by N. G. L. Hammond
[This paper first appeared in Greek, Roman and Byzantine Studies, 36, 2 p 173ff and appears here by kind permission of Professor Hammond]
Aelian informs us that fishermen used a rod and a line , each about six feet long, and an artificial fly to catch “fishes of a speckled hue” in a river called the Astraeus in Macedonia (NA 15.1). The method and the description indicate that the fish were trout. Aelian presumably chose to name the Astraeus as the river more famous for its trout, because we cannot suppose that it was unique among Macedonian rivers in harbouring trout. The only clue to the position of the river which Aelian gives us is that it lay “between Beroea and Thessalonika.”
Can we get any closer?
Leake took us farther afield in suggesting that “Astraeus” was another name for the Haliacmon;  but that river did not flow between Beroea and Thessalonica. The Loeb editor of the passage in Aelian makes the comment “presumably the Axius is intended”,  but surely Aelian knew the Axius under its given name. We need rather to note that there is a nymph Atraea, appropriate to a river of her name, “Astraeus,” and that she was selected as the nurse of the lady Beroe (Nonnus Dion. 41.212f), the patroness of Beroea (the city was sometimes called Beroe). We conclude that the Astraeus was near Beroea and that it flowed on the Thessalonika side of Beroea. The eastern neighbour of the Astraeus was the Ludias, which rose in Almopia, flowed through the Almopian city Europus and in the time of Herodotus flowed into the Haliacmon, but later its waters joined those of the Axius to form Lake Ludias by Pella.  On this evidence the Astraeus entered the Haliacmon in the time of Herodotus. It still did so ca 360 B.C. when Ps.-Scylax (66) named the Ludias as the next after the Haliacmon in his description of the coast of the Theramic Gulf. This probably continued to be the course of the Astraeus, whatever the date may be of the passage from Aelian, who wrote ca 200 but included excerpts from earlier writers.
In my study of the historical geography of Macedonia, I showed that the low-lying part of the coastal plain northeast and east of Beroea has undergone numerous changes since the time of Herodotus, when the Ludias flowed into the Haliacmon, to that of Leake, when the Ludias and the Axius "formed a joint stream."  The size and the location of a lake or lakes has also varied over the centuries.  As regards recent times, a map of 1928 showed in the centre of the plain a large lake (that of Yannitsa) into which all the waters flowing from the western foothills found their way through swamps. Thereafter the lake was drained and a “Regional Channel” was made to carry the waters from the western foothills not into the plain, but into the Haliacmon.  Because the channel follows the natural lie of the land, it is probably the line of the ancient Astraeus. That is where I showed the course of the Astraeus in my Atlas of the Greek and Roman World in Antiquity (London 1991) map 12 and in The Miracle That Was Macedonia (London 1991)
fig. 2. .
The upper courses of the rivers flowing from the foothills between Edessa and Beroea have undergone fewer changes. One was reported in antiquity: a new channel was made for the river Scirtus, which had inundated the buildings of Edessa (Procop. Arc. 18.38), then at “Longos” below the modern town. From this passage we learn that the name of the river at and to the east of Edessa, before it flowed into the lower Astraeus, was Scirtus, “the Leaper,” which cascaded over the cliffs of modern Edessa. It is now called the Vodhas and it enters the Regional Channel. In ancient times the Astraeus was joined by its northern tributary, the Scirtus, and kept its own name as it flowed on to the Haliacmon.
On 22nd April 1994 I was taken by Nikolaos Paschialides to eat the finest trout in Macedonia (he is a gourmet in regard to fish) at a restaurant in Ayios Nikolaos, a village that I visited for the first time. The Arapitsa emerges from a large "headspring" (kephalovrysi)  on the mountainside as a considerable river of which the waters are constantly cool, pellucid and bright; and it is this upper part of the river that is particularly rich in large trout. The modern saint of the headspring is St Nikolaus. In antiquity the saint was the nymph Astraea, who gave her name to the river. She was "the starry one," and her name befitted the glittering waters of the river as it emerged from the mountainside.
There was probably both a town and a district called Astraea after this nyumph, for the name recurred in Paeonia and in "Illyria,"  the latter being the area annexed by Philip II in the Lakeland, and in both areas the name seems to have been taken by the Macedonians who were transplanted from that area in the homeland. If so, we should name the foothills of Mt Bermium to the north and south of Agios Nikolaos ‘Astraea’, because they gave off the waters which made the region below so productive that it was called “the gardens of Midas” (Hdt. 8.138.2). The ancient site of a town at Gastra near Naoussa may have been either Astraea or Mieza, for the region east and below Gastra was certainly called Mieza. It included the so-called “School of Aristotle” and the three fine Macedonian-built tombs that have been excavated at Lefkadhia. 
Clare College, Cambridge
 W.M. Leake, Travels in Northern Greece (London 1835; hereafter ‘Leake’) III 293, 488.
 A.F. Schofield, tr., Aelian, On the Characteristics of Animals III (Cambridge [Mass.] 1959) 203 n.a.
 Plin. HN 4.34; Hdt. 7.127.1; Strabo 7 frr. 20, 23.
 A History of Macedonia I (Oxford 1972: hereafter ‘HM’) 142 f.
 Leake III 436; E. Oberhummer, “Pella,” RE 19.1 (1937) 343; both quoted in HM I 144f.
 Sheet “Yiannitsa 1928,” Sheets “Edessa and Yiannitsa 1944” = Sheets D 6-7, War Office version; the maps are at 1:100,000.
 I take the opportunity to correct the error in fig. 1, where the cartographer incorrectly joined the Ludias and the Astraeus. Mr. A. F. Sismore kindly pointed out this error to me.
 Such headsprings are not uncommon in limestone formations. I have seen many in Epirus: see my Epirus (Oxford 1976) 55, 172.
 See HM I 201, citing Ptol. Geog. 3.13.27 and Steph. Byz. s.v “Aστραιαwhere I suggested that ‘Iλλνρίας might be a mistake for Mακεδονίας; I have since then retracted that suggestion. I placed the Astraea by Lakes Ochrid and Prespa in The Miracle That Was Macedonia fig. 2. There are numerous springs at Izvor, derived from Lake Prespa and entering Lake Ochrid, which may have reminded the Macedonians of the headspring of the Astraeus. I visited Izvor in 1972.