The early Cambrian Carinachitidae, a family in the subclass Conulata, are intriguing and important small shelly fossils. Their gently tapering, tube-shaped skeletons consist of convex faces separated from each other by broad, deep corner sulci, and they exhibit triradial, pentaradial, or predominantly tetraradial symmetry. However, the morphology of the aperture and the modes of growth of carinachitid skeletons as well as the anatomy of their soft parts are unknown. Examination of a single new, exceptionally well-preserved specimen of tetramerousCarinachites spinatusQian, 1977, collected from the lower Cambrian Kuanchuanpu Formation in South China, reveals: (1) that its aperture is connected to a small mass of relic soft tissue and (2) that the apertural end of each of the four faces is developed into a subtriangular lappet or oral lobe that is smoothly folded toward the long axis of the tube, partially closing the tube aperture. Similarities between thorn-like spines on the faces and the oral lobes indicate that the transverse ribs were periodically displaced from the perradial portion of the aperture during formation of new ribs. In addition, the tube walls may have undergone secondary thickening during growth. The growth pattern of the tube and the spatial relationships between the tube aperture and soft parts are analogous to those of co-occurring olivooids. These findings further strengthen the previously proposed hypothesis that coeval carinachitids, olivooids, hexangulaconulariids, and Paleozoic conulariids are closely related taxa within the subphylum Medusozoa. Finally, carinachitids most likely represent an evolutionary intermediate between olivooids and hexangulaconulariids.

Various microfossils from the early Cambrian provide crucial clues for understanding the Cambrian explosion and the origin of animal phyla. However, specimens with important anatomical structures are extremely rare and the efficiency of retrieving such fossils by traditional manual selection under a microscope is quite low. Such a contradiction has hindered breakthroughs in micropaleontology for a long time. Here, we propose a solution for identifying specific taxa of Cambrian microfossils using only a few available specimens by transferring a model pre-trained on natural image datasets to the field of paleontological artificial intelligence. The method employs a 34-layer deep residual neural network as the underlying framework, migrates the ImageNet pre-trained model, freezes the low-layer network parameters and retrains the high-layer parameters to build a microfossil image recognition model. We built training sets with randomly selected images of varied number for each taxon. Our experiments show that the average recognition accuracy for specific taxa of Cambrian microfossils (50 images for each taxon) is higher than 0.97 and it can reach 0.85 with only three training samples per taxon. Comparative analyses indicate that our results are much better than those of various prevalent methods, such as the transpose convolutional neural network (TCNN). This demonstrates the feasibility of using natural images (ImageNet) for the training of microfossil recognition models and provides a promising tool for the discovery of rare fossils.