From magbeti@GPU.SRV.UALBERTA.CATue Apr 29 15:42:00 1997 Date: Sun, 23 Mar 1997 20:22:42 -0700 From: Michael Agbeti To: DIATOM-L@IUBVM.UCS.INDIANA.EDU Subject: Measuring biovolumes of diatoms other than centrics! When I posted the above request on the net some time last summer, some members on the network expressed interest in the reponses that I would and would like to have them. Below is the list of response. Although some of the responses went beyond my request, I am posting everything in event that the extra might benefit others. I would present everything where the response is short, but where it is lengthy I would summarize. (THE QUESTION SPECIFICALLY IS THIS: IN THE COURSE OF ROUTINE ALGAL ANALYSIS ONE MAY WANT TO MEASURE THE BIOVOLUME OF DIATOMS. FOR ALL DIATOMS, OTHER THAN CENTRICS, ONE NEEDS LENGTH, WIDTH AND DEPTH TO OBTAIN SUCH A MEASUREMENT. HOWEVER, FOR MANY OF THESE DIATOMS, IT NOT ALWAYS TO SEE THEM IN GIRDLE VIEW IN ORDER TO MEASURE THE DEPTH (MOST COMMONLY SEEN IN VALVE VIEW). IN CASES WHERE NO GIRDLE VIEW HAS BEEN SEEN IN ORDER FOR ONE TO DETERMINE THE DEPTH, WHAT SHOULD WE DO?) 1. This has been the taxonomist/ecologist nightmare since the beginning. They have spent whole meetings discussing the third dimension. If you look around you can find some in girdle view and you can get an approximation. It is not 100%. 2. You can actually calibrate measurement of thickness by examination of the units on your fine-focus at the high- and low-focus position on cells. This can be done by measuring any specimen of unknown dimensions, or artificial microspheres. I still do not understand the mania for biovolume alone, given the immense differences in protoplasmic portions of various taxa's cells, and the geometrically different dynamics of cell metabolisms especially especially as one gets to smaller, very active cells. 3. Most good quality research microscopes are calibrated on the fine knob to indicate the distance treaveled from high focal point on one side of the diatom to the low focal point of the opposite. 4. Measuring diatom height would be possible in the TEM. You would have to shadow cast from a known angle and measure the length of the shadow, simple trig gives the height of the diatoms (assuming of course that it is sitting flat on the microscope grid) 5. The tem biovolume is controversial because the cytoplasm does not necessarily fill the interior of the cell evenly 6. You can make measuremwnts when the diatoms are seen in the girdle view; this would be easy for the dominat forms but almost impossible for the rarely occurring ones. 7. Focus on the valve face on "top" of the cell. Look at the markings on the focusing knob (you 'll have to swivel your head around to the side of the scope), note the markings. Focus down to the "bottom" of the cell. Again check the readings. Substract the reading. Each marking should approximate those on your ocular micrometer i.e. oil immersion; each should be approximately 1 micron. This can be done with 40X or 100X objectives. 8. a) The 3-D shape od Amphora was obtained by racking the objective up and down using the scale that is on the focus knob of the objective (I dont know if there is any method of standardizing that). b) The depth dimension first doesn't have to have a relationship with the with; compare Didymosphenis with Synedra; secondly, this dimension is flexible according to the state of the cell cycle c) Have you thought of using a gel mount similar to pollen mountant, where you could try rolling your specimens by pressure on the coverslip. 9. Some work with marine phytoplankton showed that the pervalar distance of diatoms varies with the stage of cell division. Right after cell division and before the hypotheca is fully mature, the distance is smaller than it is at any other time---almost half that of a cell with fully mature hypotheca late in the cell division cycle. Many of the samples had abundant Pesudo-Nitzschia, Thalassionema and Thalassiothrix species, which are often seen in girdle view and so give an idea of the pervalar distance. If no good specimens are available in the sample or no information is available on the pervalar distance, we have used valve with equal to the average pervalar distance. This is based on the assumption (that is probably not true) of steady-state in cell division cycles. 10. Frustule depth is variable through the cell cycle (c.a. factor two for most species) and also dependent on nutrient conditions. Under growth (division) limiting conditions, some species will keep throwing!!!!! girdle bands and increase their volume without dividing. One can sometimes see a species in girdle view to measure. For things that don't fall in girdle view, you can estimate the depth, with good DIC systems. In either case, any estimate computed will be specific to the population from which it is measured, and it may not be appropriate to transfer estimates from one population to another which is growing under different conditions. Note that cell volume estimates have large inherent errors, and that cell volume estimates are not necessarily very good estimates of cell carbon or other constituents, unless you are willing to go through all the effort involved in sectioning the material and making extended morphometric measurement. a) Sicko-Goad, L., E.F. Stoermer & B.G. Ladewski. 1977. A morphometric method for correcting phytoplankton cell volume estimates. Protoplasma 93: 147-163. b) Sick-Goasd, L., C.L. Schelske & E. F. Stoermer. 1984. Estimation of carbon and silica content of diatoms from natural assemblages using morphometric techniques. Limnol. Oceanogr. 29: 1170-1178. c) Stoermer, E. F. & L. Sicko-Goad. 1985. A comparative ultrastructural and morphometric study of six species of the diatom genus Stephanodiscus. J. Plankton Res. 7: 125-135. The biovolume problem is an ancient one. The reason why it has never been resolved is that there are some fundamental estimation problems. One can refer to the following work: Kovala, P.E. & J.D. Larrance. 