Plagiodontes dentatus (Wood, 1828)

Pizá, Julia & Cazzaniga, Néstor J., 2003, Redescription, shell variability and geographic distribution of Plagiodontes dentatus (Wood, 1828) (Gastropoda: Orthalicidae: Odontostominae) from Uruguay and Argentina, Zootaxa 154, pp. 1-23 : 6-16

Plagiodontes dentatus (Wood, 1828)

Synonyms: Only original references are included; for a comprehensive list of name usage, see Richardson (1993).

Helix sowerbyana Férussac, 1821 , Tableaux systématiques des animaux mollusques (Prodrome), 2, 70, number 492 bis. (Nomen nudum.)

Helix dentata Wood, 1828 , Index Testaceologicus, Suppl., 50, pl. 8 fig. 71.

Cyclodontina brasiliana Beck, 1837 , Index molluscorum praesentis aevi musei principis augustissimi Christiani Frederici., p. 88.

Pupa labyrinthus Anton, 1839, Verzeichniss der Conchylien welche sich in der Sammlung …befinden, p. 47.

Odontostomus (Plagiodontes) teisseirei Marshall, 1931 , Proceedings of the National Museum, 78, 718.

Diagnosis: Shell 15­23 mm long, 7­10 mm wide, deeply perforated, rimate, ovate, marked with growth lines ( Fig. 6 ); 6­8 convex whorls, the spire being about 15% the length of the shell, conic, with an obtuse apex. The protoconch has waved striae ( Fig. 8 ), faintly decussated by spiral lines in young animals. Aperture 6­9 mm long, 5­8 mm wide, subvertical ovate, with 9­14 teeth (lamellae + folds) ( Fig. 14 ). An L to C­shaped parietal lamella deeply entering; the outer part of this tooth is rather square and generally unites to the small, compressed angular tooth. A spreading and partially vertical columellar lamella is the largest tooth occluding the aperture. A basal fold, in the limit between the outer (palatal) and inner (columellar) lip, is small and acutely compressed. One to three (usually two), parallel, equally or subequally sized lower­palatal folds on the right of the basal fold. A well­developed transverse lamella behind the lower­palatal folds. A twisted entering upper­palatal lamella, in the middle of the outer lip, almost as large as the columellar lamella. Two suprapalatal teeth connected at their base. Peristome thickened and narrowly reflexed in the adults. Measurements and proportions as shown in Table 1.

Remarks: Plagiodontes dentatus is similar in size and shell sculpture to typical specimens of P. patagonicus from southern Buenos Aires province. While P. dentatus and P. patagonicus lack teleoconch ribs ( Fig. 6 ), most P. multiplicatus specimens have a strong axial sculpture (6­7 ribs per millimeter on the fourth whorl; 4­5 ribs per millimeter near the peristome in adult specimens; Fig. 7 ). In some specimens of P. dentatus , the growth lines may appear more prominent than normal, in a rather regular disposition, but those shells are still distinguishable from the specimens of P. multiplicatus showing the weakest ribs, the latter being always more closely located.

TABLE 1: Mean and range of nine shell measurements for Plagiodontes dentatus from Uruguay and Argentina (n = 201).

The protoconch is striated in the three species. The axial ribs are irregularly waved in P. dentatus ( Fig. 8 ) and P. patagonicus ( Fig. 9 ); the ribs suddenly disappear, producing a clear limit with the teleoconch, which bears rough, wrinkle­like growth lines. There is a little difference between these two species, because the protoconch ribs frequently branch in P. dentatus , while they are less waved and seldom branch in P. patagonicus . The protoconch of P. multiplicatus ( Fig. 10 ) has instead mostly right, parallel axial ribs that seldom branch, and there is not a definite limit between the protoconch striation and the teleoconch sculpture, which is formed by regular costae.

Axial ribs are crossed with tiny thread lines giving the aspect of a stair in the protoconch of young animals ( Fig. 11 ). Most adult shells however are eroded ( Fig. 12 ), which may obscure the spiral sculpture of the protoconch without the use of high magnifications ( Fig. 13 ) or may even obliterate the protoconch sculpture completely.

