CALL THE WLT HOTLINE: 082 516 3602

Home2011 Action 2011 DiaryIdentification Volunteer Groups What To Do If... Toads and your pets Research WLT Conservation Committee F.A.Q. Gallerylinks


(Mainly the DIE OOG and SWAANSWYK Breeding Ponds (and the CYCLE TRACK))


The data presented here are an amalgamation of all the data collected by several teams. The area is bounded by Swaanswyk to Sweet Valley Road, and from the M3 (OuKaapse Weg) to Ladies Mile. It includes the Swaanswyk Wetlands, the Dreyersdal Farm Dam, and Die Oog.

Breeding at Dreyersdal started with a little blip in late July and early August, just over a week apart. But breeding did not materialize. Only at the end of August did the breeding migration really take place.

Some amplexis was actually observed in Starke Road in early August, and it might be well worth following up to see if there is a breeding pond near Airley Road: no amplexis was observed at any of the known breeding sites. However, by far the most amplexis was observed during the peak of the breeding cycle, almost exclusively around Die Oog.

Most dead animals were encountered on the first blip in late July. The lack of dead toads in the later peaks may well be due to the active work of the volunteers (assuming that the road-stupid toads were not eliminated). Certainly, deaths were low despite lots of toads in the roads during peak breeding, suggesting it was due to volunteer efforts.

The sexes of toads were slightly biased to males preceding the migration, but on the whole bordered on 50:50 males to females. This was not the case once breeding began. Females dominated the first day of the breeding event, but males rapidly increased to double the females on the day before the peak. On the peak day females vastly outnumbered males. Following the breeding event, females were slightly more abundant, with empty females (without eggs) being more prominent, but a female with eggs was found even on the 5th day after the breeding peak.

The Breeding Ponds in 2009 are shown below

BREEDING SITES 2009 click to enlarge

A survey of the surrounding pools revealed two new breeding ponds in Porter Estate:

NEW SITES 2009 click to enlarge


The above data were borne out by separate data from the Cycle Track at Sweet Valley, just across the M3 from Die Oog. There the peak in activity occurred a day later than at Die Oog.

However, the situation is different from east of the M3 in that at night the cycle track is deserted. The roads leading to the wetlands (the Groot River) are quiet, unlit and not frequented by traffic.

Along this path almost all the animals observed were males. These were “patrolling” parts of the path and almost all the few females seen were in amplexis. After being photographed the animals returned to their activities, and were not interested in moving to the water. We don’t know if the lack of females was because we missed the female part of the migration (we only sampled this site after the early evening activities east of the M3 had died down, so that if females moved early in the evening we would have missed them(this possibility is borne out in the late evening transects east of the M3 (- see the first 3 bars in the graphs above – the X.1 transects were done just before midnight))), or because this was a male “lek”, with males waiting for females to move through, we don’t know. It suggests though that historically males may have waited along animal paths to “ambush” females on their way to the ponds. There was an impressive snorus (snoring chorus) in the Groot River – as spectacular by night as the chorus at Die Oog, and I estimate the number of calling males to be similar in both areas.

If males naturally patrol paths to intercept females it may explain why mortality in the species might be so high: the males use roads as leks to intercept females before they get to the ponds. Unfortunately, the dead animals from the start of the Dreyersdal survey were not sexed, so we cannot determine if males are preferentially killed. Mortality during the peak breeding was too low to yield any conclusive answers (8 males and 4 females, and 3 unsexed).

Believe it or not, City of Cape Town contractors mowed the area during the peak period. How many (if any - the toads might move to the water during the day?) animals were liquidized when the grass around the tracks was mowed, and it is not known if the mowing affected their activities.

2009 Statistics:

Bergvliet 150 live & 21 dead*

Cycle Track 87 live & 0 dead

Other areas 25 live & 11 dead* (SweetValley and Tokai)

Most live photographed (over 250), most dead collected (30)

About 162 man hours spent on standardized surveys and rescue, plus 16 hours checking ponds during day.


In 2008 sex data was not collected sufficiently to come to any conclusions above sexes. Mortality peaked during the peak breeding season, although the first dead animals preceded the breeding season by two weeks. More mortality occurred in the early evening than later at night (21a/08/2008 – was done in the hour before midnight), but activity was not much different between early and late evening in 2008.

Amplexis was only observed during the peak breeding period, but there was a blip in activity the week preceding the breeding event.

Some 11 breeding sites were observed. Six potential sites were visited and found not to have any activity

BREEDING SITES 2009 click to enlarge

2008 Statistics:

Some 216 live & 25 dead*

Most live photographed, most dead collected

About 63 man hours spent on standardized surveys and rescue, plus 6 hours checking ponds during day.

2008 vs 2009 BERGVLIET

Breeding occurred much earlier in 2008 than in 2009. Much more data will be required before any trends or causation of the breeding peak or the dynamics of the peak breeding period, can be determined.

Note the difference in pattern of the peaks. However, this is because the surveys in 2009 were significantly affected by the volunteers (volunteer data are included in previous 2009 graphs, but including them here will mean that the sampling intensity is different and thus comparison is not any more valid than the graph shown here).

However, both the sample and volunteer data concur, that whereas 2008 the peak was clearly suddenly on day one, declining steadily thereafter, in 2009, the peak duration lasted much longer, and peaked on the fourth day of the peak (or the second day of the major peak) and suddenly ceased.

Note the long lead in between when activity was first noted on the streets and the breeding season. Toads feed for a few weeks to a month after activity starts before breeding. Much road mortality occurs during this time!

An attempt will be made to correct for this volunteer “interference” in 2010. No doubt a different pattern will emerge yet again. We wait and see!

* One is tempted to calculate mortality as number dead seen vs number live seen. This "mortlivity" is not valid as the dead toads cannot walk away, whereas the live toads do. Consequently, one only observes a fraction of the live toads, but every dead animal is seen (and will wait for several days to be sampled, until the crows move in).

Website design and hosting donated by Julie Anderson of J Productions
Information compiled by Tony Rebelo, September 2009.

Thanks to Alex Rebelo and James Morton who helped collect the survey data, and did all the hard work like photographing, rescuing and sexing the toads.