Lindavia comensis

Valves are disc-shaped with a tangentially undulate valve face. In smaller specimens, the undulation is difficult to discern. From 1/2 to 3/4 of the valve face is colliculate. This colliculate area is either irregular or star-like in shape. The striae are composed of two rows of areolae, with a row of small pores present between the larger areolae. The striae number 18-21 in 10 µm, based on chord count. The striae vary in length and terminate on the valve face to create an irregular circular central area. A single central fultoportula is present on the depressed half of the valve face. The rimoportula is located on the raised half of the valve face, at the end of a shortened stria. Marginal fultoportulae occur on the mantle between every 3rd to 4th striae. Spines are absent.

The specimens of Lindavia comensis shown here conform to the type (Scheffler et al. 2005). Within the species however, L. comensis shows high morphological variability within frustules (heterovalvy). The variability has been the subject of many studies and several species complexes have been identified. Names for the complexes include: the “Cyclotella comensis-pseudocomensis-costei group” (Kistenich et al. 2014), the “Cyclotella rossii-comensis-tripartita-complex” (Cremer et al. 2001), the “Cyclotella ocellata/C. comensis complex” (Duleba et al. 2015), “C. comensis morphs” (Hausmann and Lotter 2001, Werner and Smol 2006), “Cyclotella comensis and related morphotypes” (Wunsam et al. 1995), the Cyclotella comensis/var. 1/var. 2/michiganiana-comensis group (Reavie and Kireta 2015), and the dimorphic Cyclotella comensis-minima forms (Scheffler et al. 2005). Attempts to discern the value of L. comensis as an indicator have included attempts to classify morphogroups, determine the environmental variables linked to morphological variation, and utilize molecular approaches to compare phylogeny to morphological variation.

Attempts to differentiate morphological groups within the L. comensis-complex have included simple morphometric/morphologic separation based on size, striae/alveolae pattern, central area shape, central area ornamentation, and central area topography. Werner and Smol (2006) identified four common morphotypes in Ontario lakes including “comensis”, “comensis rossii”, “comensis socialis”, and “comensis fine”. Reavie and Kireta (2015) recognized four morphotypes from the Laurentian Great Lakes including Cyclotella comensis, C. comensis var. 1, C. comensis var. 2, and C. comensis michiganiana-comensis. Cremer et al. (2001) saw variability in small Greenland “Cyclotella sp. A” that encompassed variation typical of C. rossii, C.comensis, and C.tripartita. Wunsam et al. (1995) recognized Cyclotella comensis of the type and four additional morphotypes in European Alpine lakes. Still others have treated closely related members of these groups as synonymous, e.g. Cyclotella pseudocomensis as a latter synonym of C. comensis (Scheffler et al. 2005) or as separate and recognizable taxa (Houk et al. 2010). Multivariate and multi-character approaches for differentiating groups in this complex have also been used to identified up to six morphs (Hausmann and Lotter 2001, Kistenich et al. 2014, Duleba et al. 2015).

The significance of these morphotypes and morphological variability has been more difficult to resolve. Cremer et al. (2001) noted that the “C. rossii” morphotype was most common among their forms but could not determine any historical patterns among morphotypes in paleolimnological records from Greenland. Rühland et al. (2008) presented grouped data of small planktonic “Cyclotella s.l.” (sensu lato) suggesting that the unresolved group of taxa showed high responsiveness to climatic drivers. Saros et al. (2012) and Saros and Anderson (2014) showed that Cyclotella comensis, without any resolution of morphological variants, had strong indicator value for nitrogen levels and lake mixing depth. Werner and Smol (2006) noted no significant patterns in ecological distribution among their four morphotypes. Reavie and Kireta (2015) showed that, in the Great Lakes, there was little difference in TP or chloride optima between C. comensis and C. comensis var. 2, but did note that the former was a strong indicator of low anthropogenic stress in coastal environments. Hausmann and Lotter (2001) noted that alkalinity and total phosphorus were not correlated with their morphological groups, but did identify significant relationships between summer lake temperature (altitude) and some morphotypes. Molecular studies on the C. comensis group have used morphological characteristics to classify terminal taxa and molecular analysis to test those morphological groups. Kistenich et al. (2014) could not separate C. comensis, C. pseudocomensis, and C. costei based on morphology because of high morphological variability, and were similarly not able to distinguish them based on molecular analysis, thus concluding they were one taxon. Duleba et al. (2015) similarly used morphology to identify C. comensis, C. pseudocomensis, and C. costei forms in their cultures, but concluded based on molecular analyses that these taxa were conspecific and closely related to C. ocellata by a possible recent divergence.