Facebook Twitter Flickr Google + Blogger

Australasian Plant Conservation

Originally published in Australasian Plant Conservation 17(3) December 2008 – February 2009, p 6-7

Had I mixed those damned vials of smoke correctly …

Paul Gibson-Roy
Grassy Groundcover Research Project, Greening Australia/The University of Melbourne, Burnley, Vic. Email: roypg@unimelb.edu.au

Grassland species flowering prolifically within the duplicated seeding study twelve months after sowing. Photo: Paul Gibson-Roy


People (me among them), are often heard to discuss the benefits of ‘learning by one’s mistakes’. It’s not that making mistakes is particular fun; they just seem to go hand in hand with being human. They might result just in frustration and loss of time, but can also lead to serious and costly outcomes. For example, an inadvertent communication error by someone involved with my work once resulted in the destruction of a field sowing that had cost approximately $20,000 and countless hours of input from many committed people. Conversely, very interesting things can be learnt from errors, about one’s self and about the world around us.

I’ve often attended forums and workshops where participants suggest enthusiastically that we should communicate more about ‘how not to do things’ or ‘what went wrong’, as much as we communicate about ‘how to’ and ‘what went right’. While admirable, sometimes it takes a brave soul to identify, before all and sundry, past errors or mix-ups. To illustrate this point, few scientific journals appear committed to reporting errors! And, having agreed to write an article for this issue (who hasn’t made plenty of blues along the way?), I’ve been looking at a blank page for many months. Possibly there’s a degree of reluctance on my part to publicly ‘fess-up’ to being a dill. But how do you ‘report’ on what were often fleeting events, the consequences of which were not always immediately obvious?

My Story

I’ll focus on an event that occurred when undertaking my honours degree at Melbourne University. My supervisor at the time (and now dear colleague through many years) was John Delpratt. I had become captivated by John’s work with grassland species and was making my first tentative steps into the world of grassland restoration. I was also beginning to understand the level of detail and focus required to undertake research. I had begun by looking at the potential of a suite of seven indigenous grassland species to germinate and establish following direct seeding under quasi-field conditions. Collecting seed for the experiment, testing the seed prior to sowing, organising seed-lots, preparing the experimental plots for sowing and obtaining materials for the various treatments had taken months and months of effort.

The design of the experiment was relatively straightforward. I was interested in the effect of different seeding rates and altered nutrient regimes on the short- and long-term success of the sowings. At the time, there was much discussion about the possible beneficial effects of smoke in promoting germination (a topic still of considerable interest in the regeneration industry), and decided to apply smoke as a background treatment across all sowings.

At about 6.00 am on a fine Saturday morning, the time for actual sowing had arrived. I thought I’d covered every base and had lists to check-off on a regular basis. Seed-lots mixed with ‘X’ amount of smoke product, bulked up with ‘Y’ amount of vermiculite, sown onto good seed bed, pressed in, fertilized if required at ‘Z’ rate and watered in, etc etc. By 6.00 pm I was exhausted, and rang John to proudly inform him of the day’s events.

Being the meticulous person John was/is, we discussed the various aspects of the sowing in detail and I spoke of each with some confidence – that is, until a passing comment of his sent a cold shiver up my spine. We had spoken about the calculation I had used to determine the rate for applying the smoke product to a particular surface area. I realised that I had inadvertently added a zero to my figure at the point of mixing and therefore increased the application rate by a factor of ten. It was not obvious at the time because the volumes of smoke product were extremely tiny. What to do? Of course I wanted to throw up, but held my nerve. I reasoned that no-one would ever know if I didn’t mention it. To start again would take an enormous effort. I did have the seed but wasn’t sure if I had the constitution or will.
In the end, following a pretty ordinary Sunday, I did fess-up on the Monday. After some discussion with (and counselling from) John, I duplicated the experiment using the correct recommended rate of smoke product. In the following weeks and months through to the completion of the experiment, I discovered that the species I’d planted responded in very interesting ways to the particular rates of smoke that had been applied. While some species struggled in the ‘high’ smoke plots of the first sowing, one species germinated at very high rates.

What Did I Learn?

So what did I learn from my ‘error’? I certainly realised I could never be ‘sloppy’ with my maths, especially when working with concentrated substances. I also realised, however, that the seed-related outcomes of the ‘error’ were extremely interesting. After undertaking further investigation, I found that the seven grassland species I tested at varying smoke rates had differing dose responses to the smoke product. To my knowledge, there is still no commercially available smoke product that delivers a quantified amount of the active ingredients in smoke that promote germination. In lieu of this, propagators and restorationists apply smoke in various forms at essentially ‘unknown’ or ‘ball-park’ rates. This means that we probably often see species responses that differ from application to application, or from one form of smoke to another, or in results from one person to another.

Only when products are supplied with known amounts of the active smoke ingredients will we be able to confidently test the dose responses for species and understand exactly ‘how much’ (if any) promotes germination for a given species. We will then be in a better position to apply smoke with confidence in the field or propagating room (being mindful, of course, of one’s sums when mixing).


When I began this article I had a lengthy list of ‘errors’ I was going to report. However just getting this one event off my chest has taken up the word limit (and been cathartic enough!).

The comment ‘To err is human’ has been made by wise people on many occasions over a long period of time, and seems to be a fair enough statement. Thus, armed with a knowledge that we are likely to err, how do we proceed?

I learnt that undertaking research requires attention to fine detail and I must be especially careful in this respect. However, more importantly, it alerted me to the fact that errors can lead to very unexpected and interesting outcomes. In my case, at the time of realising the error, it took a huge amount of will to own up to something that no-one would have ever known about. But had I not, the duplicate experiment would never have been instigated. Despite the considerable effort the duplicate required, I was alerted to a whole new set of findings that I would not otherwise have suspected nor investigated, had I mixed those damned vials of smoke product correctly the first time.