![Thumbnail: [Volume 84, 1956-57 page 893]](/journals/rsnz/images/rsnz_84/thumbnails/rsnz_84_04_1064_0893_ac_02.gif)
Range of Genetic Rostral Variation
A collection of 16 small specimens in an early post-larval stage from Island Bay, Wellington, gives a good example of the genetic variation in rostral formulae. These specimens had carapace lengths ranging from 2.5 to 3.25 mm and showed no signs of damage or regeneration. The rostra would have been unable to regenerate completely during the short post-larval life of these specimens.
Text-fig. 5, Fig. 1, gives the range of variation of the ratio of rostral length to carapace length. The rostrum expressed as parts per 100 of the carapace varied from 112 to 141 with 40% falling above the upper limit of the sample 2 adults. This indicates that the ratio of rostrum to carapace length possibly diminishes slightly during the growth of the shrimp, but the extension to below the 100 mark (i.e., below the point where the rostrum is equal in length to the carapace) is mainly due to the fact that these rostra are incompletely regenerated and are consequently shorter than they would normally be. Thus 40% of the specimens have rostral lengths between 77 and 102 parts per 100 of the length of the carapace, and this percentage compares with that of the two minor groups (C and D) on Text-fig. 4, discussed above, which together make up 38% of sample 2.
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
These post-larval specimens were divided between the formulae 8/3, 8/4 and 9/4 with a single specimen with 10/4 (see Text-fig. 5, Fig. 2). Thus the variation cannot be expressed by a single formula, such as 8 to 10/3 to 4, but only by the double formula 8/3, 8 to 10/4.
In sample 1, 88% fall within this range, while in sample 3 the figure is 78%. This number is reduced in sample 2 to only 40%. Other samples from New Zealand and the Chatham Islands agree essentially with the scatters shown in Figs. 1 and 3 (Text-fig. 1). However, the other sample taken from the same intertidal pool in Island Bay as sample 2, has a scatter almost identical to Fig. 2, and thus differing from the others in a wider range in low numbers of dorsal teeth with 3 ventral and in the small number of specimens with 4 ventral teeth.
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
It can now be seen that a general formula such as 5 to 10/2 to 5 arrived at above has no real meaning, as it includes regeneration stages after damage, and that these stages are especially numerous in the Island Bay samples. However, from the above, a genetic rostral variation range for P. affinis of 8 to 10/3, 8 to 10/4 would be indicated. The formulae 3/9 to 10 and 10/4 are, however, quite uncommon.
Thus specimens showing a combination of most or all of the following characters, seen in the Island Bay post-larval specimens, can be presumed to be undamaged, or if formerly damaged to have completely regenerated: 8 to 10 dorsal teeth and 3 to 5 ventral; ratio of rostral length to carapace length of at least 1:1; presence of a true 1st dorsal tooth—i.e., in the almost bifid position at least; no crowding of, or reduction in size of, the anterior dorsal or ventral teeth; an attenuated, unarmed, anterior portion of the rostrum between the 1st and 2nd dorsal teeth, as opposed to a short or blunt portion.
Malformation and abnormal regeneration.
Malformation arising spontaneously or through abnormal regeneration is rare in P. affinis, but some interesting examples will be described. Text-fig. 2, Fig. 17, shows dorsal curving of the regenerating portion of the rostrum. Fig. 18 shows a more common slight malformation in the regeneration of the extreme anterior end of the.
![Thumbnail: [Volume 84, 1956-57 page 894]](/journals/rsnz/images/rsnz_84/thumbnails/rsnz_84_04_1065_0894_ac_02.gif)
rostrum. Text-fig. 3, Fig. 4 is a case of apparently spontaneous malformation, a ventral tooth has been divided into two by a long narrow split and the new integument showing inside has widened the division considerably. So a specimen with 3 ventral teeth is changed into one with 4, if this malformation is perpetuated in later moults. Duncker (1900, Plate II, Fig. 13) illustrates a similar split ventral tooth in Palaemonetes vulgaris. Text-fig. 2, Fig. 20 illustrates a malformation where a great increase in the number of teeth has resulted probably from abnormal regeneration after damage. Here subdivision of ventral teeth appears to have taken place similar to that described by Richardson (1953) for teeth on the posterior margin of the telson of the stomatopod Squilla armata from New Zealand waters.
Other Variation.
Against this background of great variation in rostral dentition, the other ornamentation and armament on the body and appendages are extremely stable, with the single exception of the spines on the dorsal surface of the telson. Nearly 10% of sample 1 had irregularities in the arrangement of these spines. The normal is two transverse rows of 2, while irregularities seen were: one member of a pair not in line; the posterior pair absent; one member from each pair on the same side absent or 3 spines absent.
Summary.
-
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
1. A considerable amount of variation in the form of the rostrum is described for Palaemon affinis. Variation in the rostral formula is wide and differs in three different samples taken; a general formula covering the 150 specimens of these three sample is 5 to 10/2 to 5 with the commonest individual formula being 8/4.
-
2. Only a small amount of variation in the shape of the rostrum, when expressed as ratio of length to depth and also degree of curvature, was found.
-
3. The ratio of the length of the rostrum to the length of the carapace varied, the rostrum being sometimes smaller but more often a little longer than the carapace.
-
4. The position of the 1st dorsal tooth varies from being so close to the tip that the rostrum is described as bifid, to a position posterior to the 1st or even the 2nd ventral tooth, themselves variable in position. It was found from a scatter diagram that the commonest position for the 1st dorsal tooth was near the tip, making the rostrum almost bifid and that rostra with only 3 ventral teeth had a much greater range of variation than those with 4. Miers (1876) is followed in placing P. quoianus M.-Edw. as a synonym of P. affinis M.-Edw., the differing positions of the 1st dorsal teeth in the two forms being within the range described for P. affinis.
-
5. A small but significant amount of unregenerated damage was seen (5% in a total of 307 specimens) and a number of individuals showing non-genetic variation are described as stages in regeneration of rostra after damage. Change in the rostral formula of an individual during regeneration is described as well as some malformations arising either spontaneously or through abnormal regeneration.
-
6. Regeneration of rostral length usually precedes regeneration of rostral ornamentation (i.e., dentition).
-
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
7. A collection of specimens in an early post-larval stage had a rostral formula range of 8/3, 8 to 10/4. These specimens showed no signs of damage or regeneration. All these specimens had the rostrum longer than the carapace and 40% were above the described upper limit for the rostrum to carapace ratio of a sample of adults from the same area.
-
[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]
8. A genetic rostral variation, as distinct from one including regeneration stages, of 8 to 10/3, 8 to 10/4 is found, with the formulae 9 to 10/3 and 10/4 uncommon.