Railroads built along ridge lines
The trick in crossing a mountain is to design enough mileage into the route to keep the rate of ascent within acceptable limits. The shortest route is rarely the best.
The Central Pacific was limited by statute to a grade of 112 feet per mile. The elevation difference between Donner and the valley is roughly 7000 feet, so the route had to run at least 62.5 miles. My recollection is that it is about 70 miles by rail from Donner to Roseville – so the CP was quite effecient. While there are countless ridges extending westward from the crest of the Sierra, few extend all the way to the floor of the Central Valley. Most end where the intervening canyons merge. The ridge the CP followed happenes to be one of the ones which runs the whole way – in this case between the American River watershed to the south and the Bear and Yuba River watershed to the north.
Nevertheless, I sincerely doubt that that remarkable ridge had anything to do with the selection of that route for the railroad. The initial goal of the movers behind the Central Pacific was to profit from the commerce then flowing between San Francisco and the Comstock Lode – and the Donner route just happened to be the one relatively direct route not yet controlled by other interests (in the form of toll road franchises). Railroads were successfully designed for other routes across the Sierra which – lacking the long ridge used by the CP – managed to fit in the necessary mileage by looping into side canyons along the way.
What Judah did was verify that one could indeed fit 70 odd miles of track between the valley and the summit along the Donner Route. I'm not sure just when – or even if – Judah realized "their" ridge was significant. Mining engineers who laid out the miles of flumes which tapped high country reservoirs to feed the hydraulic mining operations had doubtless already discovered routes with steady grades from the high country to the mines (such as those around Dutch Flat). We know that Judah's initial "survey" between Dutch Flat and the summit merely followed a previously surveyed route which Daniel Strong was trying to promote for a toll road. From Dutch Flat to the valley, Judah initially followed existing roads or railroad surveys without regard to staying on the ridge. And in the end, I'm not at all sure whether it was Theodore Judah or his brother Douglas who actually "put" the railroad onto the ridge to Auburn – rather than down in Auburn Ravine as Theodore had earlier proposed. There is just a whole lot to the final selection of the CP route that is lost to us.
Conversely, the Western Pacific required roughly 100 miles of track to make their economical descent from 5000 feet to stay within their design criteria. The ridges which extend west from their summit were considerably higher than their pass itself, and their steady descent pushed them down into the canyon fairly soon. The trick in their case was shifting from the middle fork of the Feather over to the north fork. They could indeed have followed a ridge – say between the Feather and Butte Creek – farther out into the valley, but to do so would have given them some nearly level running at a higher elevation- -where snow would have been some problem. Even so, the creation of Oroville reservoir in the 1960s forced them to relocate onto that ridge – and if memory serves, resulted in an even better grade.
As an aside, highway 50 between Placerville and the crest of the Sierra runs in the canyon of the south fork of the American River. The road is very winding. However, the little known Mormon Emigrant Trail – the first true road across the Sierra and the one followed by most overland emigrants during the gold rush – follows the ridge just to the south. Getting to it involves some twisty sections, but once on the ridge it runs relatively straight and steady for 25 miles – with some tangents running a full mile.
[from the R&LHS Newsgroup.]