Argonauta hians, Lightfoot, 1786

Finn, Julian K., 2018, Recognising variability in the shells of argonauts (Cephalopoda: Argonautidae): the key to resolving the taxonomy of the family, Memoirs of Museum Victoria 77, pp. 63-104 : 77-91

Argonauta hians [Lightfoot], 1786; the A. hians / boettgeri complex

Recognition of shell form transformations in A. nouryi provided a new perspective on shell variation in another highly variable group of small argonauts, the A. hians / boettgeri complex.

The original description of A. hians [Lightfoot], 1786, refers to a single image in Rumphius (1705): plate 18, figure B (fig. 22a), designated as a lectotype by Moolenbeek (2008) in the absence of type material. Shells of A. hians can be recognised by smooth lateral ribs and a keel that increases in width with shell growth. Inter-keel tuberculation is absent.

Argonauta hians has long been recognised as displaying considerable variation in shell form. Voss and Williamson (1971) noted that “In the series from Hong Kong the sides of the aperture at the umbilicus range from strongly eared or auriculate with very large few knobs on the keel to specimens with no trace of auriculation and with rather more numerous, smaller knobs” (p. 105). They found that “if the 30 shells are laid out graded from large few knobs and strong auricles to smaller, more numerous knobs and flat sides there is an even gradation with no breaks or sudden changes” (p. 105). They concluded that all shells “belong to the same species” (p. 105).

As part of this study, 274 A. hians shells were directly examined in museum and private collections in Australia, United States, Europe, South Africa and Japan. With knowledge gained from examining shells of A. nouryi , all shells from all sites were examined for an abrupt change in keel tubercle height or ears that had been formed or subsumed in single shells. Because inter-keel tuberculation is not expressed in argonaut shells other than A. nouryi , this character could not be used.

Two shell formation types. Shells of A. hians were found to display two clear shell formation types:

• Type 1 shell formation – few pronounced ribs, large prominent keel tubercles, formation of ears.

• Type 2 shell formation – numerous less-pronounced ribs, small and greatly reduced, keel tubercles, absence of ears.

These shell formation types are similar to those expressed in A. nouryi except that variation in the arch of the shell was not observed and inter-keel tuberculation was not present.

This variation had been noted by Voss and Williamson (1971) who stated: “The knobs on the keel are very large and prominent in the first half of the shell and may remain large on the last half or may become considerably smaller” (p. 105).

Two shells are presented as examples:

•

• A shell from the Philippines (79.6 mm ShL [P], BMNH unreg., “Cuming, i.”) (fig. 23). This shell displays a clear shift from Type 1 to Type 2 shell formation indicated by a reduction in the size and spacing of the keel tubercles, a reduction in the ratio of ribs to keel tubercles from approximately 1.5:1 to 1:1 and ears subsumed .

A shell from the North West Shelf , Western Australia (53.0 mm ShL, WAM S31510 ) (fig. 24). This shell displays a shift from Type 1 to Type 2 shell formation. This transition occurred when the shell was at a smaller size and hence the ears are less developed. The resultant aperture shape (fig. 24c) is extremely similar to that observed in Type 2 A. nouryi shells; see fig. 5c, ii for comparison .

Variation also occurs between the opposing faces of individual shells, further highlighting the plasticity of shell characters in this species. A single shell is presented here as an example:

• A shell from Madagascar (60.8 mm ShL, NMV F164734, “ Madagascar ”) displays a large ear on the right side only; the left side is earless (fig. 25).

The A. hians /boettgeri complex. Small, earless A. hians shells have regularly been attributed to the species A. boettgeri Maltzan, 1881 (fig. 22b, c). Smith (1887) outlined the diagnostic characters of A. boettgeri : “The distinguishing features of this species are the numerous ribs and tubercles, the total absence of auricular expansions at the sides, its constantly small size, and the fine granulation (a feature not remarked upon by Maltzan), which more or less covers the whole surface, producing a dull non-glossy appearance” (p. 409). Berry (1914) similarly noted that the shell of A. boettgeri “seems unique in its small size, compact coil, and the circumstance that the auricular expansion at the sides of the aperture, so frequently developed in other species of the genus, are here notable only for their entire absence” (p. 280). Robson (1932) added “the almost invariable absence of colouring on the carinal knobs” to the distinguishing characters of A. boettgeri (p. 197). While Smith (1887) concludes that “the shell of this species must not be confounded with young stages of A. hians ; the more numerous ribs and tubercles and the rougher granular surface will separate it” (p. 410). Unfortunately, this dichotomy is not so straightforward.