1996. Computation of phytoplankton cell numbers, cell volume, cell surface and plasma volume per liter, from microscopical counts. University of Washington, Dept. of Oceanography, Special Report Number 38. 11. The comarison of diatom structure to the two halves of a petri dish may not be appropriate. The two halves of the petri dish fit together while each half is touching the inner bottom of the other half. This is normally not the same with diatoms. The two halves of a diatom (epi- and hypovalve) are connected together only by their girdles, which normally overlap, but not too much. In addition, there can be lots of girdles between the two halves of a diatom cell which loosen their connection to the rest of the cell after treating the cell with acid or other chemicals, as we usually do when we want to count the cells under a microscope. These girdles can be braod and count up to 6 or 8 rings, so that the natural cell is even higher than broad (e.g. for centric diatoms). The following article may be useful. Steinman, A.D. & G.A. Lambert. 1996. Biomass and pigments of benthic algae. In Methods in Stream Ecology. pp 295-313 12. One may need some kind of regression model to estimate depth based on a width or length measurement. 13. Learn to do silicon oil grasient methods used for bacteria and cyanophytes and you will get proper intracellular volume. You will need some suitable silicon iol, tritiated water and a non-permanent solute such as sugar or 14-Inulin (not 14-Inulin carboxylate) and a verical head centrifuge (Beackman Microfuge). It takes time and a 'nack' but it works. I have been compelled to type all this on the keyboard while I logged on, because when I sent a similar stuff some time ago as an attached file it was received at some places in "funny" characters (some codes). So ignore any typos. I hope those interested find it useful. ******************************************************************************* Michael D. Agbeti, Ph.D. Bio-Limno Research & Consulting 8210-109 Street, P.O. 52197 Edmonton, Alberta T6G 2T5 CANADA OR Dept. of Biological Scienes University of Alberta Edmonton, Alberta T6G 2E9 Telephone: (403) 439-1558 Fax: (403) 439-1558 E-mail address: magbeti@gpu.srv.ualberta.ca ******************************************************************************* Date: Mon, 12 Aug 1996 09:38:15 -0700 From: David Kirschtel To: Multiple recipients of list DIATOM-L Subject: Re: Biovolume Paul M Mayer wrote: > Do methods exist for calculating biomass from cell biovolumes of cleaned > and enumerated diatoms? If you're looking to determine cell biovolumes take a look at BIOVOL: http://www.uvm.edu/~dkirscht/biovol.html the file bvolpak.exe is a self-extracting compressed file containing: biovol2.exe - the BIOVOL program bvdemo.dat - a small demo data set bvoldoc.txt - manual, plain vanilla ascii text, no graphics bvoldoc.wri - manual, Windows Write (WP applet that comes with Windoze), with graphics readme.bvl - Terms of Use, etc If you haven't already discovered them, a couple of papers that come to mind relating biovolumes to cell carbon are: Mullin,Sloan & Eppley. 1966. Relationship between carbon content cell volume and area in phytoplankton. L&O 11:307-311 Strathman. 1967. Estimating the organic carbon content of phytoplankton from cell volume or plasma volume.L&O 12:411-418 > In other words, can I estimate the biomass (cell biovolume) of > diatoms from a quantitatively prepared sample and then relate that to the > observed species diversity of that sample? In standard diversity indices (Shannon-Weaver/Weiner, Simpsons...) the p(i) has always been assumed to be prop. of individuals. I can't think of any reason why it couldn't be prop. of biomass. Given that the concept of "individual" is so nebulous with microbes it probably makes just as much sense to use biomass. There are a couple of interesting articles on microbial diversity in DL Hawkesworths Biodiversity edition. In any event using biomass would seem to be a lot more biologically robust than some of the schemes to weight biodiversity by a taxas "value" (what ever that may mean) that are floating around. -------------------------------------------------------------------------- David Kirschtel | Dept. of Botany | Sacred cows make the tastiest hamburger. Univ. of Vermont | Burlington, VT 05405-0086 | - Abbie Hoffman | tel 802.656.0429 | email dkirscht@moose.uvm.edu | From: PO3::"gab1@CORNELL.EDU" "Gail A. Blake" 25-OCT-1994 08:03:14.52 To: Multiple recipients of list DIATOM-L CC: Subj: Synopsis of biovolume responses Sorry folks - I've been dragging my toes on posting a synopsis of your responses to my inquiry about measuring biovolumes (because I've been busy measuring!). Several interesting things about overall responses: 1. Most were of the "Yeah, I'd like to know too. Tell me what you find out!" variety 2. Many people had suggestions about calculating the biovolumes once we had the measurements. 3. A few people told me how many individuals they measured. Most didn't say whether that number was for each individual "unit" (sample or rep, for instance), for a group of units (treatment or date, for instance), or for the whole experiment or survey or whatever. 4. One individual wrote to say that "it is pretty difficult to make biovolume calculations meaningful. See Sicko-Goad et al Protoplasma 93:147-163 (1977)." I haven't looked this up - but perhaps throwing up my hands would make this simpler! Suggestions on how to calculate biovolumes: David Kirschtel's BIOVOL program (thanks David!) Use the nearest geometric shape Use a digitizing tablet to calculate surface area Several references about biovolumes of phytoplankton. On how many to measure: Estimate only one biovolume for each species in an experiment or study. The number to count depends on how variable the size is. Measure more than 10 - at least 20. Measure and calculate biovolume after each measurement until the mean and std dev reach a fairly constant value. Measure 25 individuals. Thanks to all who responded! I don't feel any more certain about how many individuals to measure, but I feel less alone in that uncertainty! Gail A. Blake Entomology Department Cornell University gab1@cornell.edu