Burrhead And Silverstripe Shiner Data... I Think I Know What It Means.

Besides chasing down flame chubs, I've been working on my burrhead and silverstripe shiner reproductive ecology project. I'll be presenting a talk at the ASIH meeting in New Orleans next weekend, so I've become obsessive with this so I don't say anything stupid in front of other biologists.

The manuscript I submitted to Ecology of Freshwater Fish bounced, largely for good reason. The good news is that I received good suggestions and feedback from one of the three reviewers (the other two spent a lot of time pointing out that this project is of no global significance, and anyway what the hell's my problem for not using Imperial English rather than American English even though their guidelines say to use either). The key suggestion that I've been working with is to analyze my gonadal size relative to body size data using Analysis of Covariance (ANCOVA), which allows an allometric analysis of gonadal growth while controlling for body size. The general convention for this kind of analysis is to use Gonadosomatic Index, or GSI, which states gonadal mass as a fraction of body size. What I've worked on is to restate GSI's for each species, each month of the breeding season, by running my data through ANCOVA. This sharpens and clarifies what kind of changes the fish are going through as spawning season peaks and wanes.

Below is an image from an Excel spreadsheet contrasting female ANCOVA-transformed GSI from March through August for silverstripe (N. stilbius) and burrhead (N. asperifrons) shiners. They follow similar trends, defining a breeding season that peaks in April and has largely waned by late July. This fits in with what's known of most other Notropis species who have 3-4 month spawning seasons, depending on latitude.
Male GSI data don't seem to show a clean trend like with the females above; males have a small spike in April, and almost nothing by the end of July, and very little in August.

The more subtle result of my research comes from microscopically examining histologically stained ovarian tissues. Using a digital microscope I was able to take photographs of 4 random views of ovarian tissue from each female, and using the same software measure the size of each stage of developing oocyte. (Yes, this is work for an obsessive compulsive, I admit it.) What I found is that burrheads produce oocytes that are significantly larger than silverstripe oocytes at almost every developmental stage, almost every month over the spawning season. Also, female burrheads average about 45 mm long and produce ova about 2 mm in diameter, and female silverstripes average about 55 mm long and produce ova about 1.4 mm long. So it seems that burrheads produce larger eggs, and these eggs are usually larger during development.

The one mistake I made in this research was not to count ripe eggs in females as we dissected them. Two undergraduate students working with me did excellent work with the histology, and I didn't lean on them to count eggs. The project started off with one very good graduate student who dropped out of school just as we got started, and I didn't follow up by asking the undergraduates to do a full-fledged Master's research project; I really couldn't. So my story's not as complete as it could be. I would like to conclude that burrheads produce fewer, larger eggs than silverstripes but that's only speculation. But it's a project worth doing again with a larger data set. All I have to do is to find (another) good graduate student(!).