Where did the explorer Ensign Barrallier go in 1802? That question has tantalised bushwalkers since 1920, when surveyor and botanist Richard Cambage decided that the route went from the Nattai, past the Tonalli-Wollondilly confluence and westward to Bent Hook Swamp (on Bindook Creek), thence northward to the Kowmung and ultimately up Christys Creek to Barralliers Falls.

In the 1930s Rae Else-Mitchell, bushwalker with the Warrigals, historian, and (later) distinguished judge, decided that Barrallier reached Christys Creek by travelling up the Tonalli Valley and through Byrnes Gap.

Most subsequent work has confirmed the Else-Mitchell version, except that the probable terminus has been moved to Johnston Falls in Wheengee Whungee Creek. Those interested can read all about it in my book Blue Mountains to Bridgetown: the Life and Journeys of Barrallier.

In 1996 a new theory appeared in the book Blue Mountains Rediscovered. Its author Chris Cunningham suggested that Barrallier reached Bent Hook Swamp, as per the Cambage version; then, the next day (26 November 1802), he

• climbed a nearby mountain to obtain a view;

• headed west till he was barred by the valley of Murruin Creek (11 am);

• swung north and found the Murruin Range, which he followed to the point where Ruby Creek is overlooked;

• not perceiving that the range swung south-west, descended to Ruby Creek (about 4pm); and

• walked downstream to camp at the Burnt Hole Creek confluence.

The next day Barrallier walked up a creek - Burnt Hole Creek according to the Cunningham theory - till he met a deep pool of water with a waterfall behind, obliging him to undertake a high sidle. After regaining the creek at 8 pm he made camp, but his men took a short walk upstream and reported the existence of an impassable waterfall 30 metres high. According to Cunningham the waterfall is at GR687304 (Gurnang 1:25,000), placing Barrallier’s supposed terminus 14 km west of previous interpretations.

Deciding to check the new theory out, I went on a bushwalk with several others from Springwood Bushwalking Club. I was armed with Cunningham’s book, the relevant extracts of Barrallier’s journal, a sense of discovery and an open mind.

The findings? In short, the Burnt Hole Creek theory is completely out of the question, for the several overwhelming reasons mentioned below. I will not go over the arguments against Barrallier’s having gone to Bent Hook Swamp in the first place. Let’s just assume he got that far, and examine the ‘Burnt Hole’ theory westward from there.

(1) The distance walked on 26 November

Even if Barrallier had found his way faultlessly along the Murruin Range, the theory requires that Barrallier walked at least 27 kilometres on 26 November, not including the early sojourn up the mountain. Of this, 18 km must have been covered in a 5-hour period which would have included his normal long lunch break. This is inconceivable. Barrallier and his men were not tiger walkers.

(2) The view above Ruby Creek

Barrallier supposedly dropped into Ruby Creek gorge because he missed the south-west turn in the Murruin Range. This is most unlikely. On arriving at the edge of the gorge the Ruby Creek Falls are in clear view to the left, and it is obvious that the range swings to the plateau country behind them. Furthermore, it is virtually impossible to descend straight ahead, owing to the presence of cliffs.

(3) Ruby Creek

Let’s suppose Barrallier has made the descent, and is now heading down Ruby Creek - a tiresome undertaking, according to his journal. However, after the first 200 metres or so from the feasible descent route, Ruby Creek is easy travelling, with gentle banks and slopes on one side or the other.

(4) The Burnt Hole Creek confluence

If the theory were correct, then according to Barrallier’s journal and map, Ruby Creek would be a ‘river’, and Burnt Hole Creek would be a lesser watercourse - a ‘rather large stream’. At the confluence there would be a delta of rocks protruding into Ruby Creek, providing a place to cross it. One could also go for a short walk up the banks of Burnt Hole Creek and find some fossils in the ‘sandstones’.

But the confluence is nothing like this. Burnt Hole Creek is larger than Ruby Creek, not smaller, and there is no delta. The rock outcropping in the bed, banks and slopes at the confluence and for 3 kilometres upstream along Burnt Hole Creek, is granite. No fossils there.

On walking up Burnt Hole Creek one immediately encounters a granite canyon, in which the pools and cascades can only be negotiated by scrambling. Not a place for an evening walk, nor to embark on the morning’s journey.

(5) The sidle

have prompted the high sidle described in the journal.

The first commences immediately one leaves the confluence so, given that it took Barrallier most of the day to reach it, the sidle could not start there. The second canyon starts about 2.5 kilometres upstream (GR718277), but:

• it takes unathletic bushwalkers only 2 or 3 hours to walk up to this point from Ruby Creek, so it is still too close to the confluence;

• the only waterfall which must be sidled around is not associated with a deep pool of water; and

• because the rock is granite rather than quartzite, we do not find there the sharp points of rock described by Barrallier.

Even if Barrallier had ascended from there, the topography dictates he would have to descend and regain Burnt Hole Creek downstream from the Wattle Creek confluence.

(6) The waterfall

Chris Cunningham’s suggested terminus waterfall at GR687304 simply does not exist. There is no significant waterfall at that location, or anywhere near there.

The only waterfall worthy of the name is at GR673290, on a totally different tributary. It is only about 13 metres high. Could it be Barrallier’s waterfall? No. Apart from its small magnitude -

• It is over 4 kilometres from the end of the supposed sidle to the waterfall. The walking is not hard, but it still takes a good 2 hours: this hardly fits the journal, which implies that the walk should take only a few minutes.

• It is obvious from beneath that the waterfall can be very easily bypassed on either side. There is no imposing mountain backdrop, and no-one could claim the waterfall was impassable. The hillsides in the area are relatively gentle and not at all formidable.

Burnt Hole Creek is a pleasant spot for a wilderness bushwalk, if you ignore the feral pigs. You must also ignore the red herrings, for you can be sure that Ensign Barrallier was not there before you.