Origin of planktotrophy—evidence from early molluscs: a response to Freeman and Lundelius

Based on the study of Early Paleozoic protoconchs, we (Nutzel et al. 2006) concluded that planktotrophic mollusc larvae are unknown from the Cambrian but became common in the Ordovician. Therefore, we assumed that planktotrophy itself evolved at the Cambrian/Ordovician transition and that it is unlikely that planktotrophy represents the original state in Metazoa. This is in agreement with other authors who followed different lines of evidence (see Nutzel et al. 2006). Freeman and Lundelius (FL Riedel 2000). However, such gastropods are known from the Cretaceous onwards only. Scissurellids and some other Vetigastropoda represent another important exception. They are nonplanktotrophic (as all vetigastropods) but have a small initial shell (scissurellids are known from the Triassic onwards). However, all these exceptions do not change the general picture that at a shell length of about 100mm, most planktotrophic molluscs are smaller than most nonplanktotrophic molluscs. Even in recent gastropods, it is commonly hard or impossible to determine the boundary between embryonic shell (protoconch I) and larval shell (protoconch II) and it becomes even more difficult in fossil material that is almost always recrystallized so that fine structures are obscured. Generally, protoconch preservation is rare in the Paleozoic. Most Late Paleozoic to extant gastropods have a tightly coiled initial whorl so that it is possible to measure the diameter of the first whorl if it is preserved. Most planktotrophic gastropods hatch with about one whorl (Figs. 1A and 2J) (e.g., Bandel 1975) and therefore the diameter of the initial whorl can be used to infer the approximate size of the hatchling even if the transition from protoconch I to II is not visible (Nutzel 1998). The oldest gastropods with well-preserved larval shells have been EVOLUTION & DEVELOPMENT 9:4, 313 –318 (2007)