, the DI was graded as
follows by examining the injuries of the periostracum and underlying
prismatic layer of calcite: 0 ¼ no damage; 1 ¼ periostracum
discolouration; 2 ¼ breakage and lifting of the periostracum, prismatic
layer visible; or 3 ¼ dissolution of the prismatic layer, inner