Of the 274 A. hians shells examined, 41 can be attributed to A. boettgeri based on the above description. While it is possible to select a subset of shells possessing these characteristics, which in isolation appear distinct, examination of the entire range of material quickly dissolves the parameters on which this subset is based. All features mentioned above are variable in A. hians : ribs and keel tubercles can be numerous or scarce, pronounced or reduced, consistent across the shell or variable; ears can be present or absent, produced or subsumed, expressed on one side of the shell or both; the shell surface can be granular or smooth, pigmented or white. Two shells, displaying variation across the growth of the shell, are presented as examples:

• A shell from the British Museum (76.1 mm ShL [P], BMNH unreg., locality unknown, “B698, t.”; fig. 26). This shell displays an aperture shape and axial region consistent with the original description of A. boettgeri (fig. 22b, c) yet defies the description of A. boettgeri by showing signs of possessing ears at an earlier growth stage. While the ears have been subsumed with a shift from Type 1 to Type 2 shell formation, only the keel tubercles on the right side show a reduction in size (fig. 26b); the left keel tubercles have remained large (fig. 26a).

• A shell from Museums Victoria (25.0 mm ShL [P],

NMV F164767, locality unknown; fig. 27). This shell would historically have been attributed to A. boettgeri due to its small size and distinctive earless aperture.

This shell displays a dramatic change in keel tubercle size and spacing associated with a shift from Type 2 to

Type 1 shell formation, thus highlighting the plasticity of these characters.

In the absence of any consistent and definable diagnostic shell characters (in combination with a lack of diagnostic morphological characters or distinct distributions; see Finn, 2013), no evidence exists to justify maintaining A. boettgeri as a separate species. Consequently A. boettgeri Maltzan, 1881 , is treated here as a synonym of A. hians [Lightfoot], 1786.

Insight from whole animals. As described in the Materials and Methods section above, a single specimen lot of 73 female A. hians , most with intact shells, exist in the collections of the Western Australian Museum and Museums Victoria. On initial examination, it was found that the lot included submature, mature and spawned (i.e. females with eggs attached to the central axis of the shell) individuals. The shells of the spawned females tended to show a shift to Type 2 shell production in the last components of the shells (all other shells were composed entirely of Type 1 shell production).This led to the consideration that shell shape and transformation may be triggered by changes in reproductive stage or condition .

To understand the underlying cause of a change in shell formation type at the point of egg laying, a subset of 33 intact and measurable individuals were selected and fully measured. The subset included submature, mature and spawned individuals, with a size range of 13–27 mm DML and 21–36 mm ShL. Two larger females, also collected over the North West Shelf , were incorporated into the analysis to expand the size range ( QM Mo77789: 39.9 mm DML and 51.8 mm ShL; 28.7 mm DML and 38.9 mm ShL) .

Changes in shell morphometrics relative to animal size. Shell dimensions were plotted against DML to determine if the size of the shell relative to the size of the female changes between submature, mature and spawned individuals. Scatterplots against DML were generated for ShL, ShB, ApL, ApW, KW and EW. The scatter plots indicate a linear relationship between shell dimensions and animal size, with linear regressions returning coefficient of determination values (i.e. R 2 values) between 0.72 and 0.90 (see Table 1). No discontinuities were observed between the three maturity stages.

Ontogenetic changes in animal morphology. Dimensions and characters of the female argonauts were plotted against DML to determine if the relative proportions of the female changes between submature, mature and spawned individuals. Scatterplots against DML were generated for MW, HW, FL and AL. The scatter plots indicate a linear relationship between animal dimensions, with linear regressions returning coefficient of determination values (i.e. R 2 values) between 0.76 and 0.88 (see Table 2). No discontinuities were observed between the three maturity stages.

Ontogenetic changes in shell morphometrics. Shell dimensions and characters were plotted against ShL to determine if relative shell proportions change between submature, mature and spawned females. Scatterplots against ShL were generated for ShB, ApL, ApW, KW and EW. The scatter plots indicate a linear relationship between shell dimensions and characters, with linear regressions returning coefficient of determination values (i.e. R 2 values) between 0.74 and 0.96 (see Table 3). No discontinuities were observed between the three maturity stages.

The scatter plots provided no evidence of a change in relative shell and animal proportions between submature, mature and spawned individuals. If the examined characters underwent dramatic transformation with changes in state of maturity, it was expected that discontinuities would be observed in the plotted data. It is apparent that the visual change in shell form observed across this lot was not reflected in the relative measurements of the individuals measured.