Total number of apertural teeth (lamellae plus folds) varied from 9 to 14 in the material of P. dentatus (mode: 11; Fig. 14 ), from 9 to 13 in P. multiplicatus (mode: 10; Fig. 15 ), and from only 3 to 10 teeth in P. patagonicus (mode: 6; Fig. 16 ). All shells of P. dentatus and P. multiplicatus have a well­developed transverse lamella, such a lamella being instead absent from all the studied P. patagonicus specimens, except for a minute granule that can be seen in the position of the transverse lamella in some shells.

The position of the apertural teeth is basically the same in the three species, all specimens having well­developed columellar, parietal, and upper­palatal teeth. Although there is a considerable variability in thickness and orientation of these teeth, no regular pattern of variation could be generalized for any species.

A basal tooth is almost always present; only five shells of P. patagonicus lacked a basal tooth. Most specimens of P. multiplicatus show a rounded denticle or calcareous granule on the columellar tooth ( Fig. 15 ), which is seldom present in P. dentatus , and was never observed in P. patagonicus . The rather square, angled parietal tooth is generally united to the angular tooth in P. dentatus and P. multiplicatus ; only occasionally an independent angular tooth is observed in these two species, while this is the commonest condition in P. patagonicus ( Fig. 16 ). The suprapalatal tooth is always double in P. dentatus and P. multiplicatus , while it is usually single or absent in P. patagonicus . The upper­palatal tooth is present in all the studied shells, but it is smaller and more rounded in P. patagonicus . There are one to three (rarely four) lower­palatal teeth distributed as shown in Fig. 17.

100

P. patagonicus P. dentatus 80

)

%

( P. multiplicatus

Frequency 60 40

20

0

0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5

Lower palatal teeth number

Statistical comparisons: The four analyzed shell proportions were significantly different for the whole set ( Table 2 ). Bonferroni's comparisons between all pairs of species showed that they significantly differ from each other by the proportions of shell width to shell length, and aperture length to shell length. Plagiodontes dentatus has the most slen­ der shell, with the proportionally shortest aperture, while P. patagonicus is the most obese shell, with the longest aperture.

Comparison of shell proportions (proportion data transformed to ln (p/(1­p)) by one­way ANOVA.

Bonferroni's corrected t tests; df: degrees of freedom; corrected significance level: 0.0083 (* refers to the statistical significance of the result).

The proportion of last whorl length to shell length is significantly higher in P. patagonicus , while Plagiodontes dentatus has a significantly wider aperture than the other two species.

In a preliminary PCA on 25 samples offering no taxonomical doubts (n = 393; see Appendix), the first three principal components represented 87.5 % of the global variation (Table 3). Plagiodontes dentatus showed almost no overlap with the other two species on the plane defined by the principal components 1 and 2 ( Fig. 18), its scores being mostly influenced by its relative smaller size and its shell slenderness (as also evidenced by its higher values of major angle and lower spiral angle). Plagiodontes multiplicatus is differentiated on the plane PC1­PC3, and located on the upper right quadrant mainly because of its larger size and high number of apertural teeth. These results represent the concept of three well­differentiated morphospecies. As expected, samples from the same geographic region grouped together, each morphology corresponding thus to a discrete distribution pattern.

Identification of some problematic shells: Some shell sets that remained unidentified in the museum collections, and those that were recorded from unusual locations in the literature were reassigned by their sculpture characteristics, which were mostly congruent with the results of multivariate classification functions of their shapes. These preliminary functions were obtained by a direct MDA performed on the 25 samples mentioned in the previous paragraph, coded as three groups according to their region of origin ( Fig. 1 ). They yielded 99.19 % of correctly classified cases (equations not shown).

Two basic problems were at stake: references to P. daedaleus in Entre Ríos province ( Argentina; Fig. 1 : area 2), and P. dentatus cited in central Argentina ( Fig. 1 : area 1).

TABLE 3: Results of a principal component analysis on the sets of shells of three species of Plagiodontes that are marked with an asterisk in Appendix.

Plagiodontes daedaleus (Deshayes, 1851) is a species of large­sized (> 25 mm), obese shells from central Argentina, with a straight striated protoconch ( Pilsbry, 1902); its records in Entre Ríos were only based on three samples deposited at the La Plata Museum ( Fernández 1973), identified in the Appendix as samples 11, 12, and 13.

