Lake Cerknica in Slovenia could seem to offer several incompatible landscapes in the same place: open water, fishing grounds, grassland and cultivated land. Its disappearance and return attracted natural philosophers long before modern karst hydrology. Johann Weikhard von Valvasor’s 1687 account also recorded underground drumming, cave waters and beliefs about witches on neighbouring Slivnica. The lake’s changes are real natural phenomena. The historical mystery lies in how observers explained what they could see and hear while much of the water’s route remained hidden.
A changing lake in a karst basin
The Notranjska Regional Park describes an intermittent lake occupying the Cerknica plain between the surrounding hills and plateaus. Its outline depends on water level, so a single photograph does not represent a permanent shoreline. The park gives a maximum area of about 26 square kilometres and describes a normal wet period of roughly eight months. These are general characteristics, not a calendar that every year must follow.
The lake’s older German name, Zirknitz, appears in the English account published in Philosophical Transactions, issue 191, in December 1687. It occupies pages 411–427 and combines a description of the district, accounts of fishing, particular experiences and a proposed explanation. Valvasor is not merely repeating that a lake vanished. He identifies places, intervals and differences in water flow, giving later readers observations against which to assess his interpretation.
What Valvasor claimed to have seen
The account distinguishes personal visits from information supplied by others. Valvasor says that in June 1678 he travelled by a small fishing boat through one cavern passage in the associated river system, stopping where the water descended noisily beyond the point he dared follow. A later visit in 1684 was restricted by high water and the absence of a boat. These passages establish limits as well as exploration: he did not claim to have walked through the complete underground plumbing of the lake.
His chronology also resists a simple annual-draining story. The 1687 publication says the lake sometimes emptied two or three times in a year, while in other periods it remained full for several years. He records a drying episode in January 1685 and another beginning that August. He describes the current year, 1687, as having three dry phases. Such statements belong to his seventeenth-century account; they should not be repackaged as a modern continuous gauge record.
Fishing as a way of observing the water
Much of the report concerns named depressions in the lake bed and the sequence in which they lost water. Fish gathered in these hollows as water fell and could be taken with nets before disappearing underground. Valvasor describes a stone used to judge when fishing should begin and bells used to summon people. Fishing rights belonged to different estates, and the report specifies which communities watched the water and which catches were shared with a lord.
These details explain why local people would notice more than the broad fact that the lake was wet or dry. The timing affected access to fish, work and customary rights. Valvasor’s list gives eighteen principal fishing hollows and says a complete recession could take about twenty-five days without heavy rain. That apparent order helped him build an explanation. It also reminds us that his information came partly from an existing community of practical observers, rather than from a solitary discovery.
He describes the exposed ground being used for hay, millet and grazing, with hunting and fishing possible at different stages. He qualifies this picture: crops were not grown over the entire basin, and an early return of water could destroy them. The striking claim was therefore a conditional sequence of uses in a changing landscape. It did not require land and lake to occupy the same point at the same moment.
The underground drumming and the storm of 1685
Two hollows carried names translated as the greater and lesser drummer. Valvasor says that in 1685 he heard a drum-like response from them when thunder sounded. In a longer account of an October storm that year, he describes setting out with two experienced fishermen to examine the phenomenon. Rising water then cut off ground they had crossed dry only a short time earlier. They obtained a small boat, capsized and eventually reached safety. This is his reported experience, not a reason to recreate the journey.
The narrative moves from sound to rapidly discharging water and, more controversially, birds. He says cave outflows sometimes carried fish and featherless or poorly feathered, apparently blind waterbirds. Later he describes opening some birds and finding sand, small fish or plant-like material in their bodies. The text establishes what he reported and the inference he drew. It does not give us preserved specimens, reliable species identifications or an independently verified underground breeding population.
Folklore beside natural investigation
The same publication records local belief that witches assembled on Slivnica, encouraged by lights seen on the hill. It also describes vapours from a deep opening associated with storms and a priest’s procession and exorcism. Valvasor attributes the witches’ assemblies to country people’s belief. That wording should remain distinct from his claims of seeing water flow or hearing the drumming. A seventeenth-century description can preserve folklore without confirming the beings in it.
The park’s modern Slivnica account continues to identify the witches’ cave as part of the place’s folklore. It explains that warm, moist cave air condenses into mist in winter, contributing to the story. This offers a physical context for a striking appearance. It is not a reconstruction of every light, storm or ceremony described in the older account.
His first explanation and its limits
Near the end of the 1687 report, Valvasor expressly turns from description to his opinion about the mechanism. He proposes an underground body of water beneath the visible lake and higher reservoirs inside the Javorniki hills. Rain increases their supply beyond the capacity of their ordinary outlets, so water rises through holes and overflows into the Cerknica basin. After prolonged dry weather, outflow exceeds inflow and the surface lake sinks.
That distinction between observation and model matters. Named openings and changing water levels could be observed; the exact extent and arrangement of the proposed hidden lakes could not. He used the persistence of water beyond the visible outlets and comparisons with cave pools elsewhere to argue for underground connections. The explanation had a useful water-balance core, but its detailed subterranean geography remained a hypothesis.
Why the 1687 and 1689 versions must be separated
Historian of cave science Trevor R. Shaw made this the subject of an original study in Acta Carsologica, volume 33/2 (2004). Comparing the publications, he showed that Valvasor’s 1689 German book and Latin paper were not simply translations of the earlier English account. The later explanation added a more elaborate network of siphons, making the model more complicated and, in Shaw’s assessment, less correct. The familiar diagram of chambers and siphons belongs to that later formulation.
The correction changes how the investigation should be remembered. Valvasor did not present a single fixed solution that can be illustrated interchangeably with any diagram attached to his name. His ideas developed. A later, more detailed model need not be a better one, and recognition by a learned society does not certify every hidden feature proposed in it. Shaw’s comparison gives a concrete example of why the original editions matter.
The modern explanation
The park’s hydrological explanation describes a basin that floods when incoming water exceeds the rate at which underground outlets can carry it away. Heavy rain and snowmelt increase supply. Drainage is entirely underground, with routes towards Rakov Škocjan and towards the Ljubljanica springs. In rainless periods the water recedes over weeks. This explains an intermittent karst lake through changing inflow and drainage capacity, without requiring a supernatural agent.
Cerknica remains a useful historical mystery because its archive preserves the work of explanation: local observations, a hazardous visit, folklore, specimen claims and competing models. The dry basin was real; that reality did not make every inferred cavern or underground bird colony real. The public map point marks the verified Klejni vrh photograph viewpoint overlooking the lake. It does not locate a hidden chamber or the place of Valvasor’s storm encounter.
Related reading: Other Mysteries, Valvasor’s account of Kringa’s revenant tradition, and the Historical Cases directory. The 1687 text is a public-domain historical source; its observations and explanations are attributed here rather than treated as modern findings. Sources and image rights checked 5 November 2026.