Wed, 2018-10-03 12:10 -- Anonymous (not verified)
Agreement: 
I do not Agree
Body: 

Gregory J. Retallack, Department of Earth Sciences, University of Oregon
(October 3 2018)
Bobrovskiy et al. (1) have assembled impressive biomarker data which rules out three of five alternatives for the biological affinities of the problematic Ediacaran fossils Dickinsonia and Andiva. The cholesterols extracted from the fossils do indeed rule out affinities with lichenized fungi such as Ascomycota and Basidiomycota, and also with Rhizaria. This does not mean that Dickinsonia and Andiva were necessarily animals, because a third fungal phylum, Glomeromycota, also produces cholesterol without ergosterol (2). The living lichenized glomeromycotan, Geosiphon pyriformis, is unusual in housing the photosymbiont inside enlarged cells (3), and its fossil record may include Precambrian problematica such as Horodyskia (4) and Diskagma ranging in age back 2.2 Ga (5). Glomeromycotan fungi are also known from Ediacaran acritarchs with attached hyphae, stalked vesicles, complex wall ultrastructure, and chitin composition demonstrated by FTIR (6). A glomeromycotan lichen fragment preserved by cellular permineralization also has been described from Ediacaran rocks of China (7).
Cholesterol in Dickinsonia and Andiva permits both glomeromycotan and animal affinities, but additional observations provide a test of these alternatives. Bobrovskiy et al. (1) also found that the proportion of cholesterol relative to stigmasterol (a chlorophyte biomarker) increased in larger compared with smaller Dickinsonia. This is not what would be expected for a slow-moving or sessile animal increasingly fouled with algae as it grew, nor would such a regular decline be expected from vagaries of animal-feeding on algae. Declining stigmasterol with increasing cholesterol is compatible with building of fungal biomass by controlled populations of photosymbiotic algae. Dickinsonia and Andiva may have been glomeromycotan fungi lichenized with green algae. Undisputed Ediacaran animals trace and body fossils are small (< 5mm diameter) and vermiform with chitin or calcite skeletons, and have been characterized as Ediacaran Wormworld (8). In contrast, Dickinsonia and Andiva are part of a diverse group of large (up to 1.4 m) and unskeletonized, crustose to foliose, quilted organisms, from very different sedimentary facies (9), and could be characterized as Ediacaran Mattressland.

References and Notes
1. I. Brobovskiy, et al., Science 361, 1246-1249 (2018).
2. Fontaine et al. Lipids 36, 1357, 2001; J.D. Weete, M. Abril, M. Blackwell, PloS One 5(5), e10899 (2010).
3. A. Schüßler, M.Kluge, M., in The Mycota IX (ed. B. Hock), 151-161 (Springer, Berlin, 2000)
4. G.J. Retallack, K.L. Dunn, J. Saxby, J. Precambrian Research 126, 125–142 (2013).
5. G.J. Retallack, et al., Precambrian Research 235, 71-87 (2013).
6. G.J. Retallack, Botanica Pacifica 4(2), 19-33 (2015).
7. X. Yuan, S. Xiao, T. N. Taylor, Science 308, 1017-1020 (2005)
8. J.D. Schiffbauer et al. GSA Today 26(11), 4-11 (2016)
9. G.J. Retallack, Nature 493, 89-92 (2013), Gondwana Research 36, 94-110 (2016), Alcheringa 40, 583-600 (2016).

No competing Interests: 
Yes
The following competing Interests: 
Electronic Publication Date: 
Wednesday, October 3, 2018 - 12:07
Workflow State: 
Released
Contributors: 
<contrib xmlns="http://schema.highwire.org/NLM/Journal" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:hwp="http://schema.highwire.org/Journal" xmlns:nlm="http://schema.highwire.org/NLM/Journal"><name><surname>Retallack</surname><given-names>Gregory J.</given-names></name><email>gregr@uoregon.edu</email><role>Professor</role><aff>University of Oregon</aff></contrib>
Full Title: 

RE: Dickinsonia steroids not unique to animals.

Check this box if you would like your letter to appear anonymously::