4

P. dentatus

3 M A P. m ultiplicatus W

2 2

Component 0 1

Principal ­ ­ 2 1

P. patagonicus S A

­3

W, LW L, A W, L, M A L

­4

­4 ­3 ­2 ­1 0 1 2 3 4 Principal Component 1

4

3

P. dentatus P. m ultiplic atus

TN

3 2

ponent 1

om 0

C

ipal ­1

rinc

P ­2

W N P. patagonic us

­3

W, LW L, A W, L, A M L

­4

­4 ­3 ­2 ­1 0 1 2 3 4

P rinc ipal C om ponent 1

All specimens in set 11 have however waved striated protoconchs, with branched ribs, and irregular growth lines on the teleoconch surface, as it is typical of P. dentatus and incompatible with P. daedaleus . Their shell proportions fit within the range of the former species, so that the classification functions reassigned them to P. dentatus ( Table 4 ). According to our morphometric results on P. daedaleus (Pizá and Cazzaniga, unpublished), identification of set 11 as P. dentatus is straightforward.

The other two sets from Entre Ríos (12 and 13), also classifiable as P. dentatus by morphometry and sculpture features, bear museum labels identifying them correctly as P. dentatus (handwritten labels by M. I. Hylton­Scott and J. J. Parodiz, respectively), so that their mention as P. daedaleus by Fernández (1973) is not a taxonomic problem but probably a mere transcription error.

As a consequence, there is no evidence that P. daedaleus lives in the Entre Ríos province.

All the collections from central Argentina (Santiago del Estero, Córdoba, Catamarca, and La Rioja provinces) that appeared to be partially assignable to P. dentatus showed straight striae in the protoconch, axial ribs in the teleoconch, and teeth characteristics of P. multiplicatus , even though some of them were actually similar to P. dentatus in size and shape.

Sample MACN 9437, from Alijilán (Catamarca province), cited as P. dentatus by Fernández (1973), consists of large, obese shells identifiable as P. daedaleus , a species that will be discussed in a further paper, so that it was not included in the Appendix.

Collection 17 in the Appendix, from Santiago del Estero, consists of slender shells with axial ribs weaker than normal for P. multiplicatus , but the protoconch had typical straight ribs and the multivariate classification functions reclassified them as P. multiplicatus ( Table 4 ), and were therefore interpreted as a variety of this species.

Some other shells of P. multiplicatus were grouped with P. dentatus by the classification functions ( Table 4 ). Some of them had been labeled or mentioned as P. dentatus , while others remained unidentified, but morphometry alone is not a convincing argument in this case, because all such specimens have the same kind of straight ribs in the protoconch, and axially ribbed teleoconch allowing their identification as specimens in the lowest limit of size and obesity of P. multiplicatus .

Because typical specimens of P. dentatus were not confirmed from central Argentina, its alleged distribution over Córdoba, Santiago del Estero, La Rioja, and Catamarca provinces is weakly founded. At present, no sample can be faithfully identified as P. dentatus to the west of the Paraná River.

MDA performed on ten continuous variables from all sets, including those re­identified as P. dentatus or P. multiplicatus (n = 564) resulted in 87.41 % of correctly classified shells. Nearly all specimens of P. dentatus and P. patagonicus (96.0 % and 95.4 % respectively) were correctly identified through these functions. Only 77.2 % of the P. multiplicatus shells were instead correctly classified, 22.8 % of the specimens locating within the morphospace of P. dentatus . Not even a shell of these two species classifies as P. patagonicus through these functions.

percentage of explained variance; CC: canonical correlation.

Function Eigenvalue % cum CC After Fcn Wilks’ 2 df p

Lambda

1 0.101102 1279.873 12 0,0 0 0 0 1 4.4691 84.68 0.9040 1 0.552938 330.916 5 0,0 0 0 0

2 0.8085 100 0.6686

Standardized canonical discriminant function coefficients Discriminant Analysis: Classification functions (total correct classification: 85.64 %)

Function 1: P. dentatus (correct classification: 95.0 %)

y = 2.7091 SL ­ 0.3468 SW ­ 0.0620 LWL + 2.7727 AL ­ 0.6107 AW + 11.7726 TN ­ 98.3974

Function 2: P. multiplicatus (correct classification: 74.4 %)

y = 2.5061 SL ­ 0.2587 SW ­ 0.0376 LWL + 5.7715 AL ­ 1.7779 AW + 11.7978 TN ­ 113.6148

Function 3: P. patagonicus (correct classification: 94.5 %)

y = 0.7204 SL + 0.4919 SW ­ 0.0250 LWL + 6.5341 AL ­ 0.0672 AW + 6.6840 TN ­ 60.7845 Similar results were obtained when the analysis was repeated with only five variables ( Table 5), the resulting classification functions being therefore recommended for identifying the three morphologies.

