We present" Tracklet-less Heliocentric Orbit Recovery"(THOR), an algorithm for linking of observations of Solar System objects across multiple epochs that does not require intranight tracklets or a predefined cadence of observations within a search window. By sparsely covering regions of interest in the phase space with" test orbits," transforming nearby observations over a few nights into the corotating frame of the test orbit at each epoch, and then performing a generalized Hough transform on the transformed detections followed by orbit determination filtering, candidate clusters of observations belonging to the same objects can be recovered at moderate computational cost and with little to no constraint on cadence. We validate the effectiveness of this approach by running on simulations as well as on real data from the Zwicky Transient Facility (ZTF). Applied to a short, two-week slice of ZTF observations, we demonstrate THOR can recover 97.4% of all previously known and discoverable objects in the targeted (a> 1.7 au) population with five or more observations and with purity between 97.7% and 100%. This includes 10 likely new discoveries, and a recovery of an e∼ 1 comet C/2018 U1 (the comet would have been a ZTF discovery had THOR been running in 2018 when the data were taken). The THOR package and demo Jupyter notebooks are open source and available at https://github. com/moeyensj/thor.