Saturn's biggest moon Titan was dicovered by Christiaan Huygens in 1655 with a telescope containing two refracting lenses ground and polished by himself. The probe reaching Titan's surface in 2005 was named after him.

photograph by Huygens (contrast enhanced).



A dry primordial anvil
While life ultimately thrived in an aqueous matrix, the fundamental organic building blocks of our genes may have been forged in the young universe under non-aqueous conditions. What kind of molecular reactions can be expected on the surface of the first CEMP planets? How H, C, N, O, Mg, Si, and S might have interacted? Due to their relatively low abundance in this '2nd generation world', O and Si would have played no central role (as on the prebiotic earth). C would have dominated the scenery and created a very special hydrocarbon world under heavy UV-bombardment.
On the surface of the first Carbon-Enhanced Metal-Poor (CEMP) planets, other synthetic opportunities opened up than on the early Earth. Because CEMP systems possess high C to O ratios (> 1), oxygen is entirely locked up in carbon monoxide (CO), with little left to form water or silicates. The atmosphere is highly reducing, dominated by methane (CH4), nitrogen (N2), CO, hydrogen (H2), and some H2S (increased by eventual volcanism).The solid ground is dominated by graphite, long-chain hydrocarbons, and carbides. Mg will not find enough O and silicates to take over its usual role as divalent cation and will be available for organo-metallic reactions. 
In the absence of earth-typical clay minerals and quartz, silicon carbide (moissanite - on earth only found in asteroids) would serve as the main rock-forming mineral, accompanied by Ti- and Ca-carbides, Mg- and Ca-sulfides, and plain and simple graphite. Liquid water may have survived in a few niches, leaving most of the stage to liquid methane, ethane, and formamide. Mg lost its typical role in silicate minerals to teaching us a first lesson in Grignard metallo-organic chemistry: In the absence of water, it can insert into carbon-halogen or carbon-nitrile bonds. UV light promotes the formation of radical species, enabling reactions like R–CN + Mg + hv » R–C(Mg)=N⋅ . These intermediates act as highly reactive nucleophiles, establishing new covalent C-C bonds..
Also without the detour via metal-organic bond formation (in presence of too much moisture), Mg2+ ions readily coordinate with nitrogen in nitriles, polarizing the CN-triple bond and thereby significantly lowering the activation energy required for nucleophilic attack, allowing  nitrogen-heterocycles like purines and pyrimidines to be formed from cyanides. Under heavy UV flux, H2S and acetylene (HCCH) may react to yield small aromatic heterocycles as thiophene and thiazoles. Environmental conditions as moisture, temperature, pressure, surface, radiation intensity, and time would decide which pathways come into play.
Photolysis of the methane-rich atmosphere by stellar UV radiation yields massive quantities of hydrogen cyanide (HCN) and formaldehyde (HCHO). As first proposed by Carl Sagan, these compounds polymerize in the atmosphere into complex organic aerosols called tholins (derived from old Greek for 'muddy'; Sagan & Khare 1979), which rain down onto the surface (as on Saturn's Titan). Among other products, also diaminomaleo-nitrile (DAMN) may form. On further photochemical processing,  amino acids and nucleobases may ensue. But not always hydrocarbon strategy will follow Titan's (very cold) example. Titan's geology is dominated by stony water ice (see photograph at top), much in contrast to CEMP planets.
Organic chemistry is the strength of a hydrocarbon world under UV-influence in this act one of the Ludus vitalis. In the absence of water, long chains of simple heterocycle units may form and persist. In a hydrophilic milieu, chain elongation is often followed by fast hydrolysis (i. e. cleavage by the addition of H2O), not allowing for significant growth. In an apolar environment, first chains probably grew on supporting surfaces as graphite. Over the millenia, chains learned to grow faster along other chains, creating the first bulk amounts of pre-bio-organic matter. This probably marks the dawn of pairing certain heterocycles vis-a-vis each other. Base-pairing hydrogen-bonds would even be stronger in a hydrophobic milieu (the interior of the modern doublehelix is still free of water).
On the surface of CEMP planets, heterocycles may form in hydrocarbon seas, supported by graphite surface, Mg, and UV. But how could they form chains? In modern polynucleotides, this is accomplished by ribose and phosphate, both inaccesible in a hydrocarbon world. 'Prebiotic Peptide Nucleic Acids (PNA)' use an uncharged, pseudo-peptide backbone, chaining genetic information together without ribose or phosphate. PNAs have first been described by Nielsen et al (1991). It has been hypothesized that on the early earth a 'PNA world' has preceded an RNA world, before the advent of DNA and protein.
Probably this was the endpoint of pre-bio-evolution under 2nd-world premisses. One element indispensible for life as we know it was sparse in the primordial carbon-world: phosphorus. Phosphate and ribose as linkers of heterocycles need water. Whether single self-replicating and closed entities ('cells') are possible in a hydrophobic milieu is a matter of debate. In any case, act two in this Ludus vitalis was played out in the next (3rd) generation of solar systems, this time under 'high metal' conditions in an aequous milieu. Probably the primary organic building blocks bred in act one were robust and abundant enough to survive the turbulent change of stage.
The step-wise evolution of prebiotic material may have taken place already in the 2nd and 3rd generation of solarsystems, long before the end of the first billion years. The 1st step ended with supernovas of 2nd generation solarsystems, distributing dust containing more heavy metals than the first series of explosions, but also first pre-biotic material forged on carbon-dominated planets. The 2nd step went on with this pre-organic material in aequous worlds. The primordial cosmos with its short distances might have resulted in a universal genetic alphabet, with diversity setting in during the 2nd step leading to the first life forms.
9/26 <          MLB (10/26)          > 10/26
Nielsen PE, Egholm M, Berg RH, Buchardt O (1991). Sequence-selective recognition of DNA by strand displacement with a thymine-substituted polyamide. Science. 254, 1497–500

Sagan C, Khare B (1979) Tholins: organic chemistry of interstellar grains and gas. Nature 277, 102-7
I thank google KI for support during the preparation of this essay.