Geographic variation of Plagiodontes dentatus : MDA was performed on all sets that were eventually identified as P. dentatus (n = 201; see Appendix), including 13 shells identified as P. dentatus teisseirei in museum labels. The percentage of correctly classified cases was low (near 40 %), with wide intersections among groups, either when each collection was considered as a single group, or coded together according to different criteria.

A relatively higher degree of discrimination (58.71 %) was found when the samples were coded as three groups according to the natural areas delimited by the three main rivers within the distribution range of P. dentatus (Paraná, Uruguay, and La Plata Rivers). The La Plata River, an extremely wide estuary, separates the Oriental Republic of Uruguay from the Buenos Aires province; the province of Entre Ríos (meaning “Between Rivers”) is limited by the rivers Paraná and Uruguay, which flow a mean of 17,000 and 5,000 m 3/s respectively. The first discriminant function explains 83.24 % of the variance. Longer shells with wider apertures score higher on this function, while a short shell with a short, slender aperture gets a low score ( Fig. 19). There was not a tendency for samples labeled as P. dentatus teisseirei to group together; they were actually undistinguishable from the remaining samples. Then, even if there is a slight tendency for shells being larger towards Uruguay, most individuals clustered in a central, rather compact cloud, and no clear­cut boundaries were found, so that any subspecific statement is unwarranted.

Figures

FIGURE 1. Geographic distribution of genus Plagiodontes Doering, 1876.

FIGURES 6 ­ 7. SEM photographs of the teleoconch sculpture near the aperture lip. 6, Plagiodontes dentatus; 7, P. multiplicatus.

FIGURES 8 ­ 13. SEM photographs of the protoconch sculpture in Plagiodontes spp. 8, P. dentatus; 9, newly hatched P. patagonicus; 10, P. multiplicatus; 11, close­up of Fig. 9, showing the maximum development of the spiral lines crossing the axial striae; 12, eroded apex of P. multiplicatus; 13, close­up of figure 12, showing the remains of spiral sculpture in the protoconch. Bar “ a ” scales Figures 8, 9, 10, and 12; bar “ b ” scales Figures 11 and 13.

FIGURES 14 ­ 16. SEM photographs of the apertural teeth in Plagiodontes spp. 14, P. dentatus; 15, P. multiplicatus (arrow indicates the presence of a denticle on the columellar tooth); 16, P. patagonicus.

Tables

TABLE 2: Statistical comparison of shell proportions of three morphologically similar species of Plagiodontes: mean ± SD [minimum­maximum].

	P. dentatus (n = 186)	P. multiplicatus (n = 98)	P. patagonicus (n = 109)
100 SW / SL	45.42 ± 3.34 [36.45­52.40]	48.23 ± 3.12 [40.84­65.41]	51.17 ± 3.14 [44.90­59.81]
100 LWL / SL	64.01 ± 4.76 [58.04­71.07]	65.23 ± 3.03 [46.32­73.40]	68.34 ± 2.10 [63.92­74.15]
100 AL / SL	41.76 ± 1.93 [36.45­47.21]	43.13 ± 1.94 [36.96­47.24]	46.53 ± 2.07 [41.16­52.21]
100 AW / AL	85.05 ± 4.54 [73.58­95.04]	79.41 ± 6.06 [60.55­92.16]	81.45 ± 3.81 [70.99­91.87]

TABLE 4: Morphometric reclassification of some conflictive shell sets of Plagiodontes sp. from central Argentina, as assessed through preliminary multivariate classification functions. Set number refers to the Appendix.

Set number	n	P. dentatus	P. multiplicatus	P. patagonicus
17	22	0	22	0
18­19	34	33	1	0
21	26	19	6	1
24	20	19	1	0
25	15	7	8	0
26	20	16	3	1
28	12	9	3	0
29	6	3	3	0