The present study reports a new representative of Chlorella‐clade that was newly isolated from the Lake Vos'merka (Samara region, Russian Federation). 18S–ITS1–5.8S–ITS2 sequence analyses indicated that the studied strain ACSSI 368 and related Chlorella chlorelloides, Chlorella pulchelloides, formed an separate cluster in the Chlorella‐clade, unrelated to the holotype of the genus Chlorella – Chlorella vulgaris. The independent place of the genus was also confirmed by genetic distances. The individual species status of the studied strain ACSSI 368, compared with the sister C. chlorelloides, was confirmed by morphological differences (cell number per colony, type of chloroplast, number of autospores), genetic distances of the interspecific level by 18S–ITS1–5.8S–ITS2 fragment, alone by internal transcribed spacer 1 (ITS1) and internal transcribed spacer 2 (ITS2) sequences, one complementary base change in ITS1, and the results of species delimitation using five methods. A description of studied strain ACSSI 368 was provided as an authentic strain of the type species Aliichlorella ignota. Currently, the new genus Aliichlorella also includes Aliichlorella chlorelloides comb. nov. and Aliichlorella pulchelloides comb. nov. In general, six different delimitation algorithms were used in the present study. The ASAP (i.e. assemble species by automatic partitioning) and LocMin (i.e. ‘local minimum’ function) algorithms showed synchronous results, but their accuracy with respect to closely related species was somewhat lower. The mlPTP algorithm simultaneously aimed to combine closely related species in one and at the same time divide slightly distinguishing strains within a species into several molecular operational taxonomic units. The bPTP algorithm separated excessively Chlorella‐clade representatives into individual species. The generalized mixed Yule coalescent (GMYC) algorithm results were consistent the most with the modern understanding of the Chlorella‐clade taxonomy. However, the same method is one of the most time‐consuming because, for its implementation, it is necessary to carry out long preparatory work. The KoT results were less accurate than the GMYC results, although this is less time‐consuming because it does not require the preliminary construction of ultrametric trees.